From Ben.Lippmeier at anu.edu.au Mon Feb 1 01:37:35 2010 From: Ben.Lippmeier at anu.edu.au (Ben Lippmeier) Date: Mon Feb 1 01:09:16 2010 Subject: Exploding SpecConstr In-Reply-To: <4B630AEA.5020506@gmail.com> References: <947938FD-89E3-4B6E-9547-850824A42812@anu.edu.au> <4B630AEA.5020506@gmail.com> Message-ID: <36D67891-482F-4604-84F9-1D439ED344A0@anu.edu.au> On 30/01/2010, at 3:20 , Simon Marlow wrote: > > We decided not to try to stick with buildbot - it isn't suitable, in that it can't cope with dropped connections, and it would be too hard to fix it. There isn't a good alternative that we could find, so after due consideration we've decided to try making a replacement in Haskell. It will be pretty minimal, but it'll be clean and it'll do exactly what we need, plus we'll be able to use it for automatic remote validation of patches on multiple machines. > > Ian is currently working on it, and we hope to have something working in the next week or two (Ian?). > Sounds fabulous! Let me know if there is anything I can do to help.. Ben. From Ben.Lippmeier at anu.edu.au Mon Feb 1 01:41:33 2010 From: Ben.Lippmeier at anu.edu.au (Ben Lippmeier) Date: Mon Feb 1 01:13:13 2010 Subject: Exploding SpecConstr In-Reply-To: <404396ef1001290858i555226ccj809a3e378ff1569a@mail.gmail.com> References: <947938FD-89E3-4B6E-9547-850824A42812@anu.edu.au> <4B630AEA.5020506@gmail.com> <404396ef1001290858i555226ccj809a3e378ff1569a@mail.gmail.com> Message-ID: On 30/01/2010, at 3:58 , Neil Mitchell wrote: > I've got buildbot running reasonably reliably, but writing > configuration files in Python is a little annoying as I don't really > know what I'm doing. If you were to make your version of buildbot > decoupled from GHC, and ideally configurable in a similar manner to > XMonad (a library with which you roll your own buildbot, usually in a > couple of lines), then I would consider switching. > I wrote a simple buildbot for DDC [1] and found that most of the code was DDC specific, and not easily retargetable. It might be better to approach it as an opportunity to create a library for writing buildbots, where the one for GHC is just an instance, rather than a highly configurable all-sing-all-dancing buildbot. I haven't used the actual buildbot software though, just my own scripts... Ben. [1] http://code.haskell.org/ddc/thoth-head/ From marlowsd at gmail.com Mon Feb 1 06:19:42 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Mon Feb 1 05:51:38 2010 Subject: Exploding SpecConstr In-Reply-To: References: <947938FD-89E3-4B6E-9547-850824A42812@anu.edu.au> <4B630AEA.5020506@gmail.com> <404396ef1001290858i555226ccj809a3e378ff1569a@mail.gmail.com> Message-ID: <4B66B8CE.507@gmail.com> On 01/02/2010 06:41, Ben Lippmeier wrote: > > On 30/01/2010, at 3:58 , Neil Mitchell wrote: > >> I've got buildbot running reasonably reliably, but writing >> configuration files in Python is a little annoying as I don't really >> know what I'm doing. If you were to make your version of buildbot >> decoupled from GHC, and ideally configurable in a similar manner to >> XMonad (a library with which you roll your own buildbot, usually in a >> couple of lines), then I would consider switching. >> > > I wrote a simple buildbot for DDC [1] and found that most of the code was DDC specific, and not easily retargetable. It might be better to approach it as an opportunity to create a library for writing buildbots, where the one for GHC is just an instance, rather than a highly configurable all-sing-all-dancing buildbot. I haven't used the actual buildbot software though, just my own scripts... Correct me if I'm wrong, but your buildbot runs entirely on the client and is designed to do a (configurable) build and email the results, right? What we're trying to do is more like BuildBot: there's a server where all the configuration happens, and the clients run a very thin program that simply takes instructions from the server, executes them, and sends back the results. We want to handle an unreliable connection between server and client, which is what BuildBot doesn't do. We also want the server and client to be restartable without losing any state, as far as possible. The key concept is that the build instructions are written in a Haskell DSL on the server, e.g. do r <- buildstep "get repo" "darcs get http://darcs.haskell.org/ghc" if failed r then return something else do r <- buildstep "boot" "sh boot" ... but this needs to be restartable if the server goes down, so the idea is that we use a caching monad where the results are saved and the instructions are re-run from scratch to generate the next build step, using the technique that originally appeared in WASH. I think this is a pretty cool way to write a build server - we get to write our build instructions in Haskell, but we keep the advantages of stateless communication with the client. This part is all independent of GHC and could easily be put into a library. I expect we'll have a library of build instructions that are specific to GHC too, and it makes sense to separate those out. Cheers, Simon From ghcbuild at microsoft.com Mon Feb 1 19:58:44 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Mon Feb 1 19:58:45 2010 Subject: [nightly] 01-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100202005844.89789324076@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Mon Feb 1 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. (GHC Version 6.13.20100201) **** Building stage 1 compiler ... ok. **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Tue Feb 2 01:27:10 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Mon Feb 1 21:28:49 GMT 2010 2452 total tests, which gave rise to 13524 test cases, of which 0 caused framework failures 2787 were skipped 10382 expected passes 348 expected failures 0 unexpected passes 7 unexpected failures Unexpected failures: 3429(ghci) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) ---------------------------------------------------- Nightly run ended at Tue Feb 2 01:27:10 GMT 2010 From Ben.Lippmeier at anu.edu.au Mon Feb 1 21:35:16 2010 From: Ben.Lippmeier at anu.edu.au (Ben Lippmeier) Date: Mon Feb 1 21:06:54 2010 Subject: new buildbots - was: Exploding SpecConstr In-Reply-To: <4B66B8CE.507@gmail.com> References: <947938FD-89E3-4B6E-9547-850824A42812@anu.edu.au> <4B630AEA.5020506@gmail.com> <404396ef1001290858i555226ccj809a3e378ff1569a@mail.gmail.com> <4B66B8CE.507@gmail.com> Message-ID: On 01/02/2010, at 22:19 , Simon Marlow wrote: > Correct me if I'm wrong, but your buildbot runs entirely on the client and is designed to do a (configurable) build and email the results, right? Yes, that's all -- it's just something that I hacked together. The potential to use Ian's new buildbot for DDC as well makes me extra-keen to help with the development / testing. (And I'm guessing Neil feels the same for his own projects). The only thing that makes me wary is the potential for "complexity inversion", where the "configuration language" just becomes a poor simulacrum of the implementation language -- which is where my own code seemed to be going. Approaching it as a DSL / framework / library should help ward this off though. > What we're trying to do is more like BuildBot: there's a server where all the configuration happens, and the clients run a very thin program that simply takes instructions from the server, executes them, and sends back the results. > > We want to handle an unreliable connection between server and client, which is what BuildBot doesn't do. We also want the server and client to be restartable without losing any state, as far as possible. > > The key concept is that the build instructions are written in a Haskell DSL on the server, e.g. > > do > r <- buildstep "get repo" "darcs get http://darcs.haskell.org/ghc" > if failed r > then return something > else do > r <- buildstep "boot" "sh boot" > ... > > but this needs to be restartable if the server goes down, so the idea is that we use a caching monad where the results are saved and the instructions are re-run from scratch to generate the next build step, using the technique that originally appeared in WASH. I think this is a pretty cool way to write a build server - we get to write our build instructions in Haskell, but we keep the advantages of stateless communication with the client. > > This part is all independent of GHC and could easily be put into a library. I expect we'll have a library of build instructions that are specific to GHC too, and it makes sense to separate those out. Sounds good to me!. Yeah, such a framework sounds easily reusable. Hopefully it will also have a story about graceful exception handling. For example, if my script tries to fetch the darcs repo but code.haskell.org is down it just dies. It'd be great to have a combinator for "if you can't connect then wait 10 mins then try again, up to 5 times, before giving up for today." Ben. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/cvs-ghc/attachments/20100201/0b3b662f/attachment-0001.html From ghcbuild at microsoft.com Mon Feb 1 23:35:08 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Mon Feb 1 23:35:10 2010 Subject: [nightly] 01-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100202043508.9C92832407D@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Mon Feb 1 18:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. (GHC Version 6.13.20100201) **** Building stage 1 compiler ... ok. **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Tue Feb 2 05:03:35 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Mon Feb 1 23:09:56 GMT 2010 2452 total tests, which gave rise to 13524 test cases, of which 0 caused framework failures 2802 were skipped 10352 expected passes 350 expected failures 0 unexpected passes 20 unexpected failures Unexpected failures: annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) concprog001(ghci) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Tue Feb 2 05:03:35 GMT 2010 From simonpj at microsoft.com Tue Feb 2 02:19:50 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Tue Feb 2 01:51:24 2010 Subject: patch applied (ghc): Fix Trac #3831: blowup in SpecConstr Message-ID: <20100202071950.GA18848@abbot.galois.com> Sun Jan 31 16:24:20 PST 2010 simonpj@microsoft.com * Fix Trac #3831: blowup in SpecConstr It turned out that there were two bugs. First, we were getting an exponential number of specialisations when we had a deep nest of join points. See Note [Avoiding exponential blowup]. I fixed this by dividing sc_count (in ScEnv) by the number of specialisations when recursing. Crude but effective. Second, when making specialisations I was looking at the result of applying specExpr to the RHS of the function, whereas I should have been looking at the original RHS. See Note [Specialise original body]. There's a tantalising missed opportunity here, though. In this example (recorded as a test simplCore/should_compile/T3831), each join point has *exactly one* call pattern, so we should really just specialise for that alone, in which case there's zero code-blow-up. In particular, we don't need the *original* RHS at all. I need to think more about how to exploit this. But the blowup is now limited, so compiling terminfo with -O2 works again. M ./compiler/specialise/SpecConstr.lhs -41 +96 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100201002420-1287e-da7031609e7c01b4ed22ba2aaaa31a434894bb4f.gz From simonpj at microsoft.com Tue Feb 2 02:22:15 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Tue Feb 2 01:53:50 2010 Subject: patch applied (testsuite): Test Trac #3831: SpecConstr blowup Message-ID: <20100202072214.GA18975@abbot.galois.com> Mon Feb 1 23:24:05 PST 2010 simonpj@microsoft.com * Test Trac #3831: SpecConstr blowup A ./tests/ghc-regress/simplCore/should_compile/T3831.hs M ./tests/ghc-regress/simplCore/should_compile/all.T +1 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100202072405-1287e-725b2c5a6283b334dac64338869956f7123c7d74.gz From marlowsd at gmail.com Tue Feb 2 07:04:14 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Tue Feb 2 06:35:58 2010 Subject: patch applied (testsuite): Test Trac #3831: SpecConstr blowup In-Reply-To: <20100202072214.GA18975@abbot.galois.com> References: <20100202072214.GA18975@abbot.galois.com> Message-ID: <4B6814BE.2030902@gmail.com> On 02/02/2010 07:22, Simon Peyton Jones wrote: > Mon Feb 1 23:24:05 PST 2010 simonpj@microsoft.com > * Test Trac #3831: SpecConstr blowup > > A ./tests/ghc-regress/simplCore/should_compile/T3831.hs > M ./tests/ghc-regress/simplCore/should_compile/all.T +1 I think this test is broken: =====> T3831(normal) 590 of 2453 [0, 0, 0] cd ./simplCore/should_compile && '/64playpen/simonmar/validate/bindisttest/insta ll dir/bin/ghc' -fforce-recomp -dcore-lint -dcmm-lint -no-user-package-conf -dn o-debug-output -c T3831.hs >T3831.comp.stderr 2>&1 Compile failed (status 256) errors were: T3831.hs:15:5: `optputCap' is not a (visible) method of class `OutputCap' Cheers, Simon From marlowsd at gmail.com Tue Feb 2 07:07:16 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Tue Feb 2 06:38:49 2010 Subject: patch applied (ghc): Fix typo in error message (#3848) Message-ID: <20100202120715.GA27017@abbot.galois.com> Sat Jan 30 01:46:28 PST 2010 Simon Marlow * Fix typo in error message (#3848) M ./rts/RtsAPI.c -1 +1 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100130094628-12142-a6f6117fd78d4265b79090554ad471eed0d048de.gz From marlowsd at gmail.com Tue Feb 2 07:33:56 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Tue Feb 2 07:05:30 2010 Subject: patch applied (testsuite): accept output (#3848) Message-ID: <20100202123356.GA28105@abbot.galois.com> Tue Feb 2 04:06:26 PST 2010 Simon Marlow * accept output (#3848) M ./tests/ghc-regress/ffi/should_run/ffi020.stderr -1 +1 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100202120626-12142-5db9dc487d5e6cb3d686f2cd72b74e85496a01e2.gz From felipe.lessa at gmail.com Tue Feb 2 07:55:23 2010 From: felipe.lessa at gmail.com (Felipe Lessa) Date: Tue Feb 2 07:27:00 2010 Subject: patch applied (ghc): Fix Trac #3831: blowup in SpecConstr In-Reply-To: <20100202071950.GA18848@abbot.galois.com> References: <20100202071950.GA18848@abbot.galois.com> Message-ID: <20100202125523.GA23456@kira.casa> On Mon, Feb 01, 2010 at 11:19:50PM -0800, Simon Peyton Jones wrote: > View patch online: > http://darcs.haskell.org/ghc/_darcs/patches/20100201002420-1287e-da7031609e7c01b4ed22ba2aaaa31a434894bb4f.gz There's a small typo "oringal" on the second note, "Specialise original body". -- Felipe. From ghcbuild at microsoft.com Tue Feb 2 19:47:13 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Tue Feb 2 19:47:20 2010 Subject: [nightly] 02-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100203004713.5024032482C@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Tue Feb 2 19:00:01 GMT 2010. **** checking out new source tree ... No checkpoint. No checkpoint. warning: . already present; omitting No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. ok. (GHC Version 6.13.20100202) **** Building stage 1 compiler ... ok. **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Wed Feb 3 01:15:40 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Tue Feb 2 21:26:13 GMT 2010 2453 total tests, which gave rise to 13530 test cases, of which 0 caused framework failures 2787 were skipped 10382 expected passes 348 expected failures 0 unexpected passes 13 unexpected failures Unexpected failures: T1969(normal) T3831(normal,optc,hpc,optasm,profc,profasm) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) ---------------------------------------------------- Nightly run ended at Wed Feb 3 01:15:40 GMT 2010 From ghcbuild at microsoft.com Tue Feb 2 23:34:16 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Tue Feb 2 23:34:17 2010 Subject: [nightly] 02-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100203043416.682AF324004@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Tue Feb 2 18:00:02 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. (GHC Version 6.13.20100202) **** Building stage 1 compiler ... ok. **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Wed Feb 3 05:02:40 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Tue Feb 2 23:08:57 GMT 2010 2453 total tests, which gave rise to 13530 test cases, of which 0 caused framework failures 2802 were skipped 10351 expected passes 350 expected failures 0 unexpected passes 27 unexpected failures Unexpected failures: CPUTime001(threaded2) T3831(normal,optc,hpc,optasm,profc,profasm) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) concprog001(ghci) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Wed Feb 3 05:02:40 GMT 2010 From Ben.Lippmeier at anu.edu.au Wed Feb 3 02:02:38 2010 From: Ben.Lippmeier at anu.edu.au (Ben Lippmeier) Date: Wed Feb 3 01:34:12 2010 Subject: sparc/solaris binary distro for 6.12.1 Message-ID: <5189D7F0-0703-4516-9253-FF0DBD0BB040@anu.edu.au> I've uploaded a sparc/solaris binary distro for 6.12.1 to my home dir on darcs.haskell.org. Can someone please add this to the GHC home page? Some tests from the suite fail, which I'll look into, but this binary distro is the best solution we currently have for bootstrapping the head on this platform. Thanks, Ben. Unexpected failures: 1861(optc,profc) 3171(normal) 3586(normal) T3016(hpc,optasm,profasm) T3294(normal) arith011(profc) barton-mangler-bug(profc) break024(ghci) conc012(ghci) concprog001(ghci) concprog002(threaded2) derefnull(profc,profthreaded) divbyzero(profc,profthreaded) hClose002(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,profthreaded) hClose003(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,profthreaded) joao-circular(optc,profc) rtsflags001(normal) seward-space-leak(ghci) signals004(ghci,threaded1,threaded2,profthreaded) space_leak_001(normal,optc,hpc,optasm,ghci) testblockalloc(normal,threaded1) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,profthreaded) From igloo at earth.li Wed Feb 3 09:05:29 2010 From: igloo at earth.li (Ian Lynagh) Date: Wed Feb 3 09:05:30 2010 Subject: patch applied (/haskell/ghc): Add a 6.12.1 sparc/solaris build from Ben Lippmeier Message-ID: <20100203140529.GA19770@haskell.cs.yale.edu> Wed Feb 3 09:33:23 EST 2010 Ian Lynagh * Add a 6.12.1 sparc/solaris build from Ben Lippmeier M ./download_ghc_6_12_1.html -18 +5 From igloo at earth.li Wed Feb 3 09:06:51 2010 From: igloo at earth.li (Ian Lynagh) Date: Wed Feb 3 09:06:52 2010 Subject: patch applied (/haskell/ghc): Move the " binaries" anchor to the right place in the 6.12.1 download page Message-ID: <20100203140651.GA20000@haskell.cs.yale.edu> Wed Feb 3 09:34:52 EST 2010 Ian Lynagh * Move the "binaries" anchor to the right place in the 6.12.1 download page M ./download_ghc_6_12_1.html -2 +2 From igloo at earth.li Wed Feb 3 09:08:26 2010 From: igloo at earth.li (Ian Lynagh) Date: Wed Feb 3 09:08:27 2010 Subject: patch applied (/haskell/ghc): Improve 6.12.1 download page table of contents Message-ID: <20100203140826.GA20518@haskell.cs.yale.edu> Wed Feb 3 09:36:37 EST 2010 Ian Lynagh * Improve 6.12.1 download page table of contents M ./download_ghc_6_12_1.html -1 +2 From igloo at earth.li Wed Feb 3 09:37:39 2010 From: igloo at earth.li (Ian Lynagh) Date: Wed Feb 3 09:09:09 2010 Subject: sparc/solaris binary distro for 6.12.1 In-Reply-To: <5189D7F0-0703-4516-9253-FF0DBD0BB040@anu.edu.au> References: <5189D7F0-0703-4516-9253-FF0DBD0BB040@anu.edu.au> Message-ID: <20100203143739.GA3616@matrix.chaos.earth.li> On Wed, Feb 03, 2010 at 06:02:38PM +1100, Ben Lippmeier wrote: > > I've uploaded a sparc/solaris binary distro for 6.12.1 to my home dir on darcs.haskell.org. Thanks! > Can someone please add this to the GHC home page? Done. Thanks Ian From ben.lippmeier at anu.edu.au Wed Feb 3 18:29:55 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Wed Feb 3 18:01:25 2010 Subject: patch applied (ghc): Cast to (void*) to stifle warning about signedness Message-ID: <20100203232955.GA26630@abbot.galois.com> Tue Feb 2 18:30:29 PST 2010 benl@cse.unsw.edu.au * Cast to (void*) to stifle warning about signedness M ./rts/posix/OSMem.c -2 +2 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100203023029-578ee-23f4c29ac2758283a310b65b4004b38189662765.gz From ben.lippmeier at anu.edu.au Wed Feb 3 18:29:54 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Wed Feb 3 18:01:27 2010 Subject: patch applied (ghc): Add missing import sm_mutex, which fixes the -fvia-c build Message-ID: <20100203232954.GA26610@abbot.galois.com> Mon Feb 1 21:14:51 PST 2010 benl@cse.unsw.edu.au * Add missing import sm_mutex, which fixes the -fvia-c build M ./rts/PrimOps.cmm +1 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100202051451-578ee-7e92560040552bc1a732862b36d40eecb925c7da.gz From ben.lippmeier at anu.edu.au Wed Feb 3 18:29:57 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Wed Feb 3 18:01:31 2010 Subject: patch applied (ghc): Stifle warning about printf format strings Message-ID: <20100203232957.GA26651@abbot.galois.com> Tue Feb 2 18:31:24 PST 2010 benl@cse.unsw.edu.au * Stifle warning about printf format strings M ./rts/Hpc.c -1 +1 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100203023124-578ee-9017865e19813d4f89b768e231bfa86e171362de.gz From ben.lippmeier at anu.edu.au Wed Feb 3 18:29:59 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Wed Feb 3 18:01:32 2010 Subject: patch applied (ghc): Disable -Wcast-align when building the rts Message-ID: <20100203232958.GA26672@abbot.galois.com> Tue Feb 2 18:46:05 PST 2010 benl@cse.unsw.edu.au * Disable -Wcast-align when building the rts This flag seems buggy on GCC 4.1.2, which is the only GCC version we have that can bootstrap the SPARC build. We end up with lots of supurious warnings of the form "cast increases required alignment of target type". Some legitimate warnings can be fixed by adding an intermediate cast to (void*), but we get others in rts/sm/GCUtils.c concerning the gct var that look innocuous to me. We could enable this again once we deprecate support for registerised builds on this arch. M ./rts/ghc.mk -1 +9 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100203024605-578ee-39c8345e6bffc2dfd9ad0d3adfe3fd1c3e92eff5.gz From ben.lippmeier at anu.edu.au Wed Feb 3 18:30:00 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Wed Feb 3 18:01:33 2010 Subject: patch applied (ghc): Use bash to run boot-pkgs, the default Solaris shell doesn't support the -ot test flag Message-ID: <20100203233000.GA26692@abbot.galois.com> Tue Feb 2 19:07:48 PST 2010 benl@cse.unsw.edu.au * Use bash to run boot-pkgs, the default Solaris shell doesn't support the -ot test flag M ./boot -1 +1 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100203030748-578ee-926e54408c9d88fe907f0950575f8161d0e02fff.gz From Ben.Lippmeier at anu.edu.au Wed Feb 3 18:49:50 2010 From: Ben.Lippmeier at anu.edu.au (Ben Lippmeier) Date: Wed Feb 3 18:21:21 2010 Subject: sparc/solaris binary distro for 6.12.1 In-Reply-To: <4B69CC49.4060501@dfki.de> References: <5189D7F0-0703-4516-9253-FF0DBD0BB040@anu.edu.au> <20100203143739.GA3616@matrix.chaos.earth.li> <4B69CC49.4060501@dfki.de> Message-ID: <814093A1-2B24-4957-A98F-5EB70AC45D79@anu.edu.au> hmm. I suppose I am using the term "Solaris" fairly loosely! My build environment contains: GNU binutils 2.20 GNU coreutils 8.4 GNU make 3.81 GCC 4.1.2 GNU m4 1.4.13 Python 2.6.4 GNU sed 4.2 GNU tar 1.20 GNU grep 2.5 Alex 2.2 Happy 1.17 That's all the stuff I had to install to get it to compile, and with the same environment the head now builds with 6.12.1 as the bootstrap compiler. That's using the standard "perf" and "devel" settings in mk/build.mk as well, ie without needing to go via-c. I'll add the package versions to the developer's wiki. Ian: you might want to add these versions to the GHC page, else link to the developers wiki with a note saying that people should install this stuff first to avoid pain. Ben. On 04/02/2010, at 6:19 , Christian Maeder wrote: > Hi, > > I've tried to install this distro, because building my own failed due to > an too old gmake version. > > Installing failed, too, with: > > gmake -r --no-print-directory -f ghc.mk install BINDIST=YES > NO_INCLUDE_DEPS=YES > ghc.mk:95: Extraneous text after `else' directive > ghc.mk:96: Extraneous text after `else' directive > ghc.mk:96: *** only one `else' per conditional. Stop. > gmake: *** [install] Error 2 > > but without being warned that gmake is too old! > > Cheers Christian > > -bash-3.00$ gmake --version > GNU Make 3.80 > Copyright (C) 2002 Free Software Foundation, Inc. > > > Ian Lynagh schrieb: >> On Wed, Feb 03, 2010 at 06:02:38PM +1100, Ben Lippmeier wrote: >>> I've uploaded a sparc/solaris binary distro for 6.12.1 to my home dir on darcs.haskell.org. >> >> Thanks! >> >>> Can someone please add this to the GHC home page? >> >> Done. >> >> >> Thanks >> Ian From ghcbuild at microsoft.com Wed Feb 3 19:42:30 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Wed Feb 3 19:42:33 2010 Subject: [nightly] 03-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100204004230.9BA413241FD@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Wed Feb 3 19:00:01 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. (GHC Version 6.13.20100203) **** Building stage 1 compiler ... ok. **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... Received disconnect from 128.36.229.215: 2: Corrupted MAC on input. lost connection failed. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Thu Feb 4 01:11:02 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Wed Feb 3 21:18:46 GMT 2010 2453 total tests, which gave rise to 13530 test cases, of which 0 caused framework failures 2787 were skipped 10382 expected passes 348 expected failures 0 unexpected passes 13 unexpected failures Unexpected failures: T1969(normal) T3831(normal,optc,hpc,optasm,profc,profasm) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) ---------------------------------------------------- Nightly run ended at Thu Feb 4 01:11:02 GMT 2010 From ghcbuild at microsoft.com Wed Feb 3 23:52:46 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Wed Feb 3 23:52:48 2010 Subject: [nightly] 03-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100204045246.E73ED324149@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Wed Feb 3 18:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. ok. (GHC Version 6.13.20100203) **** Building stage 1 compiler ... ok. **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Thu Feb 4 05:21:19 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Wed Feb 3 23:16:35 GMT 2010 2453 total tests, which gave rise to 13530 test cases, of which 0 caused framework failures 2802 were skipped 10352 expected passes 350 expected failures 0 unexpected passes 26 unexpected failures Unexpected failures: T2120(profthreaded) T3831(normal,optc,hpc,optasm,profc,profasm) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Thu Feb 4 05:21:19 GMT 2010 From igloo at earth.li Thu Feb 4 06:47:04 2010 From: igloo at earth.li (Ian Lynagh) Date: Thu Feb 4 06:18:32 2010 Subject: sparc/solaris binary distro for 6.12.1 In-Reply-To: <814093A1-2B24-4957-A98F-5EB70AC45D79@anu.edu.au> References: <5189D7F0-0703-4516-9253-FF0DBD0BB040@anu.edu.au> <20100203143739.GA3616@matrix.chaos.earth.li> <4B69CC49.4060501@dfki.de> <814093A1-2B24-4957-A98F-5EB70AC45D79@anu.edu.au> Message-ID: <20100204114704.GA18250@matrix.chaos.earth.li> Hi Ben, On Thu, Feb 04, 2010 at 10:49:50AM +1100, Ben Lippmeier wrote: > > Ian: you might want to add these versions to the GHC page, else link to the developers wiki with a note saying that people should install this stuff first to avoid pain. I think that this would be best put on a WikiPage for Sparc/Solaris, linked from http://hackage.haskell.org/trac/ghc/wiki/Platforms Thanks Ian From marlowsd at gmail.com Thu Feb 4 07:50:48 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Thu Feb 4 07:22:16 2010 Subject: patch applied (ghc): add comment for srt_bitmap field Message-ID: <20100204125047.GA20026@abbot.galois.com> Wed Feb 3 02:34:31 PST 2010 Simon Marlow * add comment for srt_bitmap field M ./includes/rts/storage/InfoTables.h -1 +6 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100203103431-12142-08299a7feabfda4003c813fdc00a3cb7f187effa.gz From marlowsd at gmail.com Thu Feb 4 07:50:49 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Thu Feb 4 07:22:19 2010 Subject: patch applied (ghc): add libraries/binary Message-ID: <20100204125049.GA20046@abbot.galois.com> Wed Feb 3 05:54:21 PST 2010 Simon Marlow * add libraries/binary M ./.darcs-boring +1 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100203135421-12142-af723b3f00a833788bf290dad8ef0ae80d907261.gz From marlowsd at gmail.com Thu Feb 4 07:50:51 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Thu Feb 4 07:22:19 2010 Subject: patch applied (ghc): Implement SSE2 floating-point support in the x86 native code generator (#594) Message-ID: <20100204125051.GA20066@abbot.galois.com> Thu Feb 4 02:48:49 PST 2010 Simon Marlow * Implement SSE2 floating-point support in the x86 native code generator (#594) The new flag -msse2 enables code generation for SSE2 on x86. It results in substantially faster floating-point performance; the main reason for doing this was that our x87 code generation is appallingly bad, and since we plan to drop -fvia-C soon, we need a way to generate half-decent floating-point code. The catch is that SSE2 is only available on CPUs that support it (P4+, AMD K8+). We'll have to think hard about whether we should enable it by default for the libraries we ship. In the meantime, at least -msse2 should be an acceptable replacement for "-fvia-C -optc-ffast-math -fexcess-precision". SSE2 also has the advantage of performing all operations at the correct precision, so floating-point results are consistent with other platforms. I also tweaked the x87 code generation a bit while I was here, now it's slighlty less bad than before. M ./compiler/main/DynFlags.hs +4 M ./compiler/nativeGen/PPC/Ppr.hs +1 M ./compiler/nativeGen/PPC/Regs.hs -23 +6 M ./compiler/nativeGen/Reg.hs +5 M ./compiler/nativeGen/RegAlloc/Graph/Main.hs -15 +3 M ./compiler/nativeGen/RegAlloc/Graph/TrivColorable.hs -2 +16 M ./compiler/nativeGen/RegClass.hs +3 M ./compiler/nativeGen/SPARC/Instr.hs +3 M ./compiler/nativeGen/SPARC/Ppr.hs +1 M ./compiler/nativeGen/SPARC/Regs.hs -7 +7 M ./compiler/nativeGen/X86/CodeGen.hs -412 +380 M ./compiler/nativeGen/X86/Instr.hs -61 +27 M ./compiler/nativeGen/X86/Ppr.hs -43 +50 M ./compiler/nativeGen/X86/RegInfo.hs -16 +18 M ./compiler/nativeGen/X86/Regs.hs -165 +78 M ./docs/users_guide/flags.xml +8 M ./docs/users_guide/using.xml -1 +19 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100204104849-12142-a1cb7bcdfb7ec0aec38f612fad44ad0ab6dab294.gz From igloo at earth.li Thu Feb 4 11:50:20 2010 From: igloo at earth.li (Ian Lynagh) Date: Thu Feb 4 11:21:48 2010 Subject: patch applied (ghc-6.12/ghc): Add some missing getStablePtr()s for CAFs that the RTS refers to Message-ID: <20100204165020.GA29612@abbot.galois.com> Fri Jan 22 06:36:58 PST 2010 Simon Marlow * Add some missing getStablePtr()s for CAFs that the RTS refers to A recent patch ("Refactor CoreArity a bit") changed the arity of GHC.Conc.runSparks such that it became a CAF, and the RTS was not explicitly retaining it, which led to a crash when the CAF got GC'd. While fixing this I found a couple of other closures that the RTS refers to which weren't getting the correct CAF treatment. M ./rts/Prelude.h +2 M ./rts/RtsStartup.c -1 +8 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100122143658-12142-1a648c6beae6b0bf9ba3a1edfed67d579f53fa26.gz From igloo at earth.li Thu Feb 4 11:50:18 2010 From: igloo at earth.li (Ian Lynagh) Date: Thu Feb 4 11:21:49 2010 Subject: patch applied (ghc-6.12/ghc): Use DLL aware runIO_closure instead of base_GHCziTopHandler_runIO_closure directly Message-ID: <20100204165018.GA29592@abbot.galois.com> Fri Nov 13 22:15:59 PST 2009 Ben.Lippmeier@anu.edu.au * Use DLL aware runIO_closure instead of base_GHCziTopHandler_runIO_closure directly M ./rts/RtsStartup.c -2 +3 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20091114061559-43c66-ad46e3194d69291eb80cc72b805d42a8ca8e284a.gz From igloo at earth.li Thu Feb 4 11:50:15 2010 From: igloo at earth.li (Ian Lynagh) Date: Thu Feb 4 11:21:56 2010 Subject: patch applied (ghc-6.12/ghc): Win32 yieldThread(): use SwitchToThread() instead of Sleep(0) Message-ID: <20100204165015.GA29548@abbot.galois.com> Wed Jan 27 05:31:06 PST 2010 Simon Marlow * Win32 yieldThread(): use SwitchToThread() instead of Sleep(0) M ./rts/win32/OSThreads.c -1 +1 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100127133106-12142-7fa089880e7d015a031dbaa88f20bb69e2562043.gz From igloo at earth.li Thu Feb 4 11:50:21 2010 From: igloo at earth.li (Ian Lynagh) Date: Thu Feb 4 11:21:57 2010 Subject: patch applied (ghc-6.12/ghc): fix Windows build (GHC.Conc.runHandlers is Unix-only) Message-ID: <20100204165021.GA29632@abbot.galois.com> Wed Jan 27 06:57:18 PST 2010 Simon Marlow * fix Windows build (GHC.Conc.runHandlers is Unix-only) M ./rts/RtsStartup.c +2 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100127145718-12142-0132e02090aecdb1cc3b0034c5057922c3baba79.gz From igloo at earth.li Thu Feb 4 11:50:14 2010 From: igloo at earth.li (Ian Lynagh) Date: Thu Feb 4 11:21:58 2010 Subject: patch applied (ghc-6.12/ghc): fix warnings Message-ID: <20100204165013.GA29524@abbot.galois.com> Wed Jan 27 05:30:40 PST 2010 Simon Marlow * fix warnings M ./rts/ghc.mk -1 M ./rts/win32/Ticker.c -1 +1 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100127133040-12142-7747bc26af14567770c5a009dcfe403d62819d47.gz From igloo at earth.li Thu Feb 4 11:50:17 2010 From: igloo at earth.li (Ian Lynagh) Date: Thu Feb 4 11:21:58 2010 Subject: patch applied (ghc-6.12/ghc): RTS also uses runSparks_closure from base Message-ID: <20100204165017.GA29570@abbot.galois.com> Fri Nov 13 22:14:45 PST 2009 Ben.Lippmeier@anu.edu.au * RTS also uses runSparks_closure from base M ./rts/Prelude.h +2 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20091114061445-43c66-fb507de5075737c95096043f64cb2d8903587d36.gz From igloo at earth.li Thu Feb 4 12:42:39 2010 From: igloo at earth.li (Ian Lynagh) Date: Thu Feb 4 12:14:07 2010 Subject: patch applied (ghc-6.12/ghc): Fix typo in error message (#3848) Message-ID: <20100204174239.GA31731@abbot.galois.com> Sat Jan 30 01:46:28 PST 2010 Simon Marlow * Fix typo in error message (#3848) M ./rts/RtsAPI.c -1 +1 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100130094628-12142-a6f6117fd78d4265b79090554ad471eed0d048de.gz From igloo at earth.li Thu Feb 4 12:42:41 2010 From: igloo at earth.li (Ian Lynagh) Date: Thu Feb 4 12:14:10 2010 Subject: patch applied (ghc-6.12/ghc): Add missing import sm_mutex, which fixes the -fvia-c build Message-ID: <20100204174241.GA31751@abbot.galois.com> Mon Feb 1 21:14:51 PST 2010 benl@cse.unsw.edu.au * Add missing import sm_mutex, which fixes the -fvia-c build M ./rts/PrimOps.cmm +1 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100202051451-578ee-7e92560040552bc1a732862b36d40eecb925c7da.gz From igloo at earth.li Thu Feb 4 12:42:42 2010 From: igloo at earth.li (Ian Lynagh) Date: Thu Feb 4 12:14:10 2010 Subject: patch applied (ghc-6.12/ghc): Cast to (void*) to stifle warning about signedness Message-ID: <20100204174242.GA31771@abbot.galois.com> Tue Feb 2 18:30:29 PST 2010 benl@cse.unsw.edu.au * Cast to (void*) to stifle warning about signedness M ./rts/posix/OSMem.c -2 +2 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100203023029-578ee-23f4c29ac2758283a310b65b4004b38189662765.gz From igloo at earth.li Thu Feb 4 12:42:44 2010 From: igloo at earth.li (Ian Lynagh) Date: Thu Feb 4 12:14:14 2010 Subject: patch applied (ghc-6.12/ghc): Stifle warning about printf format strings Message-ID: <20100204174244.GA31791@abbot.galois.com> Tue Feb 2 18:31:24 PST 2010 benl@cse.unsw.edu.au * Stifle warning about printf format strings M ./rts/Hpc.c -1 +1 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100203023124-578ee-9017865e19813d4f89b768e231bfa86e171362de.gz From igloo at earth.li Thu Feb 4 12:42:45 2010 From: igloo at earth.li (Ian Lynagh) Date: Thu Feb 4 12:14:14 2010 Subject: patch applied (ghc-6.12/ghc): Disable -Wcast-align when building the rts Message-ID: <20100204174245.GA31811@abbot.galois.com> Tue Feb 2 18:46:05 PST 2010 benl@cse.unsw.edu.au * Disable -Wcast-align when building the rts This flag seems buggy on GCC 4.1.2, which is the only GCC version we have that can bootstrap the SPARC build. We end up with lots of supurious warnings of the form "cast increases required alignment of target type". Some legitimate warnings can be fixed by adding an intermediate cast to (void*), but we get others in rts/sm/GCUtils.c concerning the gct var that look innocuous to me. We could enable this again once we deprecate support for registerised builds on this arch. M ./rts/ghc.mk -1 +9 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100203024605-578ee-39c8345e6bffc2dfd9ad0d3adfe3fd1c3e92eff5.gz From igloo at earth.li Thu Feb 4 12:42:47 2010 From: igloo at earth.li (Ian Lynagh) Date: Thu Feb 4 12:14:16 2010 Subject: patch applied (ghc-6.12/ghc): Use bash to run boot-pkgs, the default Solaris shell doesn't support the -ot test flag Message-ID: <20100204174247.GA31831@abbot.galois.com> Tue Feb 2 19:07:48 PST 2010 benl@cse.unsw.edu.au * Use bash to run boot-pkgs, the default Solaris shell doesn't support the -ot test flag M ./boot -1 +1 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100203030748-578ee-926e54408c9d88fe907f0950575f8161d0e02fff.gz From igloo at earth.li Thu Feb 4 12:45:33 2010 From: igloo at earth.li (Ian Lynagh) Date: Thu Feb 4 12:17:00 2010 Subject: patch applied (ghc-6.12/testsuite): accept output (#3848) Message-ID: <20100204174533.GA32131@abbot.galois.com> Tue Feb 2 04:06:26 PST 2010 Simon Marlow * accept output (#3848) M ./tests/ghc-regress/ffi/should_run/ffi020.stderr -1 +1 View patch online: http://darcs.haskell.org/ghc-6.12/testsuite/_darcs/patches/20100202120626-12142-5db9dc487d5e6cb3d686f2cd72b74e85496a01e2.gz From ghcbuild at microsoft.com Thu Feb 4 20:05:18 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Thu Feb 4 20:05:20 2010 Subject: [nightly] 04-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100205010518.9AB9B324223@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Thu Feb 4 19:00:01 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. (GHC Version 6.13.20100204) **** Building stage 1 compiler ... ok. **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Fri Feb 5 01:33:53 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Thu Feb 4 21:19:02 GMT 2010 2453 total tests, which gave rise to 13530 test cases, of which 0 caused framework failures 2787 were skipped 10372 expected passes 348 expected failures 0 unexpected passes 23 unexpected failures Unexpected failures: 3429(ghci) T1969(normal) T3831(normal,optc,hpc,optasm,profc,profasm) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci,threaded2) ffi009(normal,hpc,optasm,profasm,threaded1,threaded2,dyn,profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) ---------------------------------------------------- Nightly run ended at Fri Feb 5 01:33:53 GMT 2010 From Ben.Lippmeier at anu.edu.au Thu Feb 4 22:46:11 2010 From: Ben.Lippmeier at anu.edu.au (Ben Lippmeier) Date: Thu Feb 4 22:17:39 2010 Subject: sparc/solaris binary distro for 6.12.1 In-Reply-To: <4B6AA200.9040909@dfki.de> References: <5189D7F0-0703-4516-9253-FF0DBD0BB040@anu.edu.au> <20100203143739.GA3616@matrix.chaos.earth.li> <4B69CC49.4060501@dfki.de> <814093A1-2B24-4957-A98F-5EB70AC45D79@anu.edu.au> <4B6AA200.9040909@dfki.de> Message-ID: <4DE359CC-808F-48D8-8F67-AC44C1A5F25C@anu.edu.au> Sorry, this looks like a problem in the build system. An absolute path to GCC was baked into the compiler where it shouldn't have been. GHC also uses GCC as an assembler, which is why ghci works but a regular compile doesn't. I've added a ticket for it, and am doing a new build. I'll upload a new distro once the build has finished. Ben. On 04/02/2010, at 21:31 , Christian Maeder wrote: > Hi Ben, > > after installing GNU make 3.81 I was able to install your distro. > ghci works, but ghc fails with: > > -bash-3.00$ ghc --make Setup.hs > [1 of 1] Compiling Main ( Setup.hs, Setup.o ) > ghc: could not execute: /export/home/benl/software/gcc-4.1.2/bin/gcc > > My environment is as follows: > > SUNWbinutils DESC: GNU binutils - Binary file utilities 2.15 > > No idea about coreutils. > > m4, sed, tar, grep must all be called as gm4, gsed, gtar and ggrep > (otherwise the version is unknown). > > Python 2.4.1 > GCC 4.2.2 > > GNU M4 1.4.8 > GNU sed version 4.1.4 > GNU tar 1.15.1 > GNU grep 2.4.2 > > Alex version 2.0.1 > Happy Version 1.15 > > Cheers Christian > > Ben Lippmeier schrieb: >> hmm. I suppose I am using the term "Solaris" fairly loosely! >> >> My build environment contains: >> GNU binutils 2.20 >> GNU coreutils 8.4 >> GNU make 3.81 >> GCC 4.1.2 >> GNU m4 1.4.13 >> Python 2.6.4 >> GNU sed 4.2 >> GNU tar 1.20 >> GNU grep 2.5 >> Alex 2.2 >> Happy 1.17 >> >> That's all the stuff I had to install to get it to compile, and with the same environment the head now builds with 6.12.1 as the bootstrap compiler. That's using the standard "perf" and "devel" settings in mk/build.mk as well, ie without needing to go via-c. >> >> I'll add the package versions to the developer's wiki. >> >> Ian: you might want to add these versions to the GHC page, else link to the developers wiki with a note saying that people should install this stuff first to avoid pain. >> >> Ben. >> >> >> On 04/02/2010, at 6:19 , Christian Maeder wrote: >> >>> Hi, >>> >>> I've tried to install this distro, because building my own failed due to >>> an too old gmake version. >>> >>> Installing failed, too, with: >>> >>> gmake -r --no-print-directory -f ghc.mk install BINDIST=YES >>> NO_INCLUDE_DEPS=YES >>> ghc.mk:95: Extraneous text after `else' directive >>> ghc.mk:96: Extraneous text after `else' directive >>> ghc.mk:96: *** only one `else' per conditional. Stop. >>> gmake: *** [install] Error 2 >>> >>> but without being warned that gmake is too old! >>> >>> Cheers Christian >>> >>> -bash-3.00$ gmake --version >>> GNU Make 3.80 >>> Copyright (C) 2002 Free Software Foundation, Inc. >>> >>> >>> Ian Lynagh schrieb: >>>> On Wed, Feb 03, 2010 at 06:02:38PM +1100, Ben Lippmeier wrote: >>>>> I've uploaded a sparc/solaris binary distro for 6.12.1 to my home dir on darcs.haskell.org. >>>> Thanks! >>>> >>>>> Can someone please add this to the GHC home page? >>>> Done. >>>> >>>> >>>> Thanks >>>> Ian >> >> From ghcbuild at microsoft.com Thu Feb 4 23:45:12 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Thu Feb 4 23:45:13 2010 Subject: [nightly] 04-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100205044512.0F967324134@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Thu Feb 4 18:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. No checkpoint. ok. (GHC Version 6.13.20100204) **** Building stage 1 compiler ... ok. **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (3 failures) **** running nofib (-O -fasm) ... ok. (3 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (3 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Fri Feb 5 05:13:47 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Thu Feb 4 23:18:14 GMT 2010 2453 total tests, which gave rise to 13530 test cases, of which 0 caused framework failures 2802 were skipped 10352 expected passes 350 expected failures 0 unexpected passes 26 unexpected failures Unexpected failures: T3831(normal,optc,hpc,optasm,profc,profasm) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) concprog001(ghci) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Fri Feb 5 05:13:47 GMT 2010 From marlowsd at gmail.com Fri Feb 5 09:21:22 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Fri Feb 5 08:52:47 2010 Subject: patch applied (ghc): Fix a bug introduced in the SSE2 support: callClobberedRegs was wrong Message-ID: <20100205142122.GA5306@abbot.galois.com> Fri Feb 5 03:29:38 PST 2010 Simon Marlow * Fix a bug introduced in the SSE2 support: callClobberedRegs was wrong M ./compiler/nativeGen/X86/Regs.hs -18 +34 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100205112938-12142-ecb4e52e239b3d108551327acede1d64ebaacf90.gz From ghcbuild at microsoft.com Fri Feb 5 19:37:48 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Fri Feb 5 19:37:50 2010 Subject: [nightly] 05-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100206003748.4237E3241CF@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Fri Feb 5 19:00:02 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. (GHC Version 6.13.20100205) **** Building stage 1 compiler ... ok. **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Sat Feb 6 01:06:25 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Fri Feb 5 21:18:59 GMT 2010 2453 total tests, which gave rise to 13530 test cases, of which 0 caused framework failures 2787 were skipped 10374 expected passes 348 expected failures 0 unexpected passes 21 unexpected failures Unexpected failures: T1969(normal) T3831(normal,optc,hpc,optasm,profc,profasm) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci) ffi009(normal,hpc,optasm,profasm,threaded1,threaded2,dyn,profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) ---------------------------------------------------- Nightly run ended at Sat Feb 6 01:06:25 GMT 2010 From ghcbuild at microsoft.com Fri Feb 5 23:29:27 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Fri Feb 5 23:29:29 2010 Subject: [nightly] 05-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100206042927.925AD3241D1@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Fri Feb 5 18:00:01 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. (GHC Version 6.13.20100205) **** Building stage 1 compiler ... ok. **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Sat Feb 6 04:58:06 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Fri Feb 5 23:03:49 GMT 2010 2453 total tests, which gave rise to 13530 test cases, of which 0 caused framework failures 2802 were skipped 10352 expected passes 350 expected failures 0 unexpected passes 26 unexpected failures Unexpected failures: T3831(normal,optc,hpc,optasm,profc,profasm) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) concprog001(ghci) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Sat Feb 6 04:58:06 GMT 2010 From ghcbuild at microsoft.com Sat Feb 6 19:40:31 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sat Feb 6 19:40:33 2010 Subject: [nightly] 06-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100207004031.C638C324117@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Sat Feb 6 19:00:01 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. (GHC Version 6.13.20100206) **** Building stage 1 compiler ... ok. **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Sun Feb 7 01:09:13 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sat Feb 6 21:19:06 GMT 2010 2453 total tests, which gave rise to 13530 test cases, of which 0 caused framework failures 2787 were skipped 10373 expected passes 348 expected failures 0 unexpected passes 22 unexpected failures Unexpected failures: 3429(ghci) T1969(normal) T3831(normal,optc,hpc,optasm,profc,profasm) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci) ffi009(normal,hpc,optasm,profasm,threaded1,threaded2,dyn,profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) ---------------------------------------------------- Nightly run ended at Sun Feb 7 01:09:13 GMT 2010 From ghcbuild at microsoft.com Sat Feb 6 23:43:40 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sat Feb 6 23:43:42 2010 Subject: [nightly] 06-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100207044340.61DAE324120@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Sat Feb 6 18:00:01 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. (GHC Version 6.13.20100206) **** Building stage 1 compiler ... ok. **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Sun Feb 7 05:12:21 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sat Feb 6 23:16:39 GMT 2010 2453 total tests, which gave rise to 13530 test cases, of which 0 caused framework failures 2802 were skipped 10353 expected passes 350 expected failures 0 unexpected passes 25 unexpected failures Unexpected failures: T3831(normal,optc,hpc,optasm,profc,profasm) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Sun Feb 7 05:12:21 GMT 2010 From ghcbuild at microsoft.com Sun Feb 7 19:42:05 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sun Feb 7 19:42:07 2010 Subject: [nightly] 07-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100208004205.CD664324165@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Sun Feb 7 19:00:01 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. (GHC Version 6.13.20100207) **** Building stage 1 compiler ... ok. **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Mon Feb 8 01:10:49 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sun Feb 7 21:18:43 GMT 2010 2453 total tests, which gave rise to 13530 test cases, of which 0 caused framework failures 2787 were skipped 10373 expected passes 348 expected failures 0 unexpected passes 22 unexpected failures Unexpected failures: 3429(ghci) T1969(normal) T3831(normal,optc,hpc,optasm,profc,profasm) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci) ffi009(normal,hpc,optasm,profasm,threaded1,threaded2,dyn,profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) ---------------------------------------------------- Nightly run ended at Mon Feb 8 01:10:50 GMT 2010 From Ben.Lippmeier at anu.edu.au Sun Feb 7 22:55:36 2010 From: Ben.Lippmeier at anu.edu.au (Ben Lippmeier) Date: Sun Feb 7 22:26:50 2010 Subject: sparc/solaris binary distro for 6.12.1 In-Reply-To: <4DE359CC-808F-48D8-8F67-AC44C1A5F25C@anu.edu.au> References: <5189D7F0-0703-4516-9253-FF0DBD0BB040@anu.edu.au> <20100203143739.GA3616@matrix.chaos.earth.li> <4B69CC49.4060501@dfki.de> <814093A1-2B24-4957-A98F-5EB70AC45D79@anu.edu.au> <4B6AA200.9040909@dfki.de> <4DE359CC-808F-48D8-8F67-AC44C1A5F25C@anu.edu.au> Message-ID: <19DFD0AD-44B2-4BC4-92C3-51C95F110C11@anu.edu.au> Ok, I've uploaded a fixed distribution to my homedir on darcs.haskell.org and to http://code.haskell.org/~benl/sparc/ghc-6.12.1-sparc-sun-solaris2-fixed.tar.bz2 I've tested it to confirm that it uses whatever GCC is in the current path. Let me know if you have anymore problems with it. Ian: can you please replace the one on the GHC download page with this new version. Thanks, Ben. On 05/02/2010, at 14:46 , Ben Lippmeier wrote: > > Sorry, this looks like a problem in the build system. An absolute path to GCC was baked into the compiler where it shouldn't have been. GHC also uses GCC as an assembler, which is why ghci works but a regular compile doesn't. > > I've added a ticket for it, and am doing a new build. I'll upload a new distro once the build has finished. > > Ben. > From ghcbuild at microsoft.com Sun Feb 7 23:34:02 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sun Feb 7 23:34:05 2010 Subject: [nightly] 07-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100208043402.6F545324128@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Sun Feb 7 18:00:01 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. (GHC Version 6.13.20100207) **** Building stage 1 compiler ... ok. **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Mon Feb 8 05:02:47 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sun Feb 7 23:02:55 GMT 2010 2453 total tests, which gave rise to 13530 test cases, of which 0 caused framework failures 2802 were skipped 10352 expected passes 350 expected failures 0 unexpected passes 26 unexpected failures Unexpected failures: T3831(normal,optc,hpc,optasm,profc,profasm) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) concprog001(ghci) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Mon Feb 8 05:02:47 GMT 2010 From simonpj at microsoft.com Mon Feb 8 04:48:32 2010 From: simonpj at microsoft.com (Simon Peyton-Jones) Date: Mon Feb 8 04:20:42 2010 Subject: build failure Message-ID: <59543203684B2244980D7E4057D5FBC10AFD28BE@DB3EX14MBXC306.europe.corp.microsoft.com> Ian My build fell over like this, after 'sh validate' Maybe a missing 'clean' target? Simon Lexer.x:11:0: Bad interface file: utils/genprimopcode/./ParserM.hi mismatched interface file versions (wanted "6121", got "6102") From igloo at earth.li Mon Feb 8 06:48:35 2010 From: igloo at earth.li (Ian Lynagh) Date: Mon Feb 8 06:19:50 2010 Subject: build failure In-Reply-To: <59543203684B2244980D7E4057D5FBC10AFD28BE@DB3EX14MBXC306.europe.corp.microsoft.com> References: <59543203684B2244980D7E4057D5FBC10AFD28BE@DB3EX14MBXC306.europe.corp.microsoft.com> Message-ID: <20100208114835.GA17690@matrix.chaos.earth.li> On Mon, Feb 08, 2010 at 09:48:32AM +0000, Simon Peyton-Jones wrote: > > My build fell over like this, after 'sh validate' > > Maybe a missing 'clean' target? > > Lexer.x:11:0: > Bad interface file: utils/genprimopcode/./ParserM.hi > mismatched interface file versions (wanted "6121", got "6102") Is this a very old tree? That file looks like a left-over from the previous build system. It might be simplest to replace the tree with a fresh one, if there's nothing important in it. Thanks Ian From simonpj at microsoft.com Mon Feb 8 07:52:39 2010 From: simonpj at microsoft.com (Simon Peyton-Jones) Date: Mon Feb 8 07:24:18 2010 Subject: build failure In-Reply-To: <20100208114835.GA17690@matrix.chaos.earth.li> References: <59543203684B2244980D7E4057D5FBC10AFD28BE@DB3EX14MBXC306.europe.corp.microsoft.com> <20100208114835.GA17690@matrix.chaos.earth.li> Message-ID: <59543203684B2244980D7E4057D5FBC10AFD2AC4@DB3EX14MBXC306.europe.corp.microsoft.com> probably old. I just deleted the .hi files and ran it again. So probably no-op S | -----Original Message----- | From: cvs-ghc-bounces@haskell.org [mailto:cvs-ghc-bounces@haskell.org] On Behalf Of | Ian Lynagh | Sent: 08 February 2010 11:49 | To: cvs-ghc@haskell.org | Subject: Re: build failure | | On Mon, Feb 08, 2010 at 09:48:32AM +0000, Simon Peyton-Jones wrote: | > | > My build fell over like this, after 'sh validate' | > | > Maybe a missing 'clean' target? | > | > Lexer.x:11:0: | > Bad interface file: utils/genprimopcode/./ParserM.hi | > mismatched interface file versions (wanted "6121", got "6102") | | Is this a very old tree? That file looks like a left-over from the | previous build system. It might be simplest to replace the tree with a | fresh one, if there's nothing important in it. | | | Thanks | Ian | | _______________________________________________ | Cvs-ghc mailing list | Cvs-ghc@haskell.org | http://www.haskell.org/mailman/listinfo/cvs-ghc From igloo at earth.li Mon Feb 8 07:47:44 2010 From: igloo at earth.li (Ian Lynagh) Date: Mon Feb 8 07:47:46 2010 Subject: patch applied (/haskell/ghc): Point at the fixed Sparc Solaris bindist Message-ID: <20100208124744.GA15384@haskell.cs.yale.edu> Mon Feb 8 08:15:45 EST 2010 Ian Lynagh * Point at the fixed Sparc Solaris bindist M ./download_ghc_6_12_1.html -1 +1 From igloo at earth.li Mon Feb 8 08:16:53 2010 From: igloo at earth.li (Ian Lynagh) Date: Mon Feb 8 07:48:08 2010 Subject: sparc/solaris binary distro for 6.12.1 In-Reply-To: <19DFD0AD-44B2-4BC4-92C3-51C95F110C11@anu.edu.au> References: <5189D7F0-0703-4516-9253-FF0DBD0BB040@anu.edu.au> <20100203143739.GA3616@matrix.chaos.earth.li> <4B69CC49.4060501@dfki.de> <814093A1-2B24-4957-A98F-5EB70AC45D79@anu.edu.au> <4B6AA200.9040909@dfki.de> <4DE359CC-808F-48D8-8F67-AC44C1A5F25C@anu.edu.au> <19DFD0AD-44B2-4BC4-92C3-51C95F110C11@anu.edu.au> Message-ID: <20100208131653.GA19999@matrix.chaos.earth.li> On Mon, Feb 08, 2010 at 02:55:36PM +1100, Ben Lippmeier wrote: > > Ok, I've uploaded a fixed distribution to my homedir on darcs.haskell.org and to http://code.haskell.org/~benl/sparc/ghc-6.12.1-sparc-sun-solaris2-fixed.tar.bz2 > > Ian: can you please replace the one on the GHC download page with this new version. Done. Thanks Ian From simonpj at microsoft.com Mon Feb 8 10:58:15 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Mon Feb 8 10:29:29 2010 Subject: patch applied (testsuite): Fix test T3831, and change to Unix coding Message-ID: <20100208155815.GA8121@abbot.galois.com> Mon Feb 8 07:57:15 PST 2010 simonpj@microsoft.com * Fix test T3831, and change to Unix coding M ./tests/ghc-regress/simplCore/should_compile/T3831.hs -107 +107 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100208155715-1287e-a58b11c10f4bc740a742245eed3315f75dabf7c6.gz From simonpj at microsoft.com Mon Feb 8 11:48:58 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Mon Feb 8 11:20:12 2010 Subject: patch applied (ghc): Fix Trac #3850 Message-ID: <20100208164858.GA11066@abbot.galois.com> Mon Feb 8 08:37:51 PST 2010 simonpj@microsoft.com * Fix Trac #3850 This patch simply avoids a needless difference in behaviour from 6.10, and one that happens to affect HList. See Note [Stupid theta]. M ./compiler/rename/RnSource.lhs -3 +13 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100208163751-1287e-efbf0c60c7459099bbb05158a4593a25c137c6d2.gz From ghcbuild at microsoft.com Mon Feb 8 13:56:00 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Mon Feb 8 13:56:02 2010 Subject: [nightly] 08-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100208185600.532C73245A3@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Mon Feb 8 19:10:01 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. **** Building stage 1 compiler ... failed; relevant barfage is below. GHC Version 6.12.1.20100203 **** building source distribution ... ok. No successfully built compilers to test; aborting. ------------------------------------------------------------------------ ------------------------------------------------------------------------ The last 30 lines of /64playpen/simonmar/nightly/STABLE-cam-04-unx/logs/x86_64-unknown-linux-stage1 are ------------------------------------------------------------------------ ------------------------------------------------------------------------ Linking utils/ghc-cabal/dist/build/tmp/ghc-cabal ... touch utils/ghc-cabal/dist/build/tmp/ghc-cabal "inplace/bin/mkdirhier" inplace/bin/ "cp" utils/ghc-cabal/dist/build/tmp/ghc-cabal inplace/bin/ghc-cabal "rm" -f utils/ghc-pkg/Version.hs echo "module Version where" >> utils/ghc-pkg/Version.hs echo "version, targetOS, targetARCH :: String" >> utils/ghc-pkg/Version.hs echo "version = \"6.12.1.20100203\"" >> utils/ghc-pkg/Version.hs echo "targetOS = \"linux\"" >> utils/ghc-pkg/Version.hs echo "targetARCH = \"x86_64\"" >> utils/ghc-pkg/Version.hs "inplace/bin/mkdirhier" bootstrapping "inplace/bin/mkdirhier" utils/ghc-pkg/dist/build "ghc-6.8.3" -H32m -O --make utils/ghc-pkg/Main.hs -o utils/ghc-pkg/dist/build/ghc-pkg \ -Wall -fno-warn-unused-imports \ -DCABAL_VERSION=1,8,0,2 \ -DBOOTSTRAPPING \ -odir bootstrapping \ -hidir bootstrapping \ -iutils/ghc-pkg \ -XCPP -XExistentialQuantification -XDeriveDataTypeable \ -ilibraries/Cabal \ -ilibraries/filepath \ -ilibraries/extensible-exceptions \ -ilibraries/hpc \ -ilibraries/ghc-binary/src \ -ilibraries/bin-package-db libraries/filepath/System/FilePath.hs:1:13: cannot parse LANGUAGE pragma gmake[1]: *** [utils/ghc-pkg/dist/build/ghc-pkg] Error 1 gmake: *** [inplace/bin/ghc-stage1] Error 2 Nightly run ended at Mon Feb 8 19:24:46 GMT 2010 From ghcbuild at microsoft.com Mon Feb 8 19:40:07 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Mon Feb 8 19:40:08 2010 Subject: [nightly] 08-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100209004007.9A7A03243E3@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Mon Feb 8 19:00:01 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100208 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Tue Feb 9 01:08:54 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Mon Feb 8 21:21:35 GMT 2010 2453 total tests, which gave rise to 13530 test cases, of which 0 caused framework failures 2787 were skipped 10379 expected passes 348 expected failures 0 unexpected passes 16 unexpected failures Unexpected failures: 3429(ghci) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci) ffi009(normal,hpc,optasm,profasm,threaded1,threaded2,dyn,profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) ---------------------------------------------------- Nightly run ended at Tue Feb 9 01:08:54 GMT 2010 From ghcbuild at microsoft.com Tue Feb 9 01:03:57 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Tue Feb 9 01:03:59 2010 Subject: [nightly] 08-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100209060357.B147832415A@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Mon Feb 8 18:00:01 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100208 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Tue Feb 9 06:32:45 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Mon Feb 8 23:28:29 GMT 2010 2453 total tests, which gave rise to 13530 test cases, of which 0 caused framework failures 2802 were skipped 10354 expected passes 350 expected failures 0 unexpected passes 24 unexpected failures Unexpected failures: 3429(threaded1,profthreaded) CPUTime001(threaded2) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) ffi002(profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) signals002(threaded2) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Tue Feb 9 06:32:45 GMT 2010 From ghcbuild at microsoft.com Tue Feb 9 01:40:26 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Tue Feb 9 01:40:28 2010 Subject: [nightly] 08-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100209064026.84626324125@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Mon Feb 8 18:10:02 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100203 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (1 failures) **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Tue Feb 9 07:09:14 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Mon Feb 8 23:39:58 GMT 2010 2367 total tests, which gave rise to 13104 test cases, of which 0 caused framework failures 2775 were skipped 9987 expected passes 317 expected failures 0 unexpected passes 25 unexpected failures Unexpected failures: 2592(profc,profasm) 3429(profthreaded) CPUTime001(threaded2) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(threaded2) drvfail006(normal) drvfail008(normal) ffi002(profthreaded) joao-circular(profc) outofmem2(normal) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Tue Feb 9 07:09:14 GMT 2010 From marlowsd at gmail.com Tue Feb 9 07:00:26 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Tue Feb 9 06:31:39 2010 Subject: patch applied (ghc): --lazy overrides --partial too Message-ID: <20100209120025.GA21780@abbot.galois.com> Tue Feb 9 01:39:27 PST 2010 Simon Marlow * --lazy overrides --partial too At least until we switch to --lazy being the default M ./darcs-all -1 +1 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100209093927-12142-049f2a66c79c0820a55780e8b707eab63185fc59.gz From marlowsd at gmail.com Tue Feb 9 07:00:27 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Tue Feb 9 06:31:41 2010 Subject: patch applied (ghc): generate slightly better code for ccall argument-pushing on x86 Message-ID: <20100209120027.GA21800@abbot.galois.com> Tue Feb 9 02:11:14 PST 2010 Simon Marlow * generate slightly better code for ccall argument-pushing on x86 M ./compiler/nativeGen/X86/CodeGen.hs -16 +14 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100209101114-12142-4536e0fa467a820f9bb1855c311bd5f3d100f18a.gz From ghcbuild at microsoft.com Tue Feb 9 19:44:05 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Tue Feb 9 19:44:07 2010 Subject: [nightly] 09-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100210004405.034CB324471@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Tue Feb 9 19:10:01 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100203 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Wed Feb 10 01:12:55 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Tue Feb 9 21:21:10 GMT 2010 2367 total tests, which gave rise to 13104 test cases, of which 0 caused framework failures 2760 were skipped 10020 expected passes 315 expected failures 0 unexpected passes 9 unexpected failures Unexpected failures: 2592(profc,profasm) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) outofmem2(normal) ---------------------------------------------------- Nightly run ended at Wed Feb 10 01:12:55 GMT 2010 From ghcbuild at microsoft.com Tue Feb 9 19:54:11 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Tue Feb 9 19:54:11 2010 Subject: [nightly] 09-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100210005411.5E58C324471@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Tue Feb 9 19:00:01 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100209 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Wed Feb 10 01:23:01 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Tue Feb 9 21:29:28 GMT 2010 2453 total tests, which gave rise to 13530 test cases, of which 0 caused framework failures 2787 were skipped 10380 expected passes 348 expected failures 0 unexpected passes 15 unexpected failures Unexpected failures: T1969(normal) annrun01(dyn) barton-mangler-bug(profc) concprog002(threaded2) ffi009(normal,hpc,optasm,profasm,threaded1,threaded2,dyn,profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) ---------------------------------------------------- Nightly run ended at Wed Feb 10 01:23:01 GMT 2010 From ghcbuild at microsoft.com Wed Feb 10 01:07:28 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Wed Feb 10 01:07:30 2010 Subject: [nightly] 09-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100210060728.A33C93242A3@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Tue Feb 9 18:00:02 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100209 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (3 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Wed Feb 10 06:36:13 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Tue Feb 9 23:30:59 GMT 2010 2453 total tests, which gave rise to 13530 test cases, of which 0 caused framework failures 2802 were skipped 10352 expected passes 350 expected failures 0 unexpected passes 26 unexpected failures Unexpected failures: 3429(threaded1) CPUTime001(threaded2) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) concprog001(ghci) ffi002(profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) signals002(threaded1,profthreaded) signals004(threaded1) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Wed Feb 10 06:36:13 GMT 2010 From ghcbuild at microsoft.com Wed Feb 10 01:55:37 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Wed Feb 10 01:55:38 2010 Subject: [nightly] 09-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100210065537.E159E324134@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Tue Feb 9 18:10:02 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100203 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (1 failures) **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. **** publishing logs ... ssh: connect to host haskell.org port 22: Bad file descriptor lost connection failed. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Wed Feb 10 07:24:14 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Tue Feb 9 23:43:05 GMT 2010 2367 total tests, which gave rise to 13104 test cases, of which 0 caused framework failures 2775 were skipped 9986 expected passes 317 expected failures 0 unexpected passes 26 unexpected failures Unexpected failures: 2592(profc,profasm) 3429(profthreaded) CPUTime001(threaded2) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) ffi002(threaded2,profthreaded) joao-circular(profc) outofmem2(normal) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Wed Feb 10 07:24:14 GMT 2010 From simonpj at microsoft.com Wed Feb 10 04:50:38 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Wed Feb 10 04:21:51 2010 Subject: patch applied (ghc): Several TH/quasiquote changes Message-ID: <20100210095038.GA6467@abbot.galois.com> Wed Feb 10 01:39:10 PST 2010 simonpj@microsoft.com * Several TH/quasiquote changes a) Added quasi-quote forms for declarations types e.g. f :: [$qq| ... |] b) Allow Template Haskell pattern quotes (but not splices) e.g. f x = [p| Int -> $x |] c) Improve pretty-printing for HsPat to remove superfluous parens. (This isn't TH related really, but it affects some of the same code.) A consequence of (a) is that when gathering and grouping declarations in RnSource.findSplice, we must expand quasiquotes as we do so. Otherwise it's all fairly straightforward. I did a little bit of refactoring in TcSplice. User-manual changes still to come. M ./compiler/deSugar/DsMeta.hs -32 +45 M ./compiler/hsSyn/HsDecls.lhs +2 M ./compiler/hsSyn/HsExpr.lhs -17 +27 M ./compiler/hsSyn/HsPat.lhs -38 +33 M ./compiler/hsSyn/HsPat.lhs-boot -4 +1 M ./compiler/hsSyn/HsTypes.lhs +25 M ./compiler/hsSyn/HsUtils.lhs -1 +1 M ./compiler/parser/Parser.y.pp -7 +11 M ./compiler/parser/RdrHsSyn.lhs -87 +10 M ./compiler/rename/RnExpr.lhs -12 +21 M ./compiler/rename/RnHsSyn.lhs +1 M ./compiler/rename/RnPat.lhs -27 +8 M ./compiler/rename/RnSource.lhs -1 +83 M ./compiler/rename/RnTypes.lhs +9 M ./compiler/typecheck/TcHsType.lhs -2 +4 M ./compiler/typecheck/TcRnDriver.lhs -3 +2 M ./compiler/typecheck/TcSplice.lhs -31 +57 M ./compiler/typecheck/TcSplice.lhs-boot -3 +5 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100210093910-1287e-cca518fcec2e326803782a967b7fe09ab3cb7efa.gz From simonpj at microsoft.com Wed Feb 10 04:50:40 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Wed Feb 10 04:21:53 2010 Subject: patch applied (ghc): Simplify syntax for quasi-quotation Message-ID: <20100210095040.GA6488@abbot.galois.com> Wed Feb 10 01:44:19 PST 2010 simonpj@microsoft.com * Simplify syntax for quasi-quotation After some discussion we decided to make a quasi-quote look like [pads| ...blah... |] rather than [$pads| ...blah... |] as before. The new syntax is quieter, although it does not signal quite as clearly that there is a splice going on. M ./compiler/parser/Lexer.x -3 +4 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100210094419-1287e-b11df0ad15832427dacbd8c8ae94c180b2b69267.gz From simonpj at microsoft.com Wed Feb 10 04:50:41 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Wed Feb 10 04:21:55 2010 Subject: patch applied (ghc): Comments only Message-ID: <20100210095041.GA6510@abbot.galois.com> Wed Feb 10 01:45:37 PST 2010 simonpj@microsoft.com * Comments only M ./compiler/specialise/SpecConstr.lhs -1 +1 M ./compiler/typecheck/TcRnTypes.lhs +4 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100210094537-1287e-2c2de1581642f8a3db9a73ab158783e241169827.gz From simonpj at microsoft.com Wed Feb 10 04:50:43 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Wed Feb 10 04:21:58 2010 Subject: patch applied (ghc): Stop fruitless ANF-ing Message-ID: <20100210095043.GA6530@abbot.galois.com> Wed Feb 10 01:47:33 PST 2010 simonpj@microsoft.com * Stop fruitless ANF-ing The simplifier is taking more iterations than it should, because we were fruitlessly ANF-ing a top-level declaration of form x = Ptr "foo"# to get x = let v = "foo"# in Ptr v and then inlining v again. This patch makes Simplify.makeTrivial top-level aware, so that it doesn't ANF if it's going to be undone. M ./compiler/simplCore/Simplify.lhs -21 +55 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100210094733-1287e-a662ecd131a28f2d9b87806468583a962e2d7bd1.gz From simonpj at microsoft.com Wed Feb 10 05:58:27 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Wed Feb 10 05:29:36 2010 Subject: patch applied (testsuite): Add quasi-quote test for declaration and type quotes Message-ID: <20100210105827.GA8848@abbot.galois.com> Wed Feb 10 02:58:22 PST 2010 simonpj@microsoft.com * Add quasi-quote test for declaration and type quotes A ./tests/ghc-regress/quasiquotation/qq007/ A ./tests/ghc-regress/quasiquotation/qq007/Makefile A ./tests/ghc-regress/quasiquotation/qq007/QQ.hs A ./tests/ghc-regress/quasiquotation/qq007/Test.hs A ./tests/ghc-regress/quasiquotation/qq007/test.T View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100210105822-1287e-7d1a15ffeff530d7962e6195433d769b22755765.gz From simonpj at microsoft.com Wed Feb 10 05:58:28 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Wed Feb 10 05:29:38 2010 Subject: patch applied (testsuite): New syntax for quasi-quotes, and record field names for QuasiQuoter type Message-ID: <20100210105828.GA8868@abbot.galois.com> Wed Feb 10 02:59:11 PST 2010 simonpj@microsoft.com * New syntax for quasi-quotes, and record field names for QuasiQuoter type M ./tests/ghc-regress/quasiquotation/qq001/qq001.hs -1 +1 M ./tests/ghc-regress/quasiquotation/qq002/qq002.hs -1 +1 M ./tests/ghc-regress/quasiquotation/qq003/qq003.hs -1 +1 M ./tests/ghc-regress/quasiquotation/qq004/qq004.hs -1 +1 M ./tests/ghc-regress/quasiquotation/qq005/Expr.hs -1 +1 M ./tests/ghc-regress/quasiquotation/qq005/Main.hs -4 +4 M ./tests/ghc-regress/quasiquotation/qq006/Expr.hs -1 +1 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100210105911-1287e-8ac2e4a055e8ddfc1a86f96e30d8883ae49c26e1.gz From simonpj at microsoft.com Wed Feb 10 05:58:29 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Wed Feb 10 05:29:41 2010 Subject: patch applied (testsuite): Fewer parens when printing HsPat Message-ID: <20100210105829.GA8888@abbot.galois.com> Wed Feb 10 03:00:24 PST 2010 simonpj@microsoft.com * Fewer parens when printing HsPat M ./tests/ghc-regress/deSugar/should_compile/ds020.stderr-ghc -18 +18 M ./tests/ghc-regress/gadt/gadt21.stderr -10 +10 M ./tests/ghc-regress/typecheck/should_fail/tcfail124.stderr -18 +18 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100210110024-1287e-797bea9549b7a30dd164e01a487b73b0fdc06b50.gz From simonpj at microsoft.com Wed Feb 10 09:35:32 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Wed Feb 10 09:06:42 2010 Subject: patch applied (ghc): Remove redundant import Message-ID: <20100210143532.GA17029@abbot.galois.com> Mon Feb 8 08:57:56 PST 2010 simonpj@microsoft.com * Remove redundant import M ./compiler/utils/Util.lhs -1 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100208165756-1287e-1922c13f651bf3dbffd03d28a7c44befef2812ba.gz From simonpj at microsoft.com Wed Feb 10 09:35:34 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Wed Feb 10 09:06:44 2010 Subject: patch applied (ghc): Documentation for changes to Template Haskell and quasi-quotation Message-ID: <20100210143534.GA17049@abbot.galois.com> Wed Feb 10 03:42:13 PST 2010 simonpj@microsoft.com * Documentation for changes to Template Haskell and quasi-quotation M ./docs/users_guide/glasgow_exts.xml -30 +74 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100210114213-1287e-33e7c335d6e9fcae01cf528a3f5f401a15e46b7a.gz From simonpj at microsoft.com Wed Feb 10 09:52:59 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Wed Feb 10 09:24:07 2010 Subject: patch applied (ghc): Keep track of explicit kinding in HsTyVarBndr; plus fix Trac #3845 Message-ID: <20100210145259.GA17695@abbot.galois.com> Wed Feb 10 06:51:55 PST 2010 simonpj@microsoft.com * Keep track of explicit kinding in HsTyVarBndr; plus fix Trac #3845 To print HsTypes correctly we should remember whether the Kind on a HsTyVarBndr came from type inference, or was put there by the user. See Note [Printing KindedTyVars] in HsTypes. So instead of changing a UserTyVar to a KindedTyVar during kind checking, we simply add a PostTcKind to the UserTyVar. The change was provoked by Trac #3830, although other changes mean that #3830 gets a diferent and better error message now. So this patch is simply doing the Right Thing for the future. This patch also fixes Trac #3845, which was caused by a *type splice* not remembering the free *term variables* mentioned in it. Result was that we build a 'let' when it should have been 'letrec'. Hence a new FreeVars field in HsSpliceTy. While I was at it, I got rid of HsSpliceTyOut and use a PostTcKind on HsSpliceTy instead, just like on the UserTyVar. M ./compiler/deSugar/DsMeta.hs -6 +7 M ./compiler/hsSyn/Convert.lhs -1 +1 M ./compiler/hsSyn/HsDecls.lhs -1 +1 M ./compiler/hsSyn/HsTypes.lhs -31 +53 M ./compiler/hsSyn/HsUtils.lhs -1 +5 M ./compiler/parser/Parser.y.pp -8 +6 M ./compiler/parser/RdrHsSyn.lhs -4 +2 M ./compiler/rename/RnHsSyn.lhs -3 +2 M ./compiler/rename/RnTypes.lhs -5 +4 M ./compiler/typecheck/TcClassDcl.lhs -1 +2 M ./compiler/typecheck/TcEnv.lhs -6 +3 M ./compiler/typecheck/TcGenDeriv.lhs -1 +2 M ./compiler/typecheck/TcHsType.lhs -45 +14 M ./compiler/typecheck/TcSplice.lhs -7 +5 M ./compiler/typecheck/TcSplice.lhs-boot -1 +2 M ./compiler/typecheck/TcTyClsDecls.lhs -16 +16 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100210145155-1287e-31783b96a46852a3100f1e8a98db053f9e8c2511.gz From simonpj at microsoft.com Wed Feb 10 09:53:01 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Wed Feb 10 09:24:10 2010 Subject: patch applied (ghc): Improve error dump in TcEnv Message-ID: <20100210145301.GA17716@abbot.galois.com> Wed Feb 10 06:52:10 PST 2010 simonpj@microsoft.com * Improve error dump in TcEnv M ./compiler/typecheck/TcEnv.lhs -7 +9 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100210145210-1287e-b1553c07607cc894b30d48021bd018691ec2f1f6.gz From simonpj at microsoft.com Wed Feb 10 10:38:21 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Wed Feb 10 10:09:35 2010 Subject: patch applied (testsuite): Test Trac #3845 Message-ID: <20100210153821.GA19516@abbot.galois.com> Wed Feb 10 07:37:41 PST 2010 simonpj@microsoft.com * Test Trac #3845 A ./tests/ghc-regress/th/T3845.hs M ./tests/ghc-regress/th/all.T +1 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100210153741-1287e-eddd498a571e3514d155c69a19cfc0b09a15b7ed.gz From simonpj at microsoft.com Wed Feb 10 10:42:25 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Wed Feb 10 10:13:35 2010 Subject: patch applied (ghc-new-tc/ghc): Comments Message-ID: <20100210154225.GA19697@abbot.galois.com> Wed Feb 10 07:44:03 PST 2010 simonpj@microsoft.com * Comments M ./compiler/typecheck/TcSimplify.lhs +17 View patch online: http://darcs.haskell.org/ghc-new-tc/ghc/_darcs/patches/20100210154403-1287e-75c8de8efb4bd6c5c43ab52b1f641df0f48bd9e0.gz From simonpj at microsoft.com Wed Feb 10 12:53:23 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Wed Feb 10 12:24:34 2010 Subject: patch applied (ghc-new-tc/ghc): Small changes with Dimitrios Message-ID: <20100210175323.GA24575@abbot.galois.com> Wed Feb 10 09:52:57 PST 2010 simonpj@microsoft.com * Small changes with Dimitrios M ./compiler/basicTypes/Var.lhs -2 +2 M ./compiler/hsSyn/HsBinds.lhs -2 +4 M ./compiler/typecheck/TcSimplify.lhs -7 +20 View patch online: http://darcs.haskell.org/ghc-new-tc/ghc/_darcs/patches/20100210175257-1287e-9af344165071d828abbdedd123db284c501208e2.gz From ghcbuild at microsoft.com Wed Feb 10 20:02:43 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Wed Feb 10 20:02:44 2010 Subject: [nightly] 10-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100211010243.33E3A3242CC@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Wed Feb 10 19:00:01 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100210 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Thu Feb 11 01:31:34 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Wed Feb 10 21:26:50 GMT 2010 2455 total tests, which gave rise to 13542 test cases, of which 0 caused framework failures 2792 were skipped 10385 expected passes 348 expected failures 0 unexpected passes 17 unexpected failures Unexpected failures: T1969(normal) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci) ffi009(normal,hpc,optasm,profasm,threaded1,threaded2,dyn,profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) qq007(profc,profasm) ---------------------------------------------------- Nightly run ended at Thu Feb 11 01:31:34 GMT 2010 From ghcbuild at microsoft.com Wed Feb 10 20:11:32 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Wed Feb 10 20:11:35 2010 Subject: [nightly] 10-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100211011132.8CD5D3242D5@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Wed Feb 10 19:10:01 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100203 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Thu Feb 11 01:40:24 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Wed Feb 10 21:21:25 GMT 2010 2367 total tests, which gave rise to 13104 test cases, of which 0 caused framework failures 2760 were skipped 10020 expected passes 315 expected failures 0 unexpected passes 9 unexpected failures Unexpected failures: 2592(profc,profasm) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) outofmem2(normal) ---------------------------------------------------- Nightly run ended at Thu Feb 11 01:40:24 GMT 2010 From ghcbuild at microsoft.com Thu Feb 11 01:04:23 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Thu Feb 11 01:04:24 2010 Subject: [nightly] 10-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100211060423.4767B3241CE@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Wed Feb 10 18:00:01 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100210 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Thu Feb 11 06:33:17 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Wed Feb 10 23:23:07 GMT 2010 2455 total tests, which gave rise to 13542 test cases, of which 0 caused framework failures 2807 were skipped 10359 expected passes 350 expected failures 0 unexpected passes 26 unexpected failures Unexpected failures: 3429(threaded1,profthreaded) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) qq007(profc,profasm) signals002(threaded1,threaded2) signals004(threaded1) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Thu Feb 11 06:33:17 GMT 2010 From ghcbuild at microsoft.com Thu Feb 11 01:51:56 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Thu Feb 11 01:51:57 2010 Subject: [nightly] 10-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100211065156.7A802324326@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Wed Feb 10 18:10:01 GMT 2010. **** checking out new source tree ... --partial: hashed or darcs-2 repository detected, using --lazy instead warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100203 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (1 failures) **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Thu Feb 11 07:20:50 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Wed Feb 10 23:40:19 GMT 2010 2367 total tests, which gave rise to 13104 test cases, of which 0 caused framework failures 2775 were skipped 9987 expected passes 317 expected failures 0 unexpected passes 25 unexpected failures Unexpected failures: 2592(profc,profasm) 3429(threaded1,profthreaded) CPUTime001(threaded2) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) joao-circular(profc) outofmem2(normal) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Thu Feb 11 07:20:50 GMT 2010 From marlowsd at gmail.com Thu Feb 11 08:19:17 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Thu Feb 11 07:50:23 2010 Subject: patch applied (ghc): don't constant fold division that would result in negative zero (#3676) Message-ID: <20100211131917.GA3031@abbot.galois.com> Thu Feb 11 05:15:43 PST 2010 Simon Marlow * don't constant fold division that would result in negative zero (#3676) M ./compiler/prelude/PrelRules.lhs -2 +10 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100211131543-12142-81c0bae72afd112800fc0a4aea8ede80b6dc77ab.gz From marlowsd at gmail.com Thu Feb 11 08:36:05 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Thu Feb 11 08:07:10 2010 Subject: patch applied (testsuite): add test for #3676 Message-ID: <20100211133605.GA4004@abbot.galois.com> Thu Feb 11 05:16:39 PST 2010 Simon Marlow * add test for #3676 A ./tests/ghc-regress/numeric/should_run/3676.hs A ./tests/ghc-regress/numeric/should_run/3676.stdout M ./tests/ghc-regress/numeric/should_run/all.T +1 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100211131639-12142-0c40f27c1d4d2b062cb4ebf0571d975738bf3d47.gz From ghcbuild at microsoft.com Thu Feb 11 19:53:42 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Thu Feb 11 19:53:44 2010 Subject: [nightly] 11-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100212005342.66E06324933@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Thu Feb 11 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100211 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Fri Feb 12 01:22:37 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Thu Feb 11 21:27:33 GMT 2010 2456 total tests, which gave rise to 13553 test cases, of which 0 caused framework failures 2792 were skipped 10396 expected passes 348 expected failures 0 unexpected passes 17 unexpected failures Unexpected failures: T1969(normal) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci) ffi009(normal,hpc,optasm,profasm,threaded1,threaded2,dyn,profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) qq007(profc,profasm) ---------------------------------------------------- Nightly run ended at Fri Feb 12 01:22:37 GMT 2010 From ghcbuild at microsoft.com Thu Feb 11 19:55:09 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Thu Feb 11 19:55:10 2010 Subject: [nightly] 11-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100212005509.BAE7B324933@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Thu Feb 11 19:10:02 GMT 2010. **** checking out new source tree ... warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100203 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Fri Feb 12 01:24:03 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Thu Feb 11 21:21:21 GMT 2010 2367 total tests, which gave rise to 13104 test cases, of which 0 caused framework failures 2760 were skipped 10020 expected passes 315 expected failures 0 unexpected passes 9 unexpected failures Unexpected failures: 2592(profc,profasm) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) outofmem2(normal) ---------------------------------------------------- Nightly run ended at Fri Feb 12 01:24:03 GMT 2010 From ghcbuild at microsoft.com Fri Feb 12 01:05:47 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Fri Feb 12 01:05:49 2010 Subject: [nightly] 11-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100212060547.BF419324035@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Thu Feb 11 18:00:02 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100211 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Fri Feb 12 06:34:44 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Thu Feb 11 23:23:55 GMT 2010 2456 total tests, which gave rise to 13553 test cases, of which 0 caused framework failures 2807 were skipped 10370 expected passes 350 expected failures 0 unexpected passes 26 unexpected failures Unexpected failures: annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) ffi002(threaded1) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) qq007(profc,profasm) signals002(threaded1,threaded2,profthreaded) signals004(threaded1) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Fri Feb 12 06:34:44 GMT 2010 From ghcbuild at microsoft.com Fri Feb 12 01:41:06 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Fri Feb 12 01:41:08 2010 Subject: [nightly] 11-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100212064106.A8011324930@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Thu Feb 11 18:10:02 GMT 2010. **** checking out new source tree ... warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100203 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (1 failures) **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Fri Feb 12 07:10:03 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Thu Feb 11 23:41:01 GMT 2010 2367 total tests, which gave rise to 13104 test cases, of which 0 caused framework failures 2775 were skipped 9986 expected passes 317 expected failures 0 unexpected passes 26 unexpected failures Unexpected failures: 2592(profc,profasm) 3429(threaded1) CPUTime001(threaded2) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) conc019(profthreaded) concprog001(ghci) drvfail006(normal) drvfail008(normal) ffi002(profthreaded) joao-circular(profc) outofmem2(normal) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Fri Feb 12 07:10:03 GMT 2010 From simonpj at microsoft.com Fri Feb 12 10:05:43 2010 From: simonpj at microsoft.com (Simon Peyton Jones) Date: Fri Feb 12 09:36:44 2010 Subject: patch applied (ghc-new-tc/ghc): More changes Message-ID: <20100212150543.GA27895@abbot.galois.com> Fri Feb 12 07:04:55 PST 2010 simonpj@microsoft.com * More changes M ./compiler/typecheck/Constraints.lhs -1 +2 M ./compiler/typecheck/TcInteract.lhs -28 +51 M ./compiler/typecheck/TcSimplify.lhs -3 +3 M ./compiler/utils/Bag.lhs -1 +7 View patch online: http://darcs.haskell.org/ghc-new-tc/ghc/_darcs/patches/20100212150455-1287e-223b3e5aba55a23d724a9e9aa64a49db70ea93e5.gz From igloo at earth.li Fri Feb 12 12:15:16 2010 From: igloo at earth.li (Ian Lynagh) Date: Fri Feb 12 11:46:19 2010 Subject: new buildbots - was: Exploding SpecConstr In-Reply-To: References: <947938FD-89E3-4B6E-9547-850824A42812@anu.edu.au> <4B630AEA.5020506@gmail.com> <404396ef1001290858i555226ccj809a3e378ff1569a@mail.gmail.com> <4B66B8CE.507@gmail.com> Message-ID: <20100212171516.GA21868@matrix.chaos.earth.li> I've uploaded the current code to community: http://code.haskell.org/builder/ There are some brief docs in the doc/ directory. It is very much barebones at the moment, and doesn't yet have anything like the DSL that Simon was talking about. But it is able to do a GHC build, and produces webpages like these: http://projects.haskell.org/builder/example/builders/ghcBuilder/2.html http://projects.haskell.org/builder/example/builders/ghcBuilder/2/9.html My TODO list contains: * Protocol version * Some sort of locking so the same client can't connect twice * More exception handling * "validate this bundle please" feature * e-mail reporting * IRC reporting * twitter * RSS Thanks Ian From igloo at earth.li Fri Feb 12 16:39:13 2010 From: igloo at earth.li (Ian Lynagh) Date: Fri Feb 12 16:10:14 2010 Subject: 1969 In-Reply-To: <59543203684B2244980D7E4057D5FBC10AF767D3@DB3EX14MBXC310.europe.corp.microsoft.com> References: <59543203684B2244980D7E4057D5FBC10AF767D3@DB3EX14MBXC310.europe.corp.microsoft.com> Message-ID: <20100212213913.GA26436@matrix.chaos.earth.li> On Fri, Dec 18, 2009 at 09:22:07AM +0000, Simon Peyton-Jones wrote: > > I think #1969 is fixed, at least as well as we can for now. Do we need to keep it open? I've closed it. Thanks Ian From igloo at earth.li Fri Feb 12 17:27:52 2010 From: igloo at earth.li (Ian Lynagh) Date: Fri Feb 12 16:58:54 2010 Subject: patch applied (ghc-6.12/ghc): simplification/optimisation: update tso->bound->tso when scavenging the TSO Message-ID: <20100212222752.GA11392@abbot.galois.com> Tue Dec 8 00:57:39 PST 2009 Simon Marlow * simplification/optimisation: update tso->bound->tso when scavenging the TSO M ./rts/sm/GC.c -37 M ./rts/sm/Scav.c +5 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20091208085739-12142-d6396a69c99ed766a5c5c1f42bafc43eb9a7d256.gz From igloo at earth.li Fri Feb 12 17:27:54 2010 From: igloo at earth.li (Ian Lynagh) Date: Fri Feb 12 16:58:57 2010 Subject: patch applied (ghc-6.12/ghc): --lazy overrides --partial too Message-ID: <20100212222754.GA11413@abbot.galois.com> Tue Feb 9 01:39:27 PST 2010 Simon Marlow * --lazy overrides --partial too At least until we switch to --lazy being the default M ./darcs-all -1 +1 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100209093927-12142-049f2a66c79c0820a55780e8b707eab63185fc59.gz From igloo at earth.li Fri Feb 12 17:29:12 2010 From: igloo at earth.li (Ian Lynagh) Date: Fri Feb 12 17:00:13 2010 Subject: patch applied (ghc): simplification/optimisation: update tso->bound->tso when scavenging the TSO In-Reply-To: <4B1E2992.1040200@gmail.com> References: <20091208102334.GA12406@haskell.galois.com> <4B1E2992.1040200@gmail.com> Message-ID: <20100212222912.GA338@matrix.chaos.earth.li> On Tue, Dec 08, 2009 at 10:25:22AM +0000, Simon Marlow wrote: > On 08/12/2009 10:23, Simon Marlow wrote: >> Tue Dec 8 00:57:39 PST 2009 Simon Marlow >> * simplification/optimisation: update tso->bound->tso when scavenging the TSO >> Ignore-this: 401e2c67e42de9671191ba9d18c3fcf7 >> >> M ./rts/sm/GC.c -37 >> M ./rts/sm/Scav.c +5 > > I think this may fix a bug. Ian, let's merge post-6.12.1. Merged. Thanks Ian From marlowsd at gmail.com Fri Feb 12 17:52:54 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Fri Feb 12 17:23:59 2010 Subject: new buildbots - was: Exploding SpecConstr In-Reply-To: <20100212171516.GA21868@matrix.chaos.earth.li> References: <947938FD-89E3-4B6E-9547-850824A42812@anu.edu.au> <4B630AEA.5020506@gmail.com> <404396ef1001290858i555226ccj809a3e378ff1569a@mail.gmail.com> <4B66B8CE.507@gmail.com> <20100212171516.GA21868@matrix.chaos.earth.li> Message-ID: <4B75DBC6.4020407@gmail.com> On 12/02/10 17:15, Ian Lynagh wrote: > > I've uploaded the current code to community: > > http://code.haskell.org/builder/ > > There are some brief docs in the doc/ directory. > > It is very much barebones at the moment, and doesn't yet have anything > like the DSL that Simon was talking about. boo :-( Oh well, better to concentrate on the other functionality I suppose. Looks good so far! Cheers, Simon From ghcbuild at microsoft.com Fri Feb 12 20:06:31 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Fri Feb 12 20:06:32 2010 Subject: [nightly] 12-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100213010631.337C2324A37@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Fri Feb 12 19:10:01 GMT 2010. **** checking out new source tree ... warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100203 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Sat Feb 13 01:35:29 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Fri Feb 12 21:20:14 GMT 2010 2367 total tests, which gave rise to 13104 test cases, of which 0 caused framework failures 2760 were skipped 10020 expected passes 315 expected failures 0 unexpected passes 9 unexpected failures Unexpected failures: 2592(profc,profasm) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) outofmem2(normal) ---------------------------------------------------- Nightly run ended at Sat Feb 13 01:35:29 GMT 2010 From ghcbuild at microsoft.com Fri Feb 12 20:16:03 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Fri Feb 12 20:16:05 2010 Subject: [nightly] 12-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100213011603.07AB6324192@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Fri Feb 12 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100211 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Sat Feb 13 01:45:02 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Fri Feb 12 21:26:41 GMT 2010 2456 total tests, which gave rise to 13553 test cases, of which 0 caused framework failures 2792 were skipped 10396 expected passes 348 expected failures 0 unexpected passes 17 unexpected failures Unexpected failures: T1969(normal) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci) ffi009(normal,hpc,optasm,profasm,threaded1,threaded2,dyn,profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) qq007(profc,profasm) ---------------------------------------------------- Nightly run ended at Sat Feb 13 01:45:02 GMT 2010 From ghcbuild at microsoft.com Sat Feb 13 01:23:14 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sat Feb 13 01:23:16 2010 Subject: [nightly] 12-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100213062314.F240B32420B@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Fri Feb 12 18:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100211 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (3 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Sat Feb 13 06:52:15 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Fri Feb 12 23:23:36 GMT 2010 2456 total tests, which gave rise to 13553 test cases, of which 0 caused framework failures 2807 were skipped 10368 expected passes 350 expected failures 0 unexpected passes 28 unexpected failures Unexpected failures: 3429(profthreaded) CPUTime001(threaded2) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) ffi002(threaded1,profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) qq007(profc,profasm) signals002(threaded1,threaded2,profthreaded) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Sat Feb 13 06:52:15 GMT 2010 From ghcbuild at microsoft.com Sat Feb 13 01:49:51 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sat Feb 13 01:49:52 2010 Subject: [nightly] 12-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100213064951.4474632413F@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Fri Feb 12 18:10:01 GMT 2010. **** checking out new source tree ... warning: adding --partial, to override use --complete warning: . already present; omitting --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead --partial: hashed or darcs-2 repository detected, using --lazy instead ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100203 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (1 failures) **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Sat Feb 13 07:18:50 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Fri Feb 12 23:40:34 GMT 2010 2367 total tests, which gave rise to 13104 test cases, of which 0 caused framework failures 2775 were skipped 9989 expected passes 317 expected failures 0 unexpected passes 23 unexpected failures Unexpected failures: 2592(profc,profasm) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) ffi002(threaded2) joao-circular(profc) outofmem2(normal) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Sat Feb 13 07:18:50 GMT 2010 From ghcbuild at microsoft.com Sat Feb 13 13:55:52 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sat Feb 13 13:55:54 2010 Subject: [nightly] 13-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100213185552.3ACF9324170@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Sat Feb 13 19:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... failed; relevant barfage is below. GHC Version 6.12.1.20100209 **** building source distribution ... ok. No successfully built compilers to test; aborting. ------------------------------------------------------------------------ ------------------------------------------------------------------------ The last 30 lines of /64playpen/simonmar/nightly/STABLE-cam-04-unx/logs/x86_64-unknown-linux-stage1 are ------------------------------------------------------------------------ ------------------------------------------------------------------------ Linking utils/ghc-cabal/dist/build/tmp/ghc-cabal ... touch utils/ghc-cabal/dist/build/tmp/ghc-cabal "inplace/bin/mkdirhier" inplace/bin/ "cp" utils/ghc-cabal/dist/build/tmp/ghc-cabal inplace/bin/ghc-cabal "rm" -f utils/ghc-pkg/Version.hs echo "module Version where" >> utils/ghc-pkg/Version.hs echo "version, targetOS, targetARCH :: String" >> utils/ghc-pkg/Version.hs echo "version = \"6.12.1.20100209\"" >> utils/ghc-pkg/Version.hs echo "targetOS = \"linux\"" >> utils/ghc-pkg/Version.hs echo "targetARCH = \"x86_64\"" >> utils/ghc-pkg/Version.hs "inplace/bin/mkdirhier" bootstrapping "inplace/bin/mkdirhier" utils/ghc-pkg/dist/build "ghc-6.8.3" -H32m -O --make utils/ghc-pkg/Main.hs -o utils/ghc-pkg/dist/build/ghc-pkg \ -Wall -fno-warn-unused-imports \ -DCABAL_VERSION=1,8,0,2 \ -DBOOTSTRAPPING \ -odir bootstrapping \ -hidir bootstrapping \ -iutils/ghc-pkg \ -XCPP -XExistentialQuantification -XDeriveDataTypeable \ -ilibraries/Cabal \ -ilibraries/filepath \ -ilibraries/extensible-exceptions \ -ilibraries/hpc \ -ilibraries/ghc-binary/src \ -ilibraries/bin-package-db libraries/filepath/System/FilePath.hs:1:13: cannot parse LANGUAGE pragma gmake[1]: *** [utils/ghc-pkg/dist/build/ghc-pkg] Error 1 gmake: *** [inplace/bin/ghc-stage1] Error 2 Nightly run ended at Sat Feb 13 19:24:54 GMT 2010 From ghcbuild at microsoft.com Sat Feb 13 20:18:44 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sat Feb 13 20:18:45 2010 Subject: [nightly] 13-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100214011844.3051232415C@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Sat Feb 13 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100211 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Sun Feb 14 01:47:46 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sat Feb 13 21:18:59 GMT 2010 2456 total tests, which gave rise to 13553 test cases, of which 0 caused framework failures 2792 were skipped 10395 expected passes 348 expected failures 0 unexpected passes 18 unexpected failures Unexpected failures: 3429(ghci) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci) ffi009(normal,hpc,optasm,profasm,threaded1,threaded2,dyn,profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) qq007(profc,profasm) ---------------------------------------------------- Nightly run ended at Sun Feb 14 01:47:46 GMT 2010 From ghcbuild at microsoft.com Sun Feb 14 01:35:51 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sun Feb 14 01:35:52 2010 Subject: [nightly] 13-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100214063551.089AD324169@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Sat Feb 13 18:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100211 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Sun Feb 14 07:04:54 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sat Feb 13 23:27:16 GMT 2010 2456 total tests, which gave rise to 13553 test cases, of which 0 caused framework failures 2807 were skipped 10367 expected passes 350 expected failures 0 unexpected passes 29 unexpected failures Unexpected failures: 3429(threaded1,profthreaded) CPUTime001(threaded2) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) conc019(threaded1) ffi002(profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) qq007(profc,profasm) signals002(threaded2,profthreaded) signals004(profthreaded) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Sun Feb 14 07:04:54 GMT 2010 From ghcbuild at microsoft.com Sun Feb 14 01:44:20 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sun Feb 14 01:44:22 2010 Subject: [nightly] 13-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100214064420.417CC3241E7@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Sat Feb 13 18:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100209 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (1 failures) **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Sun Feb 14 07:13:23 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sat Feb 13 23:43:23 GMT 2010 2367 total tests, which gave rise to 13104 test cases, of which 0 caused framework failures 2775 were skipped 9984 expected passes 317 expected failures 0 unexpected passes 28 unexpected failures Unexpected failures: 2592(profc,profasm) 3429(threaded1) CPUTime001(threaded2) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) ffi002(threaded1,profthreaded) joao-circular(profc) outofmem2(normal) signals002(profthreaded) signals004(profthreaded) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Sun Feb 14 07:13:23 GMT 2010 From igloo at earth.li Sun Feb 14 14:13:23 2010 From: igloo at earth.li (Ian Lynagh) Date: Sun Feb 14 13:44:19 2010 Subject: patch applied (ghc-6.12/ghc): Fix Trac #3850 Message-ID: <20100214191322.GA2444@abbot.galois.com> Mon Feb 8 08:37:51 PST 2010 simonpj@microsoft.com * Fix Trac #3850 This patch simply avoids a needless difference in behaviour from 6.10, and one that happens to affect HList. See Note [Stupid theta]. M ./compiler/rename/RnSource.lhs -3 +13 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100208163751-1287e-efbf0c60c7459099bbb05158a4593a25c137c6d2.gz From igloo at earth.li Sun Feb 14 17:58:54 2010 From: igloo at earth.li (Ian Lynagh) Date: Sun Feb 14 17:29:52 2010 Subject: patch applied (ghc): Don't rely on tar supporting -z; trac #3841 Message-ID: <20100214225854.GA11053@abbot.galois.com> Sun Feb 14 14:04:50 PST 2010 Ian Lynagh * Don't rely on tar supporting -z; trac #3841 M ./libffi/ghc.mk -1 +1 M ./mk/config.mk.in +2 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100214220450-3fd76-4948b72ada05fbfe6c8d1121a4fddcbb7a8585a0.gz From igloo at earth.li Sun Feb 14 18:31:04 2010 From: igloo at earth.li (Ian Lynagh) Date: Sun Feb 14 18:02:00 2010 Subject: patch applied (ghc-6.12/ghc): Don't rely on tar supporting -z; trac #3841 Message-ID: <20100214233104.GA12356@abbot.galois.com> Sun Feb 14 14:04:50 PST 2010 Ian Lynagh * Don't rely on tar supporting -z; trac #3841 M ./libffi/ghc.mk -1 +1 M ./mk/config.mk.in +2 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100214220450-3fd76-4948b72ada05fbfe6c8d1121a4fddcbb7a8585a0.gz From ghcbuild at microsoft.com Sun Feb 14 19:43:09 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sun Feb 14 19:43:11 2010 Subject: [nightly] 14-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100215004309.7A6783245E1@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Sun Feb 14 19:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100209 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Mon Feb 15 01:12:15 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sun Feb 14 21:20:22 GMT 2010 2367 total tests, which gave rise to 13104 test cases, of which 0 caused framework failures 2760 were skipped 10018 expected passes 315 expected failures 0 unexpected passes 11 unexpected failures Unexpected failures: 2592(profc,profasm) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci,threaded2,profthreaded) drvfail006(normal) drvfail008(normal) outofmem2(normal) ---------------------------------------------------- Nightly run ended at Mon Feb 15 01:12:15 GMT 2010 From ghcbuild at microsoft.com Sun Feb 14 19:58:40 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sun Feb 14 19:58:42 2010 Subject: [nightly] 14-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100215005840.657E5324441@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Sun Feb 14 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100211 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Mon Feb 15 01:27:45 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sun Feb 14 21:26:17 GMT 2010 2456 total tests, which gave rise to 13553 test cases, of which 0 caused framework failures 2792 were skipped 10395 expected passes 348 expected failures 0 unexpected passes 18 unexpected failures Unexpected failures: 3429(ghci) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci) ffi009(normal,hpc,optasm,profasm,threaded1,threaded2,dyn,profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) qq007(profc,profasm) ---------------------------------------------------- Nightly run ended at Mon Feb 15 01:27:45 GMT 2010 From Ben.Lippmeier at anu.edu.au Sun Feb 14 21:54:31 2010 From: Ben.Lippmeier at anu.edu.au (Ben Lippmeier) Date: Sun Feb 14 21:25:13 2010 Subject: Head broken on OSX Message-ID: echo "libraries/base_dist-install_depfile_c_asm_EXISTS = YES" >> libraries/base/dist-install/build/.depend-v.c_asm.tmp mv libraries/base/dist-install/build/.depend-v.c_asm.tmp libraries/base/dist-install/build/.depend-v.c_asm "inplace/bin/mkdirhier" libraries/base/dist-install/build/System//. rm -rf gmp-4.2.4 libraries/integer-gmp/gmp/gmpbuild libraries/integer-gmp/gmp/objs cat libraries/integer-gmp/gmp/tarball/gmp-4.2.4-nodoc-patched.tar.bz2 | bzip2 -d { cd libraries/integer-gmp/gmp && /usr/bin/gnutar -xf - ; } /bin/sh: -c: line 0: syntax error near unexpected token `}' /bin/sh: -c: line 0: `cat libraries/integer-gmp/gmp/tarball/gmp-4.2.4-nodoc-patched.tar.bz2 | bzip2 -d { cd libraries/integer-gmp/gmp && /usr/bin/gnutar -xf - ; }' make[1]: *** [libraries/integer-gmp/gmp/gmp.h] Error 2 make[1]: *** Waiting for unfinished jobs.... unused rules: 2 shift/reduce conflicts: 38 reduce/reduce conflicts: 1 make: *** [all] Error 2 Looks tar related. Ben. From Ben.Lippmeier at anu.edu.au Sun Feb 14 21:59:20 2010 From: Ben.Lippmeier at anu.edu.au (Ben Lippmeier) Date: Sun Feb 14 21:30:03 2010 Subject: Head broken on OSX In-Reply-To: References: Message-ID: Unpulling the following patch from libraries/integer-gmp fixes it: Sun Feb 14 14:05:56 PST 2010 Ian Lynagh * Don't rely on tar supporting -j; trac #3841 M ./gmp/ghc.mk -1 +1 On 15/02/2010, at 13:54 , Ben Lippmeier wrote: > > echo "libraries/base_dist-install_depfile_c_asm_EXISTS = YES" >> libraries/base/dist-install/build/.depend-v.c_asm.tmp > mv libraries/base/dist-install/build/.depend-v.c_asm.tmp libraries/base/dist-install/build/.depend-v.c_asm > "inplace/bin/mkdirhier" libraries/base/dist-install/build/System//. > rm -rf gmp-4.2.4 libraries/integer-gmp/gmp/gmpbuild libraries/integer-gmp/gmp/objs > cat libraries/integer-gmp/gmp/tarball/gmp-4.2.4-nodoc-patched.tar.bz2 | bzip2 -d { cd libraries/integer-gmp/gmp && /usr/bin/gnutar -xf - ; } > /bin/sh: -c: line 0: syntax error near unexpected token `}' > /bin/sh: -c: line 0: `cat libraries/integer-gmp/gmp/tarball/gmp-4.2.4-nodoc-patched.tar.bz2 | bzip2 -d { cd libraries/integer-gmp/gmp && /usr/bin/gnutar -xf - ; }' > make[1]: *** [libraries/integer-gmp/gmp/gmp.h] Error 2 > make[1]: *** Waiting for unfinished jobs.... > unused rules: 2 > shift/reduce conflicts: 38 > reduce/reduce conflicts: 1 > make: *** [all] Error 2 > > > Looks tar related. > > Ben. > > _______________________________________________ > Cvs-ghc mailing list > Cvs-ghc@haskell.org > http://www.haskell.org/mailman/listinfo/cvs-ghc -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/cvs-ghc/attachments/20100214/2de0e2c2/attachment.html From rl at cse.unsw.edu.au Sun Feb 14 22:41:59 2010 From: rl at cse.unsw.edu.au (Roman Leshchinskiy) Date: Sun Feb 14 22:12:56 2010 Subject: patch applied (ghc): Spot ForceSpecConstr arguments in polymorphic loops Message-ID: <20100215034159.GA20738@abbot.galois.com> Sun Feb 14 18:53:08 PST 2010 Roman Leshchinskiy * Spot ForceSpecConstr arguments in polymorphic loops M ./compiler/specialise/SpecConstr.lhs -1 +1 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100215025308-b2b0a-b46dbb1f80702311feaf03f3de2b43fd796cfbb2.gz From rl at cse.unsw.edu.au Sun Feb 14 22:42:01 2010 From: rl at cse.unsw.edu.au (Roman Leshchinskiy) Date: Sun Feb 14 22:12:56 2010 Subject: patch applied (ghc): Add comments about the ForceSpecConstr mechanism Message-ID: <20100215034200.GA20758@abbot.galois.com> Sun Feb 14 19:01:03 PST 2010 Roman Leshchinskiy * Add comments about the ForceSpecConstr mechanism M ./compiler/specialise/SpecConstr.lhs +39 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100215030103-b2b0a-541cca24842f2af43612e026a5204a4e68a9101e.gz From ben.lippmeier at anu.edu.au Mon Feb 15 00:24:50 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Sun Feb 14 23:55:46 2010 Subject: patch applied (ghc): Fix build for non-x86 architectures Message-ID: <20100215052450.GA23836@abbot.galois.com> Sun Feb 14 17:47:19 PST 2010 Ben.Lippmeier@anu.edu.au * Fix build for non-x86 architectures M ./compiler/nativeGen/X86/RegInfo.hs +2 M ./compiler/nativeGen/X86/Regs.hs +3 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100215014719-43c66-3f11cc65b3a585f0ac9752a793617bbb3cd9d759.gz From Ben.Lippmeier at anu.edu.au Mon Feb 15 01:04:30 2010 From: Ben.Lippmeier at anu.edu.au (Ben Lippmeier) Date: Mon Feb 15 00:35:27 2010 Subject: new buildbots - was: Exploding SpecConstr In-Reply-To: <20100212171516.GA21868@matrix.chaos.earth.li> References: <947938FD-89E3-4B6E-9547-850824A42812@anu.edu.au> <4B630AEA.5020506@gmail.com> <404396ef1001290858i555226ccj809a3e378ff1569a@mail.gmail.com> <4B66B8CE.507@gmail.com> <20100212171516.GA21868@matrix.chaos.earth.li> Message-ID: <3C68C2D8-14AC-41E6-A596-2B45098BA6A0@anu.edu.au> On 13/02/2010, at 4:15 , Ian Lynagh wrote: > I've uploaded the current code to community: Looks good so far! > http://projects.haskell.org/builder/example/builders/ghcBuilder/2.html > http://projects.haskell.org/builder/example/builders/ghcBuilder/2/9.html > > My TODO list contains: > > * Protocol version > * Some sort of locking so the same client can't connect twice > * More exception handling > * "validate this bundle please" feature I'm very interested in the "validate this bundle please" feature. As functional programmers we spend a lot of our time talking up the benefits of strong typing and well designed data structures. I always liked the quote "functional languages make bad programs unrepresentable". If I spend half my day thinking about advanced type systems but the other half fighting stupid build problems then I end up feeling fairly inadequate. If bad programs should be unrepresentable, then bad patches should be unpushable. Let me know if there's anything I can do to help with the new buildbot. I'd like to get it running on our SPARC/Solaris machine as soon as possible. Ben. From ghcbuild at microsoft.com Mon Feb 15 01:28:41 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Mon Feb 15 01:28:44 2010 Subject: [nightly] 14-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100215062841.D11AF3241EC@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Sun Feb 14 18:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100211 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Mon Feb 15 06:57:47 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sun Feb 14 23:23:55 GMT 2010 2456 total tests, which gave rise to 13553 test cases, of which 0 caused framework failures 2807 were skipped 10368 expected passes 350 expected failures 0 unexpected passes 28 unexpected failures Unexpected failures: annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) ffi002(threaded1,profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) qq007(profc,profasm) signals002(threaded1,threaded2,profthreaded) signals004(threaded1) syn-perf(normal) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Mon Feb 15 06:57:47 GMT 2010 From ghcbuild at microsoft.com Mon Feb 15 02:13:54 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Mon Feb 15 02:13:55 2010 Subject: [nightly] 14-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100215071354.98A8D32413B@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Sun Feb 14 18:10:02 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100209 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (1 failures) **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Mon Feb 15 07:43:01 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sun Feb 14 23:41:09 GMT 2010 2367 total tests, which gave rise to 13104 test cases, of which 0 caused framework failures 2775 were skipped 9982 expected passes 317 expected failures 0 unexpected passes 30 unexpected failures Unexpected failures: 2592(profc,profasm) 3429(threaded1,profthreaded) CPUTime001(threaded2) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) conc019(threaded1) concprog001(ghci,threaded2) drvfail006(normal) drvfail008(normal) ffi002(threaded1) joao-circular(profc) outofmem2(normal) signals002(profthreaded) signals004(threaded1) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Mon Feb 15 07:43:01 GMT 2010 From marlowsd at gmail.com Mon Feb 15 08:41:58 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Mon Feb 15 08:12:54 2010 Subject: patch applied (ghc): fix lost context switches in GHCi (fixes test 3429(ghci)) Message-ID: <20100215134158.GA8441@abbot.galois.com> Mon Feb 15 01:32:05 PST 2010 Simon Marlow * fix lost context switches in GHCi (fixes test 3429(ghci)) M ./rts/Schedule.c -17 +13 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100215093205-12142-2da4478584e3401c9dd201a185baaa71e8aaf648.gz From marlowsd at gmail.com Mon Feb 15 08:42:00 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Mon Feb 15 08:12:55 2010 Subject: patch applied (ghc): Represent the free register set using Word64 on x86-64 (fixes ffi009) Message-ID: <20100215134200.GA8463@abbot.galois.com> Mon Feb 15 05:01:02 PST 2010 Simon Marlow * Represent the free register set using Word64 on x86-64 (fixes ffi009) Following recent changes to the numbering of registers, we overflowed Word32 on x86-64, with the result that xmm8 and later we not being allocated. M ./compiler/nativeGen/RegAlloc/Linear/X86/FreeRegs.hs +4 M ./compiler/nativeGen/X86/Ppr.hs -1 +1 M ./compiler/nativeGen/X86/Regs.hs -6 +13 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100215130102-12142-2d6f4f20a6427a03f65837e3f57e8f95273741cb.gz From igloo at earth.li Mon Feb 15 11:56:01 2010 From: igloo at earth.li (Ian Lynagh) Date: Mon Feb 15 11:26:58 2010 Subject: Head broken on OSX In-Reply-To: References: Message-ID: <20100215165601.GA24552@matrix.chaos.earth.li> On Mon, Feb 15, 2010 at 01:54:31PM +1100, Ben Lippmeier wrote: > > cat libraries/integer-gmp/gmp/tarball/gmp-4.2.4-nodoc-patched.tar.bz2 | bzip2 -d { cd libraries/integer-gmp/gmp && /usr/bin/gnutar -xf - ; } > /bin/sh: -c: line 0: syntax error near unexpected token `}' Should be fixed now. Please let me know if you still have problems. Thanks Ian From ghcbuild at microsoft.com Mon Feb 15 19:51:34 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Mon Feb 15 19:51:36 2010 Subject: [nightly] 15-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100216005134.0384A3244DE@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Mon Feb 15 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100215 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Tue Feb 16 01:20:40 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Mon Feb 15 21:26:50 GMT 2010 2456 total tests, which gave rise to 13553 test cases, of which 0 caused framework failures 2792 were skipped 10404 expected passes 348 expected failures 0 unexpected passes 9 unexpected failures Unexpected failures: T1969(normal) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) qq007(profc,profasm) ---------------------------------------------------- Nightly run ended at Tue Feb 16 01:20:40 GMT 2010 From ghcbuild at microsoft.com Mon Feb 15 19:53:07 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Mon Feb 15 19:53:07 2010 Subject: [nightly] 15-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100216005307.1270B3244DE@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Mon Feb 15 19:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100214 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Tue Feb 16 01:22:15 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Mon Feb 15 21:20:14 GMT 2010 2367 total tests, which gave rise to 13104 test cases, of which 0 caused framework failures 2760 were skipped 10020 expected passes 315 expected failures 0 unexpected passes 9 unexpected failures Unexpected failures: 2592(profc,profasm) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) outofmem2(normal) ---------------------------------------------------- Nightly run ended at Tue Feb 16 01:22:15 GMT 2010 From chak at cse.unsw.edu.au Mon Feb 15 23:42:35 2010 From: chak at cse.unsw.edu.au (Manuel M T Chakravarty) Date: Mon Feb 15 23:13:38 2010 Subject: new buildbots - was: Exploding SpecConstr In-Reply-To: <3C68C2D8-14AC-41E6-A596-2B45098BA6A0@anu.edu.au> References: <947938FD-89E3-4B6E-9547-850824A42812@anu.edu.au> <4B630AEA.5020506@gmail.com> <404396ef1001290858i555226ccj809a3e378ff1569a@mail.gmail.com> <4B66B8CE.507@gmail.com> <20100212171516.GA21868@matrix.chaos.earth.li> <3C68C2D8-14AC-41E6-A596-2B45098BA6A0@anu.edu.au> Message-ID: Ben Lippmeier: > If bad programs should be unrepresentable, then bad patches should be unpushable. Here is a great slogan! Manuel From ghcbuild at microsoft.com Tue Feb 16 01:09:21 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Tue Feb 16 01:09:22 2010 Subject: [nightly] 15-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100216060921.F3AF8324221@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Mon Feb 15 18:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100215 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Tue Feb 16 06:38:30 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Mon Feb 15 23:24:49 GMT 2010 2456 total tests, which gave rise to 13553 test cases, of which 0 caused framework failures 2807 were skipped 10365 expected passes 350 expected failures 0 unexpected passes 31 unexpected failures Unexpected failures: 3429(threaded1) CPUTime001(threaded2) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) conc019(threaded1) concprog001(ghci) ffi002(threaded1,threaded2,profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) qq007(profc,profasm) signals002(profthreaded) signals004(threaded1,profthreaded) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Tue Feb 16 06:38:30 GMT 2010 From ghcbuild at microsoft.com Tue Feb 16 01:40:37 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Tue Feb 16 01:40:38 2010 Subject: [nightly] 15-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100216064037.D5286324221@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Mon Feb 15 18:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100214 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (1 failures) **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Tue Feb 16 07:09:47 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Mon Feb 15 23:41:55 GMT 2010 2367 total tests, which gave rise to 13104 test cases, of which 0 caused framework failures 2775 were skipped 9985 expected passes 317 expected failures 0 unexpected passes 27 unexpected failures Unexpected failures: 2592(profc,profasm) 3429(threaded1,profthreaded) CPUTime001(threaded2) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) ffi002(threaded1,profthreaded) joao-circular(profc) outofmem2(normal) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Tue Feb 16 07:09:47 GMT 2010 From marlowsd at gmail.com Tue Feb 16 09:59:30 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Tue Feb 16 09:30:21 2010 Subject: patch applied (ghc): Fix a bug that can lead to noDuplicate# not working sometimes. Message-ID: <20100216145929.GA31112@abbot.galois.com> Tue Feb 16 04:34:11 PST 2010 Simon Marlow * Fix a bug that can lead to noDuplicate# not working sometimes. The symptom is that under some rare conditions when running in parallel, an unsafePerformIO or unsafeInterleaveIO computation might be duplicated, so e.g. lazy I/O might give the wrong answer (the stream might appear to have duplicate parts or parts missing). I have a program that demonstrates it -N3 or more, some lazy I/O, and a lot of shared mutable state. See the comment with stg_noDuplicatezh in PrimOps.cmm that explains the problem and the fix. This took me about a day to find :-( M ./rts/PrimOps.cmm -8 +71 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100216123411-12142-ef26bfaf1f8faf38c895eda0db121bd91c7a773a.gz From marlowsd at gmail.com Tue Feb 16 09:59:31 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Tue Feb 16 09:30:23 2010 Subject: patch applied (ghc): disable a false assertion, with a comment to explain why Message-ID: <20100216145931.GA31132@abbot.galois.com> Tue Feb 16 04:35:44 PST 2010 Simon Marlow * disable a false assertion, with a comment to explain why M ./rts/Capability.h -1 +2 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100216123544-12142-40431277f5cb331d8c23336970b617f956ff208c.gz From marlowsd at gmail.com Tue Feb 16 09:59:33 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Tue Feb 16 09:30:24 2010 Subject: patch applied (ghc): Write the binary cache file atomically Message-ID: <20100216145933.GA31152@abbot.galois.com> Tue Feb 16 05:48:41 PST 2010 Simon Marlow * Write the binary cache file atomically Should fix an occasional build error of the form ghc-pkg: too few bytes. Failed reading at byte position 8 M ./utils/ghc-pkg/Main.hs -5 +13 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100216134841-12142-9fa8f442cdbee9dca82b4a020470641d6f55f8cf.gz From marlowsd at gmail.com Tue Feb 16 10:01:48 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Tue Feb 16 09:32:38 2010 Subject: patch applied (testsuite): add a test for a bug in noDuplicate# Message-ID: <20100216150147.GA31507@abbot.galois.com> Tue Feb 16 04:32:10 PST 2010 Simon Marlow * add a test for a bug in noDuplicate# A ./tests/ghc-regress/concurrent/prog003/ A ./tests/ghc-regress/concurrent/prog003/BackList2.lhs A ./tests/ghc-regress/concurrent/prog003/CASList.hs A ./tests/ghc-regress/concurrent/prog003/Collate.hs A ./tests/ghc-regress/concurrent/prog003/Collection.hs A ./tests/ghc-regress/concurrent/prog003/IOList.lhs A ./tests/ghc-regress/concurrent/prog003/ImmList.hs A ./tests/ghc-regress/concurrent/prog003/MVarListLockCoupling.hs A ./tests/ghc-regress/concurrent/prog003/Main.lhs A ./tests/ghc-regress/concurrent/prog003/MainMVarList.lhs A ./tests/ghc-regress/concurrent/prog003/Makefile A ./tests/ghc-regress/concurrent/prog003/RefInterface.hs A ./tests/ghc-regress/concurrent/prog003/TestData.hs A ./tests/ghc-regress/concurrent/prog003/TestDataParser.hs A ./tests/ghc-regress/concurrent/prog003/TestRun.hs A ./tests/ghc-regress/concurrent/prog003/all.T A ./tests/ghc-regress/concurrent/prog003/concprog003.stdout A ./tests/ghc-regress/concurrent/prog003/test-8-3000-3000-2-1-4 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100216123210-12142-3259d8eb53ded96e1bc6cfa5fcdcab871671a056.gz From marlowsd at gmail.com Tue Feb 16 10:01:49 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Tue Feb 16 09:32:42 2010 Subject: patch applied (testsuite): enable the asm ways for ffi009 on i386 if -msse2 is available Message-ID: <20100216150149.GA31527@abbot.galois.com> Tue Feb 16 04:45:14 PST 2010 Simon Marlow * enable the asm ways for ffi009 on i386 if -msse2 is available M ./tests/ghc-regress/ffi/should_run/all.T -11 +11 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100216124514-12142-25b192afa929b873c71ce57cb4818bacf585a140.gz From igloo at earth.li Tue Feb 16 11:03:20 2010 From: igloo at earth.li (Ian Lynagh) Date: Tue Feb 16 10:34:11 2010 Subject: patch applied (ghc-6.12/ghc): don't sanity check the whole stack when switching interp<->compiled Message-ID: <20100216160320.GA1555@abbot.galois.com> Wed Dec 2 05:41:21 PST 2009 Simon Marlow * don't sanity check the whole stack when switching interp<->compiled M ./rts/Schedule.c -3 +4 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20091202134121-12142-7a6b1c3611c742af0e24b648d482b76e2ea2b3db.gz From igloo at earth.li Tue Feb 16 11:03:22 2010 From: igloo at earth.li (Ian Lynagh) Date: Tue Feb 16 10:34:12 2010 Subject: patch applied (ghc-6.12/ghc): fix lost context switches in GHCi (fixes test 3429(ghci)) Message-ID: <20100216160322.GA1575@abbot.galois.com> Mon Feb 15 01:32:05 PST 2010 Simon Marlow * fix lost context switches in GHCi (fixes test 3429(ghci)) M ./rts/Schedule.c -17 +13 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100215093205-12142-2da4478584e3401c9dd201a185baaa71e8aaf648.gz From marlowsd at gmail.com Tue Feb 16 11:38:01 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Tue Feb 16 11:08:53 2010 Subject: patch applied (ghc): Beef up cmmMiniInline a tiny bit Message-ID: <20100216163801.GA3194@abbot.galois.com> Tue Feb 16 07:05:06 PST 2010 Simon Marlow * Beef up cmmMiniInline a tiny bit Allow a temporary assignment to be pushed past an assignment to a global if the global is not mentioned in the rhs of the assignment we are inlining. This fixes up some bad code. We should make sure we're doing something equivalent in the new backend in due course. M ./compiler/cmm/CmmExpr.hs +11 M ./compiler/cmm/CmmOpt.hs -18 +19 M ./compiler/codeGen/CgUtils.hs -7 M ./compiler/codeGen/StgCmmUtils.hs -2 +1 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100216150506-12142-3f76173e3d74f38ddc4494a8a1da988eab6e995a.gz From ghcbuild at microsoft.com Tue Feb 16 19:44:19 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Tue Feb 16 19:44:20 2010 Subject: [nightly] 16-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100217004419.B3493324136@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Tue Feb 16 19:10:02 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100215 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Wed Feb 17 01:13:31 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Tue Feb 16 21:20:54 GMT 2010 2367 total tests, which gave rise to 13104 test cases, of which 0 caused framework failures 2760 were skipped 10019 expected passes 315 expected failures 0 unexpected passes 10 unexpected failures Unexpected failures: 2592(profc,profasm) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci,threaded2) drvfail006(normal) drvfail008(normal) outofmem2(normal) ---------------------------------------------------- Nightly run ended at Wed Feb 17 01:13:31 GMT 2010 From ghcbuild at microsoft.com Tue Feb 16 19:52:51 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Tue Feb 16 19:52:53 2010 Subject: [nightly] 16-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100217005251.5400832415D@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Tue Feb 16 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100216 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Wed Feb 17 01:21:57 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Tue Feb 16 21:27:25 GMT 2010 2457 total tests, which gave rise to 13564 test cases, of which 0 caused framework failures 2792 were skipped 10415 expected passes 348 expected failures 0 unexpected passes 9 unexpected failures Unexpected failures: T1969(normal) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) qq007(profc,profasm) ---------------------------------------------------- Nightly run ended at Wed Feb 17 01:21:57 GMT 2010 From ghcbuild at microsoft.com Wed Feb 17 01:12:40 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Wed Feb 17 01:12:42 2010 Subject: [nightly] 16-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100217061240.EF73D32411A@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Tue Feb 16 18:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100216 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Wed Feb 17 06:41:53 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Tue Feb 16 23:25:13 GMT 2010 2457 total tests, which gave rise to 13564 test cases, of which 0 caused framework failures 2799 were skipped 10388 expected passes 350 expected failures 0 unexpected passes 27 unexpected failures Unexpected failures: 3429(threaded1,profthreaded) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) concprog001(ghci) ffi002(profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) qq007(profc,profasm) signals002(profthreaded) signals004(profthreaded) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Wed Feb 17 06:41:53 GMT 2010 From ghcbuild at microsoft.com Wed Feb 17 01:45:37 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Wed Feb 17 01:45:38 2010 Subject: [nightly] 16-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100217064537.D0C27324076@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Tue Feb 16 18:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100215 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (1 failures) **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Wed Feb 17 07:14:50 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Tue Feb 16 23:43:10 GMT 2010 2367 total tests, which gave rise to 13104 test cases, of which 0 caused framework failures 2775 were skipped 9987 expected passes 317 expected failures 0 unexpected passes 25 unexpected failures Unexpected failures: 2592(profc,profasm) 3429(threaded1) CPUTime001(threaded2) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) joao-circular(profc) outofmem2(normal) signals002(threaded1) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Wed Feb 17 07:14:50 GMT 2010 From marlowsd at gmail.com Wed Feb 17 09:30:03 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Wed Feb 17 09:00:51 2010 Subject: patch applied (ghc): Allow extra packages to be added to the build, by setting $(EXTRA_PACKAGES) Message-ID: <20100217143003.GA20265@abbot.galois.com> Wed Feb 17 02:31:34 PST 2010 Simon Marlow * Allow extra packages to be added to the build, by setting $(EXTRA_PACKAGES) M ./ghc.mk +2 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100217103134-12142-46571f6196fb4c0773ed71b34ffe94e955c00609.gz From marlowsd at gmail.com Wed Feb 17 09:30:05 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Wed Feb 17 09:00:53 2010 Subject: patch applied (ghc): darcs-all --extra get: gets a few extra packages Message-ID: <20100217143005.GA20292@abbot.galois.com> Wed Feb 17 02:46:34 PST 2010 Simon Marlow * darcs-all --extra get: gets a few extra packages currently deepseq, parallel, stm. M ./darcs-all +5 M ./packages +3 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100217104634-12142-c2798c309dfc9219f3a3939a46a090ac2539ca82.gz From marlowsd at gmail.com Wed Feb 17 09:30:06 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Wed Feb 17 09:00:54 2010 Subject: patch applied (ghc): Fix a recompilation checking bug when a package dependency changes Message-ID: <20100217143006.GA20313@abbot.galois.com> Wed Feb 17 05:32:50 PST 2010 Simon Marlow * Fix a recompilation checking bug when a package dependency changes We weren't forcing enough recompilationg when package dependencies changed, with the result that bumping a package version could lead to linking failures or other problems later. The problem/solutation are described on the wiki: http://hackage.haskell.org/trac/ghc/wiki/Commentary/Compiler/RecompilationAvoidance#Packageversionchanges M ./compiler/iface/MkIface.lhs -1 +6 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100217133250-12142-5fd51608b310ae75bfa8dd488ad720a3d5e18a9f.gz From marlowsd at gmail.com Wed Feb 17 09:32:17 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Wed Feb 17 09:03:04 2010 Subject: patch applied (testsuite): Add a test for recompilation when a package version changes Message-ID: <20100217143217.GA20649@abbot.galois.com> Wed Feb 17 05:53:21 PST 2010 Simon Marlow * Add a test for recompilation when a package version changes A ./tests/ghc-regress/driver/recomp007/ A ./tests/ghc-regress/driver/recomp007/Makefile A ./tests/ghc-regress/driver/recomp007/Setup.hs A ./tests/ghc-regress/driver/recomp007/a1/ A ./tests/ghc-regress/driver/recomp007/a1/A.hs A ./tests/ghc-regress/driver/recomp007/a1/a.cabal A ./tests/ghc-regress/driver/recomp007/a2/ A ./tests/ghc-regress/driver/recomp007/a2/A.hs A ./tests/ghc-regress/driver/recomp007/a2/a.cabal A ./tests/ghc-regress/driver/recomp007/all.T A ./tests/ghc-regress/driver/recomp007/b/ A ./tests/ghc-regress/driver/recomp007/b/B.hs A ./tests/ghc-regress/driver/recomp007/b/Main.hs A ./tests/ghc-regress/driver/recomp007/b/b.cabal A ./tests/ghc-regress/driver/recomp007/recomp007.stderr A ./tests/ghc-regress/driver/recomp007/recomp007.stdout View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100217135321-12142-9a18844a3f41f4bb24131fdcae667e126e6dbb45.gz From kili at outback.escape.de Wed Feb 17 15:29:39 2010 From: kili at outback.escape.de (Matthias Kilian) Date: Wed Feb 17 15:00:48 2010 Subject: packages/parallel disappeared from darcs.haskell.org Message-ID: <20100217202938.GA11630@nutty.outback.escape.de> Hi, has the packages/parallel repository been removed from darcs.haskell.org by intention? Ciao, Kili From igloo at earth.li Wed Feb 17 18:04:23 2010 From: igloo at earth.li (Ian Lynagh) Date: Wed Feb 17 17:35:13 2010 Subject: patch applied (ghc-6.12/ghc): Fix a bug that can lead to noDuplicate# not working sometimes. Message-ID: <20100217230422.GA9657@abbot.galois.com> Tue Feb 16 04:34:11 PST 2010 Simon Marlow * Fix a bug that can lead to noDuplicate# not working sometimes. The symptom is that under some rare conditions when running in parallel, an unsafePerformIO or unsafeInterleaveIO computation might be duplicated, so e.g. lazy I/O might give the wrong answer (the stream might appear to have duplicate parts or parts missing). I have a program that demonstrates it -N3 or more, some lazy I/O, and a lot of shared mutable state. See the comment with stg_noDuplicatezh in PrimOps.cmm that explains the problem and the fix. This took me about a day to find :-( M ./rts/PrimOps.cmm -8 +71 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100216123411-12142-ef26bfaf1f8faf38c895eda0db121bd91c7a773a.gz From igloo at earth.li Wed Feb 17 18:04:24 2010 From: igloo at earth.li (Ian Lynagh) Date: Wed Feb 17 17:35:14 2010 Subject: patch applied (ghc-6.12/ghc): disable a false assertion, with a comment to explain why Message-ID: <20100217230424.GA9677@abbot.galois.com> Tue Feb 16 04:35:44 PST 2010 Simon Marlow * disable a false assertion, with a comment to explain why M ./rts/Capability.h -1 +2 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100216123544-12142-40431277f5cb331d8c23336970b617f956ff208c.gz From igloo at earth.li Wed Feb 17 18:04:27 2010 From: igloo at earth.li (Ian Lynagh) Date: Wed Feb 17 17:35:17 2010 Subject: patch applied (ghc-6.12/ghc): Fix a recompilation checking bug when a package dependency changes Message-ID: <20100217230427.GA9719@abbot.galois.com> Wed Feb 17 05:32:50 PST 2010 Simon Marlow * Fix a recompilation checking bug when a package dependency changes We weren't forcing enough recompilationg when package dependencies changed, with the result that bumping a package version could lead to linking failures or other problems later. The problem/solutation are described on the wiki: http://hackage.haskell.org/trac/ghc/wiki/Commentary/Compiler/RecompilationAvoidance#Packageversionchanges M ./compiler/iface/MkIface.lhs -1 +6 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100217133250-12142-5fd51608b310ae75bfa8dd488ad720a3d5e18a9f.gz From igloo at earth.li Wed Feb 17 18:04:26 2010 From: igloo at earth.li (Ian Lynagh) Date: Wed Feb 17 17:35:18 2010 Subject: patch applied (ghc-6.12/ghc): Write the binary cache file atomically Message-ID: <20100217230426.GA9698@abbot.galois.com> Tue Feb 16 05:48:41 PST 2010 Simon Marlow * Write the binary cache file atomically Should fix an occasional build error of the form ghc-pkg: too few bytes. Failed reading at byte position 8 M ./utils/ghc-pkg/Main.hs -5 +13 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100216134841-12142-9fa8f442cdbee9dca82b4a020470641d6f55f8cf.gz From igloo at earth.li Wed Feb 17 18:07:25 2010 From: igloo at earth.li (Ian Lynagh) Date: Wed Feb 17 17:38:12 2010 Subject: patch applied (ghc-6.12/testsuite): add a test for a bug in noDuplicate# Message-ID: <20100217230725.GA9944@abbot.galois.com> Tue Feb 16 04:32:10 PST 2010 Simon Marlow * add a test for a bug in noDuplicate# A ./tests/ghc-regress/concurrent/prog003/ A ./tests/ghc-regress/concurrent/prog003/BackList2.lhs A ./tests/ghc-regress/concurrent/prog003/CASList.hs A ./tests/ghc-regress/concurrent/prog003/Collate.hs A ./tests/ghc-regress/concurrent/prog003/Collection.hs A ./tests/ghc-regress/concurrent/prog003/IOList.lhs A ./tests/ghc-regress/concurrent/prog003/ImmList.hs A ./tests/ghc-regress/concurrent/prog003/MVarListLockCoupling.hs A ./tests/ghc-regress/concurrent/prog003/Main.lhs A ./tests/ghc-regress/concurrent/prog003/MainMVarList.lhs A ./tests/ghc-regress/concurrent/prog003/Makefile A ./tests/ghc-regress/concurrent/prog003/RefInterface.hs A ./tests/ghc-regress/concurrent/prog003/TestData.hs A ./tests/ghc-regress/concurrent/prog003/TestDataParser.hs A ./tests/ghc-regress/concurrent/prog003/TestRun.hs A ./tests/ghc-regress/concurrent/prog003/all.T A ./tests/ghc-regress/concurrent/prog003/concprog003.stdout A ./tests/ghc-regress/concurrent/prog003/test-8-3000-3000-2-1-4 View patch online: http://darcs.haskell.org/ghc-6.12/testsuite/_darcs/patches/20100216123210-12142-3259d8eb53ded96e1bc6cfa5fcdcab871671a056.gz From igloo at earth.li Wed Feb 17 18:07:26 2010 From: igloo at earth.li (Ian Lynagh) Date: Wed Feb 17 17:38:13 2010 Subject: patch applied (ghc-6.12/testsuite): Add a test for recompilation when a package version changes Message-ID: <20100217230726.GA9964@abbot.galois.com> Wed Feb 17 05:53:21 PST 2010 Simon Marlow * Add a test for recompilation when a package version changes A ./tests/ghc-regress/driver/recomp007/ A ./tests/ghc-regress/driver/recomp007/Makefile A ./tests/ghc-regress/driver/recomp007/Setup.hs A ./tests/ghc-regress/driver/recomp007/a1/ A ./tests/ghc-regress/driver/recomp007/a1/A.hs A ./tests/ghc-regress/driver/recomp007/a1/a.cabal A ./tests/ghc-regress/driver/recomp007/a2/ A ./tests/ghc-regress/driver/recomp007/a2/A.hs A ./tests/ghc-regress/driver/recomp007/a2/a.cabal A ./tests/ghc-regress/driver/recomp007/all.T A ./tests/ghc-regress/driver/recomp007/b/ A ./tests/ghc-regress/driver/recomp007/b/B.hs A ./tests/ghc-regress/driver/recomp007/b/Main.hs A ./tests/ghc-regress/driver/recomp007/b/b.cabal A ./tests/ghc-regress/driver/recomp007/recomp007.stderr A ./tests/ghc-regress/driver/recomp007/recomp007.stdout View patch online: http://darcs.haskell.org/ghc-6.12/testsuite/_darcs/patches/20100217135321-12142-9a18844a3f41f4bb24131fdcae667e126e6dbb45.gz From ghcbuild at microsoft.com Wed Feb 17 19:51:53 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Wed Feb 17 19:51:56 2010 Subject: [nightly] 17-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100218005153.88AB932407F@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Wed Feb 17 19:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100215 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Thu Feb 18 01:21:03 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Wed Feb 17 21:20:57 GMT 2010 2367 total tests, which gave rise to 13104 test cases, of which 0 caused framework failures 2760 were skipped 10019 expected passes 315 expected failures 0 unexpected passes 10 unexpected failures Unexpected failures: 2592(profc,profasm) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci,threaded2) drvfail006(normal) drvfail008(normal) outofmem2(normal) ---------------------------------------------------- Nightly run ended at Thu Feb 18 01:21:03 GMT 2010 From ghcbuild at microsoft.com Wed Feb 17 19:55:49 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Wed Feb 17 19:55:51 2010 Subject: [nightly] 17-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100218005549.56AE03242C9@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Wed Feb 17 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100217 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Thu Feb 18 01:25:03 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Wed Feb 17 21:26:41 GMT 2010 2458 total tests, which gave rise to 13565 test cases, of which 0 caused framework failures 2792 were skipped 10415 expected passes 348 expected failures 0 unexpected passes 10 unexpected failures Unexpected failures: T1969(normal) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci,threaded2) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) qq007(profc,profasm) ---------------------------------------------------- Nightly run ended at Thu Feb 18 01:25:03 GMT 2010 From chak at cse.unsw.edu.au Wed Feb 17 20:53:44 2010 From: chak at cse.unsw.edu.au (Manuel M T Chakravarty) Date: Wed Feb 17 20:24:34 2010 Subject: patch applied (ghc): darcs-all --extra get: gets a few extra packages In-Reply-To: <20100217143005.GA20292@abbot.galois.com> References: <20100217143005.GA20292@abbot.galois.com> Message-ID: <33936BFC-F436-470B-B452-FD1754FD9FA6@cse.unsw.edu.au> Simon Marlow: > Wed Feb 17 02:46:34 PST 2010 Simon Marlow > * darcs-all --extra get: gets a few extra packages > currently deepseq, parallel, stm. That's a very useful addition, but unfortunately it doesn't seem to work (for me): > limitingfactor chak 5 (.../Code/ghc-test): ./darcs-all --extra get > warning: adding --partial, to override use --complete > warning: . already present; omitting > warning: ghc-tarballs already present; omitting > warning: utils/hsc2hs already present; omitting > warning: utils/haddock already present; omitting > warning: libraries/array already present; omitting > warning: libraries/base already present; omitting > warning: libraries/binary already present; omitting > warning: libraries/Cabal already present; omitting > warning: libraries/containers already present; omitting > warning: libraries/directory already present; omitting > warning: libraries/dph already present; omitting > warning: libraries/extensible-exceptions already present; omitting > warning: libraries/filepath already present; omitting > warning: libraries/ghc-prim already present; omitting > warning: libraries/haskeline already present; omitting > warning: libraries/haskell98 already present; omitting > warning: libraries/hpc already present; omitting > warning: libraries/integer-gmp already present; omitting > warning: libraries/integer-simple already present; omitting > warning: libraries/mtl already present; omitting > warning: libraries/old-locale already present; omitting > warning: libraries/old-time already present; omitting > warning: libraries/pretty already present; omitting > warning: libraries/process already present; omitting > warning: libraries/random already present; omitting > warning: libraries/template-haskell already present; omitting > warning: libraries/terminfo already present; omitting > warning: libraries/unix already present; omitting > warning: libraries/utf8-string already present; omitting > warning: libraries/Win32 already present; omitting > == running darcs get --partial http://darcs.haskell.org/packages/deepseq libraries/deepseq > --partial: hashed or darcs-2 repository detected, using --lazy instead > Finished getting. > == running darcs get --partial http://darcs.haskell.org/packages/parallel libraries/parallel > > darcs failed: Not a repository: http://darcs.haskell.org/packages/parallel (Failed to download URL http://darcs.haskell.org/packages/parallel/_darcs/inventory: HTTP response code said error) > darcs failed: 512 at ./darcs-all line 148. > == Checking for old bytestring repo > == Checking for unpulled tarball patches > limitingfactor chak 6 (.../Code/ghc-test): Package deepseq is indeed on darcs.haskell.org, but parallel and stm are on code.haskell.org. Cheers, Manuel From ghcbuild at microsoft.com Thu Feb 18 00:56:28 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Thu Feb 18 00:56:32 2010 Subject: [nightly] 17-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100218055628.D3335324829@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Wed Feb 17 18:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100217 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Thu Feb 18 06:25:42 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Wed Feb 17 23:24:12 GMT 2010 2458 total tests, which gave rise to 13565 test cases, of which 0 caused framework failures 2799 were skipped 10389 expected passes 350 expected failures 0 unexpected passes 27 unexpected failures Unexpected failures: 3429(threaded1) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) concprog001(ghci) ffi002(profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) qq007(profc,profasm) signals002(threaded1,threaded2) signals004(profthreaded) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Thu Feb 18 06:25:42 GMT 2010 From ghcbuild at microsoft.com Thu Feb 18 01:57:21 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Thu Feb 18 01:57:22 2010 Subject: [nightly] 17-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100218065721.75A3432407F@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Wed Feb 17 18:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100215 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (1 failures) **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Thu Feb 18 07:26:37 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Wed Feb 17 23:41:02 GMT 2010 2367 total tests, which gave rise to 13104 test cases, of which 0 caused framework failures 2775 were skipped 9985 expected passes 317 expected failures 0 unexpected passes 27 unexpected failures Unexpected failures: 2592(profc,profasm) 3429(threaded1) CPUTime001(threaded2) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) conc070(threaded1) concprog001(ghci,threaded2) drvfail006(normal) drvfail008(normal) ffi002(threaded1) joao-circular(profc) outofmem2(normal) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Thu Feb 18 07:26:37 GMT 2010 From marlowsd at gmail.com Thu Feb 18 04:46:47 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Thu Feb 18 04:17:36 2010 Subject: patch applied (ghc): darcs-all --extra get: gets a few extra packages In-Reply-To: <33936BFC-F436-470B-B452-FD1754FD9FA6@cse.unsw.edu.au> References: <20100217143005.GA20292@abbot.galois.com> <33936BFC-F436-470B-B452-FD1754FD9FA6@cse.unsw.edu.au> Message-ID: <4B7D0C87.2020001@gmail.com> On 18/02/2010 01:53, Manuel M T Chakravarty wrote: > Simon Marlow: >> Wed Feb 17 02:46:34 PST 2010 Simon Marlow >> * darcs-all --extra get: gets a few extra packages >> currently deepseq, parallel, stm. > > That's a very useful addition, but unfortunately it doesn't seem to work (for me): Yes I know, I need to shift some repositories around. I hadn't realised that Ian has moved many of the repositories that used to be at http://darcs.haskell.org/packages to code.haskell.org recently. I think many of the .cabal files still refer to the old locations so those need updating too. Cheers, Simon From marlowsd at gmail.com Thu Feb 18 05:34:13 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Thu Feb 18 05:05:05 2010 Subject: packages/parallel disappeared from darcs.haskell.org In-Reply-To: <20100217202938.GA11630@nutty.outback.escape.de> References: <20100217202938.GA11630@nutty.outback.escape.de> Message-ID: <4B7D17A5.2030507@gmail.com> On 17/02/2010 20:29, Matthias Kilian wrote: > has the packages/parallel repository been removed from darcs.haskell.org > by intention? I've put it back now, along with deepseq and stm. Cheers, Simon From davidterei at gmail.com Thu Feb 18 18:55:08 2010 From: davidterei at gmail.com (David Terei) Date: Thu Feb 18 18:25:53 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B7CF343.3020608@gmail.com> References: <4B7CF343.3020608@gmail.com> Message-ID: <4B7DD35C.2010408@gmail.com> Hi all, Over the last 6 months I've been working on a new code generator for GHC which targets the LLVM compiler infrastructure. Most of the work was done as part of an honours thesis at the University of New South Wales under the supervision of Manuel Chakravarty. This ended at the start of November but I have continued to work on the code generator since (although at a much reduced pace since I'm back at full time work). Its now at a stage where I feel pretty confident in its correctness and would love to see it merged into GHC mainline. The patch for the llvm back-end can be found here (should apply cleanly to GHC head): http://www.cse.unsw.edu.au/~davidt/downloads/ghc-llvmbackend-full.gz This is what I would like/am requesting merged into GHC head. The thesis paper which offers a detailed performance evaluation, as well as the motivation and design of the back-end can be found at: http://www.cse.unsw.edu.au/~pls/thesis/davidt-thesis.pdf Below I'll quickly detail out the important points though. There are also instructions on how to get started with the back-end. Finally there are also some issues that I think may need to be sorted out before a merge could be done. They are at the end. Performance ----------- (All done on linux/x86-32) A quick summary of the results are that for the 'nofib' benchmark suite, the llvm code generator was 3.8% slower than the NCG (the C code generator was 6.9% slower than the NCG). The DPH project includes a benchmark suite which I also ran and for this type of code using the llvm back-end shortened the runtime by an average of 25% compared to the NCG. Also, while not included in my thesis paper as I ran out of time, I did do some benchmarking with the 'nobench' benchmark suite. It gave performance ratios for the back-ends of around: NCG : 1.11 C : 1.05 LLVM : 1.14 Supported Platforms & 'Correctness' ----------------------------------- Linux x86-32/x86-64 are currently well supported. The back-end can pass the test suite and build a working version of GHC (bootstrap test). Mac OS X 10.5 currently has a rather nasty bug with any dynamic lib calls (all libffi stuff) [due to the stack not being 16byte aligned when the calls are made as required by OSX ABI for the curious]. Test suite passes except for most the ffi tests. Other platforms haven't been tested at all. As using the back-end with a registered build of GHC requires a modified version of LLVM, people wanting to try it out on those platforms will need to either make the needed changes to LLVM themselves, or use an unregistered build of GHC which will work with a vanilla install of LLVM. (A patch for LLVM for x86 is linked to below.) Validate -------- I've validated my GHC patch to make sure it won't break anything. This is just compiling and running GHC normally but with the llvm back-end code included. It doesn't actually test the llvm code generator, just makes sure it hasn't broken the NCG or C code generator. Linux/x86-32: OVERALL SUMMARY for test run started at Do 18. Feb 11:21:48 EST 2010 2457 total tests, which gave rise to 9738 test cases, of which 0 caused framework failures 7573 were skipped 2088 expected passes 76 expected failures 0 unexpected passes 1 unexpected failures Unexpected failures: user001(normal) Linux/x86-64: OVERALL SUMMARY for test run started at Thu 18 Feb 15:28:32 EST 2010 2458 total tests, which gave rise to 9739 test cases, of which 0 caused framework failures 7574 were skipped 2087 expected passes 77 expected failures 0 unexpected passes 1 unexpected failures Unexpected failures: T1969(normal) Mac OS X 10.5/x86-32: OVERALL SUMMARY for test run started at Thu Feb 18 12:35:49 EST 2010 2458 total tests, which gave rise to 9122 test cases, of which 0 caused framework failures 6959 were skipped 2085 expected passes 76 expected failures 0 unexpected passes 2 unexpected failures Unexpected failures: T1969(normal) ffi005(optc) All of the test failures fail for me with a unmodified GHC head build as well as when the llvm patch is included, so the llvm patch isn't introducing any new failures. Installing ---------- http://www.cse.unsw.edu.au/~davidt/downloads/ghc-llvmbackend-full.gz Apply the darcs patch linked above to GHC head. This will make some changes across GHC, with the bulk of the new code ending up in 'compiler/llvmGen'. To build GHC you need to add two flags to build.mk, they are: GhcWithLlvmCodeGen = YES GhcEnableTablesNextToCode = NO The llvm code generator doesn't support at this time the TABLES_NEXT_TO_CODE optimisation due to limitations with LLVM. You will also need LLVM installed on your computer to use the back-end. Version 2.6 or SVN trunk is supported. If you want to use the back-end in an unregistered ghc build, then you can use a vanilla build of LLVM. However if you want to use a registered ghc build (very likely) then you need to patch LLVM for this to work. The patch for llvm can be found here: http://www.cse.unsw.edu.au/~davidt/downloads/llvm-ghc.patch LLVM is very easy to build and install. It can be done as follows: $ svn co http://llvm.org/svn/llvm-project/llvm/trunk llvm $ cd llvm $ patch -p0 -i ~/llvm-ghc.patch $ ./configure --enable-optimized # probably also want to set --prefix $ make $ make install Just make sure this modified version of LLVM is on your path and takes precedence over any other builds. Using ----- Once GHC is built, you can trigger GHC to use the LLVM back-end with the '-fllvm' flag. There is also a new '-ddump-llvm' which will dump out the llvm IR code generated. (or use the '-keep-tmp-files' flag). 'ghc --info' should also now report that it includes the llvm code generator. Issues ------ Issues that might need to be resolved before merging the patch: 1. Developed in isolation by 1 person with no Haskell knowledge at first. So usual issues with that may apply, misused data structures, bad style... ect. Criticisms of the code are very welcome. There are some specific notes on what I think may be wrong with the code atm in 'compiler/llvmGen/NOTES'. 2. The back-end has a LLVM binding of sorts, this binding is similar in design to say the Cmm representation used in GHC. It represents the LLVM Assembly language using a collection of data types and can pretty print it out correctly. This binding lives in the 'compiler/llvmGen/Llvm' folder. Should this binding be split out into a separate library? 3. As mentioned above, LLVM needs to be patched to work with a registered build of GHC. If the llvm back-end was merged, how would this be handled? I would suggest simply carrying the patch with some instructions on how to use it in the GHC repo. People using GHC head could be expected to grab the LLVM source code and apply the patch themselves at this stage. 4. Finally this email is long. I need to put all this info into a web page as well. Cheers, David From ghcbuild at microsoft.com Thu Feb 18 19:39:04 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Thu Feb 18 19:39:06 2010 Subject: [nightly] 18-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100219003904.8A46032433E@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Thu Feb 18 19:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100217 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Fri Feb 19 01:08:22 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Thu Feb 18 21:20:50 GMT 2010 2369 total tests, which gave rise to 13116 test cases, of which 0 caused framework failures 2760 were skipped 10031 expected passes 315 expected failures 0 unexpected passes 10 unexpected failures Unexpected failures: 2592(profc,profasm) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) conc019(threaded1) concprog001(ghci) drvfail006(normal) drvfail008(normal) outofmem2(normal) ---------------------------------------------------- Nightly run ended at Fri Feb 19 01:08:22 GMT 2010 From ghcbuild at microsoft.com Thu Feb 18 19:55:00 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Thu Feb 18 19:55:02 2010 Subject: [nightly] 18-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100219005500.1FC9432492F@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Thu Feb 18 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100217 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Fri Feb 19 01:24:18 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Thu Feb 18 21:26:27 GMT 2010 2458 total tests, which gave rise to 13565 test cases, of which 0 caused framework failures 2792 were skipped 10416 expected passes 348 expected failures 0 unexpected passes 9 unexpected failures Unexpected failures: T1969(normal) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) qq007(profc,profasm) ---------------------------------------------------- Nightly run ended at Fri Feb 19 01:24:18 GMT 2010 From chak at cse.unsw.edu.au Thu Feb 18 21:25:58 2010 From: chak at cse.unsw.edu.au (Manuel M T Chakravarty) Date: Thu Feb 18 20:56:46 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B7DD35C.2010408@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> Message-ID: <55A9DA74-C9B8-4988-A91E-2A150C48F150@cse.unsw.edu.au> David wrote, > Issues > ------ > Issues that might need to be resolved before merging the patch: > > 1. Developed in isolation by 1 person with no Haskell knowledge at first. So usual issues with that may apply, misused data structures, bad style... ect. Criticisms of the code are very welcome. There are some specific notes on what I think may be wrong with the code atm in 'compiler/llvmGen/NOTES'. I have seen worse code in GHC, *much* worse ;) > 2. The back-end has a LLVM binding of sorts, this binding is similar in design to say the Cmm representation used in GHC. It represents the LLVM Assembly language using a collection of data types and can pretty print it out correctly. This binding lives in the 'compiler/llvmGen/Llvm' folder. Should this binding be split out into a separate library? That library would be a build requirement for GHC, then. It may be useful to separate it out for other projects that want to use it, but for GHC, I don't think it would help anything, so I wouldn't worry about that until somebody actually wants to use it for something else. > 3. As mentioned above, LLVM needs to be patched to work with a registered build of GHC. If the llvm back-end was merged, how would this be handled? I would suggest simply carrying the patch with some instructions on how to use it in the GHC repo. People using GHC head could be expected to grab the LLVM source code and apply the patch themselves at this stage. I think, we should bundle LLVM with GHC and auto apply the patch at build boot time (in the same manner that gmp is bundled and patched and libffi is bundeled). There is also an obvious place where the LLVM tarbar would go, namely http://darcs.haskell.org/ghc-tarballs/ Why should we bundle? Because we want to make life for current and prospective GHC devs as easy as possible. A build dependency that you have to download, patch, and install manually complicates the build instructions and is annoying. Given it has to be patched, package managers ?such as those for various Linux distros? won't help either. Moreover, the day will come where LLVM gets updated and GHC requires a tweak to work with the new LLVM and people will download the wrong version and spam the support list. No fun. My vote is for merging the LLVM backend as it is (with an LLVM tarball in ghc-tarballs/). It needs to be an optional build option for now, though, as it requires to switch off TABLES_NEXT_TO_CODE. Once it is in, it's much easier for people to try it, play with it, and improve it. Manuel From felipe.lessa at gmail.com Thu Feb 18 21:53:17 2010 From: felipe.lessa at gmail.com (Felipe Lessa) Date: Thu Feb 18 21:23:59 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B7DD35C.2010408@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> Message-ID: <20100219025316.GA18876@kira.casa> On Fri, Feb 19, 2010 at 10:55:08AM +1100, David Terei wrote: > http://www.cse.unsw.edu.au/~davidt/downloads/llvm-ghc.patch I glanced over your patch and it seems you have broken LLVM's backend. I'm not a LLVM expert by no means, but if GHC's needs were implemented as a subtarget, wouldn't there be the possibility of upstream LLVM accepting the patch? If the patch didn't break the rest of LLVM then at least it should be possible for the same patched LLVM be used by GHC and e.g. clang. By the way, nice work! :) Cheers, -- Felipe. From davidterei at gmail.com Fri Feb 19 00:18:46 2010 From: davidterei at gmail.com (David Terei) Date: Thu Feb 18 23:49:29 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <20100219025316.GA18876@kira.casa> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <20100219025316.GA18876@kira.casa> Message-ID: <4B7E1F36.1080608@gmail.com> Felipe Lessa wrote: > I glanced over your patch and it seems you have broken LLVM's > backend. I'm not a LLVM expert by no means, but if GHC's needs > were implemented as a subtarget, wouldn't there be the > possibility of upstream LLVM accepting the patch? If the patch > didn't break the rest of LLVM then at least it should be possible > for the same patched LLVM be used by GHC and e.g. clang. Yeah sorry, should have included this in issues but I just forgot all about it. The LLVM patch defiantly need to be changed so that it doesn't break other llvm users like clang. As for if LLVM would accept these patches? Hard to say, I've had a draft email sitting around for a while now asking them this question but never finished it off as I wanted to gain a clearer idea of how I would need to change my patch before sending it. I'll try to look into it this weekend if time permits and send that email finally. A 2.7 release of LLVM is happening soon with the code freeze scheduled for March 7th, so the earliest any changes needed for GHC would be included would be 2.8, which should be around the end of October 2010 (LLVM has 6 month releases). From Ben.Lippmeier at anu.edu.au Fri Feb 19 00:31:33 2010 From: Ben.Lippmeier at anu.edu.au (Ben Lippmeier) Date: Fri Feb 19 00:02:17 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B7DD35C.2010408@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> Message-ID: <877FBC5F-3229-4CB4-98B6-4851516C29BB@anu.edu.au> I've built this fine on X86/Linux, and I'll check it on SPARC/Solaris after I fix the latest head breakage (and after that I'm forming the official "I hate #ifdef" club...) One question: When you set GhcWithLlvmCodeGen = YES should that enable -fllvm for the stage 1 compiler as well? I ask because I tried to use -fllvm to bootstrap another ghc-head build, but it died like this: "inplace/bin/ghc-stage1" -M -include-pkg-deps -dep-makefile libraries/dph/dph-par/dist-install/build/.depend-v-p-dyn.haskell.tmp -dep-suffix p -dep-suffix dyn -O -H64m -fllvm -package-name dph-par-0.4.0 -hide-all-packages -i -ilibraries/dph/dph-par/../dph-common .... ghc-stage1: unrecognised flags: -fllvm The bootstrapping compiler supports -fllvm, but the stage 1 compiler built with it doesn't seem to... The mk/build.mk file I'm using says: GhcLibWays = v SRC_HC_OPTS = -O -H64m -fllvm GhcStage1HcOpts = -O GhcStage2HcOpts = -O2 GhcHcOpts = -Rghc-timing GhcLibHcOpts = -O2 -XGenerics GhcLibWays += p ifeq "$(PlatformSupportsSharedLibs)" "YES" GhcLibWays += dyn endif GhcWithLlvmCodeGen = YES GhcEnableTablesNextToCode = NO Also, a minor point on commenting the code: With something like: -- | Blocks consist of -- * label: The code label for this block -- * stmts: A list of LlvmStatement's representing the code for this block. -- This list must end with a control flow. A return, tail call or -- branch to another LlvmBlock within the current function scope. data LlvmBlock = LlvmBlock { blockLabel :: LlvmBlockId, blockStmts :: [LlvmStatement] } You should use haddock style comments to describe the field directly, eg -- | A block of LLVM code. data LlvmBlock = LlvmBlock { -- | The code label for this block blockLabel :: LlvmBlockId, -- | A list of LlvmStatement's representing the code for this block. -- This list must end with a control flow. A return, tail call or -- branch to another LlvmBlock within the current function scope. blockStmts :: [LlvmStatement] } This can be fixed later though, so I don't think it should hold up the commit. Ben.  From ghcbuild at microsoft.com Fri Feb 19 01:30:13 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Fri Feb 19 01:30:15 2010 Subject: [nightly] 18-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100219063013.EFBB1324035@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Thu Feb 18 18:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100217 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Fri Feb 19 06:59:32 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Fri Feb 19 00:08:52 GMT 2010 2458 total tests, which gave rise to 13565 test cases, of which 0 caused framework failures 2799 were skipped 10391 expected passes 350 expected failures 0 unexpected passes 25 unexpected failures Unexpected failures: 3429(threaded1,profthreaded) CPUTime001(threaded2) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) concprog001(ghci) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) qq007(profc,profasm) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Fri Feb 19 06:59:32 GMT 2010 From ghcbuild at microsoft.com Fri Feb 19 01:39:03 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Fri Feb 19 01:39:03 2010 Subject: [nightly] 18-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100219063903.33311324035@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Thu Feb 18 18:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100217 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (1 failures) **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Fri Feb 19 07:08:21 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Thu Feb 18 23:39:44 GMT 2010 2369 total tests, which gave rise to 13116 test cases, of which 0 caused framework failures 2775 were skipped 9997 expected passes 317 expected failures 0 unexpected passes 27 unexpected failures Unexpected failures: 2592(profc,profasm) 3429(profthreaded) CPUTime001(threaded2) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) ffi002(profthreaded) joao-circular(profc) outofmem2(normal) signals002(profthreaded) signals004(threaded1) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Fri Feb 19 07:08:21 GMT 2010 From kr.angelov at gmail.com Fri Feb 19 08:30:59 2010 From: kr.angelov at gmail.com (Krasimir Angelov) Date: Fri Feb 19 08:01:40 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B7DD35C.2010408@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> Message-ID: I am just curious. What had to be changed in LLVM? Could it be useful for other projects using LLVM? On Fri, Feb 19, 2010 at 12:55 AM, David Terei wrote: > Hi all, > > Over the last 6 months I've been working on a new code generator for GHC > which targets the LLVM compiler infrastructure. Most of the work was done as > part of an honours thesis at the University of New South Wales under the > supervision of Manuel Chakravarty. This ended at the start of November but I > have continued to work on the code generator since (although at a much > reduced pace since I'm back at full time work). Its now at a stage where I > feel pretty confident in its correctness and would love to see it merged > into GHC mainline. > > The patch for the llvm back-end can be found here (should apply cleanly to > GHC head): > > http://www.cse.unsw.edu.au/~davidt/downloads/ghc-llvmbackend-full.gz > > This is what I would like/am requesting merged into GHC head. > > The thesis paper which offers a detailed performance evaluation, as well as > the motivation and design of the back-end can be found at: > > http://www.cse.unsw.edu.au/~pls/thesis/davidt-thesis.pdf > > Below I'll quickly detail out the important points though. There are also > instructions on how to get started with the back-end. > > Finally there are also some issues that I think may need to be sorted out > before a merge could be done. They are at the end. > > Performance > ----------- > (All done on linux/x86-32) > > A quick summary of the results are that for the 'nofib' benchmark suite, the > llvm code generator was 3.8% slower than the NCG (the C code generator was > 6.9% slower than the NCG). The DPH project includes a benchmark suite which > I also ran and for this type of code using the llvm back-end shortened the > runtime by an average of 25% compared to the NCG. Also, while not included > in my thesis paper as I ran out of time, I did do some benchmarking with the > 'nobench' benchmark suite. It gave performance ratios for the back-ends of > around: > > NCG : 1.11 > C : 1.05 > LLVM : 1.14 > > > Supported Platforms & 'Correctness' > ----------------------------------- > > Linux x86-32/x86-64 are currently well supported. The back-end can pass the > test suite and build a working version of GHC (bootstrap test). > > Mac OS X 10.5 currently has a rather nasty bug with any dynamic lib calls > (all libffi stuff) [due to the stack not being 16byte aligned when the calls > are made as required by OSX ABI for the curious]. Test suite passes except > for most the ffi tests. > > Other platforms haven't been tested at all. As using the back-end with a > registered build of GHC requires a modified version of LLVM, people wanting > to try it out on those platforms will need to either make the needed changes > to LLVM themselves, or use an unregistered build of GHC which will work with > a vanilla install of LLVM. (A patch for LLVM for x86 is linked to below.) > > Validate > -------- > > I've validated my GHC patch to make sure it won't break anything. This is > just compiling and running GHC normally but with the llvm back-end code > included. It doesn't actually test the llvm code generator, just makes sure > it hasn't broken the NCG or C code generator. > > Linux/x86-32: > > OVERALL SUMMARY for test run started at Do 18. Feb 11:21:48 EST 2010 > 2457 total tests, which gave rise to > 9738 test cases, of which > 0 caused framework failures > 7573 were skipped > > 2088 expected passes > 76 expected failures > 0 unexpected passes > 1 unexpected failures > > Unexpected failures: > user001(normal) > > Linux/x86-64: > > OVERALL SUMMARY for test run started at Thu 18 Feb 15:28:32 EST 2010 > 2458 total tests, which gave rise to > 9739 test cases, of which > 0 caused framework failures > 7574 were skipped > > 2087 expected passes > 77 expected failures > 0 unexpected passes > 1 unexpected failures > > Unexpected failures: > T1969(normal) > > Mac OS X 10.5/x86-32: > > OVERALL SUMMARY for test run started at Thu Feb 18 12:35:49 EST 2010 > 2458 total tests, which gave rise to > 9122 test cases, of which > 0 caused framework failures > 6959 were skipped > > 2085 expected passes > 76 expected failures > 0 unexpected passes > 2 unexpected failures > > Unexpected failures: > T1969(normal) > ffi005(optc) > > All of the test failures fail for me with a unmodified GHC head build as > well as when the llvm patch is included, so the llvm patch isn't introducing > any new failures. > > Installing > ---------- > > http://www.cse.unsw.edu.au/~davidt/downloads/ghc-llvmbackend-full.gz > > Apply the darcs patch linked above to GHC head. This will make some changes > across GHC, with the bulk of the new code ending up in 'compiler/llvmGen'. > > To build GHC you need to add two flags to build.mk, they are: > > GhcWithLlvmCodeGen = YES > GhcEnableTablesNextToCode = NO > > The llvm code generator doesn't support at this time the TABLES_NEXT_TO_CODE > optimisation due to limitations with LLVM. > > You will also need LLVM installed on your computer to use the back-end. > Version 2.6 or SVN trunk is supported. If you want to use the back-end in an > unregistered ghc build, then you can use a vanilla build of LLVM. However if > you want to use a registered ghc build (very likely) then you need to patch > LLVM for this to work. The patch for llvm can be found here: > > http://www.cse.unsw.edu.au/~davidt/downloads/llvm-ghc.patch > > LLVM is very easy to build and install. It can be done as follows: > > $ svn co http://llvm.org/svn/llvm-project/llvm/trunk llvm > $ cd llvm > $ patch -p0 -i ~/llvm-ghc.patch > $ ./configure --enable-optimized # probably also want to set --prefix > $ make > $ make install > > Just make sure this modified version of LLVM is on your path and takes > precedence over any other builds. > > Using > ----- > > Once GHC is built, you can trigger GHC to use the LLVM back-end with the > '-fllvm' flag. There is also a new '-ddump-llvm' which will dump out the > llvm IR code generated. (or use the '-keep-tmp-files' flag). > > 'ghc --info' should also now report that it includes the llvm code > generator. > > Issues > ------ > Issues that might need to be resolved before merging the patch: > > 1. Developed in isolation by 1 person with no Haskell knowledge at first. So > usual issues with that may apply, misused data structures, bad style... ect. > Criticisms of the code are very welcome. There are some specific notes on > what I think may be wrong with the code atm in 'compiler/llvmGen/NOTES'. > > 2. The back-end has a LLVM binding of sorts, this binding is similar in > design to say the Cmm representation used in GHC. It represents the LLVM > Assembly language using a collection of data types and can pretty print it > out correctly. This binding lives in the 'compiler/llvmGen/Llvm' folder. > Should this binding be split out into a separate library? > > 3. As mentioned above, LLVM needs to be patched to work with a registered > build of GHC. If the llvm back-end was merged, how would this be handled? I > would suggest simply carrying the patch with some instructions on how to use > it in the GHC repo. People using GHC head could be expected to grab the LLVM > source code and apply the patch themselves at this stage. > > 4. Finally this email is long. I need to put all this info into a web page > as well. > > Cheers, > David > > _______________________________________________ > Cvs-ghc mailing list > Cvs-ghc@haskell.org > http://www.haskell.org/mailman/listinfo/cvs-ghc > From igloo at earth.li Fri Feb 19 18:42:56 2010 From: igloo at earth.li (Ian Lynagh) Date: Fri Feb 19 18:13:38 2010 Subject: patch applied (ghc): Use the shared C wrapper code in ghci.c too Message-ID: <20100219234256.GA25605@abbot.galois.com> Thu Feb 18 09:17:16 PST 2010 Ian Lynagh * Use the shared C wrapper code in ghci.c too M ./driver/ghci/ghc.mk -1 +2 M ./driver/ghci/ghci.c -158 +18 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100218171716-3fd76-8f0e3b19c7427196c30e0a9bbdea19ebfe127c96.gz From igloo at earth.li Fri Feb 19 18:42:55 2010 From: igloo at earth.li (Ian Lynagh) Date: Fri Feb 19 18:13:40 2010 Subject: patch applied (ghc): Refactor gcc.c, pulling out the reusable code Message-ID: <20100219234255.GA25583@abbot.galois.com> Thu Feb 18 09:09:31 PST 2010 Ian Lynagh * Refactor gcc.c, pulling out the reusable code M ./configure.ac -1 +1 M ./driver/gcc/gcc.c -81 +8 A ./driver/utils/cwrapper.c A ./driver/utils/cwrapper.h View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100218170931-3fd76-80743f6fced887f31e70d93ae2ca803b19fc4f52.gz From igloo at earth.li Fri Feb 19 18:42:58 2010 From: igloo at earth.li (Ian Lynagh) Date: Fri Feb 19 18:13:41 2010 Subject: patch applied (ghc): Fix build on Windows Message-ID: <20100219234258.GA25625@abbot.galois.com> Fri Feb 19 05:42:22 PST 2010 Ian Lynagh * Fix build on Windows M ./driver/utils/cwrapper.c -2 +1 M ./driver/utils/cwrapper.h -1 +1 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100219134222-3fd76-41d0046472006f377b4c5bb6cd8cbb7daa96acf4.gz From ghcbuild at microsoft.com Fri Feb 19 20:10:59 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Fri Feb 19 20:11:01 2010 Subject: [nightly] 19-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100220011059.19BD9324250@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Fri Feb 19 19:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100217 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Sat Feb 20 01:40:19 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Fri Feb 19 21:21:24 GMT 2010 2369 total tests, which gave rise to 13116 test cases, of which 0 caused framework failures 2760 were skipped 10031 expected passes 315 expected failures 0 unexpected passes 10 unexpected failures Unexpected failures: 2592(profc,profasm) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) concprog002(threaded2) drvfail006(normal) drvfail008(normal) outofmem2(normal) ---------------------------------------------------- Nightly run ended at Sat Feb 20 01:40:20 GMT 2010 From ghcbuild at microsoft.com Fri Feb 19 20:12:14 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Fri Feb 19 20:12:16 2010 Subject: [nightly] 19-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100220011214.9A7C5324250@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Fri Feb 19 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100217 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... Received disconnect from 128.36.229.215: 2: Corrupted MAC on input. lost connection failed. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Sat Feb 20 01:41:34 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Fri Feb 19 21:26:51 GMT 2010 2458 total tests, which gave rise to 13565 test cases, of which 0 caused framework failures 2792 were skipped 10416 expected passes 348 expected failures 0 unexpected passes 9 unexpected failures Unexpected failures: T1969(normal) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) qq007(profc,profasm) ---------------------------------------------------- Nightly run ended at Sat Feb 20 01:41:34 GMT 2010 From ghcbuild at microsoft.com Sat Feb 20 01:44:30 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sat Feb 20 01:44:31 2010 Subject: [nightly] 19-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100220064430.3722032418C@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Fri Feb 19 18:00:02 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100217 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Sat Feb 20 07:13:51 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sat Feb 20 00:08:57 GMT 2010 2458 total tests, which gave rise to 13565 test cases, of which 0 caused framework failures 2799 were skipped 10391 expected passes 350 expected failures 0 unexpected passes 25 unexpected failures Unexpected failures: CPUTime001(threaded2) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) concprog001(ghci) ffi002(profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) qq007(profc,profasm) signals002(threaded1) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Sat Feb 20 07:13:51 GMT 2010 From ghcbuild at microsoft.com Sat Feb 20 02:12:28 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sat Feb 20 02:12:29 2010 Subject: [nightly] 19-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100220071228.4061D32418C@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Fri Feb 19 18:10:02 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100217 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (1 failures) **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Sat Feb 20 07:41:49 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Fri Feb 19 23:38:59 GMT 2010 2369 total tests, which gave rise to 13116 test cases, of which 0 caused framework failures 2775 were skipped 9994 expected passes 317 expected failures 0 unexpected passes 30 unexpected failures Unexpected failures: 2592(profc,profasm) 3429(threaded1,profthreaded) CPUTime001(threaded2) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) ffi002(threaded1,profthreaded) hGetLine001(threaded2) joao-circular(profc) outofmem2(normal) signals002(profthreaded) signals004(threaded1) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Sat Feb 20 07:41:49 GMT 2010 From igloo at earth.li Sat Feb 20 08:24:09 2010 From: igloo at earth.li (Ian Lynagh) Date: Sat Feb 20 07:54:49 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <55A9DA74-C9B8-4988-A91E-2A150C48F150@cse.unsw.edu.au> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <55A9DA74-C9B8-4988-A91E-2A150C48F150@cse.unsw.edu.au> Message-ID: <20100220132409.GA20095@matrix.chaos.earth.li> On Fri, Feb 19, 2010 at 01:25:58PM +1100, Manuel M T Chakravarty wrote: > David wrote, > > > 2. The back-end has a LLVM binding of sorts, this binding is similar in design to say the Cmm representation used in GHC. It represents the LLVM Assembly language using a collection of data types and can pretty print it out correctly. This binding lives in the 'compiler/llvmGen/Llvm' folder. Should this binding be split out into a separate library? > > That library would be a build requirement for GHC, then. It may be useful to separate it out for other projects that want to use it, but for GHC, I don't think it would help anything, so I wouldn't worry about that until somebody actually wants to use it for something else. I would say that if the LLVM binding is currently easy to split out then it would be worth doing so. Otherwise it'll probably end up getting entangled with GHC-specific code. Also, I suspect someone else wanting an LLVM binding is more likely to make their own than to try to separate out GHC's. > > 3. As mentioned above, LLVM needs to be patched to work with a registered build of GHC. If the llvm back-end was merged, how would this be handled? I would suggest simply carrying the patch with some instructions on how to use it in the GHC repo. People using GHC head could be expected to grab the LLVM source code and apply the patch themselves at this stage. > > I think, we should bundle LLVM with GHC and auto apply the patch at build boot time That sounds like it would be a pain to maintain when new LLVM releases come out. Is this patch something that LLVM upstream are likely to accept? Thanks Ian From igloo at earth.li Sat Feb 20 08:38:26 2010 From: igloo at earth.li (Ian Lynagh) Date: Sat Feb 20 08:09:06 2010 Subject: {darcs,hackage,cvs}.haskell.org, monk.galois.com Message-ID: <20100220133826.GA20833@matrix.chaos.earth.li> Hi all, monk is just days away from being turned off, so if you want to get anything off of it before that happens, you need to do it now. More information here: http://www.haskell.org/pipermail/haskell/2010-January/021861.html Thanks Ian From dmp at rice.edu Sat Feb 20 12:17:09 2010 From: dmp at rice.edu (David Peixotto) Date: Sat Feb 20 11:47:48 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B7DD35C.2010408@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> Message-ID: <57C4205A-F3F1-4F36-8B91-9D14CB77ED65@rice.edu> Hi David, This looks like great work! Thanks for making this contribution available. I followed your instructions and was able to get the llvm backend up and running. Just a few comments. 1) It would be nice if -ddump-llvm implies -fllvm. My first test was to try -ddump-llvm and it produced no output (without the -fllvm flag). 2) Do you expect the llvm backend to work with the -dynamic ghc flag? I was not able to get it to work on my system (linux x86_64). It looks like it is having some linking issues. A sample of the linking errors when compiling bernoulli from nofib is shown below. [dave@haskell bernouilli]$ ~/local/ghc-llvm/bin/ghc -fforce-recomp -fllvm -dynamic Main.hs /usr/bin/ld: dynamic variable `stg_INTLIKE_closure' is zero size /usr/bin/ld: dynamic variable `base_GHCziList_zzip_closure' is zero size /usr/bin/ld: dynamic variable `base_GHCziBase_zgzgze_entry' is zero size /usr/bin/ld: dynamic variable `base_GHCziNum_fromInteger_entry' is zero size /usr/bin/ld: dynamic variable `stg_ap_ppp_fast' is zero size /usr/bin/ld: dynamic variable `__stginit_base_Prelude_dyn' is zero size ...... /usr/bin/ld: Main.o(.text+0x97): unresolvable R_X86_64_32S relocation against symbol `stg_CAF_BLACKHOLE_info' /usr/bin/ld: Main.o(.text+0xc4): unresolvable R_X86_64_32S relocation against symbol `stg_IND_STATIC_info' /usr/bin/ld: Main.o(.text+0xd1): unresolvable R_X86_64_32S relocation against symbol `stg_upd_frame_info' /usr/bin/ld: Main.o(.text+0xec): unresolvable R_X86_64_32S relocation against symbol `base_GHCziReal_zdfIntegralInt_closure' /usr/bin/ld: Main.o(.text+0x267): unresolvable R_X86_64_32S relocation against symbol `stg_CAF_BLACKHOLE_info' ...... 3) I started a page on the GHC wiki for the LLVM backend that copies the relevant info from your initial email. Hope this small head start makes it easier for everyone to add more documentation in the future. You can find it at: http://hackage.haskell.org/trac/ghc/wiki/Commentary/Compiler/Backends/LLVM Thanks again! -David On Feb 18, 2010, at 5:55 PM, David Terei wrote: > Hi all, > > Over the last 6 months I've been working on a new code generator for GHC which targets the LLVM compiler infrastructure. Most of the work was done as part of an honours thesis at the University of New South Wales under the supervision of Manuel Chakravarty. This ended at the start of November but I have continued to work on the code generator since (although at a much reduced pace since I'm back at full time work). Its now at a stage where I feel pretty confident in its correctness and would love to see it merged into GHC mainline. > > The patch for the llvm back-end can be found here (should apply cleanly to GHC head): > > http://www.cse.unsw.edu.au/~davidt/downloads/ghc-llvmbackend-full.gz > > This is what I would like/am requesting merged into GHC head. > > The thesis paper which offers a detailed performance evaluation, as well as the motivation and design of the back-end can be found at: > > http://www.cse.unsw.edu.au/~pls/thesis/davidt-thesis.pdf > > Below I'll quickly detail out the important points though. There are also instructions on how to get started with the back-end. > > Finally there are also some issues that I think may need to be sorted out before a merge could be done. They are at the end. > > Performance > ----------- > (All done on linux/x86-32) > > A quick summary of the results are that for the 'nofib' benchmark suite, the llvm code generator was 3.8% slower than the NCG (the C code generator was 6.9% slower than the NCG). The DPH project includes a benchmark suite which I also ran and for this type of code using the llvm back-end shortened the runtime by an average of 25% compared to the NCG. Also, while not included in my thesis paper as I ran out of time, I did do some benchmarking with the 'nobench' benchmark suite. It gave performance ratios for the back-ends of around: > > NCG : 1.11 > C : 1.05 > LLVM : 1.14 > > > Supported Platforms & 'Correctness' > ----------------------------------- > > Linux x86-32/x86-64 are currently well supported. The back-end can pass the test suite and build a working version of GHC (bootstrap test). > > Mac OS X 10.5 currently has a rather nasty bug with any dynamic lib calls (all libffi stuff) [due to the stack not being 16byte aligned when the calls are made as required by OSX ABI for the curious]. Test suite passes except for most the ffi tests. > > Other platforms haven't been tested at all. As using the back-end with a registered build of GHC requires a modified version of LLVM, people wanting to try it out on those platforms will need to either make the needed changes to LLVM themselves, or use an unregistered build of GHC which will work with a vanilla install of LLVM. (A patch for LLVM for x86 is linked to below.) > > Validate > -------- > > I've validated my GHC patch to make sure it won't break anything. This is just compiling and running GHC normally but with the llvm back-end code included. It doesn't actually test the llvm code generator, just makes sure it hasn't broken the NCG or C code generator. > > Linux/x86-32: > > OVERALL SUMMARY for test run started at Do 18. Feb 11:21:48 EST 2010 > 2457 total tests, which gave rise to > 9738 test cases, of which > 0 caused framework failures > 7573 were skipped > > 2088 expected passes > 76 expected failures > 0 unexpected passes > 1 unexpected failures > > Unexpected failures: > user001(normal) > > Linux/x86-64: > > OVERALL SUMMARY for test run started at Thu 18 Feb 15:28:32 EST 2010 > 2458 total tests, which gave rise to > 9739 test cases, of which > 0 caused framework failures > 7574 were skipped > > 2087 expected passes > 77 expected failures > 0 unexpected passes > 1 unexpected failures > > Unexpected failures: > T1969(normal) > > Mac OS X 10.5/x86-32: > > OVERALL SUMMARY for test run started at Thu Feb 18 12:35:49 EST 2010 > 2458 total tests, which gave rise to > 9122 test cases, of which > 0 caused framework failures > 6959 were skipped > > 2085 expected passes > 76 expected failures > 0 unexpected passes > 2 unexpected failures > > Unexpected failures: > T1969(normal) > ffi005(optc) > > All of the test failures fail for me with a unmodified GHC head build as well as when the llvm patch is included, so the llvm patch isn't introducing any new failures. > > Installing > ---------- > > http://www.cse.unsw.edu.au/~davidt/downloads/ghc-llvmbackend-full.gz > > Apply the darcs patch linked above to GHC head. This will make some changes across GHC, with the bulk of the new code ending up in 'compiler/llvmGen'. > > To build GHC you need to add two flags to build.mk, they are: > > GhcWithLlvmCodeGen = YES > GhcEnableTablesNextToCode = NO > > The llvm code generator doesn't support at this time the TABLES_NEXT_TO_CODE optimisation due to limitations with LLVM. > > You will also need LLVM installed on your computer to use the back-end. Version 2.6 or SVN trunk is supported. If you want to use the back-end in an unregistered ghc build, then you can use a vanilla build of LLVM. However if you want to use a registered ghc build (very likely) then you need to patch LLVM for this to work. The patch for llvm can be found here: > > http://www.cse.unsw.edu.au/~davidt/downloads/llvm-ghc.patch > > LLVM is very easy to build and install. It can be done as follows: > > $ svn co http://llvm.org/svn/llvm-project/llvm/trunk llvm > $ cd llvm > $ patch -p0 -i ~/llvm-ghc.patch > $ ./configure --enable-optimized # probably also want to set --prefix > $ make > $ make install > > Just make sure this modified version of LLVM is on your path and takes precedence over any other builds. > > Using > ----- > > Once GHC is built, you can trigger GHC to use the LLVM back-end with the '-fllvm' flag. There is also a new '-ddump-llvm' which will dump out the llvm IR code generated. (or use the '-keep-tmp-files' flag). > > 'ghc --info' should also now report that it includes the llvm code generator. > > Issues > ------ > Issues that might need to be resolved before merging the patch: > > 1. Developed in isolation by 1 person with no Haskell knowledge at first. So usual issues with that may apply, misused data structures, bad style... ect. Criticisms of the code are very welcome. There are some specific notes on what I think may be wrong with the code atm in 'compiler/llvmGen/NOTES'. > > 2. The back-end has a LLVM binding of sorts, this binding is similar in design to say the Cmm representation used in GHC. It represents the LLVM Assembly language using a collection of data types and can pretty print it out correctly. This binding lives in the 'compiler/llvmGen/Llvm' folder. Should this binding be split out into a separate library? > > 3. As mentioned above, LLVM needs to be patched to work with a registered build of GHC. If the llvm back-end was merged, how would this be handled? I would suggest simply carrying the patch with some instructions on how to use it in the GHC repo. People using GHC head could be expected to grab the LLVM source code and apply the patch themselves at this stage. > > 4. Finally this email is long. I need to put all this info into a web page as well. > > Cheers, > David > > _______________________________________________ > Cvs-ghc mailing list > Cvs-ghc@haskell.org > http://www.haskell.org/mailman/listinfo/cvs-ghc > From igloo at earth.li Sat Feb 20 17:02:56 2010 From: igloo at earth.li (Ian Lynagh) Date: Sat Feb 20 16:33:39 2010 Subject: patch applied (ghc): Check for suitable versions of make in bindists; fixes trac #3860 Message-ID: <20100220220256.GA6194@abbot.galois.com> Sat Feb 20 13:11:57 PST 2010 Ian Lynagh * Check for suitable versions of make in bindists; fixes trac #3860 We already had a check in configure of the actual build. This patch adds it to the configure of the bindists too. M ./distrib/configure.ac.in +19 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100220211157-3fd76-5c4e0cc474c71c3dd92b4edea5124318c70f44ef.gz From igloo at earth.li Sat Feb 20 17:38:02 2010 From: igloo at earth.li (Ian Lynagh) Date: Sat Feb 20 17:08:42 2010 Subject: patch applied (ghc-6.12/ghc): Check for suitable versions of make in bindists; fixes trac #3860 Message-ID: <20100220223802.GA7594@abbot.galois.com> Sat Feb 20 13:11:57 PST 2010 Ian Lynagh * Check for suitable versions of make in bindists; fixes trac #3860 We already had a check in configure of the actual build. This patch adds it to the configure of the bindists too. M ./distrib/configure.ac.in +19 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100220211157-3fd76-5c4e0cc474c71c3dd92b4edea5124318c70f44ef.gz From ghcbuild at microsoft.com Sat Feb 20 19:45:18 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sat Feb 20 19:45:20 2010 Subject: [nightly] 20-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100221004518.DF9BE32400E@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Sat Feb 20 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100219 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... failed. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Sun Feb 21 01:14:42 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sat Feb 20 21:27:13 GMT 2010 2458 total tests, which gave rise to 13565 test cases, of which 0 caused framework failures 2792 were skipped 10416 expected passes 348 expected failures 0 unexpected passes 9 unexpected failures Unexpected failures: T1969(normal) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) qq007(profc,profasm) ---------------------------------------------------- Nightly run ended at Sun Feb 21 01:14:42 GMT 2010 From ghcbuild at microsoft.com Sat Feb 20 19:58:30 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sat Feb 20 19:58:32 2010 Subject: [nightly] 20-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100221005830.4D25D32400E@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Sat Feb 20 19:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100217 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Sun Feb 21 01:27:54 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sat Feb 20 21:21:47 GMT 2010 2369 total tests, which gave rise to 13116 test cases, of which 0 caused framework failures 2760 were skipped 10030 expected passes 315 expected failures 0 unexpected passes 11 unexpected failures Unexpected failures: 2592(profc,profasm) T2701(normal) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci,threaded2) drvfail006(normal) drvfail008(normal) outofmem2(normal) ---------------------------------------------------- Nightly run ended at Sun Feb 21 01:27:54 GMT 2010 From ghcbuild at microsoft.com Sun Feb 21 01:28:49 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sun Feb 21 01:28:51 2010 Subject: [nightly] 20-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100221062849.ADD67324223@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Sat Feb 20 18:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100219 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... Received disconnect from 128.36.229.215: 2: Corrupted MAC on input. lost connection failed. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Sun Feb 21 06:58:14 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sun Feb 21 00:11:22 GMT 2010 2458 total tests, which gave rise to 13565 test cases, of which 0 caused framework failures 2799 were skipped 10389 expected passes 350 expected failures 0 unexpected passes 27 unexpected failures Unexpected failures: 3429(threaded1,profthreaded) CPUTime001(threaded2) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) concprog001(ghci) ffi002(profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) qq007(profc,profasm) signals002(threaded2) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Sun Feb 21 06:58:14 GMT 2010 From ghcbuild at microsoft.com Sun Feb 21 02:19:17 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sun Feb 21 02:19:18 2010 Subject: [nightly] 20-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100221071917.8C5FB32412A@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Sat Feb 20 18:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100217 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (1 failures) **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Sun Feb 21 07:48:42 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sat Feb 20 23:41:05 GMT 2010 2369 total tests, which gave rise to 13116 test cases, of which 0 caused framework failures 2775 were skipped 9994 expected passes 317 expected failures 0 unexpected passes 30 unexpected failures Unexpected failures: 2452(ghci) 2592(profc,profasm) 3429(profthreaded) CPUTime001(threaded2) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) ffi002(threaded1,threaded2,profthreaded) joao-circular(profc) outofmem2(normal) signals002(threaded1,profthreaded) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Sun Feb 21 07:48:42 GMT 2010 From malcolm.wallace at cs.york.ac.uk Sun Feb 21 08:09:43 2010 From: malcolm.wallace at cs.york.ac.uk (Malcolm Wallace) Date: Sun Feb 21 07:40:18 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <20100220132409.GA20095@matrix.chaos.earth.li> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <55A9DA74-C9B8-4988-A91E-2A150C48F150@cse.unsw.edu.au> <20100220132409.GA20095@matrix.chaos.earth.li> Message-ID: <5C21E849-BA2C-42DC-97B2-13F8DF07F356@cs.york.ac.uk> >>> 2. The back-end has a LLVM binding of sorts, ... Should this >>> binding be split out into a separate library? There is already a published Haskell binding for LLVM on Hackage. Does your new binding cover the same territory? Or different aspects? Could it be converted to use the existing binding? Or could your features be merged with theirs? Regards, Malcolm From igloo at earth.li Sun Feb 21 12:28:57 2010 From: igloo at earth.li (Ian Lynagh) Date: Sun Feb 21 11:59:33 2010 Subject: patch applied (ghc): Remove the old hstags util Message-ID: <20100221172857.GA15611@abbot.galois.com> Sat Feb 20 15:59:09 PST 2010 Ian Lynagh * Remove the old hstags util R ./utils/hstags/ R ./utils/hstags/Makefile R ./utils/hstags/README R ./utils/hstags/hstags-help.c R ./utils/hstags/hstags.prl R ./utils/hstags/prefix.txt View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100220235909-3fd76-b1d052423bd687b4c18b5407c70e2f3ffeefd7e3.gz From igloo at earth.li Sun Feb 21 12:28:58 2010 From: igloo at earth.li (Ian Lynagh) Date: Sun Feb 21 11:59:34 2010 Subject: patch applied (ghc): Make "make tags" work in the new build system Message-ID: <20100221172858.GA15637@abbot.galois.com> Sun Feb 21 08:44:32 PST 2010 Ian Lynagh * Make "make tags" work in the new build system ./utils/ghctags/GhcTags.hs -> ./utils/ghctags/Main.hs M ./compiler/ghc.mk +3 M ./ghc.mk +6 A ./rules/tags-package.mk M ./utils/ghctags/Main.hs -49 +39 R ./utils/ghctags/Makefile View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100221164432-3fd76-28766847d8803cdc5899feeea4fa24cbb8bd5ca5.gz From igloo at earth.li Sun Feb 21 13:54:32 2010 From: igloo at earth.li (Ian Lynagh) Date: Sun Feb 21 13:25:16 2010 Subject: patch applied (ghc): Add utils/ghctags/ghc.mk Message-ID: <20100221185432.GA18717@abbot.galois.com> Sun Feb 21 10:52:00 PST 2010 Ian Lynagh * Add utils/ghctags/ghc.mk A ./utils/ghctags/ghc.mk View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100221185200-3fd76-5a640efdb119536d350e253f6ca6cd5e12d76615.gz From ghcbuild at microsoft.com Sun Feb 21 13:35:04 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sun Feb 21 13:35:07 2010 Subject: [nightly] 21-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100221183504.5F3EE324621@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Sun Feb 21 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... failed; relevant barfage is below. GHC Version 6.13.20100221 **** building source distribution ... failed. No successfully built compilers to test; aborting. ------------------------------------------------------------------------ ------------------------------------------------------------------------ The last 30 lines of /64playpen/simonmar/nightly/HEAD-cam-04-unx/logs/x86_64-unknown-linux-stage1 are ------------------------------------------------------------------------ ------------------------------------------------------------------------ dph-prim-seq-0.4.0: file Data/Array/Parallel/Unlifted/Sequential/Segmented.hi is missing (ignoring) dph-prim-seq-0.4.0: cannot find libHSdph-prim-seq-0.4.0.a on library path (ignoring) "inplace/bin/ghc-cabal" configure --with-ghc="/64playpen/simonmar/nightly/HEAD-cam-04-unx/x86_64-unknown-linux/inplace/bin/dummy-ghc" --with-ghc-pkg="/64playpen/simonmar/nightly/HEAD-cam-04-unx/x86_64-unknown-linux/inplace/bin/ghc-pkg" --with-gcc="/usr/bin/gcc" --configure-option=--with-cc="/usr/bin/gcc" --enable-library-profiling --enable-shared --with-hscolour="/home/simonmar/fp/bin/x86_64-unknown-linux/HsColour" --configure-option=CFLAGS=" " --configure-option=LDFLAGS=" " -- dist-install libraries/dph/dph-prim-par Configuring dph-prim-par-0.4.0... "/64playpen/simonmar/nightly/HEAD-cam-04-unx/x86_64-unknown-linux/inplace/bin/ghc-pkg" update --force libraries/dph/dph-prim-par/dist-install/inplace-pkg-config Reading package info from "libraries/dph/dph-prim-par/dist-install/inplace-pkg-config" ... done. dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted/Distributed.hi is missing (ignoring) dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted/Parallel.hi is missing (ignoring) dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted.hi is missing (ignoring) dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted/Gabi.hi is missing (ignoring) dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted/Distributed/Gang.hi is missing (ignoring) dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted/Distributed/TheGang.hi is missing (ignoring) dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted/Distributed/DistST.hi is missing (ignoring) dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted/Distributed/Types.hi is missing (ignoring) dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted/Distributed/Combinators.hi is missing (ignoring) dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted/Distributed/Scalars.hi is missing (ignoring) dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted/Distributed/Arrays.hi is missing (ignoring) dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted/Distributed/Basics.hi is missing (ignoring) dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted/Parallel/Combinators.hi is missing (ignoring) dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted/Parallel/Sums.hi is missing (ignoring) dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted/Parallel/Basics.hi is missing (ignoring) dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted/Parallel/Permute.hi is missing (ignoring) dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted/Parallel/Enum.hi is missing (ignoring) dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted/Parallel/Segmented.hi is missing (ignoring) dph-prim-par-0.4.0: file Data/Array/Parallel/Unlifted/Parallel/Subarrays.hi is missing (ignoring) dph-prim-par-0.4.0: cannot find libHSdph-prim-par-0.4.0.a on library path (ignoring) "inplace/bin/ghc-cabal" configure --with-ghc="/64playpen/simonmar/nightly/HEAD-cam-04-unx/x86_64-unknown-linux/inplace/bin/dummy-ghc" --with-ghc-pkg="/64playpen/simonmar/nightly/HEAD-cam-04-unx/x86_64-unknown-linux/inplace/bin/ghc-pkg" --with-gcc="/usr/bin/gcc" --configure-option=--with-cc="/usr/bin/gcc" --enable-library-profiling --enable-shared --with-hscolour="/home/simonmar/fp/bin/x86_64-unknown-linux/HsColour" --configure-option=CFLAGS=" " --configure-option=LDFLAGS=" " -- dist utils/compare_sizes Configuring compareSizes-0.1.0.0... gmake[1]: *** No rule to make target `utils/ghctags/ghc.mk'. Stop. gmake: *** [inplace/bin/ghc-stage1] Error 2 Nightly run ended at Sun Feb 21 19:04:27 GMT 2010 From ghcbuild at microsoft.com Sun Feb 21 19:24:02 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sun Feb 21 19:24:04 2010 Subject: [nightly] 21-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100222002402.BCD463241AC@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Sun Feb 21 19:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100220 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Mon Feb 22 00:53:29 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sun Feb 21 21:12:20 GMT 2010 2369 total tests, which gave rise to 13116 test cases, of which 0 caused framework failures 2760 were skipped 10032 expected passes 315 expected failures 0 unexpected passes 9 unexpected failures Unexpected failures: 2592(profc,profasm) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) outofmem2(normal) ---------------------------------------------------- Nightly run ended at Mon Feb 22 00:53:29 GMT 2010 From ghcbuild at microsoft.com Mon Feb 22 01:46:10 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Mon Feb 22 01:46:13 2010 Subject: [nightly] 21-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100222064610.CA969324572@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Sun Feb 21 18:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100220 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Mon Feb 22 07:15:34 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Mon Feb 22 00:10:48 GMT 2010 2458 total tests, which gave rise to 13565 test cases, of which 0 caused framework failures 2799 were skipped 10391 expected passes 350 expected failures 0 unexpected passes 25 unexpected failures Unexpected failures: 3429(profthreaded) CPUTime001(threaded2) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) concprog001(ghci) concprog003(ghci) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) qq007(profc,profasm) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Mon Feb 22 07:15:34 GMT 2010 From ghcbuild at microsoft.com Mon Feb 22 02:06:27 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Mon Feb 22 02:06:28 2010 Subject: [nightly] 21-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100222070627.087DE3240C0@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Sun Feb 21 18:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100220 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (1 failures) **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Mon Feb 22 07:35:53 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sun Feb 21 23:40:21 GMT 2010 2369 total tests, which gave rise to 13116 test cases, of which 0 caused framework failures 2775 were skipped 9996 expected passes 317 expected failures 0 unexpected passes 28 unexpected failures Unexpected failures: 2592(profc,profasm) 3429(threaded1,profthreaded) CPUTime001(threaded2) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) ffi002(threaded1) joao-circular(profc) outofmem2(normal) signals002(profthreaded) signals004(threaded2) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Mon Feb 22 07:35:53 GMT 2010 From ben.lippmeier at anu.edu.au Mon Feb 22 02:55:54 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Mon Feb 22 02:26:27 2010 Subject: patch applied (ghc): Fix #3875: Crash in parallel GC, wrong pointer was being tested. Message-ID: <20100222075554.GA12155@abbot.galois.com> Sun Feb 21 19:16:27 PST 2010 Ben.Lippmeier@anu.edu.au * Fix #3875: Crash in parallel GC, wrong pointer was being tested. M ./rts/sm/Evac.c -1 +1 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100222031627-43c66-fe592a4ab3ace65dce5e30a9091b57d7db668667.gz From marlowsd at gmail.com Mon Feb 22 06:59:59 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Mon Feb 22 06:30:56 2010 Subject: patch applied (ghc): Fix #3875: Crash in parallel GC, wrong pointer was being tested. In-Reply-To: <20100222075554.GA12155@abbot.galois.com> References: <20100222075554.GA12155@abbot.galois.com> Message-ID: <4B8271BF.9010305@gmail.com> On 22/02/2010 07:55, Ben Lippmeier wrote: > Sun Feb 21 19:16:27 PST 2010 Ben.Lippmeier@anu.edu.au > * Fix #3875: Crash in parallel GC, wrong pointer was being tested. > > M ./rts/sm/Evac.c -1 +1 Good catch! That must have been a tricky one to find. Cheers, Simon From davidterei at gmail.com Mon Feb 22 07:07:12 2010 From: davidterei at gmail.com (David Terei) Date: Mon Feb 22 06:38:05 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <877FBC5F-3229-4CB4-98B6-4851516C29BB@anu.edu.au> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <877FBC5F-3229-4CB4-98B6-4851516C29BB@anu.edu.au> Message-ID: <150b8f421002220407l5173add4i9b18cf5c7c25a91b@mail.gmail.com> On 19 February 2010 16:31, Ben Lippmeier wrote: ]> One question: > When you set GhcWithLlvmCodeGen = YES > should that enable -fllvm for the stage 1 compiler as well? Yes. > "inplace/bin/ghc-stage1" -M -include-pkg-deps -dep-makefile libraries/dph/dph-par/dist-install/build/.depend-v-p-dyn.haskell.tmp -dep-suffix p -dep-suffix dyn -O -H64m -fllvm -package-name dph-par-0.4.0 -hide-all-packages -i -ilibraries/dph/dph-par/../dph-common .... > ghc-stage1: unrecognised flags: -fllvm > I'm not sure how this would happen if I understood your setup correctly. When doing development work I generally do development work with simply a stage1 build (e.g cd compiler; make 1), only doing a full build every once in a while for validating and benchmarking. Are you sure that 'GhcWithLlvmCodeGen = YES' is in the build.mk of the ghc being built? I use a similar build.mk when building Ghc with the LLVM back-end so should be fine. From marlowsd at gmail.com Mon Feb 22 07:34:06 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Mon Feb 22 07:04:58 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B7DD35C.2010408@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> Message-ID: <4B8279BE.6000909@gmail.com> First of all, congrats to David on getting this to the point where it can be merged. Nice work! It's clear we want this in GHC, so let's look at what needs to be done, and the potential sticking points. We should make LLVM optional for the time being, since it entails turning off TABLES_NEXT_TO_CODE which impacts performance of the -fasm backend. This is the main sticking point I believe: we have to resolve this question one way or the other. Either we stop using TABLES_NEXT_TO_CODE altogether, or we add support to for it to LLVM. My preference would be the latter, especially since we're already patching LLVM, but then I know very little about LLVM or whether they'd be likely to accept patches to do TNTC. If the LLVM backend did TNTC, then it would probably consistently beat the NCG (albeit with longer compile times). I'm currently running some benchmarks to see how much impact turning off TNTC has on the -fasm backend. For the time being we can pull LLVM into ghc-tarballs as Manuel suggested, and modify GHC to use the local LLVM. This is fine for experimentation, but not as a long-term solution. It will cause problems for distros like Debian which have a policy that disallows bundled copies of libs, so the long-term strategy should be to remove the bundled LLVM, which means that any changes we make to LLVM should be potentially acceptable upstream. I presume the calling-convention changes can be made in an acceptable way, by defining a new calling convention rather than modifying the exisitng C one? Does LLVM allow new calling conventions to be defined dynamically, without patchig LLVM itself? Cheers, Simon On 18/02/2010 23:55, David Terei wrote: > Hi all, > > Over the last 6 months I've been working on a new code generator for GHC > which targets the LLVM compiler infrastructure. Most of the work was > done as part of an honours thesis at the University of New South Wales > under the supervision of Manuel Chakravarty. This ended at the start of > November but I have continued to work on the code generator since > (although at a much reduced pace since I'm back at full time work). Its > now at a stage where I feel pretty confident in its correctness and > would love to see it merged into GHC mainline. > > The patch for the llvm back-end can be found here (should apply cleanly > to GHC head): > > http://www.cse.unsw.edu.au/~davidt/downloads/ghc-llvmbackend-full.gz > > This is what I would like/am requesting merged into GHC head. > > The thesis paper which offers a detailed performance evaluation, as well > as the motivation and design of the back-end can be found at: > > http://www.cse.unsw.edu.au/~pls/thesis/davidt-thesis.pdf > > Below I'll quickly detail out the important points though. There are > also instructions on how to get started with the back-end. > > Finally there are also some issues that I think may need to be sorted > out before a merge could be done. They are at the end. > > Performance > ----------- > (All done on linux/x86-32) > > A quick summary of the results are that for the 'nofib' benchmark suite, > the llvm code generator was 3.8% slower than the NCG (the C code > generator was 6.9% slower than the NCG). The DPH project includes a > benchmark suite which I also ran and for this type of code using the > llvm back-end shortened the runtime by an average of 25% compared to the > NCG. Also, while not included in my thesis paper as I ran out of time, I > did do some benchmarking with the 'nobench' benchmark suite. It gave > performance ratios for the back-ends of around: > > NCG : 1.11 > C : 1.05 > LLVM : 1.14 > > > Supported Platforms & 'Correctness' > ----------------------------------- > > Linux x86-32/x86-64 are currently well supported. The back-end can pass > the test suite and build a working version of GHC (bootstrap test). > > Mac OS X 10.5 currently has a rather nasty bug with any dynamic lib > calls (all libffi stuff) [due to the stack not being 16byte aligned when > the calls are made as required by OSX ABI for the curious]. Test suite > passes except for most the ffi tests. > > Other platforms haven't been tested at all. As using the back-end with a > registered build of GHC requires a modified version of LLVM, people > wanting to try it out on those platforms will need to either make the > needed changes to LLVM themselves, or use an unregistered build of GHC > which will work with a vanilla install of LLVM. (A patch for LLVM for > x86 is linked to below.) > > Validate > -------- > > I've validated my GHC patch to make sure it won't break anything. This > is just compiling and running GHC normally but with the llvm back-end > code included. It doesn't actually test the llvm code generator, just > makes sure it hasn't broken the NCG or C code generator. > > Linux/x86-32: > > OVERALL SUMMARY for test run started at Do 18. Feb 11:21:48 EST 2010 > 2457 total tests, which gave rise to > 9738 test cases, of which > 0 caused framework failures > 7573 were skipped > > 2088 expected passes > 76 expected failures > 0 unexpected passes > 1 unexpected failures > > Unexpected failures: > user001(normal) > > Linux/x86-64: > > OVERALL SUMMARY for test run started at Thu 18 Feb 15:28:32 EST 2010 > 2458 total tests, which gave rise to > 9739 test cases, of which > 0 caused framework failures > 7574 were skipped > > 2087 expected passes > 77 expected failures > 0 unexpected passes > 1 unexpected failures > > Unexpected failures: > T1969(normal) > > Mac OS X 10.5/x86-32: > > OVERALL SUMMARY for test run started at Thu Feb 18 12:35:49 EST 2010 > 2458 total tests, which gave rise to > 9122 test cases, of which > 0 caused framework failures > 6959 were skipped > > 2085 expected passes > 76 expected failures > 0 unexpected passes > 2 unexpected failures > > Unexpected failures: > T1969(normal) > ffi005(optc) > > All of the test failures fail for me with a unmodified GHC head build as > well as when the llvm patch is included, so the llvm patch isn't > introducing any new failures. > > Installing > ---------- > > http://www.cse.unsw.edu.au/~davidt/downloads/ghc-llvmbackend-full.gz > > Apply the darcs patch linked above to GHC head. This will make some > changes across GHC, with the bulk of the new code ending up in > 'compiler/llvmGen'. > > To build GHC you need to add two flags to build.mk, they are: > > GhcWithLlvmCodeGen = YES > GhcEnableTablesNextToCode = NO > > The llvm code generator doesn't support at this time the > TABLES_NEXT_TO_CODE optimisation due to limitations with LLVM. > > You will also need LLVM installed on your computer to use the back-end. > Version 2.6 or SVN trunk is supported. If you want to use the back-end > in an unregistered ghc build, then you can use a vanilla build of LLVM. > However if you want to use a registered ghc build (very likely) then you > need to patch LLVM for this to work. The patch for llvm can be found here: > > http://www.cse.unsw.edu.au/~davidt/downloads/llvm-ghc.patch > > LLVM is very easy to build and install. It can be done as follows: > > $ svn co http://llvm.org/svn/llvm-project/llvm/trunk llvm > $ cd llvm > $ patch -p0 -i ~/llvm-ghc.patch > $ ./configure --enable-optimized # probably also want to set --prefix > $ make > $ make install > > Just make sure this modified version of LLVM is on your path and takes > precedence over any other builds. > > Using > ----- > > Once GHC is built, you can trigger GHC to use the LLVM back-end with the > '-fllvm' flag. There is also a new '-ddump-llvm' which will dump out the > llvm IR code generated. (or use the '-keep-tmp-files' flag). > > 'ghc --info' should also now report that it includes the llvm code > generator. > > Issues > ------ > Issues that might need to be resolved before merging the patch: > > 1. Developed in isolation by 1 person with no Haskell knowledge at > first. So usual issues with that may apply, misused data structures, bad > style... ect. Criticisms of the code are very welcome. There are some > specific notes on what I think may be wrong with the code atm in > 'compiler/llvmGen/NOTES'. > > 2. The back-end has a LLVM binding of sorts, this binding is similar in > design to say the Cmm representation used in GHC. It represents the LLVM > Assembly language using a collection of data types and can pretty print > it out correctly. This binding lives in the 'compiler/llvmGen/Llvm' > folder. Should this binding be split out into a separate library? > > 3. As mentioned above, LLVM needs to be patched to work with a > registered build of GHC. If the llvm back-end was merged, how would this > be handled? I would suggest simply carrying the patch with some > instructions on how to use it in the GHC repo. People using GHC head > could be expected to grab the LLVM source code and apply the patch > themselves at this stage. > > 4. Finally this email is long. I need to put all this info into a web > page as well. > > Cheers, > David > > _______________________________________________ > Cvs-ghc mailing list > Cvs-ghc@haskell.org > http://www.haskell.org/mailman/listinfo/cvs-ghc From davidterei at gmail.com Mon Feb 22 09:40:34 2010 From: davidterei at gmail.com (David Terei) Date: Mon Feb 22 09:11:26 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> Message-ID: <150b8f421002220640u7c8c8d1m5d7a01b6cfc821f3@mail.gmail.com> On 20 February 2010 00:30, Krasimir Angelov wrote: > I am just curious. What had to be changed in LLVM? Could it be useful > for other projects using LLVM? In the C and native code generator back-end GHC stores some variables permanently in registers for speed reasons due to how frequently they are accessed ('pinned registers'). The variables stored are the STG virtual registers (so pinning them turns them into real registers) which define a stack pointer, heap pointer.. ect for Haskells execution model. LLVM doesn't support reserving a register in this way for exclusive use. Instead of changing LLVM to support this feature directly, I added a new calling convention which passes the arguments in the same hardware registers that GHC normally uses. This has the advantage that it while it keeps the STG virtual registers in the correct hardware registers across call boundaries, inside a call LLVM can allocate registers in the most efficient manner (which may be simply leaving them in the registers and never spilling). It's also compatible with the other code generators. If you want more info Chris Lattner (main LLVM guy) wrote some notes up on it based on a brief email conversation the two of us had: http://www.nondot.org/sabre/LLVMNotes/GlobalRegisterVariables.txt I don't know if this would be useful for other projects, its fairly specific so probably not although I have seen other people asking about it on the LLVM mailing list though and referring to the approach that I took. From simonpj at microsoft.com Mon Feb 22 11:47:02 2010 From: simonpj at microsoft.com (Simon Peyton-Jones) Date: Mon Feb 22 11:17:54 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <57C4205A-F3F1-4F36-8B91-9D14CB77ED65@rice.edu> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <57C4205A-F3F1-4F36-8B91-9D14CB77ED65@rice.edu> Message-ID: <59543203684B2244980D7E4057D5FBC10B070D67@DB3EX14MBXC306.europe.corp.microsoft.com> | 3) I started a page on the GHC wiki for the LLVM backend that copies the relevant | info from your initial email. Hope this small head start makes it easier for | everyone to add more documentation in the future. You can find it at: | http://hackage.haskell.org/trac/ghc/wiki/Commentary/Compiler/Backends/LLVM David, thanks for doing this! But please link it from the Commentary http://hackage.haskell.org/trac/ghc/wiki/Commentary under "Contributed documentation". I started to do this for you, but in doing so I found a link to another LLVM page: http://hackage.haskell.org/trac/ghc/wiki/LlvmBackend No point in having two! Perhaps you can connect them together somehow? Simon From marlowsd at gmail.com Mon Feb 22 11:49:07 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Mon Feb 22 11:21:08 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B8279BE.6000909@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <4B8279BE.6000909@gmail.com> Message-ID: <4B82B583.5040103@gmail.com> On 22/02/2010 12:34, Simon Marlow wrote: > I'm currently running some benchmarks to see how much impact turning off > TNTC has on the -fasm backend. Here are the results on x86-64/Linux: -------------------------------------------------------------------------------- Program Size Allocs Runtime Elapsed -------------------------------------------------------------------------------- anna +8.9% +0.0% 0.16 0.17 ansi +5.9% +0.0% 0.00 0.00 atom +6.1% +0.0% +2.5% +1.5% awards +5.8% +0.0% 0.00 0.00 banner +4.7% +0.0% 0.00 0.00 bernouilli +5.8% +0.0% +0.5% +0.0% boyer +5.9% +0.0% 0.06 0.07 boyer2 +5.6% +0.0% 0.01 0.01 bspt +7.2% +0.0% 0.02 0.02 cacheprof +7.1% +0.0% +8.5% +8.6% calendar +5.9% +0.0% 0.00 0.00 cichelli +5.7% -0.0% 0.15 0.16 circsim +6.3% +0.0% +2.1% +1.7% clausify +5.9% +0.0% 0.06 0.06 comp_lab_zift +6.2% +0.0% +6.7% +6.5% compress +6.1% +0.0% +11.9% +12.9% compress2 +6.1% +0.0% +1.8% +0.0% constraints +6.2% +0.0% +2.9% +3.0% cryptarithm1 +5.3% +0.0% +12.8% +12.8% cryptarithm2 +5.6% +0.0% 0.02 0.02 cse +5.4% +0.0% 0.00 0.00 eliza +4.9% +0.0% 0.00 0.00 event +5.9% +0.0% +2.8% +1.9% exp3_8 +5.8% +0.0% +3.7% +0.0% expert +5.6% +0.0% 0.00 0.00 fem +7.3% +0.0% 0.03 0.04 fft +6.2% +0.0% 0.06 0.06 fft2 +6.5% +0.0% 0.12 0.13 fibheaps +5.9% +0.0% 0.05 0.05 fish +5.4% +0.0% 0.03 0.03 fluid +8.1% +0.0% 0.01 0.01 fulsom +7.5% +0.0% +5.9% +6.1% gamteb +6.5% -0.0% 0.10 0.10 gcd +5.8% +0.0% 0.05 0.05 gen_regexps +5.5% +0.0% 0.00 0.00 genfft +5.8% +0.0% 0.05 0.05 gg +6.9% +0.0% 0.02 0.02 grep +6.2% +0.0% 0.00 0.00 hidden +7.5% +0.0% +14.8% +13.8% hpg +7.0% +0.0% +3.3% +2.7% ida +6.0% +0.0% 0.17 0.18 infer +6.7% +0.0% 0.08 0.10 integer +5.8% +0.0% +2.8% +3.0% integrate +6.2% +0.0% +5.8% +4.6% knights +6.2% +0.0% 0.01 0.01 lcss +5.9% +0.0% +2.6% +0.3% life +5.4% +0.0% +4.9% +4.3% lift +6.2% +0.0% 0.00 0.00 listcompr +5.3% +0.0% 0.13 0.15 listcopy +5.3% +0.0% 0.14 0.16 maillist +5.6% -0.0% 0.09 0.13 mandel +6.2% +0.0% 0.10 0.11 mandel2 +5.3% +0.0% 0.01 0.01 minimax +5.8% +0.0% 0.00 0.01 mkhprog +5.8% +0.0% 0.00 0.00 multiplier +6.1% +0.0% 0.20 0.20 nucleic2 +5.6% +0.0% 0.11 0.11 para +6.1% +0.0% +7.4% +6.6% paraffins +5.9% +0.0% 0.19 +0.0% parser +6.7% +0.0% 0.05 0.05 parstof +6.4% +0.0% 0.00 0.01 pic +6.7% +0.0% 0.02 0.02 power +6.1% +0.0% +1.5% +1.8% pretty +6.1% +0.0% 0.00 0.00 primes +5.8% +0.0% 0.10 0.11 primetest +6.0% +0.0% +0.0% +0.0% prolog +5.7% +0.0% 0.00 0.00 puzzle +5.6% +0.0% +5.2% +4.2% queens +5.7% +0.0% 0.03 0.03 reptile +6.6% +0.0% 0.02 0.02 rewrite +6.3% +0.0% 0.02 0.02 rfib +6.1% +0.0% 0.03 0.03 rsa +5.9% +0.0% 0.05 0.06 scc +5.2% +0.0% 0.00 0.00 sched +5.9% +0.0% 0.03 0.03 scs +7.5% +0.0% +3.6% +3.5% simple +8.4% -0.0% +6.3% +5.4% solid +6.2% +0.0% +0.9% -1.7% sorting +5.6% +0.0% 0.00 0.00 sphere +6.7% +0.0% 0.13 0.14 symalg +7.1% +0.0% 0.02 0.03 tak +5.7% +0.0% 0.02 0.02 transform +6.6% +0.0% +5.8% +7.5% treejoin +5.1% +0.0% +0.0% +1.3% typecheck +6.0% +0.0% +16.9% +13.5% veritas +8.3% +0.0% 0.00 0.00 wang +6.2% +0.0% 0.16 0.18 wave4main +6.0% +0.0% +7.8% +7.9% wheel-sieve1 +5.8% +0.0% +2.5% +2.9% wheel-sieve2 +5.8% +0.0% -0.6% -0.5% x2n1 +6.3% +0.0% 0.03 0.03 -------------------------------------------------------------------------------- Min +4.7% -0.0% -0.6% -1.7% Max +8.9% +0.0% +16.9% +13.8% Geometric Mean +6.1% -0.0% +4.9% +4.2% So, nofib programs are on average 4-5% slower, with a 6% increase in binary sizes (this is with -split-objs). What's interesting is that there are some outliers here: programs that go 12-17% slower without TNTC. Looking at David's thesis the results are quite different. Perhaps it is processor dependent? I have to admit, the results I'm getting are what I would expect, so its conceivable that some subconscious experimental bias has crept in somewhere :-) This is an Intel Xeon E5320, 1.8GHz. I'll try to get some measurements on our 32-bit box too. Cheers, Simon From igloo at earth.li Mon Feb 22 17:23:08 2010 From: igloo at earth.li (Ian Lynagh) Date: Mon Feb 22 16:53:40 2010 Subject: patch applied (ghc-6.12/ghc): Fix #3875: Crash in parallel GC, wrong pointer was being tested. Message-ID: <20100222222307.GA21666@abbot.galois.com> Sun Feb 21 19:16:27 PST 2010 Ben.Lippmeier@anu.edu.au * Fix #3875: Crash in parallel GC, wrong pointer was being tested. M ./rts/sm/Evac.c -1 +1 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100222031627-43c66-fe592a4ab3ace65dce5e30a9091b57d7db668667.gz From marlowsd at gmail.com Mon Feb 22 17:53:45 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Mon Feb 22 17:24:19 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <59543203684B2244980D7E4057D5FBC10B070D67@DB3EX14MBXC306.europe.corp.microsoft.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <57C4205A-F3F1-4F36-8B91-9D14CB77ED65@rice.edu> <59543203684B2244980D7E4057D5FBC10B070D67@DB3EX14MBXC306.europe.corp.microsoft.com> Message-ID: <4B830AF9.6010902@gmail.com> On 22/02/10 16:47, Simon Peyton-Jones wrote: > | 3) I started a page on the GHC wiki for the LLVM backend that copies the relevant > | info from your initial email. Hope this small head start makes it easier for > | everyone to add more documentation in the future. You can find it at: > | http://hackage.haskell.org/trac/ghc/wiki/Commentary/Compiler/Backends/LLVM > > David, thanks for doing this! > > But please link it from the Commentary > http://hackage.haskell.org/trac/ghc/wiki/Commentary > under "Contributed documentation". I see it's linked from http://hackage.haskell.org/trac/ghc/wiki/Commentary/Compiler under Backends, which is the right place, I think. > I started to do this for you, but in doing so I found a link to another LLVM page: > http://hackage.haskell.org/trac/ghc/wiki/LlvmBackend > > No point in having two! Perhaps you can connect them together somehow? Agreed, there's lots of useful stuff in that page. Cheers, Simon From ghcbuild at microsoft.com Mon Feb 22 19:55:31 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Mon Feb 22 19:55:32 2010 Subject: [nightly] 22-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100223005531.B24C93242BE@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Mon Feb 22 19:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100220 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Tue Feb 23 01:25:01 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Mon Feb 22 21:20:13 GMT 2010 2369 total tests, which gave rise to 13116 test cases, of which 0 caused framework failures 2760 were skipped 10032 expected passes 315 expected failures 0 unexpected passes 9 unexpected failures Unexpected failures: 2592(profc,profasm) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) outofmem2(normal) ---------------------------------------------------- Nightly run ended at Tue Feb 23 01:25:01 GMT 2010 From ghcbuild at microsoft.com Mon Feb 22 19:59:29 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Mon Feb 22 19:59:31 2010 Subject: [nightly] 22-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100223005929.8D47332421A@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Mon Feb 22 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100222 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... failed. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Tue Feb 23 01:28:59 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Mon Feb 22 21:26:12 GMT 2010 2458 total tests, which gave rise to 13565 test cases, of which 0 caused framework failures 2792 were skipped 10416 expected passes 348 expected failures 0 unexpected passes 9 unexpected failures Unexpected failures: T1969(normal) annrun01(dyn) barton-mangler-bug(profc) concprog001(ghci) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) qq007(profc,profasm) ---------------------------------------------------- Nightly run ended at Tue Feb 23 01:28:59 GMT 2010 From dmp at rice.edu Mon Feb 22 22:14:19 2010 From: dmp at rice.edu (David Peixotto) Date: Mon Feb 22 21:44:49 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B830AF9.6010902@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <57C4205A-F3F1-4F36-8B91-9D14CB77ED65@rice.edu> <59543203684B2244980D7E4057D5FBC10B070D67@DB3EX14MBXC306.europe.corp.microsoft.com> <4B830AF9.6010902@gmail.com> Message-ID: <3E212AD2-9A71-4195-B114-95A359233596@rice.edu> On Feb 22, 2010, at 4:53 PM, Simon Marlow wrote: >> I started to do this for you, but in doing so I found a link to another LLVM page: >> http://hackage.haskell.org/trac/ghc/wiki/LlvmBackend >> >> No point in having two! Perhaps you can connect them together somehow? > > Agreed, there's lots of useful stuff in that page. > > Cheers, > Simon > Thanks for pointing out the LlvmBackend page, I had totally missed it. It looks like the page contained info about the design of the backend, but had not been updated in a while. I merged what I thought was the relevant info and updated it based what I understood from David's thesis and by browsing the code. I'm sure David could improve the description and check for any glaring omissions. I "deprecated" the LlvmBackend page in favor of Commentary/Backends/LLVM by putting a forwarding reference to the latter. Otherwise I left the current LlvmBackend page intact, but can remove it once David agrees the important info has been transferred. From ghcbuild at microsoft.com Tue Feb 23 01:47:36 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Tue Feb 23 01:47:38 2010 Subject: [nightly] 22-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100223064736.8312C3241A4@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Mon Feb 22 18:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100222 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** running nofib (-unreg) ... failed. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Tue Feb 23 07:17:07 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Tue Feb 23 00:10:57 GMT 2010 2458 total tests, which gave rise to 13565 test cases, of which 0 caused framework failures 2799 were skipped 10391 expected passes 350 expected failures 0 unexpected passes 25 unexpected failures Unexpected failures: 3429(threaded1) annrun01(dyn) arith008(profasm,profthreaded) barton-mangler-bug(profc) concprog001(ghci) ffi002(threaded1) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) qq007(profc,profasm) signals002(threaded2) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Tue Feb 23 07:17:07 GMT 2010 From ghcbuild at microsoft.com Tue Feb 23 02:11:26 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Tue Feb 23 02:11:28 2010 Subject: [nightly] 22-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100223071126.DDE1D32471F@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Mon Feb 22 18:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100220 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (1 failures) **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Tue Feb 23 07:40:57 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Mon Feb 22 23:40:10 GMT 2010 2369 total tests, which gave rise to 13116 test cases, of which 0 caused framework failures 2775 were skipped 9996 expected passes 317 expected failures 0 unexpected passes 28 unexpected failures Unexpected failures: 2592(profc,profasm) CPUTime001(threaded2) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) ffi002(threaded1,profthreaded) joao-circular(profc) outofmem2(normal) signals002(threaded1,threaded2,profthreaded) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Tue Feb 23 07:40:57 GMT 2010 From marlowsd at gmail.com Tue Feb 23 06:05:17 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Tue Feb 23 05:35:47 2010 Subject: patch applied (testsuite): omit profilng ways Message-ID: <20100223110517.GA18944@abbot.galois.com> Tue Feb 23 01:40:06 PST 2010 Simon Marlow * omit profilng ways M ./tests/ghc-regress/quasiquotation/qq007/test.T +4 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100223094006-12142-d9f954d9ea995d0b841205835615d95c92c18c20.gz From marlowsd at gmail.com Tue Feb 23 06:05:18 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Tue Feb 23 05:35:50 2010 Subject: patch applied (testsuite): omit via-C ways for this test, gcc takes too long Message-ID: <20100223110518.GA18964@abbot.galois.com> Tue Feb 23 01:42:41 PST 2010 Simon Marlow * omit via-C ways for this test, gcc takes too long M ./tests/ghc-regress/programs/barton-mangler-bug/test.T +2 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100223094241-12142-36682a70ceaada429d8c9ab7b59a6bb4efe2b29c.gz From marlowsd at gmail.com Tue Feb 23 06:05:21 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Tue Feb 23 05:35:53 2010 Subject: patch applied (testsuite): expect_broken(3676) Message-ID: <20100223110521.GA19006@abbot.galois.com> Tue Feb 23 02:59:35 PST 2010 Simon Marlow * expect_broken(3676) M ./tests/ghc-regress/numeric/should_run/all.T -1 +1 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100223105935-12142-3a61a74f8afd65480009443bd958b86cd31410ed.gz From marlowsd at gmail.com Tue Feb 23 06:05:20 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Tue Feb 23 05:35:53 2010 Subject: patch applied (testsuite): disable annrun01(dyn) (see comments) Message-ID: <20100223110520.GA18984@abbot.galois.com> Tue Feb 23 01:48:35 PST 2010 Simon Marlow * disable annrun01(dyn) (see comments) M ./tests/ghc-regress/annotations/should_run/all.T -2 +2 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100223094835-12142-d18be71be77984af07a0c27d6b8c8cdd031dff34.gz From simonpj at microsoft.com Tue Feb 23 09:31:37 2010 From: simonpj at microsoft.com (Simon Peyton-Jones) Date: Tue Feb 23 09:02:25 2010 Subject: FW: Installing syb(-0.1.03) package in head version of Haskell Message-ID: <59543203684B2244980D7E4057D5FBC10B08D018@DB3EX14MBXC312.europe.corp.microsoft.com> Friends Can anyone help Kathleen out? She has a cabal-install issue. I think she's on a Mac. Many thanks Simon From: Kathleen Fisher [mailto:kathleen.fisher@gmail.com] Sent: 16 February 2010 22:54 To: Simon Peyton-Jones Subject: Fwd: Installing syb(-0.1.03) package in head version of Haskell Hi Simon, I'm trying to get the PADS file we were working on compiling under the HEAD version of Haskell, but I'm having trouble getting the Data.Generics module to install. I sent the question below to the libraries@haskell.org last week, but I haven't seen any replies. Is that the right address? Or is there a different place to ask such questions? Thanks for any pointers! Kathleen Begin forwarded message: From: Kathleen Fisher > Date: February 11, 2010 2:28:20 PM PST To: libraries@haskell.org Subject: Installing syb(-0.1.03) package in head version of Haskell Hi all, I'm trying to get a program that uses the Data.Generics module to compile under the head version of GHC, but I am having trouble getting the syb package installed. I believe I am using the most recent version of cabal: kfisher-laptop:/Users/kfisher/pads/dirpads/language-c-quote>~/.cabal/bin/cabal --version cabal-install version 0.8.0 using version 1.8.0.1 of the Cabal library When I ask cabal to install the syb package, I get the following error: kfisher-laptop:/Users/kfisher/pads/dirpads/language-c-quote>~/.cabal/bin/cabal install -w ~/sw/ghc-head/bin/ghc Resolving dependencies... Downloading mtl-1.1.0.2... Configuring mtl-1.1.0.2... Preprocessing library mtl-1.1.0.2... ... Registering syb-0.1.0.3... cabal: ghc-pkg: : hGetContents: invalid argument (invalid UTF-8 byte sequence) cabal: Error: some packages failed to install: language-c-quote-0.1 depends on syb-0.1.0.3 which failed to install. syb-0.1.0.3 failed during the building phase. The exception was: exit: ExitFailure 1 The full trace is included below. Here's the version information for the ghc I am running. Kfisher-laptop:/Users/kfisher/pads/dirpads/language-c-quote>~/sw/ghc-head/bin/ghc --version The Glorious Glasgow Haskell Compilation System, version 6.13.20100210 The cabal website suggests sending questions to this list... Any suggestions for how to fix this problem? Kathleen kfisher-laptop:/Users/kfisher/pads/dirpads/language-c-quote>~/.cabal/bin/cabal install -w ~/sw/ghc-head/bin/ghc Resolving dependencies... Downloading mtl-1.1.0.2... Configuring mtl-1.1.0.2... Preprocessing library mtl-1.1.0.2... Building mtl-1.1.0.2... ...stuff deleted... src/Data/Generics.hs:40:1: Warning: The import of `Prelude' is redundant except perhaps to import instances from `Prelude' To import instances alone, use: import Prelude() Registering syb-0.1.0.3... cabal: ghc-pkg: : hGetContents: invalid argument (invalid UTF-8 byte sequence) cabal: Error: some packages failed to install: language-c-quote-0.1 depends on syb-0.1.0.3 which failed to install. syb-0.1.0.3 failed during the building phase. The exception was: exit: ExitFailure 1 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/cvs-ghc/attachments/20100223/8baa3216/attachment-0001.html From davidterei at gmail.com Tue Feb 23 09:38:04 2010 From: davidterei at gmail.com (David Terei) Date: Tue Feb 23 09:08:54 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <5C21E849-BA2C-42DC-97B2-13F8DF07F356@cs.york.ac.uk> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <55A9DA74-C9B8-4988-A91E-2A150C48F150@cse.unsw.edu.au> <20100220132409.GA20095@matrix.chaos.earth.li> <5C21E849-BA2C-42DC-97B2-13F8DF07F356@cs.york.ac.uk> Message-ID: <150b8f421002230638o242add99u3e9dd66e32a2f8bd@mail.gmail.com> On 22 February 2010 00:09, Malcolm Wallace wrote: > There is already a published Haskell binding for LLVM on Hackage. ?Does your > new binding cover the same territory? ?Or different aspects? ?Could it be > converted to use the existing binding? ?Or could your features be merged > with theirs? A good intro to that binding can be found at http://augustss.blogspot.com/2009/01/llvm-llvm-low-level-virtual-machine-is.html. My binding is fairly straight forward. It provides a way to represent LLVM assembly code in Haskell and the ability to write it to a file. Once in a file it can then be processed by calling the various llvm tools (e.g llvm-as, opt, llc). The other way to interact with llvm is through a C++ API that llvm provides which allows you to build up llvm code programmatically and then do a variety of things on it such as optiming, compiler (inlcuing JIT), interpretitng.. ect. Bryan and Lennart's Haskell LLVM binding works this way, providing a Haskell binding to the C++ API as well as a higher level type safe interface to it. So the two bindings cover very different territory. They can be used for the some of the same things but the approach is very different and there wouldn't really be anything to be gained from trying to merge them. From davidterei at gmail.com Tue Feb 23 09:42:29 2010 From: davidterei at gmail.com (David Terei) Date: Tue Feb 23 09:14:34 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <57C4205A-F3F1-4F36-8B91-9D14CB77ED65@rice.edu> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <57C4205A-F3F1-4F36-8B91-9D14CB77ED65@rice.edu> Message-ID: <150b8f421002230642m4889b891l7910a2ed70cac5ae@mail.gmail.com> On 21 February 2010 04:17, David Peixotto wrote: > 1) It would be nice if -ddump-llvm implies -fllvm. My first test was to try -ddump-llvm and it produced no output (without the -fllvm flag). Yep sure, thanks for the pick up. > 2) Do you expect the llvm backend to work with the -dynamic ghc flag? I was not able to get it to work on my system (linux x86_64). It looks like it is having some linking issues. A sample of the linking errors when compiling bernoulli from nofib is shown below. I have never tried to use '-dynamic' before, so this is probably an actual problem, not your mistake. Will add to list of todo. I can't check on my machine right now how it goes sorry as despite the number of builds lying around none of my builds have dynamic enabled. > 3) I started a page on the GHC wiki for the LLVM backend that copies the relevant info from your initial email. Hope this small head start makes it easier for everyone to add more documentation in the future. You can find it at: http://hackage.haskell.org/trac/ghc/wiki/Commentary/Compiler/Backends/LLVM Great thanks for this. Will have a browse over and add anything that needs adding. I've kept a personal wiki during my thesis work so I'll also see if any of that stuff should be moved across. From jpm at cs.uu.nl Tue Feb 23 10:10:05 2010 From: jpm at cs.uu.nl (=?ISO-8859-1?Q?Jos=E9_Pedro_Magalh=E3es?=) Date: Tue Feb 23 09:40:55 2010 Subject: Installing syb(-0.1.03) package in head version of Haskell In-Reply-To: <125EACD0CAE4D24ABDB4D148C4593DA911026775@GBLONXMB02.corp.amvescap.net> References: <59543203684B2244980D7E4057D5FBC10B08D018@DB3EX14MBXC312.europe.corp.microsoft.com> <125EACD0CAE4D24ABDB4D148C4593DA911026775@GBLONXMB02.corp.amvescap.net> Message-ID: <52f14b211002230710o3c701353w2ec32aeb76c41e79@mail.gmail.com> I'm guessing that is indeed the problem, though I cannot reproduce the error (but I'm using Cabal-1.8.0.2). Assuming this is the cause of the problem, what is the best correction? Encoding the cabal file as UTF-8 or removing the accent? Do earlier versions of cabal deal well with UTF-8?... Cheers, Pedro On Tue, Feb 23, 2010 at 16:05, Bayley, Alistair wrote: > Just a wild guess, but the package description has this non-ascii text: > > author: Ralf L?mmel, Simon Peyton Jones > > It could well be Latin-1 encoded, rather than UTF8. > > Try editing the author field to this: > > author: Ralf Laemmel, Simon Peyton Jones > > and rebuilding/installing. > > > -----Original Message----- > > From: glasgow-haskell-users-bounces@haskell.org > > [mailto:glasgow-haskell-users-bounces@haskell.org] On Behalf > > Of Simon Peyton-Jones > > Sent: 23 February 2010 14:32 > > To: GHC users > > Cc: cvs-ghc@haskell.org > > Subject: FW: Installing syb(-0.1.03) package in head version > > of Haskell > > > > Friends > > > > > > > > Can anyone help Kathleen out? She has a cabal-install issue. > > I think she's on a Mac. > > > > > > > > Many thanks > > > > > > Simon > > > > > > > > From: Kathleen Fisher [mailto:kathleen.fisher@gmail.com] > > Sent: 16 February 2010 22:54 > > To: Simon Peyton-Jones > > Subject: Fwd: Installing syb(-0.1.03) package in head version > > of Haskell > > > > > > > > Hi Simon, > > > > > > > > I'm trying to get the PADS file we were working on compiling > > under the HEAD version of Haskell, but I'm having trouble > > getting the Data.Generics module to install. I sent the > > question below to the libraries@haskell.org last week, but I > > haven't seen any replies. Is that the right address? Or is > > there a different place to ask such questions? > > > > > > > > Thanks for any pointers! > > > > > > > > Kathleen > > > > > > > > Begin forwarded message: > > > > > > > > > > > > From: Kathleen Fisher > > > > Date: February 11, 2010 2:28:20 PM PST > > > > To: libraries@haskell.org > > > > Subject: Installing syb(-0.1.03) package in head version of Haskell > > > > > > > > Hi all, > > > > I'm trying to get a program that uses the Data.Generics > > module to compile under the head version of GHC, but I am > > having trouble getting the syb package installed. > > > > I believe I am using the most recent version of cabal: > > > > > > > > > > kfisher-laptop:/Users/kfisher/pads/dirpads/language-c-quote>~/ > > .cabal/bin/cabal --version > > > > cabal-install version 0.8.0 > > > > using version 1.8.0.1 of the Cabal library > > > > > > When I ask cabal to install the syb package, I get the > > following error: > > > > > > > > > > kfisher-laptop:/Users/kfisher/pads/dirpads/language-c-quote>~/ > > .cabal/bin/cabal install -w ~/sw/ghc-head/bin/ghc > > > > Resolving dependencies... > > > > Downloading mtl-1.1.0.2... > > > > Configuring mtl-1.1.0.2... > > > > Preprocessing library mtl-1.1.0.2... > > > > ... > > > > Registering syb-0.1.0.3... > > > > cabal: ghc-pkg: : hGetContents: invalid argument > > (invalid UTF-8 byte > > > > sequence) > > > > cabal: Error: some packages failed to install: > > > > language-c-quote-0.1 depends on syb-0.1.0.3 which > > failed to install. > > > > syb-0.1.0.3 failed during the building phase. The exception was: > > > > exit: ExitFailure 1 > > > > > > > > The full trace is included below. > > > > Here's the version information for the ghc I am running. > > > > > > > > > > Kfisher-laptop:/Users/kfisher/pads/dirpads/language-c-quote>~/ > > sw/ghc-head/bin/ghc --version > > > > The Glorious Glasgow Haskell Compilation System, > > version 6.13.20100210 > > > > > > The cabal website suggests sending questions to this list... > > > > Any suggestions for how to fix this problem? > > > > Kathleen > > > > > > > > > > > > kfisher-laptop:/Users/kfisher/pads/dirpads/language-c-quote>~/ > > .cabal/bin/cabal install -w ~/sw/ghc-head/bin/ghc > > > > Resolving dependencies... > > > > Downloading mtl-1.1.0.2... > > > > Configuring mtl-1.1.0.2... > > > > Preprocessing library mtl-1.1.0.2... > > > > Building mtl-1.1.0.2... > > > > > > > > ...stuff deleted... > > > > > > > > src/Data/Generics.hs:40:1: > > > > Warning: The import of `Prelude' is redundant > > > > except perhaps to import instances from `Prelude' > > > > To import instances alone, use: import Prelude() > > > > Registering syb-0.1.0.3... > > > > cabal: ghc-pkg: : hGetContents: invalid argument > > (invalid UTF-8 byte > > > > sequence) > > > > cabal: Error: some packages failed to install: > > > > language-c-quote-0.1 depends on syb-0.1.0.3 which > > failed to install. > > > > syb-0.1.0.3 failed during the building phase. The exception was: > > > > exit: ExitFailure 1 > > > > > > > > > ***************************************************************** > Confidentiality Note: The information contained in this message, > and any attachments, may contain confidential and/or privileged > material. It is intended solely for the person(s) or entity to > which it is addressed. Any review, retransmission, dissemination, > or taking of any action in reliance upon this information by > persons or entities other than the intended recipient(s) is > prohibited. If you received this in error, please contact the > sender and delete the material from any computer. > ***************************************************************** > > _______________________________________________ > Glasgow-haskell-users mailing list > Glasgow-haskell-users@haskell.org > http://www.haskell.org/mailman/listinfo/glasgow-haskell-users > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/cvs-ghc/attachments/20100223/cc000eb8/attachment.html From simonpj at microsoft.com Tue Feb 23 10:15:53 2010 From: simonpj at microsoft.com (Simon Peyton-Jones) Date: Tue Feb 23 09:46:36 2010 Subject: Extra packages in GHC In-Reply-To: <20100217143003.GA20265@abbot.galois.com> References: <20100217143003.GA20265@abbot.galois.com> Message-ID: <59543203684B2244980D7E4057D5FBC10B08E30F@DB3EX14MBXC312.europe.corp.microsoft.com> Folks, esp DPH folk and Ian, | Wed Feb 17 02:31:34 PST 2010 Simon Marlow | * Allow extra packages to be added to the build, by setting $(EXTRA_PACKAGES) | | M ./ghc.mk +2 | | Wed Feb 17 02:46:34 PST 2010 Simon Marlow | * darcs-all --extra get: gets a few extra packages | currently deepseq, parallel, stm. Simon and I had a chat this afternoon. Here's a summary: * Simon added these 'extra-packages' stuff to help automate the process of downloading and building packages - that are not boot packages - but which we want to test regularly * DPH is another such package. It's not really a boot package, in the sense that (a) it's not necessary to build GHC and (b) it can only be built with a stage-2 compiler. The main reason it's treated as a boot package today is so that new patches are validated against it; it is a particularly good stress test. But it seems a bit overkill for it to masquerade as a boot package for that reason alone. * The current extra-packages story is a bit of a mess: - The extra packages are listed in $(TOP)/packages - But they are, quite separately listed in a make variable EXTRA_PACKAGES, and then stuffed into PACKAGES by 'addPackage' in $(TOP)/ghc.mk. - A series of addPackage calls in ghc.mk duplicates the list of boot packages in $(TOP)/packages. - Moreover, the list in EXTRA_PACKAGES must be in dependency order. We propose the following plan * Make the list in $(TOP)/packages the authoritative list. * Make 'sh boot' construct the list $(PACKAGES) from $(TOP)/packages and stuff it somewhere appropriate. ['sh boot' has to iterate over the packages anyhow, to boot them, but doesn't currently consult $(TOP)/packages to do so; it looks a the file system, which is less good.] * Make DPH into an extra package * And/or add a tag 'validated' to mean "this is an extra package, not a boot package, but it should be validated before pushing patches". That doesn't necessarily apply to all extra packages (eg stm). * The list in $(TOP)/packages is really a list of *repositories* not a list of *packages*. For example, utils/hcs2hs isn't a package. We probably want a tag ("util"?) to distinguish non-packages from packages. * DPH is unusual because it is a *repository* that contains several *packages*. We can tell this because it has a file ghc-packages in its root directory, that lists the packages it contains. 'sh boot' must understand this when constructing $(PACKAGES); indeed it already does understand this. * Packages must be built in dependency order, and the build system has no way of figuring out the right order. But it would be easy to stipulate that $(TOP)/packages (and the ghc-packages file, if any, in the repo) should list the packages in dependency order. We don't need to execute this plan right away, but would be a very nice clean-up if anyone felt up to doing something about it. Simon From simonpj at microsoft.com Tue Feb 23 10:18:55 2010 From: simonpj at microsoft.com (Simon Peyton-Jones) Date: Tue Feb 23 09:50:05 2010 Subject: Installing syb(-0.1.03) package in head version of Haskell In-Reply-To: <52f14b211002230710o3c701353w2ec32aeb76c41e79@mail.gmail.com> References: <59543203684B2244980D7E4057D5FBC10B08D018@DB3EX14MBXC312.europe.corp.microsoft.com> <125EACD0CAE4D24ABDB4D148C4593DA911026775@GBLONXMB02.corp.amvescap.net> <52f14b211002230710o3c701353w2ec32aeb76c41e79@mail.gmail.com> Message-ID: <59543203684B2244980D7E4057D5FBC10B08E351@DB3EX14MBXC312.europe.corp.microsoft.com> Thanks! I forgot to cc Kathleen herself, so I'm now doing so. Assuming that's the problem, does it reflect a bug in Cabal, or GHC, or what? And if so, can someone make a ticket?] Thanks Simon From: Jos? Pedro Magalh?es [mailto:jpm@cs.uu.nl] Sent: 23 February 2010 15:10 To: Bayley, Alistair Cc: Simon Peyton-Jones; GHC users; cvs-ghc@haskell.org Subject: Re: Installing syb(-0.1.03) package in head version of Haskell I'm guessing that is indeed the problem, though I cannot reproduce the error (but I'm using Cabal-1.8.0.2). Assuming this is the cause of the problem, what is the best correction? Encoding the cabal file as UTF-8 or removing the accent? Do earlier versions of cabal deal well with UTF-8?... Cheers, Pedro On Tue, Feb 23, 2010 at 16:05, Bayley, Alistair > wrote: Just a wild guess, but the package description has this non-ascii text: author: Ralf L?mmel, Simon Peyton Jones It could well be Latin-1 encoded, rather than UTF8. Try editing the author field to this: author: Ralf Laemmel, Simon Peyton Jones and rebuilding/installing. > -----Original Message----- > From: glasgow-haskell-users-bounces@haskell.org > [mailto:glasgow-haskell-users-bounces@haskell.org] On Behalf > Of Simon Peyton-Jones > Sent: 23 February 2010 14:32 > To: GHC users > Cc: cvs-ghc@haskell.org > Subject: FW: Installing syb(-0.1.03) package in head version > of Haskell > > Friends > > > > Can anyone help Kathleen out? She has a cabal-install issue. > I think she's on a Mac. > > > > Many thanks > > > Simon > > > > From: Kathleen Fisher [mailto:kathleen.fisher@gmail.com] > Sent: 16 February 2010 22:54 > To: Simon Peyton-Jones > Subject: Fwd: Installing syb(-0.1.03) package in head version > of Haskell > > > > Hi Simon, > > > > I'm trying to get the PADS file we were working on compiling > under the HEAD version of Haskell, but I'm having trouble > getting the Data.Generics module to install. I sent the > question below to the libraries@haskell.org last week, but I > haven't seen any replies. Is that the right address? Or is > there a different place to ask such questions? > > > > Thanks for any pointers! > > > > Kathleen > > > > Begin forwarded message: > > > > > > From: Kathleen Fisher > > > Date: February 11, 2010 2:28:20 PM PST > > To: libraries@haskell.org > > Subject: Installing syb(-0.1.03) package in head version of Haskell > > > > Hi all, > > I'm trying to get a program that uses the Data.Generics > module to compile under the head version of GHC, but I am > having trouble getting the syb package installed. > > I believe I am using the most recent version of cabal: > > > > > kfisher-laptop:/Users/kfisher/pads/dirpads/language-c-quote>~/ > .cabal/bin/cabal --version > > cabal-install version 0.8.0 > > using version 1.8.0.1 of the Cabal library > > > When I ask cabal to install the syb package, I get the > following error: > > > > > kfisher-laptop:/Users/kfisher/pads/dirpads/language-c-quote>~/ > .cabal/bin/cabal install -w ~/sw/ghc-head/bin/ghc > > Resolving dependencies... > > Downloading mtl-1.1.0.2... > > Configuring mtl-1.1.0.2... > > Preprocessing library mtl-1.1.0.2... > > ... > > Registering syb-0.1.0.3... > > cabal: ghc-pkg: : hGetContents: invalid argument > (invalid UTF-8 byte > > sequence) > > cabal: Error: some packages failed to install: > > language-c-quote-0.1 depends on syb-0.1.0.3 which > failed to install. > > syb-0.1.0.3 failed during the building phase. The exception was: > > exit: ExitFailure 1 > > > > The full trace is included below. > > Here's the version information for the ghc I am running. > > > > > Kfisher-laptop:/Users/kfisher/pads/dirpads/language-c-quote>~/ > sw/ghc-head/bin/ghc --version > > The Glorious Glasgow Haskell Compilation System, > version 6.13.20100210 > > > The cabal website suggests sending questions to this list... > > Any suggestions for how to fix this problem? > > Kathleen > > > > > > kfisher-laptop:/Users/kfisher/pads/dirpads/language-c-quote>~/ > .cabal/bin/cabal install -w ~/sw/ghc-head/bin/ghc > > Resolving dependencies... > > Downloading mtl-1.1.0.2... > > Configuring mtl-1.1.0.2... > > Preprocessing library mtl-1.1.0.2... > > Building mtl-1.1.0.2... > > > > ...stuff deleted... > > > > src/Data/Generics.hs:40:1: > > Warning: The import of `Prelude' is redundant > > except perhaps to import instances from `Prelude' > > To import instances alone, use: import Prelude() > > Registering syb-0.1.0.3... > > cabal: ghc-pkg: : hGetContents: invalid argument > (invalid UTF-8 byte > > sequence) > > cabal: Error: some packages failed to install: > > language-c-quote-0.1 depends on syb-0.1.0.3 which > failed to install. > > syb-0.1.0.3 failed during the building phase. The exception was: > > exit: ExitFailure 1 > > > > ***************************************************************** Confidentiality Note: The information contained in this message, and any attachments, may contain confidential and/or privileged material. It is intended solely for the person(s) or entity to which it is addressed. Any review, retransmission, dissemination, or taking of any action in reliance upon this information by persons or entities other than the intended recipient(s) is prohibited. If you received this in error, please contact the sender and delete the material from any computer. ***************************************************************** _______________________________________________ Glasgow-haskell-users mailing list Glasgow-haskell-users@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-users -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/cvs-ghc/attachments/20100223/98563988/attachment-0001.html From marlowsd at gmail.com Tue Feb 23 10:21:55 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Tue Feb 23 09:52:41 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B7DD35C.2010408@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> Message-ID: <4B83F293.6020807@gmail.com> On 18/02/2010 23:55, David Terei wrote: > Hi all, > > Over the last 6 months I've been working on a new code generator for GHC > which targets the LLVM compiler infrastructure. Most of the work was > done as part of an honours thesis at the University of New South Wales > under the supervision of Manuel Chakravarty. This ended at the start of > November but I have continued to work on the code generator since > (although at a much reduced pace since I'm back at full time work). Its > now at a stage where I feel pretty confident in its correctness and > would love to see it merged into GHC mainline. > > The patch for the llvm back-end can be found here (should apply cleanly > to GHC head): > > http://www.cse.unsw.edu.au/~davidt/downloads/ghc-llvmbackend-full.gz Here's a patch review. I think one round of feedback should be enough: most of my suggestions are cleanups or optimisations, there are very few XXX's. On the whole the patch looks fine. I haven't done a deep review of the code - there's quite a lot of it, and I'm not an LLVM expert, so this is a high-level review. In fact, I'm surprised at how much there is here. The LLVM backend has more in common with the native code generators than the C backend, but I was expecting something closer to a straightforward pretty-printer. I wonder if having a complete datatype for LLVM is really buying us anything since we don't actually do anything to it other than print it out. As a quick example of what I mean, what would be lost if we replace data LlvmCastOp = LM_Trunc | LM_Zext ... lots of constructors ... deriving (Eq) instance Show LlvmCastOp where show LM_Trunc = "trunc" show LM_Zext = "zext" ... with type LlvmCastOp = FastString lM_Trunc = fsLit "trunc" ... ? Do we envisage doing anything complex to the Llvm code inside GHC in the future? I can't think of a reason to. Some of the cleanups and optimisations are quite significant. Highlights: * I think the module LlvmCodeGen.Regs can mostly go away. * There are some obvious representation changes that would improve performance (String -> FastString, [a] -> OrdList a). Review key: XXX This needs to be addressed before the patch goes in CLEAN A potential cleanup (some of these are quite major) OPT A possible optimisation Patch review follows: hunk ./compiler/ghc.cabal.in 44 +Flag llvm + Description: Build the LLVM code generator. + Default: False + Manual: True CLEAN Is there a good reason to want to disable the LLVM support, or can it always be compiled in? +Flag unreg + Description: Build an unregistered version. + Default: False + Manual: True + XXX Is this necessary? As far as I'm aware an unregisterised build currently works. hunk ./compiler/ghc.cabal.in 101 + if flag(llvm) + CPP-Options: -DLLVM + if flag(unreg) + CPP-Options: -DNO_REGS -DUSE_MINIINTERPRETER XXX Why is this necessary? These CPP flags are added by GHC itself (StaticFlagParser.unregFlags). +type LMString = String OPT You probably want to use something better than String, perhaps FastString. +-- | Llvm variables +data LlvmVar + -- references to variables with a global scope. + = LMGlobalVar LMString LlvmType LlvmLinkageType + -- references to variables local for a function or parameters. + | LMLocalVar LMString LlvmType + -- a constant variable + | LMLitVar LlvmLit OPT Rather than LMString in LMLocalVar, probably Unique would be better. I think you always generate these from a Unique (but I could be wrong) in which case it would be better to do that conversion when printing. I'm not sure whether this applies to LMGlobalVar or not. + deriving (Eq) +fixAssignsTop :: RawCmmTop -> UniqSM RawCmmTop + +fixAssignsTop top@(CmmData _ _) = returnUs top + +fixAssignsTop (CmmProc info lbl params (ListGraph blocks)) = + mapUs fixAssignsBlock blocks `thenUs` \ blocks' -> + returnUs (CmmProc info lbl params (ListGraph blocks')) + CLEAN I think your local fixAssigns can go away (see comment with LlvmCodeGen.Regs below): use the one from the NCG. +cmmToCmm :: RawCmmTop -> RawCmmTop +cmmToCmm top@(CmmData _ _) = top +cmmToCmm (CmmProc info lbl params (ListGraph blocks)) = + let blocks' = map cmmBlockConFold (cmmMiniInline blocks) + blocks'' = map cmmAddReturn blocks' + in CmmProc info lbl params (ListGraph $ blocks'') + CLEAN I think your local cmmToCmm can go away (see comment with LlvmCodeGen.Regs below): use the one from the NCG. We should also look into not doing any optimisation at the C-- level at all, since LLVM will do it all, however we do need to turn GlobalRegs into BaseReg offsets where necessary, which is done by cmmToCmm. +type LlvmEnvMap = Map.Map LMString LlvmType OPT A Map of Strings is very inefficient compared with a UniqFM of FastStrings. +type StmtData = (LlvmEnv, [LlvmStatement], [LlvmCmmTop]) OPT Use OrdList rather than lists here, as you're doing a lot of appending. Append is O(1) in OrdList, O(n) with lists. +type ExprData = (LlvmEnv, LlvmVar, LlvmStatements, [LlvmCmmTop]) OPT Again, LlvmStatements should probably be an OrdList. addfile ./compiler/llvmGen/LlvmCodeGen/Regs.hs hunk ./compiler/llvmGen/LlvmCodeGen/Regs.hs 1 +-- ---------------------------------------------------------------------------- +-- Deal with Cmm registers +-- +-- Specifically, force use of register table (in memory), not pinned +-- hardware registers as the LLVM back-end doesn't support this. +-- + +module LlvmCodeGen.Regs ( + + RealReg, realRegsOrdered, getRealRegReg, getRealRegArg, getGlobalRegAddr + + ) where CLEAN Unless I'm mistaken, most of this module can go away. 1. Use fixAssigns from the NCG. 2. Print out GlobalRegs as "stg_whatever..." Then you don't need to bake in the register mapping (a maintenance headache), and a lot of this code can go away: the RealReg type and everything using it. addfile ./compiler/llvmGen/NOTES hunk ./compiler/llvmGen/NOTES 1 +LLVM Back-end for GHC. + +Author: David Terei CLEAN Useful-looking docs. I suggest moving them into either (a) the wiki commentary or (b) the code as appropriate. hunk ./compiler/main/CodeOutput.lhs 16 +#ifdef LLVM +#ifdef OMIT_NATIVE_CODEGEN +import UniqSupply ( mkSplitUniqSupply ) +#endif +import qualified LlvmCodeGen ( llvmCodeGen ) +#endif + hunk ./compiler/main/CodeOutput.lhs 90 + HscLlvm -> outputLlvm dflags filenm flat_abstractC; hunk ./compiler/main/CodeOutput.lhs 176 +%************************************************************************ +%* * +\subsection{LLVM} +%* * +%************************************************************************ + +\begin{code} +outputLlvm :: DynFlags -> FilePath -> [RawCmm] -> IO () + +#ifdef LLVM + +outputLlvm dflags filenm flat_absC + = do ncg_uniqs <- mkSplitUniqSupply 'n' + doOutput filenm $ \f -> + LlvmCodeGen.llvmCodeGen dflags f ncg_uniqs flat_absC + +#else /* !LLVM */ CLEAN suggest removing the #ifdefs and enabling LLVM unconditionally, as above. From igloo at earth.li Tue Feb 23 10:22:49 2010 From: igloo at earth.li (Ian Lynagh) Date: Tue Feb 23 09:53:18 2010 Subject: patch applied (ghc-6.12/ghc): don't constant fold division that would result in negative zero (#3676) Message-ID: <20100223152249.GA29461@abbot.galois.com> Thu Feb 11 05:15:43 PST 2010 Simon Marlow * don't constant fold division that would result in negative zero (#3676) M ./compiler/prelude/PrelRules.lhs -2 +10 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100211131543-12142-81c0bae72afd112800fc0a4aea8ede80b6dc77ab.gz From igloo at earth.li Tue Feb 23 12:23:29 2010 From: igloo at earth.li (Ian Lynagh) Date: Tue Feb 23 11:53:59 2010 Subject: patch applied (ghc): Fix the links to the base docs from the user guide Message-ID: <20100223172329.GA2035@abbot.galois.com> Tue Feb 23 07:05:22 PST 2010 Ian Lynagh * Fix the links to the base docs from the user guide ./docs/users_guide/ug-ent.xml -> ./docs/users_guide/ug-ent.xml.in M ./aclocal.m4 +8 M ./configure.ac -1 +3 M ./docs/users_guide/ffi-chap.xml -2 +2 M ./docs/users_guide/glasgow_exts.xml -5 +5 M ./docs/users_guide/parallel.xml -1 +1 M ./docs/users_guide/ug-ent.xml.in +1 M ./docs/users_guide/using.xml -1 +1 M ./docs/users_guide/win32-dlls.xml -1 +1 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100223150522-3fd76-9cd53098fb1952ff73ead1b895b3d7a643e8d3b5.gz From igloo at earth.li Tue Feb 23 12:23:27 2010 From: igloo at earth.li (Ian Lynagh) Date: Tue Feb 23 11:54:00 2010 Subject: patch applied (ghc): Remove old, unused release notes Message-ID: <20100223172327.GA2012@abbot.galois.com> Tue Feb 23 07:03:27 PST 2010 Ian Lynagh * Remove old, unused release notes R ./docs/users_guide/5-00-notes.xml R ./docs/users_guide/5-02-notes.xml R ./docs/users_guide/5-04-notes.xml R ./docs/users_guide/6.0-notes.xml View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100223150327-3fd76-a53c983f16c69214441a91a00b21a93fc1b30286.gz From igloo at earth.li Tue Feb 23 12:23:30 2010 From: igloo at earth.li (Ian Lynagh) Date: Tue Feb 23 11:54:03 2010 Subject: patch applied (ghc): Fix more library links in the user guide Message-ID: <20100223172330.GA2058@abbot.galois.com> Tue Feb 23 08:45:51 PST 2010 Ian Lynagh * Fix more library links in the user guide M ./aclocal.m4 -3 +4 M ./configure.ac +2 M ./docs/users_guide/glasgow_exts.xml -3 +3 M ./docs/users_guide/packages.xml -3 +3 M ./docs/users_guide/parallel.xml -7 +8 M ./docs/users_guide/ug-ent.xml.in -1 +3 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100223164551-3fd76-e18a366d06a99c049ffdc04df69e13cc42d77c81.gz From ml at isaac.cedarswampstudios.org Tue Feb 23 12:31:20 2010 From: ml at isaac.cedarswampstudios.org (Isaac Dupree) Date: Tue Feb 23 12:01:51 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B83F293.6020807@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <4B83F293.6020807@gmail.com> Message-ID: <4B8410E8.2070000@isaac.cedarswampstudios.org> On 02/23/10 10:21, Simon Marlow wrote: > +Flag unreg > + Description: Build an unregistered version. it's called "unregisterized", is it not? (vs. "unregistered" ). Yes, "unregisterised" is not a word outside the world of GHC (and I'm not sure if we use "s" or "z") -Isaac From igloo at earth.li Tue Feb 23 13:54:42 2010 From: igloo at earth.li (Ian Lynagh) Date: Tue Feb 23 13:25:14 2010 Subject: patch applied (ghc): Fix the link to the ghc docs in libraries/prologue.txt Message-ID: <20100223185441.GA5519@abbot.galois.com> Tue Feb 23 10:16:44 PST 2010 Ian Lynagh * Fix the link to the ghc docs in libraries/prologue.txt ./libraries/prologue.txt -> ./libraries/prologue.txt.in M ./configure.ac -1 +3 M ./libraries/prologue.txt.in -1 +1 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100223181644-3fd76-fb799ce25887f5af29e5e98c697cb6058c38b363.gz From ghcbuild at microsoft.com Tue Feb 23 13:50:43 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Tue Feb 23 13:50:44 2010 Subject: [nightly] 23-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100223185043.1FF7A3241A4@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Tue Feb 23 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100223 **** Building stage 2 compiler ... failed; relevant barfage is below. **** building source distribution ... failed. **** building testsuite tools ... failed. **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (91 failures) **** running nofib (-O -fvia-C) ... ok. (91 failures) **** running nofib (-O -prof -auto-all) ... ok. (91 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (91 failures) **** running nofib (-fasm) ... ok. (91 failures) **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Tue Feb 23 19:20:15 GMT 2010 ------------------------------------------------------------------------ ------------------------------------------------------------------------ The last 30 lines of /64playpen/simonmar/nightly/HEAD-cam-04-unx/logs/x86_64-unknown-linux-stage2 are ------------------------------------------------------------------------ ------------------------------------------------------------------------ ^ /64playpen/simonmar/nightly/HEAD-cam-04-unx/fptools/docs/users_guide/ug-book.xml:18: parser error : Entity 'ffi-chap' not defined &ffi-chap; ^ /64playpen/simonmar/nightly/HEAD-cam-04-unx/fptools/docs/users_guide/ug-book.xml:19: parser error : Entity 'wrong' not defined &wrong; ^ /64playpen/simonmar/nightly/HEAD-cam-04-unx/fptools/docs/users_guide/ug-book.xml:20: parser error : Entity 'utils' not defined &utils; ^ /64playpen/simonmar/nightly/HEAD-cam-04-unx/fptools/docs/users_guide/ug-book.xml:21: parser error : Entity 'win32-dll' not defined &win32-dll; ^ /64playpen/simonmar/nightly/HEAD-cam-04-unx/fptools/docs/users_guide/ug-book.xml:22: parser error : Entity 'bugs' not defined &bugs; ^ /64playpen/simonmar/nightly/HEAD-cam-04-unx/fptools/docs/users_guide/users_guide.xml:10: parser error : Failure to process entity ug-book &ug-book; ^ /64playpen/simonmar/nightly/HEAD-cam-04-unx/fptools/docs/users_guide/users_guide.xml:10: parser error : Entity 'ug-book' not defined &ug-book; ^ unable to parse /64playpen/simonmar/nightly/HEAD-cam-04-unx/fptools/docs/users_guide/users_guide.xml Error: xsltproc failed gmake[1]: *** [docs/users_guide/users_guide.ps] Error 1 gmake: *** [all] Error 2 real 0m49.694s user 0m48.010s sys 0m1.417s Nightly run ended at Tue Feb 23 19:20:15 GMT 2010 From igloo at earth.li Tue Feb 23 14:34:53 2010 From: igloo at earth.li (Ian Lynagh) Date: Tue Feb 23 14:05:24 2010 Subject: patch applied (ghc-6.12/ghc): Fix the links to the base docs from the user guide Message-ID: <20100223193452.GA7175@abbot.galois.com> Tue Feb 23 07:05:22 PST 2010 Ian Lynagh * Fix the links to the base docs from the user guide ./docs/users_guide/ug-ent.xml -> ./docs/users_guide/ug-ent.xml.in M ./aclocal.m4 +8 M ./configure.ac -1 +3 M ./docs/users_guide/ffi-chap.xml -2 +2 M ./docs/users_guide/glasgow_exts.xml -5 +5 M ./docs/users_guide/parallel.xml -1 +1 M ./docs/users_guide/ug-ent.xml.in +1 M ./docs/users_guide/using.xml -1 +1 M ./docs/users_guide/win32-dlls.xml -1 +1 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100223150522-3fd76-9cd53098fb1952ff73ead1b895b3d7a643e8d3b5.gz From igloo at earth.li Tue Feb 23 14:34:54 2010 From: igloo at earth.li (Ian Lynagh) Date: Tue Feb 23 14:05:25 2010 Subject: patch applied (ghc-6.12/ghc): Fix more library links in the user guide Message-ID: <20100223193454.GA7198@abbot.galois.com> Tue Feb 23 08:45:51 PST 2010 Ian Lynagh * Fix more library links in the user guide M ./aclocal.m4 -3 +4 M ./configure.ac +2 M ./docs/users_guide/glasgow_exts.xml -3 +3 M ./docs/users_guide/packages.xml -3 +3 M ./docs/users_guide/parallel.xml -7 +8 M ./docs/users_guide/ug-ent.xml.in -1 +3 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100223164551-3fd76-e18a366d06a99c049ffdc04df69e13cc42d77c81.gz From igloo at earth.li Tue Feb 23 14:34:56 2010 From: igloo at earth.li (Ian Lynagh) Date: Tue Feb 23 14:05:27 2010 Subject: patch applied (ghc-6.12/ghc): Fix the link to the ghc docs in libraries/prologue.txt Message-ID: <20100223193456.GA7220@abbot.galois.com> Tue Feb 23 10:16:44 PST 2010 Ian Lynagh * Fix the link to the ghc docs in libraries/prologue.txt ./libraries/prologue.txt -> ./libraries/prologue.txt.in M ./configure.ac -1 +3 M ./libraries/prologue.txt.in -1 +1 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100223181644-3fd76-fb799ce25887f5af29e5e98c697cb6058c38b363.gz From marlowsd at gmail.com Tue Feb 23 15:35:27 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Tue Feb 23 15:06:06 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B8410E8.2070000@isaac.cedarswampstudios.org> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <4B83F293.6020807@gmail.com> <4B8410E8.2070000@isaac.cedarswampstudios.org> Message-ID: <4B843C0F.2090407@gmail.com> On 23/02/10 17:31, Isaac Dupree wrote: > On 02/23/10 10:21, Simon Marlow wrote: >> +Flag unreg >> + Description: Build an unregistered version. > > it's called > "unregisterized", is it not? (vs. > "unregistered" > ). Yes, > "unregisterised" > is not a word outside the world of GHC (and I'm not sure if we use "s" > or "z") Yes, it's called "unregisterised". Cheers, Simon From davidterei at gmail.com Tue Feb 23 19:31:20 2010 From: davidterei at gmail.com (David Terei) Date: Tue Feb 23 19:01:49 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <3E212AD2-9A71-4195-B114-95A359233596@rice.edu> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <57C4205A-F3F1-4F36-8B91-9D14CB77ED65@rice.edu> <59543203684B2244980D7E4057D5FBC10B070D67@DB3EX14MBXC306.europe.corp.microsoft.com> <4B830AF9.6010902@gmail.com> <3E212AD2-9A71-4195-B114-95A359233596@rice.edu> Message-ID: <4B847358.8070104@gmail.com> David Peixotto wrote: > I "deprecated" the LlvmBackend page in favor of Commentary/Backends/LLVM by putting a forwarding reference to the latter. Otherwise I left the current LlvmBackend page intact, but can remove it once David agrees the important info has been transferred. OK, have checked over the new page and its all fine, good to remove the old one. From ghcbuild at microsoft.com Tue Feb 23 19:34:38 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Tue Feb 23 19:34:40 2010 Subject: [nightly] 23-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100224003438.A0A0E32416F@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Tue Feb 23 19:10:02 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100222 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Wed Feb 24 01:04:11 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Tue Feb 23 21:13:23 GMT 2010 2369 total tests, which gave rise to 13116 test cases, of which 0 caused framework failures 2760 were skipped 10031 expected passes 315 expected failures 0 unexpected passes 10 unexpected failures Unexpected failures: 2592(profc,profasm) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci,threaded2) drvfail006(normal) drvfail008(normal) outofmem2(normal) ---------------------------------------------------- Nightly run ended at Wed Feb 24 01:04:11 GMT 2010 From dons at galois.com Tue Feb 23 20:38:31 2010 From: dons at galois.com (Don Stewart) Date: Tue Feb 23 20:09:00 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B847358.8070104@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <57C4205A-F3F1-4F36-8B91-9D14CB77ED65@rice.edu> <59543203684B2244980D7E4057D5FBC10B070D67@DB3EX14MBXC306.europe.corp.microsoft.com> <4B830AF9.6010902@gmail.com> <3E212AD2-9A71-4195-B114-95A359233596@rice.edu> <4B847358.8070104@gmail.com> Message-ID: <20100224013831.GA15752@whirlpool.galois.com> David, Did you see my patch to fix the parsing of -optlo and friends? http://www.galois.com/~dons/add-new-llvm-code-generator-to-ghc_.dpatch Order of the flags appears to matter to GetOpt. -- Don davidterei: > David Peixotto wrote: >> I "deprecated" the LlvmBackend page in favor of Commentary/Backends/LLVM by putting a forwarding reference to the latter. Otherwise I left the current LlvmBackend page intact, but can remove it once David agrees the important info has been transferred. > > OK, have checked over the new page and its all fine, good to remove the > old one. > > _______________________________________________ > Cvs-ghc mailing list > Cvs-ghc@haskell.org > http://www.haskell.org/mailman/listinfo/cvs-ghc From ben.lippmeier at anu.edu.au Wed Feb 24 02:18:35 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Wed Feb 24 01:49:04 2010 Subject: patch applied (testsuite): Add DPH quickhull test Message-ID: <20100224071835.GA32339@abbot.galois.com> Tue Feb 23 21:29:16 PST 2010 benl@ouroborus.net * Add DPH quickhull test A ./tests/ghc-regress/dph/ A ./tests/ghc-regress/dph/Makefile A ./tests/ghc-regress/dph/quickhull/ A ./tests/ghc-regress/dph/quickhull/Main.hs A ./tests/ghc-regress/dph/quickhull/Makefile A ./tests/ghc-regress/dph/quickhull/QuickHullVect.hs A ./tests/ghc-regress/dph/quickhull/SVG.hs A ./tests/ghc-regress/dph/quickhull/TestData.hs A ./tests/ghc-regress/dph/quickhull/Types.hs A ./tests/ghc-regress/dph/quickhull/quickhull.T View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100224052916-bf82d-82b31686cac6bd0643db080aef3851b0119aaa71.gz From ben.lippmeier at anu.edu.au Wed Feb 24 02:18:36 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Wed Feb 24 01:49:06 2010 Subject: patch applied (testsuite): Add quickhull test output Message-ID: <20100224071836.GA32359@abbot.galois.com> Tue Feb 23 23:12:33 PST 2010 benl@ouroborus.net * Add quickhull test output A ./tests/ghc-regress/dph/quickhull/quickhull.stdout View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100224071233-bf82d-28f1c11ffe98ce7870d0592d544eb008e9a33515.gz From ben.lippmeier at anu.edu.au Wed Feb 24 02:20:40 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Wed Feb 24 01:51:07 2010 Subject: patch applied (ghc): Fix build for non-x86 arches again. Message-ID: <20100224072040.GA32524@abbot.galois.com> Sun Feb 21 23:51:44 PST 2010 Ben.Lippmeier@anu.edu.au * Fix build for non-x86 arches again. M ./compiler/nativeGen/X86/Ppr.hs +2 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100222075144-43c66-55935bfe880e9e70ffdc7630346e445ff5da12a0.gz From ghcbuild at microsoft.com Wed Feb 24 02:18:43 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Wed Feb 24 02:18:45 2010 Subject: [nightly] 23-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100224071843.6F3AE324118@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Tue Feb 23 18:10:02 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100222 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (1 failures) **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Wed Feb 24 07:48:17 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Tue Feb 23 23:43:16 GMT 2010 2369 total tests, which gave rise to 13116 test cases, of which 0 caused framework failures 2775 were skipped 9992 expected passes 317 expected failures 0 unexpected passes 32 unexpected failures Unexpected failures: 2592(profc,profasm) 3429(profthreaded) CPUTime001(threaded2) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) conc019(profthreaded) concprog001(ghci,threaded2) concprog002(threaded2) drvfail006(normal) drvfail008(normal) ffi002(threaded1,profthreaded) joao-circular(profc) outofmem2(normal) signals002(threaded1,profthreaded) signals004(threaded1) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Wed Feb 24 07:48:17 GMT 2010 From ghcbuild at microsoft.com Wed Feb 24 02:39:29 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Wed Feb 24 02:39:31 2010 Subject: [nightly] 23-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100224073929.C38FB324035@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Tue Feb 23 18:00:02 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100222 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (2 failures) **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (2 failures) **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Wed Feb 24 08:09:03 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Wed Feb 24 00:10:30 GMT 2010 2458 total tests, which gave rise to 13565 test cases, of which 0 caused framework failures 2804 were skipped 10375 expected passes 361 expected failures 0 unexpected passes 25 unexpected failures Unexpected failures: 3429(threaded1,profthreaded) CPUTime001(threaded2) arith008(profasm,profthreaded) concprog001(ghci) ffi002(threaded1,threaded2) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) signals002(threaded1,profthreaded) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Wed Feb 24 08:09:03 GMT 2010 From simonpj at microsoft.com Wed Feb 24 06:57:51 2010 From: simonpj at microsoft.com (Simon Peyton-Jones) Date: Wed Feb 24 06:28:41 2010 Subject: patch applied (ghc): Make "make tags" work in the new build system In-Reply-To: <20100221172858.GA15637@abbot.galois.com> References: <20100221172858.GA15637@abbot.galois.com> Message-ID: <59543203684B2244980D7E4057D5FBC10B08F069@DB3EX14MBXC312.europe.corp.microsoft.com> Ian | Sun Feb 21 08:44:32 PST 2010 Ian Lynagh | * Make "make tags" work in the new build system Thanks. This does construct a file 'tags' in $(TOP), snippet below. Alas emacs says 'not a valid tags table'. The old TAGS file I have looks like the second snippet below. Any ideas? Simon ========= New 'tags' file made by 'make tags' ============= getCoreModuleName compiler/parser/ParserCoreUtils.hs 21 failP compiler/parser/ParserCoreUtils.hs 18 returnP compiler/parser/ParserCoreUtils.hs 15 thenP compiler/parser/ParserCoreUtils.hs 9 Token compiler/parser/ParserCoreUtils.hs 38 TKEOF compiler/parser/ParserCoreUtils.hs 75 TKchar compiler/parser/ParserCoreUtils.hs 74 TKstring compiler/parser/ParserCoreUtils.hs 73 TKrational compiler/parser/ParserCoreUtils.hs 72 TKinteger compiler/parser/ParserCoreUtils.hs 71 ========= Old TAGS file make by 'hasktags' (I think) ========= (there are some Ctrl-L's in here, and Ctrl-?) parser/Parser.y.pp,6920 module Parser17,18 1 for ambiguity in '{-# RULES "name" forall = ...145,146 1 for ambiguity when the source file starts with "-- | doc".153,154 | 'newtype'181,182 | qcon355,356 | qvarop356,357 | qconop357,358 | '('358,359 | body2375,376 | {-383,384 | {-390,391 | vocurly394,395 | missing_module_keyword398,399 | importdecls402,403 | cvtopdecls403,404 | missing_module_keyword416,417 | vocurly423,424 | {-430,431 | exportlist1434,435 | expdoclist438,439 | expdoclist439,440 | {-443,444 | docnamed447,448 | docnext448,449 | oqtycon454,455 | oqtycon455,456 | oqtycon456,457 | oqtycon457,458 | 'module'458,459 | qcname_ext462,463 | 'type'467,468 | qcon474,475 | importdecls484,485 | importdecl485,486 | {-486,487 | {-494,495 | {-498,499 | {-502,503 | {-506,507 | 'hiding'510,511 | INTEGER517,518 | 'infixl'521,522 | 'infixr'522,523 | op526,527 | topdecls533,534 | topdecl534,535 | ty_decl538,539 | 'instance'539,540 L0,163,164 L0, L1, LL163,164 '::'264,265 %name344,345 { let (binds, sigs, ats, _) = cvBindsAndSigs540,541 | stand_alone_deriving543,544 | 'default'544,545 | 'foreign'545,546 | '{-#546,547 | '{-#547,548 | decl548,549 | '$('551,552 | TH_ID_SPLICE552,553 identifier353,354 module :: {370,371 maybedocheader381,382 missing_module_keyword385,386 maybemoddeprec388,389 body392,393 body2396,397 top400,401 From davidterei at gmail.com Wed Feb 24 09:01:45 2010 From: davidterei at gmail.com (David Terei) Date: Wed Feb 24 08:32:35 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B8279BE.6000909@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <4B8279BE.6000909@gmail.com> Message-ID: <150b8f421002240601q16db1e94o80387b43547d51d1@mail.gmail.com> On 22 February 2010 23:34, Simon Marlow wrote: > I presume the calling-convention changes can be made in an acceptable way, > by defining a new calling convention rather than modifying the exisitng C > one? ?Does LLVM allow new calling conventions to be defined dynamically, > without patchig LLVM itself? No it doesn't, only really optimiser passes can be loaded in dynamically. I've just finished changing the patch though so that it defines a new calling convention instead of stealing an existing C one for that purpose, I've tested it and its working well with GHC, now need to check it does what its meant to and doesn't break the rest of LLVM. Will post it to this list soon and also post it to LLVM to get a discussion going about inclusion. From igloo at earth.li Wed Feb 24 09:05:11 2010 From: igloo at earth.li (Ian Lynagh) Date: Wed Feb 24 08:35:36 2010 Subject: patch applied (ghc): Make "make tags" work in the new build system In-Reply-To: <59543203684B2244980D7E4057D5FBC10B08F069@DB3EX14MBXC312.europe.corp.microsoft.com> References: <20100221172858.GA15637@abbot.galois.com> <59543203684B2244980D7E4057D5FBC10B08F069@DB3EX14MBXC312.europe.corp.microsoft.com> Message-ID: <20100224140511.GA2994@matrix.chaos.earth.li> On Wed, Feb 24, 2010 at 11:57:51AM +0000, Simon Peyton-Jones wrote: > Ian > > | Sun Feb 21 08:44:32 PST 2010 Ian Lynagh > | * Make "make tags" work in the new build system > > Thanks. This does construct a file 'tags' in $(TOP), snippet below. > > Alas emacs says 'not a valid tags table'. > > The old TAGS file I have looks like the second snippet below. > > Any ideas? It should make both a tags file for vim and a TAGS file for emacs. Maybe I should have called the target something else, but I'm not sure what. Thanks Ian From davidterei at gmail.com Wed Feb 24 09:05:13 2010 From: davidterei at gmail.com (David Terei) Date: Wed Feb 24 08:35:57 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <20100224013831.GA15752@whirlpool.galois.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <57C4205A-F3F1-4F36-8B91-9D14CB77ED65@rice.edu> <59543203684B2244980D7E4057D5FBC10B070D67@DB3EX14MBXC306.europe.corp.microsoft.com> <4B830AF9.6010902@gmail.com> <3E212AD2-9A71-4195-B114-95A359233596@rice.edu> <4B847358.8070104@gmail.com> <20100224013831.GA15752@whirlpool.galois.com> Message-ID: <150b8f421002240605h71d0bdb9v9d7e68d3a75cce42@mail.gmail.com> On 24 February 2010 12:38, Don Stewart wrote: > David, > > Did you see my patch to fix the parsing of -optlo and friends? > > ? ?http://www.galois.com/~dons/add-new-llvm-code-generator-to-ghc_.dpatch > > Order of the flags appears to matter to GetOpt. > Great! Thanks for that. I remember fixing that issue myself at one stage but I think it got lost in a temporary branch as I can't find it lying around anywhere. Where was I meant to 'see' it though other then this email? Because no, this is the first I've heard of it. From simonpj at microsoft.com Wed Feb 24 10:18:25 2010 From: simonpj at microsoft.com (Simon Peyton-Jones) Date: Wed Feb 24 09:48:57 2010 Subject: patch applied (ghc): Make "make tags" work in the new build system In-Reply-To: <20100224140511.GA2994@matrix.chaos.earth.li> References: <20100221172858.GA15637@abbot.galois.com> <59543203684B2244980D7E4057D5FBC10B08F069@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224140511.GA2994@matrix.chaos.earth.li> Message-ID: <59543203684B2244980D7E4057D5FBC10B08F3BB@DB3EX14MBXC312.europe.corp.microsoft.com> | It should make both a tags file for vim and a TAGS file for emacs. Maybe | I should have called the target something else, but I'm not sure what. Ah yes so it does. Thanks. What is 'tags' for? S From igloo at earth.li Wed Feb 24 10:45:40 2010 From: igloo at earth.li (Ian Lynagh) Date: Wed Feb 24 10:16:05 2010 Subject: patch applied (ghc): Make "make tags" work in the new build system In-Reply-To: <59543203684B2244980D7E4057D5FBC10B08F3BB@DB3EX14MBXC312.europe.corp.microsoft.com> References: <20100221172858.GA15637@abbot.galois.com> <59543203684B2244980D7E4057D5FBC10B08F069@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224140511.GA2994@matrix.chaos.earth.li> <59543203684B2244980D7E4057D5FBC10B08F3BB@DB3EX14MBXC312.europe.corp.microsoft.com> Message-ID: <20100224154540.GB5117@matrix.chaos.earth.li> On Wed, Feb 24, 2010 at 03:18:25PM +0000, Simon Peyton-Jones wrote: > | It should make both a tags file for vim and a TAGS file for emacs. Maybe > | I should have called the target something else, but I'm not sure what. > > Ah yes so it does. Thanks. What is 'tags' for? tags is the vim equivalent of TAGS. Thanks Ian From Malcolm.Wallace at cs.york.ac.uk Wed Feb 24 11:03:16 2010 From: Malcolm.Wallace at cs.york.ac.uk (Malcolm Wallace) Date: Wed Feb 24 10:35:20 2010 Subject: patch applied (ghc): Make "make tags" work in the new build system In-Reply-To: <20100224154540.GB5117@matrix.chaos.earth.li> References: <20100221172858.GA15637@abbot.galois.com> <59543203684B2244980D7E4057D5FBC10B08F069@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224140511.GA2994@matrix.chaos.earth.li> <59543203684B2244980D7E4057D5FBC10B08F3BB@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224154540.GB5117@matrix.chaos.earth.li> Message-ID: <6147FF6B-2779-4066-BC88-F0584B4FC27D@cs.york.ac.uk> > tags is the vim equivalent of TAGS. Unfortunately, on a case-insensitive filesystem, such as the defaults on Windows and MacOS, there is no way to have both a 'tags' and a 'TAGS' file. Regards, Malcolm From simonpj at microsoft.com Wed Feb 24 11:26:23 2010 From: simonpj at microsoft.com (Simon Peyton-Jones) Date: Wed Feb 24 10:57:11 2010 Subject: patch applied (ghc): Make "make tags" work in the new build system In-Reply-To: <6147FF6B-2779-4066-BC88-F0584B4FC27D@cs.york.ac.uk> References: <20100221172858.GA15637@abbot.galois.com> <59543203684B2244980D7E4057D5FBC10B08F069@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224140511.GA2994@matrix.chaos.earth.li> <59543203684B2244980D7E4057D5FBC10B08F3BB@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224154540.GB5117@matrix.chaos.earth.li> <6147FF6B-2779-4066-BC88-F0584B4FC27D@cs.york.ac.uk> Message-ID: <59543203684B2244980D7E4057D5FBC10B08F4B3@DB3EX14MBXC312.europe.corp.microsoft.com> good point. Could we call them 'tags-emacs' and 'tags-vim'? Simoin | -----Original Message----- | From: cvs-ghc-bounces@haskell.org [mailto:cvs-ghc-bounces@haskell.org] On Behalf Of | Malcolm Wallace | Sent: 24 February 2010 16:03 | To: cvs-ghc@haskell.org | Subject: Re: patch applied (ghc): Make "make tags" work in the new build system | | > tags is the vim equivalent of TAGS. | | Unfortunately, on a case-insensitive filesystem, such as the defaults | on Windows and MacOS, there is no way to have both a 'tags' and a | 'TAGS' file. | | Regards, | Malcolm | | | | | _______________________________________________ | Cvs-ghc mailing list | Cvs-ghc@haskell.org | http://www.haskell.org/mailman/listinfo/cvs-ghc From marlowsd at gmail.com Wed Feb 24 11:40:44 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Wed Feb 24 11:11:22 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B82B583.5040103@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <4B8279BE.6000909@gmail.com> <4B82B583.5040103@gmail.com> Message-ID: <4B85568C.6020306@gmail.com> On 22/02/2010 16:49, Simon Marlow wrote: > On 22/02/2010 12:34, Simon Marlow wrote: > >> I'm currently running some benchmarks to see how much impact turning off >> TNTC has on the -fasm backend. > > Here are the results on x86-64/Linux: [ snip ] > -------------------------------------------------------------------------------- > > Mi +4.7% -0.0% -0.6% -1.7% > Max +8.9% +0.0% +16.9% +13.8% > Geometric Mean +6.1% -0.0% +4.9% +4.2% and here are the results on x86/Linux: -------------------------------------------------------------------------------- Program Size Allocs Runtime Elapsed -------------------------------------------------------------------------------- anna +6.9% +0.0% +7.1% +7.4% ansi +4.3% +0.0% 0.00 0.00 atom +4.5% +0.0% +23.6% +21.7% awards +4.2% +0.0% 0.00 0.00 banner +3.5% +0.0% 0.00 0.00 bernouilli +4.2% +0.0% +2.7% +1.8% boyer +4.3% +0.0% 0.10 0.11 boyer2 +4.1% +0.0% 0.01 0.02 bspt +5.5% +0.0% 0.02 0.02 cacheprof +5.3% +0.0% +3.1% +3.0% calendar +4.2% +0.0% 0.00 0.00 cichelli +4.2% +0.0% 0.19 0.22 circsim +4.6% +0.0% +3.3% +2.5% clausify +4.3% +0.0% 0.07 0.09 comp_lab_zift +4.5% +0.0% +15.3% +14.4% compress +4.4% +0.0% +4.1% +4.3% compress2 +4.3% +0.0% +0.5% +0.4% constraints +4.5% +0.0% +6.4% +5.9% cryptarithm1 +3.8% +0.0% +5.3% +3.3% cryptarithm2 +4.0% +0.0% 0.03 0.03 cse +3.9% +0.0% 0.00 0.00 eliza +3.6% +0.0% 0.00 0.00 event +4.3% +0.0% +7.9% +7.5% exp3_8 +4.2% +0.0% +17.8% +13.3% expert +4.1% +0.0% 0.00 0.00 fem +5.5% +0.0% 0.06 0.06 fft +4.6% +0.0% 0.09 0.10 fft2 +4.9% +0.0% 0.22 +12.3% fibheaps +4.3% +0.0% 0.08 0.08 fish +4.0% +0.0% 0.05 0.06 fluid +6.3% +0.0% 0.02 0.02 fulsom +6.1% +0.0% +3.4% +3.2% gamteb +5.0% +0.0% 0.19 0.21 gcd +4.2% +0.0% 0.06 0.07 gen_regexps +4.0% +0.0% 0.00 0.00 genfft +4.2% +0.0% 0.09 0.10 gg +5.1% +0.0% 0.03 0.03 grep +4.5% +0.0% 0.00 0.00 hidden +5.7% (stdout) (stdout) (stdout) hpg +5.2% +0.0% +6.1% +2.0% ida +4.4% +0.0% +10.2% +6.6% infer +4.9% +0.0% 0.13 0.14 integer +4.2% +0.0% +1.2% -0.2% integrate +4.6% +0.0% +4.9% +5.0% knights +4.6% +0.0% 0.01 0.01 lcss +4.2% +0.0% +8.5% +7.7% life +3.8% +0.0% +23.8% +19.5% lift +4.5% +0.0% 0.00 0.00 listcompr +3.8% +0.0% +5.3% +4.7% listcopy +3.8% +0.0% +5.7% +6.3% maillist +4.0% +0.0% 0.15 +6.1% mandel +4.5% +0.0% -0.6% -2.4% mandel2 +3.9% +0.0% 0.02 0.02 minimax +4.2% +0.0% 0.01 0.01 mkhprog +4.2% +0.0% 0.00 0.01 multiplier +4.4% +0.0% +10.0% +10.6% nucleic2 +4.6% +0.0% +16.8% +15.0% para +4.4% +0.0% +11.7% +9.7% paraffins +4.3% +0.0% -1.9% +0.8% parser +5.0% +0.0% 0.08 0.08 parstof +4.8% +0.0% 0.02 0.02 pic +5.0% +0.0% 0.03 0.03 power +4.4% +0.0% +2.7% +2.7% pretty +4.4% +0.0% 0.00 0.00 primes +4.2% +0.0% 0.12 0.13 primetest +4.3% +0.0% -0.9% +0.5% prolog +4.2% +0.0% 0.00 0.00 puzzle +4.1% +0.0% +8.7% +7.8% queens +4.2% +0.0% 0.03 0.03 reptile +5.1% +0.0% 0.03 0.04 rewrite +4.6% +0.0% 0.02 0.03 rfib +4.5% +0.0% 0.12 0.12 rsa +4.3% +0.0% 0.17 0.18 scc +3.7% +0.0% 0.00 0.00 sched +4.3% +0.0% 0.05 0.05 scs +5.7% +0.0% +2.3% +1.3% simple +6.8% +0.0% +5.6% +5.8% solid +4.5% +0.0% +11.1% +6.6% sorting +4.0% +0.0% 0.00 0.00 sphere +5.3% +0.0% +17.2% +12.9% symalg +5.3% +0.0% 0.10 0.10 tak +4.2% +0.0% 0.02 0.02 transform +4.9% +0.0% +2.2% +2.1% treejoin +3.7% +0.0% -0.4% +2.7% typecheck +4.3% +0.0% -23.8% -24.1% veritas +6.5% +0.0% 0.00 0.00 wang +4.6% +0.0% +8.0% +7.7% wave4main +4.4% +0.0% +5.2% +5.3% wheel-sieve1 +4.2% +0.0% +10.0% +8.8% wheel-sieve2 +4.2% +0.0% +2.1% +2.2% x2n1 +4.6% +0.0% 0.06 0.06 -------------------------------------------------------------------------------- Min +3.5% +0.0% -23.8% -24.1% Max +6.9% +0.0% +23.8% +21.7% Geometric Mean +4.5% -0.0% +6.0% +5.3% Slightly worse than the x86_64 results, though this is an older processor. The result for typecheck is very odd. It's repeatable, but only on this machine - I suspect a bad cache interaction or similar. I should probably re-run the tests on a machine with a more recent processor. While I was at it, I measured the -fvia-C backend against the NCG: -------------------------------------------------------------------------------- Program Size Allocs Runtime Elapsed -------------------------------------------------------------------------------- Min -7.7% -47.3% -33.0% -30.2% Max -3.8% +0.0% +29.5% +28.8% Geometric Mean -5.0% -0.7% -6.7% -5.7% while we weren't looking, the via-C backend has regressed a lot, at least on these "average" Haskell programs. The +29.5% outlier is typecheck agaain, since I'm using the same set of results for -fasm as above. I think the main reason for the regression is code like this: movl $stg_ap_n_fast, %eax .L3: jmp *%eax .L2: movl $8, 112(%edx) movl -8(%ebx), %eax jmp .L3 gcc is being too clever in commoning up the indirect jump. Conclusion: don't use -fvia-C, even in 6.12, unless you are sure it speeds things up. I'm turning it off for our builds. =============== So here's a crazy idea. Why don't we post-process the assembly code coming out of LLVM? Before you throw up your hands in horror, consider that - it's a simple transformation, just re-ordering blocks of code - we can do it in Haskell using ByteStrings, it would probably amount to a couple of hundred lines of code at the most. Perhaps an Alex lexer would be the quickest way to split into blocks, then a bit of Haskell to glue them back into the correct order. We may have to fiddle with the .aligns a bit. - we don't care too much about compile-time performance, since LLVM is a -O2 thing, we have the NCG for generating code fast - at the same time we can talk with the LLVM folks about adding support for TNTC, but we'd have a way to generate code in the meantime. Just a thought... Cheers, Simon From ghcbuild at microsoft.com Wed Feb 24 13:08:36 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Wed Feb 24 13:08:39 2010 Subject: [nightly] 24-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100224180836.0396D324293@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Wed Feb 24 18:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100223 **** Building stage 2 compiler ... failed; relevant barfage is below. **** building testsuite tools ... failed. **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (91 failures) **** running nofib (-O -fvia-C) ... ok. (91 failures) **** running nofib (-O -prof -auto-all) ... ok. (91 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (91 failures) **** running nofib (-fasm) ... ok. (91 failures) **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Wed Feb 24 18:38:12 GMT 2010 ------------------------------------------------------------------------ ------------------------------------------------------------------------ The last 30 lines of /playpen/simonmar/nightly/HEAD/logs/i386-unknown-linux-stage2 are ------------------------------------------------------------------------ ------------------------------------------------------------------------ ^ /playpen/simonmar/nightly/HEAD/fptools/docs/users_guide/ug-book.xml:18: parser error : Entity 'ffi-chap' not defined &ffi-chap; ^ /playpen/simonmar/nightly/HEAD/fptools/docs/users_guide/ug-book.xml:19: parser error : Entity 'wrong' not defined &wrong; ^ /playpen/simonmar/nightly/HEAD/fptools/docs/users_guide/ug-book.xml:20: parser error : Entity 'utils' not defined &utils; ^ /playpen/simonmar/nightly/HEAD/fptools/docs/users_guide/ug-book.xml:21: parser error : Entity 'win32-dll' not defined &win32-dll; ^ /playpen/simonmar/nightly/HEAD/fptools/docs/users_guide/ug-book.xml:22: parser error : Entity 'bugs' not defined &bugs; ^ /playpen/simonmar/nightly/HEAD/fptools/docs/users_guide/users_guide.xml:10: parser error : Failure to process entity ug-book &ug-book; ^ /playpen/simonmar/nightly/HEAD/fptools/docs/users_guide/users_guide.xml:10: parser error : Entity 'ug-book' not defined &ug-book; ^ unable to parse /playpen/simonmar/nightly/HEAD/fptools/docs/users_guide/users_guide.xml Error: xsltproc failed gmake[1]: *** [docs/users_guide/users_guide.ps] Error 1 gmake: *** [all] Error 2 real 1m26.658s user 1m21.759s sys 0m3.017s Nightly run ended at Wed Feb 24 18:38:12 GMT 2010 From ghcbuild at microsoft.com Wed Feb 24 13:15:05 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Wed Feb 24 13:15:07 2010 Subject: [nightly] 24-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100224181505.BEA15324293@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Wed Feb 24 18:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100223 **** Building stage 2 compiler ... failed; relevant barfage is below. **** building testsuite tools ... failed. **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (91 failures) **** running nofib (-O -fvia-C) ... ok. (91 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (91 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. (91 failures) **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Wed Feb 24 18:44:40 GMT 2010 ------------------------------------------------------------------------ ------------------------------------------------------------------------ The last 30 lines of /playpen/simonmar/nightly/STABLE/logs/i386-unknown-linux-stage2 are ------------------------------------------------------------------------ ------------------------------------------------------------------------ ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/ug-book.xml:18: parser error : Entity 'ffi-chap' not defined &ffi-chap; ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/ug-book.xml:19: parser error : Entity 'wrong' not defined &wrong; ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/ug-book.xml:20: parser error : Entity 'utils' not defined &utils; ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/ug-book.xml:21: parser error : Entity 'win32-dll' not defined &win32-dll; ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/ug-book.xml:22: parser error : Entity 'bugs' not defined &bugs; ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/users_guide.xml:10: parser error : Failure to process entity ug-book &ug-book; ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/users_guide.xml:10: parser error : Entity 'ug-book' not defined &ug-book; ^ unable to parse /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/users_guide.xml Error: xsltproc failed gmake[1]: *** [docs/users_guide/users_guide.ps] Error 1 gmake: *** [all] Error 2 real 1m19.446s user 1m16.266s sys 0m2.901s Nightly run ended at Wed Feb 24 18:44:40 GMT 2010 From igloo at earth.li Wed Feb 24 13:56:36 2010 From: igloo at earth.li (Ian Lynagh) Date: Wed Feb 24 13:27:01 2010 Subject: patch applied (ghc): Make "make tags" work in the new build system In-Reply-To: <59543203684B2244980D7E4057D5FBC10B08F4B3@DB3EX14MBXC312.europe.corp.microsoft.com> References: <20100221172858.GA15637@abbot.galois.com> <59543203684B2244980D7E4057D5FBC10B08F069@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224140511.GA2994@matrix.chaos.earth.li> <59543203684B2244980D7E4057D5FBC10B08F3BB@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224154540.GB5117@matrix.chaos.earth.li> <6147FF6B-2779-4066-BC88-F0584B4FC27D@cs.york.ac.uk> <59543203684B2244980D7E4057D5FBC10B08F4B3@DB3EX14MBXC312.europe.corp.microsoft.com> Message-ID: <20100224185636.GA28070@matrix.chaos.earth.li> On Wed, Feb 24, 2010 at 04:26:23PM +0000, Simon Peyton-Jones wrote: > | > tags is the vim equivalent of TAGS. > | > | Unfortunately, on a case-insensitive filesystem, such as the defaults > | on Windows and MacOS, there is no way to have both a 'tags' and a > | 'TAGS' file. > > good point. Could we call them 'tags-emacs' and 'tags-vim'? We could do, but tags and TAGS are the standard names. We could also have separate "make tags" and "make TAGS" commands, which generate only the named file. Whatever the decision, implementing it will be easy. Thanks Ian From ghcbuild at microsoft.com Wed Feb 24 13:49:02 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Wed Feb 24 13:49:02 2010 Subject: [nightly] 24-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100224184902.2044032413D@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Wed Feb 24 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100223 **** Building stage 2 compiler ... failed; relevant barfage is below. **** building source distribution ... failed. **** building testsuite tools ... failed. **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (91 failures) **** running nofib (-O -fvia-C) ... ok. (91 failures) **** running nofib (-O -prof -auto-all) ... ok. (91 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (91 failures) **** running nofib (-fasm) ... ok. (91 failures) **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Wed Feb 24 19:18:37 GMT 2010 ------------------------------------------------------------------------ ------------------------------------------------------------------------ The last 30 lines of /64playpen/simonmar/nightly/HEAD-cam-04-unx/logs/x86_64-unknown-linux-stage2 are ------------------------------------------------------------------------ ------------------------------------------------------------------------ ^ /64playpen/simonmar/nightly/HEAD-cam-04-unx/fptools/docs/users_guide/ug-book.xml:18: parser error : Entity 'ffi-chap' not defined &ffi-chap; ^ /64playpen/simonmar/nightly/HEAD-cam-04-unx/fptools/docs/users_guide/ug-book.xml:19: parser error : Entity 'wrong' not defined &wrong; ^ /64playpen/simonmar/nightly/HEAD-cam-04-unx/fptools/docs/users_guide/ug-book.xml:20: parser error : Entity 'utils' not defined &utils; ^ /64playpen/simonmar/nightly/HEAD-cam-04-unx/fptools/docs/users_guide/ug-book.xml:21: parser error : Entity 'win32-dll' not defined &win32-dll; ^ /64playpen/simonmar/nightly/HEAD-cam-04-unx/fptools/docs/users_guide/ug-book.xml:22: parser error : Entity 'bugs' not defined &bugs; ^ /64playpen/simonmar/nightly/HEAD-cam-04-unx/fptools/docs/users_guide/users_guide.xml:10: parser error : Failure to process entity ug-book &ug-book; ^ /64playpen/simonmar/nightly/HEAD-cam-04-unx/fptools/docs/users_guide/users_guide.xml:10: parser error : Entity 'ug-book' not defined &ug-book; ^ unable to parse /64playpen/simonmar/nightly/HEAD-cam-04-unx/fptools/docs/users_guide/users_guide.xml Error: xsltproc failed gmake[1]: *** [docs/users_guide/users_guide.ps] Error 1 gmake: *** [all] Error 2 real 0m49.135s user 0m47.696s sys 0m1.422s Nightly run ended at Wed Feb 24 19:18:37 GMT 2010 From ghcbuild at microsoft.com Wed Feb 24 13:59:16 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Wed Feb 24 13:59:18 2010 Subject: [nightly] 24-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100224185916.A259632438F@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Wed Feb 24 19:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100223 **** Building stage 2 compiler ... failed; relevant barfage is below. **** building source distribution ... ok. **** building testsuite tools ... failed. **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (91 failures) **** running nofib (-O -fvia-C) ... ok. (91 failures) **** running nofib (-O -prof -auto-all) ... ok. (91 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (91 failures) **** running nofib (-fasm) ... ok. (91 failures) **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Wed Feb 24 19:28:51 GMT 2010 ------------------------------------------------------------------------ ------------------------------------------------------------------------ The last 30 lines of /64playpen/simonmar/nightly/STABLE-cam-04-unx/logs/x86_64-unknown-linux-stage2 are ------------------------------------------------------------------------ ------------------------------------------------------------------------ ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/ug-book.xml:18: parser error : Entity 'ffi-chap' not defined &ffi-chap; ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/ug-book.xml:19: parser error : Entity 'wrong' not defined &wrong; ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/ug-book.xml:20: parser error : Entity 'utils' not defined &utils; ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/ug-book.xml:21: parser error : Entity 'win32-dll' not defined &win32-dll; ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/ug-book.xml:22: parser error : Entity 'bugs' not defined &bugs; ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/users_guide.xml:10: parser error : Failure to process entity ug-book &ug-book; ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/users_guide.xml:10: parser error : Entity 'ug-book' not defined &ug-book; ^ unable to parse /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/users_guide.xml Error: xsltproc failed gmake[1]: *** [docs/users_guide/users_guide.ps] Error 1 gmake: *** [all] Error 2 real 0m42.664s user 0m41.148s sys 0m1.517s Nightly run ended at Wed Feb 24 19:28:51 GMT 2010 From dons at galois.com Wed Feb 24 17:02:20 2010 From: dons at galois.com (Don Stewart) Date: Wed Feb 24 16:32:54 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <150b8f421002240605h71d0bdb9v9d7e68d3a75cce42@mail.gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <57C4205A-F3F1-4F36-8B91-9D14CB77ED65@rice.edu> <59543203684B2244980D7E4057D5FBC10B070D67@DB3EX14MBXC306.europe.corp.microsoft.com> <4B830AF9.6010902@gmail.com> <3E212AD2-9A71-4195-B114-95A359233596@rice.edu> <4B847358.8070104@gmail.com> <20100224013831.GA15752@whirlpool.galois.com> <150b8f421002240605h71d0bdb9v9d7e68d3a75cce42@mail.gmail.com> Message-ID: <20100224220220.GJ19176@whirlpool.galois.com> davidterei: > On 24 February 2010 12:38, Don Stewart wrote: > > David, > > > > Did you see my patch to fix the parsing of -optlo and friends? > > > > ? ?http://www.galois.com/~dons/add-new-llvm-code-generator-to-ghc_.dpatch > > > > Order of the flags appears to matter to GetOpt. > > > > Great! Thanks for that. I remember fixing that issue myself at one > stage but I think it got lost in a temporary branch as I can't find it > lying around anywhere. Where was I meant to 'see' it though other then > this email? Because no, this is the first I've heard of it. > Ah , I think it got blocked by the mailing list filter (patch too large). -- Don From davidterei at gmail.com Wed Feb 24 19:08:13 2010 From: davidterei at gmail.com (David Terei) Date: Wed Feb 24 18:38:38 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B85568C.6020306@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <4B8279BE.6000909@gmail.com> <4B82B583.5040103@gmail.com> <4B85568C.6020306@gmail.com> Message-ID: <4B85BF6D.4020100@gmail.com> Simon Marlow wrote: > So, nofib programs are on average 4-5% slower, with a 6% increase in > binary sizes (this is with -split-objs). > > What's interesting is that there are some outliers here: programs that > go 12-17% slower without TNTC. Looking at David's thesis the results are > quite different. Perhaps it is processor dependent? I ran nofib again and am getting results similar to yours. I have all the nofib runs laying around from my thesis and they give the results I put in it. Seems to perhaps be a mistake from me that the TNTC enabled runs weren't optimal due to some background process I didn't notice. Also there were some nofib programs I didn't include at that time due to problems getting them running, the missing ones seems to have reduced the advantage of TNTC. > I have to admit, the results I'm getting are what I would expect, so its > conceivable that some subconscious experimental bias has crept in > somewhere :-) Actually its more of an unconscious self sabotage. When I did the benchmarking of TNTC I was hoping to get the 4-6% value that I had seen talked of before since that's basically the margin by which the LLVM back-end is slower then the NCG. I think my body was punishing me from lack of sleep :). > [snip/new email] > So here's a crazy idea. Why don't we post-process the assembly code > coming out of LLVM? Before you throw up your hands in horror, consider that I think it should work pretty well, especially if its intentioned to just be a stopgap solution until we get support for TNTC from LLVM. > > Cheers, > Simon From chak at cse.unsw.edu.au Wed Feb 24 20:15:24 2010 From: chak at cse.unsw.edu.au (Manuel M T Chakravarty) Date: Wed Feb 24 19:45:52 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B85568C.6020306@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <4B8279BE.6000909@gmail.com> <4B82B583.5040103@gmail.com> <4B85568C.6020306@gmail.com> Message-ID: <71FA1768-BC25-40B3-8017-0244439E4B16@cse.unsw.edu.au> Simon Marlow: > On 22/02/2010 16:49, Simon Marlow wrote: >> On 22/02/2010 12:34, Simon Marlow wrote: >> >>> I'm currently running some benchmarks to see how much impact turning off >>> TNTC has on the -fasm backend. >> >> Here are the results on x86-64/Linux: > [ snip ] >> -------------------------------------------------------------------------------- >> >> Mi +4.7% -0.0% -0.6% -1.7% >> Max +8.9% +0.0% +16.9% +13.8% >> Geometric Mean +6.1% -0.0% +4.9% +4.2% > > and here are the results on x86/Linux: > > -------------------------------------------------------------------------------- > Program Size Allocs Runtime Elapsed > -------------------------------------------------------------------------------- > anna +6.9% +0.0% +7.1% +7.4% > ansi +4.3% +0.0% 0.00 0.00 > atom +4.5% +0.0% +23.6% +21.7% > awards +4.2% +0.0% 0.00 0.00 > banner +3.5% +0.0% 0.00 0.00 > bernouilli +4.2% +0.0% +2.7% +1.8% > boyer +4.3% +0.0% 0.10 0.11 > boyer2 +4.1% +0.0% 0.01 0.02 > bspt +5.5% +0.0% 0.02 0.02 > cacheprof +5.3% +0.0% +3.1% +3.0% > calendar +4.2% +0.0% 0.00 0.00 > cichelli +4.2% +0.0% 0.19 0.22 > circsim +4.6% +0.0% +3.3% +2.5% > clausify +4.3% +0.0% 0.07 0.09 > comp_lab_zift +4.5% +0.0% +15.3% +14.4% > compress +4.4% +0.0% +4.1% +4.3% > compress2 +4.3% +0.0% +0.5% +0.4% > constraints +4.5% +0.0% +6.4% +5.9% > cryptarithm1 +3.8% +0.0% +5.3% +3.3% > cryptarithm2 +4.0% +0.0% 0.03 0.03 > cse +3.9% +0.0% 0.00 0.00 > eliza +3.6% +0.0% 0.00 0.00 > event +4.3% +0.0% +7.9% +7.5% > exp3_8 +4.2% +0.0% +17.8% +13.3% > expert +4.1% +0.0% 0.00 0.00 > fem +5.5% +0.0% 0.06 0.06 > fft +4.6% +0.0% 0.09 0.10 > fft2 +4.9% +0.0% 0.22 +12.3% > fibheaps +4.3% +0.0% 0.08 0.08 > fish +4.0% +0.0% 0.05 0.06 > fluid +6.3% +0.0% 0.02 0.02 > fulsom +6.1% +0.0% +3.4% +3.2% > gamteb +5.0% +0.0% 0.19 0.21 > gcd +4.2% +0.0% 0.06 0.07 > gen_regexps +4.0% +0.0% 0.00 0.00 > genfft +4.2% +0.0% 0.09 0.10 > gg +5.1% +0.0% 0.03 0.03 > grep +4.5% +0.0% 0.00 0.00 > hidden +5.7% (stdout) (stdout) (stdout) > hpg +5.2% +0.0% +6.1% +2.0% > ida +4.4% +0.0% +10.2% +6.6% > infer +4.9% +0.0% 0.13 0.14 > integer +4.2% +0.0% +1.2% -0.2% > integrate +4.6% +0.0% +4.9% +5.0% > knights +4.6% +0.0% 0.01 0.01 > lcss +4.2% +0.0% +8.5% +7.7% > life +3.8% +0.0% +23.8% +19.5% > lift +4.5% +0.0% 0.00 0.00 > listcompr +3.8% +0.0% +5.3% +4.7% > listcopy +3.8% +0.0% +5.7% +6.3% > maillist +4.0% +0.0% 0.15 +6.1% > mandel +4.5% +0.0% -0.6% -2.4% > mandel2 +3.9% +0.0% 0.02 0.02 > minimax +4.2% +0.0% 0.01 0.01 > mkhprog +4.2% +0.0% 0.00 0.01 > multiplier +4.4% +0.0% +10.0% +10.6% > nucleic2 +4.6% +0.0% +16.8% +15.0% > para +4.4% +0.0% +11.7% +9.7% > paraffins +4.3% +0.0% -1.9% +0.8% > parser +5.0% +0.0% 0.08 0.08 > parstof +4.8% +0.0% 0.02 0.02 > pic +5.0% +0.0% 0.03 0.03 > power +4.4% +0.0% +2.7% +2.7% > pretty +4.4% +0.0% 0.00 0.00 > primes +4.2% +0.0% 0.12 0.13 > primetest +4.3% +0.0% -0.9% +0.5% > prolog +4.2% +0.0% 0.00 0.00 > puzzle +4.1% +0.0% +8.7% +7.8% > queens +4.2% +0.0% 0.03 0.03 > reptile +5.1% +0.0% 0.03 0.04 > rewrite +4.6% +0.0% 0.02 0.03 > rfib +4.5% +0.0% 0.12 0.12 > rsa +4.3% +0.0% 0.17 0.18 > scc +3.7% +0.0% 0.00 0.00 > sched +4.3% +0.0% 0.05 0.05 > scs +5.7% +0.0% +2.3% +1.3% > simple +6.8% +0.0% +5.6% +5.8% > solid +4.5% +0.0% +11.1% +6.6% > sorting +4.0% +0.0% 0.00 0.00 > sphere +5.3% +0.0% +17.2% +12.9% > symalg +5.3% +0.0% 0.10 0.10 > tak +4.2% +0.0% 0.02 0.02 > transform +4.9% +0.0% +2.2% +2.1% > treejoin +3.7% +0.0% -0.4% +2.7% > typecheck +4.3% +0.0% -23.8% -24.1% > veritas +6.5% +0.0% 0.00 0.00 > wang +4.6% +0.0% +8.0% +7.7% > wave4main +4.4% +0.0% +5.2% +5.3% > wheel-sieve1 +4.2% +0.0% +10.0% +8.8% > wheel-sieve2 +4.2% +0.0% +2.1% +2.2% > x2n1 +4.6% +0.0% 0.06 0.06 > -------------------------------------------------------------------------------- > Min +3.5% +0.0% -23.8% -24.1% > Max +6.9% +0.0% +23.8% +21.7% > Geometric Mean +4.5% -0.0% +6.0% +5.3% > > Slightly worse than the x86_64 results, though this is an older processor. > > The result for typecheck is very odd. It's repeatable, but only on this machine - I suspect a bad cache interaction or similar. I should probably re-run the tests on a machine with a more recent processor. Here the results on a E5472 (QuadCore Harpertown @ 3GHz) with x86/MacOS and not all tests worked for me. -------------------------------------------------------------------------------- Program Size Allocs Runtime Elapsed -------------------------------------------------------------------------------- ansi +1.4% +0.0% 0.00 0.00 atom +1.3% +0.0% -4.8% +0.0% awards +0.7% +0.0% 0.00 0.00 banner +1.6% +0.0% 0.00 0.00 bernouilli +1.4% +0.0% +0.0% +0.0% boyer +0.7% +0.0% 0.03 0.03 boyer2 +0.8% +0.0% 0.00 0.00 calendar +2.1% +0.0% 0.00 0.00 cichelli +1.7% +0.0% 0.08 0.08 circsim +1.9% +0.0% +4.4% +3.4% clausify +1.4% +0.0% 0.04 0.04 comp_lab_zift +2.0% +0.0% 0.17 0.17 constraints +1.4% +0.0% +1.3% +1.7% cryptarithm1 +0.9% +0.0% +5.3% +5.0% cryptarithm2 +1.8% +0.0% 0.01 0.01 cse +1.8% +0.0% 0.00 0.00 eliza +0.7% +0.0% 0.00 0.00 exp3_8 +0.7% +0.0% 0.10 0.11 expert +1.5% +0.0% 0.00 0.00 fft2 +1.2% +0.0% 0.06 0.06 fibheaps +1.4% +0.0% 0.03 0.03 fish +0.9% +0.0% 0.02 0.02 gcd +1.4% +0.0% 0.02 0.02 gen_regexps +0.9% +0.0% 0.00 0.00 integrate +1.9% +0.0% +2.7% +3.1% nucleic2 +1.4% (stdout) (stdout) (stdout) paraffins +2.0% +0.0% 0.08 0.10 primes +1.4% +0.0% 0.05 0.05 queens +1.4% +0.0% 0.02 0.02 rfib +1.3% +0.0% 0.05 0.05 sched +2.1% +0.0% 0.02 0.02 solid +1.2% +0.0% 0.11 0.14 tak +1.4% +0.0% 0.01 0.01 transform +1.2% +0.0% +7.4% +3.4% typecheck +1.4% +0.0% 0.21 0.21 wang +1.9% +0.0% 0.08 0.09 wave4main +1.3% +0.0% +6.2% +3.0% wheel-sieve1 +1.4% +0.0% +22.7% +22.7% wheel-sieve2 +1.4% +0.0% 0.15 0.19 x2n1 +1.9% +0.0% 0.02 0.02 -------------------------------------------------------------------------------- Min +0.7% +0.0% -4.8% +0.0% Max +2.1% +0.0% +22.7% +22.7% Geometric Mean +1.4% -0.0% +4.8% +4.5% The most interesting observation seems to be that the results are largely inconsistent (look at 'atom', for example). Even wheel-sieve1, where our numbers go in the same direction, varies between +10% and +22.7%. My numbers are from an older HEAD (as I had nofib handy there), but I'm still wondering where this difference comes from. Is it just that different processors are more or less sensible to TNTC? But then it doesn't seem that one processor is consistently less affected, but it also seems to depend on the program. > =============== > > So here's a crazy idea. Why don't we post-process the assembly code coming out of LLVM? Before you throw up your hands in horror, consider that > > - it's a simple transformation, just re-ordering blocks of code > > - we can do it in Haskell using ByteStrings, it would probably > amount to a couple of hundred lines of code at the most. Perhaps > an Alex lexer would be the quickest way to split into blocks, then > a bit of Haskell to glue them back into the correct order. We may > have to fiddle with the .aligns a bit. > > - we don't care too much about compile-time performance, since LLVM is > a -O2 thing, we have the NCG for generating code fast > > - at the same time we can talk with the LLVM folks about adding > support for TNTC, but we'd have a way to generate code in the > meantime. > > Just a thought... Maybe it's not that crazy, at least as an interim solution. Manuel From davidterei at gmail.com Wed Feb 24 21:00:53 2010 From: davidterei at gmail.com (David Terei) Date: Wed Feb 24 20:31:18 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B83F293.6020807@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <4B83F293.6020807@gmail.com> Message-ID: <4B85D9D5.5090804@gmail.com> Simon Marlow wrote: > In fact, I'm surprised at how much there is here. The LLVM backend has > more in common with the native code generators than the C backend, but I > was expecting something closer to a straightforward pretty-printer. I > wonder if having a complete datatype for LLVM is really buying us > anything since we don't actually do anything to it other than print it > out. As a quick example of what I mean, what would be lost if we replace > > data LlvmCastOp > = LM_Trunc > | LM_Zext > ... lots of constructors ... > deriving (Eq) > > instance Show LlvmCastOp where > show LM_Trunc = "trunc" > show LM_Zext = "zext" > ... > > with > > type LlvmCastOp = FastString > lM_Trunc = fsLit "trunc" > ... > At the moment, nothing. > > Do we envisage doing anything complex to the Llvm code inside GHC in the > future? I can't think of a reason to. One thing I would like to investigate at some point is to have the LLVM binding use the C++ API of LLVM instead of printing out LLVM Assembly. This should potentially speed up the compilation process since it would replace the current 4 passes (generate LLVM Assembly, convert to bitcode [llvm-as], optimise [opt], compile to native code [llc]). With just one. I believe the current design would be easier to extend for this purpose then the simplified one you propose. After trying this out I'd be happy to simplify the LLVM binding if we stick with the current approach of generating LLVM assembly and calling the tools. > > Some of the cleanups and optimisations are quite significant. Highlights: > > * I think the module LlvmCodeGen.Regs can mostly go away. > * There are some obvious representation changes that would improve > performance (String -> FastString, [a] -> OrdList a). > > Review key: > > XXX This needs to be addressed before the patch goes in > CLEAN A potential cleanup (some of these are quite major) > OPT A possible optimisation > > Patch review follows: > > hunk ./compiler/ghc.cabal.in 44 > +Flag llvm > + Description: Build the LLVM code generator. > + Default: False > + Manual: True > > CLEAN Is there a good reason to want to disable the LLVM support, or > can it always be compiled in? Since the LLVM back-end requires TNTC to be disabled, yes it should be disable by default. > > +Flag unreg > + Description: Build an unregistered version. > + Default: False > + Manual: True > + > > XXX Is this necessary? As far as I'm aware an unregisterised build > currently works. > > hunk ./compiler/ghc.cabal.in 101 > + if flag(llvm) > + CPP-Options: -DLLVM > + if flag(unreg) > + CPP-Options: -DNO_REGS -DUSE_MINIINTERPRETER > > XXX Why is this necessary? These CPP flags are added by GHC itself > (StaticFlagParser.unregFlags). I did this at the time as I was finding that unregistered builds of GHC still used the BaseReg. All other registers were disabled but the BaseReg was still being used. > > +type LMString = String > > OPT You probably want to use something better than String, perhaps > FastString. > > +-- | Llvm variables > +data LlvmVar > + -- references to variables with a global scope. > + = LMGlobalVar LMString LlvmType LlvmLinkageType > + -- references to variables local for a function or parameters. > + | LMLocalVar LMString LlvmType > + -- a constant variable > + | LMLitVar LlvmLit > > OPT Rather than LMString in LMLocalVar, probably Unique would be > better. I think you always generate these from a Unique (but I could > be wrong) in which case it would be better to do that conversion when > printing. I'm not sure whether this applies to LMGlobalVar or not. > Yes except for one case. The STG virtual registers names are hard coded. E.g -------- define fastcc void @Main_main2_entry(i32 %stg_terei_baseArg,i32 %stg_terei_spArg,i32 %stg_terei_hpArg,i32 %stg_terei_r1Arg) nounwind { cLD: %stg_terei_baseReg = alloca i32, i32 1 %stg_terei_spReg = alloca i32, i32 1 %stg_terei_hpReg = alloca i32, i32 1 %stg_terei_r1Reg = alloca i32, i32 1 store i32 %stg_terei_baseArg, i32* %stg_terei_baseReg store i32 %stg_terei_spArg, i32* %stg_terei_spReg store i32 %stg_terei_hpArg, i32* %stg_terei_hpReg store i32 %stg_terei_r1Arg, i32* %stg_terei_r1Reg -------- Although this can be handled very easily. > + deriving (Eq) > +fixAssignsTop :: RawCmmTop -> UniqSM RawCmmTop > + > +fixAssignsTop top@(CmmData _ _) = returnUs top > + > +fixAssignsTop (CmmProc info lbl params (ListGraph blocks)) = > + mapUs fixAssignsBlock blocks `thenUs` \ blocks' -> > + returnUs (CmmProc info lbl params (ListGraph blocks')) > + > > CLEAN I think your local fixAssigns can go away (see comment with > LlvmCodeGen.Regs below): use the one from the NCG. > > +cmmToCmm :: RawCmmTop -> RawCmmTop > +cmmToCmm top@(CmmData _ _) = top > +cmmToCmm (CmmProc info lbl params (ListGraph blocks)) = > + let blocks' = map cmmBlockConFold (cmmMiniInline blocks) > + blocks'' = map cmmAddReturn blocks' > + in CmmProc info lbl params (ListGraph $ blocks'') > + > > CLEAN I think your local cmmToCmm can go away (see comment with > LlvmCodeGen.Regs below): use the one from the NCG. We should also > look into not doing any optimisation at the C-- level at all, since > LLVM will do it all, however we do need to turn GlobalRegs into > BaseReg offsets where necessary, which is done by cmmToCmm. Sure. I just copied the code from the NCG originally as I wanted to keep my work as self contained as possible to avoid any issues with tracking GHC head. Most the changes you suggest are ones that I was aware of but my guiding principle was that developer time was much more important then compile time due to me having a hard deadline for the work. > > +type LlvmEnvMap = Map.Map LMString LlvmType > > OPT A Map of Strings is very inefficient compared with a UniqFM of > FastStrings. > > +type StmtData = (LlvmEnv, [LlvmStatement], [LlvmCmmTop]) > > OPT Use OrdList rather than lists here, as you're doing a lot of > appending. Append is O(1) in OrdList, O(n) with lists. > > +type ExprData = (LlvmEnv, LlvmVar, LlvmStatements, [LlvmCmmTop]) > > OPT Again, LlvmStatements should probably be an OrdList. Yep. Thought this was the case but went with what I knew for developer speed. > > addfile ./compiler/llvmGen/LlvmCodeGen/Regs.hs > hunk ./compiler/llvmGen/LlvmCodeGen/Regs.hs 1 > +-- > ---------------------------------------------------------------------------- > > +-- Deal with Cmm registers > +-- > +-- Specifically, force use of register table (in memory), not pinned > +-- hardware registers as the LLVM back-end doesn't support this. > +-- > + > +module LlvmCodeGen.Regs ( > + > + RealReg, realRegsOrdered, getRealRegReg, getRealRegArg, getGlobalRegAddr > + > + ) where > > CLEAN Unless I'm mistaken, most of this module can go away. > > 1. Use fixAssigns from the NCG. > 2. Print out GlobalRegs as "stg_whatever..." > > Then you don't need to bake in the register mapping (a maintenance > headache), and a lot of this code can go away: the RealReg type and > everything using it. > I have to think about this one. Regs can defiantly be cleaned up, no question I'm just not sure about removing RealReg. Its purpose right now is three things: 1) 'getGlobalRegAddr' makes STG registers into real registers or table offsets. 2) STG registers as said have hard coded names in the code the LLVM back-end generates. 'getRealRegReg', 'getRealRegArg'. These methods return the correctly named LLVM local register for a particular STG register. 3) realRegsOrdered encodes the calling convention for this target. E.g for x86-32 = [RR_Base, RR_Sp, RR_Hp, RR_R1]. This encodes the number and order of the registers which are part of the calling convention. Here is the LLVM code for x86-32 CCIfType<[i32], CCAssignToReg<[EBX, EBP, EDI, ESI]>>, So its quite simple really, just the number of registers pinned and the order they should be used as arguments. The first two purposes can be removed. I'm just not so sure about the third at the moment. Will look into this more, have to check what's already provided by NCG again. > > addfile ./compiler/llvmGen/NOTES > hunk ./compiler/llvmGen/NOTES 1 > +LLVM Back-end for GHC. > + > +Author: David Terei > > CLEAN Useful-looking docs. I suggest moving them into either (a) the > wiki commentary or (b) the code as appropriate. > > hunk ./compiler/main/CodeOutput.lhs 16 > +#ifdef LLVM > +#ifdef OMIT_NATIVE_CODEGEN > +import UniqSupply ( mkSplitUniqSupply ) > +#endif > +import qualified LlvmCodeGen ( llvmCodeGen ) > +#endif > + > hunk ./compiler/main/CodeOutput.lhs 90 > + HscLlvm -> outputLlvm dflags filenm flat_abstractC; > hunk ./compiler/main/CodeOutput.lhs 176 > +%************************************************************************ > +%* * > +\subsection{LLVM} > +%* * > +%************************************************************************ > + > +\begin{code} > +outputLlvm :: DynFlags -> FilePath -> [RawCmm] -> IO () > + > +#ifdef LLVM > + > +outputLlvm dflags filenm flat_absC > + = do ncg_uniqs <- mkSplitUniqSupply 'n' > + doOutput filenm $ \f -> > + LlvmCodeGen.llvmCodeGen dflags f ncg_uniqs flat_absC > + > +#else /* !LLVM */ > > CLEAN suggest removing the #ifdefs and enabling LLVM unconditionally, as > above. As above, TNTC problem means LLVM should be disabled by default. Thanks for the feedback, will start working on this when I have time, away this weekend so I expect will take to the weekend following or so. Cheers, David From davidterei at gmail.com Wed Feb 24 21:47:14 2010 From: davidterei at gmail.com (David Terei) Date: Wed Feb 24 21:17:38 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B83F293.6020807@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <4B83F293.6020807@gmail.com> Message-ID: <4B85E4B2.9090101@gmail.com> Simon Marlow wrote: > Here's a patch review. Oh there is one other issue I keep forgetting about. LLVM requires that all basic blocks end in a control flow statement. Cmm basic blocks have this property for everything except the hand written cmm. e.g rts/Apply.cmm has the code: INFO_TABLE(stg_PAP,/*special layout*/0,0,PAP,"PAP","PAP") { foreign "C" barf("PAP object entered!") never returns; } This get passed to back-end as on x86-64: stg_PAP_entry() { has static closure: False update_frame: type: 0 desc: 0 tag: 26 ptrs: 0 nptrs: 0 srt: _no_srt_ } cp: I64[BaseReg + 16] = R3; I64[BaseReg + 24] = R4; I64[BaseReg + 32] = R5; I64[BaseReg + 40] = R6; F32[BaseReg + 80] = F1; F32[BaseReg + 84] = F2; F32[BaseReg + 88] = F3; F32[BaseReg + 92] = F4; F64[BaseReg + 96] = D1; F64[BaseReg + 104] = D2; foreign "ccall" barf((cn_str, PtrHint))[_unsafe_call_] never returns; R3 = I64[BaseReg + 16]; R4 = I64[BaseReg + 24]; R5 = I64[BaseReg + 32]; R6 = I64[BaseReg + 40]; F1 = F32[BaseReg + 80]; F2 = F32[BaseReg + 84]; F3 = F32[BaseReg + 88]; F4 = F32[BaseReg + 92]; D1 = F64[BaseReg + 96]; D2 = F64[BaseReg + 104]; } This case is the only one that occurs where a Cmm block doesn't end with a control flow statement and its only really since it does a 'never returns' call before hand with exits the function for good. This can be handled in two ways: 1) Have a pass in the LLVM back-end which checks each basic block and adds an assumed 'return void' at the end if it doesn't end with a control flow statement. 2) Modify 'compiler/codeGen/CgForeignCall.hs', changing the function emitForeignCall' as so: -- alternative entry point, used by CmmParse emitForeignCall' :: Safety -> HintedCmmFormals -- where to put the results -> CmmCallTarget -- the op -> [CmmHinted CmmExpr] -- arguments -> Maybe [GlobalReg] -- live vars, in case we save them -> C_SRT -- the SRT of the calls continuation -> CmmReturnInfo -> Code emitForeignCall' safety results target args vols _srt ret | not (playSafe safety) = do temp_args <- load_args_into_temps args let (caller_save, caller_load) = callerSaveVolatileRegs vols + let caller_load' = if ret == CmmNeverReturns then [] else caller_load stmtsC caller_save stmtC (CmmCall target results temp_args CmmUnsafe ret) - stmtsC caller_load + stmtsC caller_load' This stops caller save registers being restored if the call is never meant to return, which should be fine since the code is dead code anyway. For the moment I've handled it in the first way as I felt my patch had a better chance of being merged if it changed existing code in GHC as little as possible. I prefer the second way though as adding in 'return void' to cmm basic blocks feels like a hack and the second approach give Cmm the property that all blocks end in a control flow statement which seems pretty useful to me. So should I change to the second approach, keep the first or start all over again? Cheers, David From simonpj at microsoft.com Thu Feb 25 04:14:46 2010 From: simonpj at microsoft.com (Simon Peyton-Jones) Date: Thu Feb 25 03:45:29 2010 Subject: patch applied (ghc): Make "make tags" work in the new build system In-Reply-To: <20100224185636.GA28070@matrix.chaos.earth.li> References: <20100221172858.GA15637@abbot.galois.com> <59543203684B2244980D7E4057D5FBC10B08F069@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224140511.GA2994@matrix.chaos.earth.li> <59543203684B2244980D7E4057D5FBC10B08F3BB@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224154540.GB5117@matrix.chaos.earth.li> <6147FF6B-2779-4066-BC88-F0584B4FC27D@cs.york.ac.uk> <59543203684B2244980D7E4057D5FBC10B08F4B3@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224185636.GA28070@matrix.chaos.earth.li> Message-ID: <59543203684B2244980D7E4057D5FBC10B08FA68@DB3EX14MBXC312.europe.corp.microsoft.com> | > good point. Could we call them 'tags-emacs' and 'tags-vim'? | | We could do, but tags and TAGS are the standard names. | | We could also have separate "make tags" and "make TAGS" commands, which | generate only the named file. | | Whatever the decision, implementing it will be easy. | Does it matter whether it's standard? Esp if the standard doesn't work on Windows and MacOS. The file names are self-describing, and the 'make' target could emit a message to say what the outputs are called. It'd be good to decide one way or t'other and clear it away. Does anyone else care? (This can't be a new problem.) S From tora at zonetora.co.uk Thu Feb 25 04:24:59 2010 From: tora at zonetora.co.uk (Tristan Allwood) Date: Thu Feb 25 03:55:21 2010 Subject: patch applied (ghc): Make "make tags" work in the new build system In-Reply-To: <59543203684B2244980D7E4057D5FBC10B08FA68@DB3EX14MBXC312.europe.corp.microsoft.com> References: <20100221172858.GA15637@abbot.galois.com> <59543203684B2244980D7E4057D5FBC10B08F069@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224140511.GA2994@matrix.chaos.earth.li> <59543203684B2244980D7E4057D5FBC10B08F3BB@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224154540.GB5117@matrix.chaos.earth.li> <6147FF6B-2779-4066-BC88-F0584B4FC27D@cs.york.ac.uk> <59543203684B2244980D7E4057D5FBC10B08F4B3@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224185636.GA28070@matrix.chaos.earth.li> <59543203684B2244980D7E4057D5FBC10B08FA68@DB3EX14MBXC312.europe.corp.microsoft.com> Message-ID: <20100225092459.GA8088@doc.ic.ac.uk> +1 for having separate 'make tags' and 'make TAGS'. Cheers, Tris On Thu, Feb 25, 2010 at 09:14:46AM +0000, Simon Peyton-Jones wrote: > | > good point. Could we call them 'tags-emacs' and 'tags-vim'? > | > | We could do, but tags and TAGS are the standard names. > | > | We could also have separate "make tags" and "make TAGS" commands, which > | generate only the named file. > | > | Whatever the decision, implementing it will be easy. > | > > Does it matter whether it's standard? Esp if the standard doesn't work on Windows and MacOS. The file names are self-describing, and the 'make' target could emit a message to say what the outputs are called. > > It'd be good to decide one way or t'other and clear it away. Does anyone else care? > > (This can't be a new problem.) > > S > > _______________________________________________ > Cvs-ghc mailing list > Cvs-ghc@haskell.org > http://www.haskell.org/mailman/listinfo/cvs-ghc -- Tristan Allwood PhD Student Department of Computing Imperial College London From marlowsd at gmail.com Thu Feb 25 05:02:52 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Thu Feb 25 04:33:29 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B85BF6D.4020100@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <4B8279BE.6000909@gmail.com> <4B82B583.5040103@gmail.com> <4B85568C.6020306@gmail.com> <4B85BF6D.4020100@gmail.com> Message-ID: <4B864ACC.3020905@gmail.com> On 25/02/2010 00:08, David Terei wrote: > Simon Marlow wrote: > > So, nofib programs are on average 4-5% slower, with a 6% increase in > > binary sizes (this is with -split-objs). > > > > What's interesting is that there are some outliers here: programs that > > go 12-17% slower without TNTC. Looking at David's thesis the results are > > quite different. Perhaps it is processor dependent? > > I ran nofib again and am getting results similar to yours. I have > all the nofib runs laying around from my thesis and they give the > results I put in it. Seems to perhaps be a mistake from me that the TNTC > enabled runs weren't optimal due to some background process I didn't > notice. Also there were some nofib programs I didn't include at that > time due to problems getting them running, the missing ones seems to > have reduced the advantage of TNTC. > > > I have to admit, the results I'm getting are what I would expect, so its > > conceivable that some subconscious experimental bias has crept in > > somewhere :-) > > Actually its more of an unconscious self sabotage. When I did the > benchmarking of TNTC I was hoping to get the 4-6% value that I had seen > talked of before since that's basically the margin by which the LLVM > back-end is slower then the NCG. I think my body was punishing me from > lack of sleep :). Ok, at least that's cleared up. > > [snip/new email] >> So here's a crazy idea. Why don't we post-process the assembly code >> coming out of LLVM? Before you throw up your hands in horror, consider >> that > > I think it should work pretty well, especially if its intentioned to > just be a stopgap solution until we get support for TNTC from LLVM. Great! Cheers, Simon From marlowsd at gmail.com Thu Feb 25 05:08:47 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Thu Feb 25 04:39:22 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <71FA1768-BC25-40B3-8017-0244439E4B16@cse.unsw.edu.au> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <4B8279BE.6000909@gmail.com> <4B82B583.5040103@gmail.com> <4B85568C.6020306@gmail.com> <71FA1768-BC25-40B3-8017-0244439E4B16@cse.unsw.edu.au> Message-ID: <4B864C2F.80804@gmail.com> On 25/02/2010 01:15, Manuel M T Chakravarty wrote: > The most interesting observation seems to be that the results are > largely inconsistent (look at 'atom', for example). Even > wheel-sieve1, where our numbers go in the same direction, varies > between +10% and +22.7%. My numbers are from an older HEAD (as I had > nofib handy there), but I'm still wondering where this difference > comes from. Is it just that different processors are more or less > sensible to TNTC? But then it doesn't seem that one processor is > consistently less affected, but it also seems to depend on the > program. I should mention that for my tests I had the libraries compiled with -O2 -fasm, but the programs themselves compiled with -O (for no particular reason). Presumably you're using different optimisation settings, which may account for some of the difference. Also, I expect the effect depends a lot on how well the processor's branch-prediction engine is able to cope with the extra indirect jumps that arise when TNTC is turned off. Predicting indirect jumps is one area where processors still differ quite a lot. Many of these programs have fairly small inner loops, and the effect of consistently mispredicting or predicting one jump could have a large effect. Cheers, Simon From marlowsd at gmail.com Thu Feb 25 05:24:46 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Thu Feb 25 04:55:20 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B85D9D5.5090804@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <4B83F293.6020807@gmail.com> <4B85D9D5.5090804@gmail.com> Message-ID: <4B864FEE.6070102@gmail.com> On 25/02/2010 02:00, David Terei wrote: >> Do we envisage doing anything complex to the Llvm code inside GHC in the >> future? I can't think of a reason to. > > One thing I would like to investigate at some point is to have the LLVM > binding use the C++ API of LLVM instead of printing out LLVM Assembly. > This should potentially speed up the compilation process since it would > replace the current 4 passes (generate LLVM Assembly, convert to bitcode > [llvm-as], optimise [opt], compile to native code [llc]). With just one. > I believe the current design would be easier to extend for this purpose > then the simplified one you propose. After trying this out I'd be happy > to simplify the LLVM binding if we stick with the current approach of > generating LLVM assembly and calling the tools. That sounds like a good plan. But in that case wouldn't it make sense to use the LLVM binding from Hackage? >> hunk ./compiler/ghc.cabal.in 44 >> +Flag llvm >> + Description: Build the LLVM code generator. >> + Default: False >> + Manual: True >> >> CLEAN Is there a good reason to want to disable the LLVM support, or >> can it always be compiled in? > > Since the LLVM back-end requires TNTC to be disabled, yes it should be > disable by default. The point I wanted to make was that there's no harm in compiling in the LLVM support, even if it can't be used (the flag itself should be disabled, of course). The benefit is that compilation errors in the LLVM backend code can't creep in, because it's part of validate. We do the same thing for the native backends, incedentally. >> >> +Flag unreg >> + Description: Build an unregistered version. >> + Default: False >> + Manual: True >> + >> >> XXX Is this necessary? As far as I'm aware an unregisterised build >> currently works. >> >> hunk ./compiler/ghc.cabal.in 101 >> + if flag(llvm) >> + CPP-Options: -DLLVM >> + if flag(unreg) >> + CPP-Options: -DNO_REGS -DUSE_MINIINTERPRETER >> >> XXX Why is this necessary? These CPP flags are added by GHC itself >> (StaticFlagParser.unregFlags). > > I did this at the time as I was finding that unregistered builds of GHC > still used the BaseReg. All other registers were disabled but the > BaseReg was still being used. That shouldn't happen. In any case, let's leave out this unreg stuff and investigate the problem separately. >> +-- Deal with Cmm registers >> +-- >> +-- Specifically, force use of register table (in memory), not pinned >> +-- hardware registers as the LLVM back-end doesn't support this. >> +-- >> + >> +module LlvmCodeGen.Regs ( >> + >> + RealReg, realRegsOrdered, getRealRegReg, getRealRegArg, >> getGlobalRegAddr >> + >> + ) where >> >> CLEAN Unless I'm mistaken, most of this module can go away. >> >> 1. Use fixAssigns from the NCG. >> 2. Print out GlobalRegs as "stg_whatever..." >> >> Then you don't need to bake in the register mapping (a maintenance >> headache), and a lot of this code can go away: the RealReg type and >> everything using it. >> > > I have to think about this one. Regs can defiantly be cleaned up, no > question I'm just not sure about removing RealReg. Its purpose right now > is three things: > > 1) 'getGlobalRegAddr' makes STG registers into real registers or table > offsets. > > 2) STG registers as said have hard coded names in the code the LLVM > back-end generates. 'getRealRegReg', 'getRealRegArg'. These methods > return the correctly named LLVM local register for a particular STG > register. > > 3) realRegsOrdered encodes the calling convention for this target. > > E.g for x86-32 = [RR_Base, RR_Sp, RR_Hp, RR_R1]. > > This encodes the number and order of the registers which are part of the > calling convention. Here is the LLVM code for x86-32 > > CCIfType<[i32], CCAssignToReg<[EBX, EBP, EDI, ESI]>>, > > So its quite simple really, just the number of registers pinned and the > order they should be used as arguments. > > The first two purposes can be removed. I'm just not so sure about the > third at the moment. Will look into this more, have to check what's > already provided by NCG again. Ok, so let me be a bit clearer about what I'm proposing. If you use share fixAssigns and cmmToCmm with the NCG, then any GlobalReg which is not assigned to a machine register on the target architecture gets transformed into a BaseReg[n] access. So you're left with Cmm code containing references to only machine-reg GlobalRegs. In the LLVM backend you just apply a simple transformation to turn GlobalRegs into LLVM variables, e.g. CmmGlobal Sp -> %stg_terei_spReg You're right that function definitions still need to list the registers in the architecture-specific ordering, but I don't think you need your own RealReg type: just use GlobalReg. Cheers, Simon From marlowsd at gmail.com Thu Feb 25 05:36:41 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Thu Feb 25 05:07:15 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B85E4B2.9090101@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <4B83F293.6020807@gmail.com> <4B85E4B2.9090101@gmail.com> Message-ID: <4B8652B9.70403@gmail.com> On 25/02/2010 02:47, David Terei wrote: > Simon Marlow wrote: >> Here's a patch review. > > Oh there is one other issue I keep forgetting about. LLVM requires that > all basic blocks end in a control flow statement. Cmm basic blocks have > this property for everything except the hand written cmm. > > e.g rts/Apply.cmm has the code: > > INFO_TABLE(stg_PAP,/*special layout*/0,0,PAP,"PAP","PAP") > { foreign "C" barf("PAP object entered!") never returns; } > > This get passed to back-end as on x86-64: > > stg_PAP_entry() > { has static closure: False update_frame: > type: 0 > desc: 0 > tag: 26 > ptrs: 0 > nptrs: 0 > srt: _no_srt_ > } > cp: I64[BaseReg + 16] = R3; > I64[BaseReg + 24] = R4; > I64[BaseReg + 32] = R5; > I64[BaseReg + 40] = R6; > F32[BaseReg + 80] = F1; > F32[BaseReg + 84] = F2; > F32[BaseReg + 88] = F3; > F32[BaseReg + 92] = F4; > F64[BaseReg + 96] = D1; > F64[BaseReg + 104] = D2; > foreign "ccall" > barf((cn_str, PtrHint))[_unsafe_call_] > never returns; > R3 = I64[BaseReg + 16]; > R4 = I64[BaseReg + 24]; > R5 = I64[BaseReg + 32]; > R6 = I64[BaseReg + 40]; > F1 = F32[BaseReg + 80]; > F2 = F32[BaseReg + 84]; > F3 = F32[BaseReg + 88]; > F4 = F32[BaseReg + 92]; > D1 = F64[BaseReg + 96]; > D2 = F64[BaseReg + 104]; > } > > This case is the only one that occurs where a Cmm block doesn't end with > a control flow statement and its only really since it does a 'never > returns' call before hand with exits the function for good. I believe the new C-- is going to "fix" this, because it has a similar constraint to LLVM. You can see the new form of C-- in compiler/cmm/ZipCfgCmmRep.hs. The plan is to switch over to this new form for the input to the NCG, and presumably the LLVM backend at some point. > This can be handled in two ways: > > 1) Have a pass in the LLVM back-end which checks each basic block and > adds an assumed 'return void' at the end if it doesn't end with a > control flow statement. > > 2) Modify 'compiler/codeGen/CgForeignCall.hs', changing the function > emitForeignCall' as so: > > -- alternative entry point, used by CmmParse > emitForeignCall' > :: Safety > -> HintedCmmFormals -- where to put the results > -> CmmCallTarget -- the op > -> [CmmHinted CmmExpr] -- arguments > -> Maybe [GlobalReg] -- live vars, in case we save them > -> C_SRT -- the SRT of the calls continuation > -> CmmReturnInfo > -> Code > emitForeignCall' safety results target args vols _srt ret > | not (playSafe safety) = do > temp_args <- load_args_into_temps args > let (caller_save, caller_load) = callerSaveVolatileRegs vols > + let caller_load' = if ret == CmmNeverReturns then [] else caller_load > stmtsC caller_save > stmtC (CmmCall target results temp_args CmmUnsafe ret) > - stmtsC caller_load > + stmtsC caller_load' > > This stops caller save registers being restored if the call is never > meant to return, which should be fine since the code is dead code anyway. > > For the moment I've handled it in the first way as I felt my patch had a > better chance of being merged if it changed existing code in GHC as > little as possible. I prefer the second way though as adding in 'return > void' to cmm basic blocks feels like a hack and the second approach give > Cmm the property that all blocks end in a control flow statement which > seems pretty useful to me. > > So should I change to the second approach, keep the first or start all > over again? Either approach is fine for now, I wouldn't worry about changing it, I think the new code generator handles this slightly differently, by chopping off the code after the call later. It appears to be done by CmmCvt.toZGraph. Cheers, Simon From marlowsd at gmail.com Thu Feb 25 05:41:59 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Thu Feb 25 05:12:33 2010 Subject: patch applied (ghc): Make "make tags" work in the new build system In-Reply-To: <59543203684B2244980D7E4057D5FBC10B08FA68@DB3EX14MBXC312.europe.corp.microsoft.com> References: <20100221172858.GA15637@abbot.galois.com> <59543203684B2244980D7E4057D5FBC10B08F069@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224140511.GA2994@matrix.chaos.earth.li> <59543203684B2244980D7E4057D5FBC10B08F3BB@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224154540.GB5117@matrix.chaos.earth.li> <6147FF6B-2779-4066-BC88-F0584B4FC27D@cs.york.ac.uk> <59543203684B2244980D7E4057D5FBC10B08F4B3@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224185636.GA28070@matrix.chaos.earth.li> <59543203684B2244980D7E4057D5FBC10B08FA68@DB3EX14MBXC312.europe.corp.microsoft.com> Message-ID: <4B8653F7.3070800@gmail.com> On 25/02/2010 09:14, Simon Peyton-Jones wrote: > |> good point. Could we call them 'tags-emacs' and 'tags-vim'? > | > | We could do, but tags and TAGS are the standard names. > | > | We could also have separate "make tags" and "make TAGS" commands, which > | generate only the named file. > | > | Whatever the decision, implementing it will be easy. > | > > Does it matter whether it's standard? Esp if the standard doesn't work on Windows and MacOS. The file names are self-describing, and the 'make' target could emit a message to say what the outputs are called. One advantage of using the standard name is that it's the default in emacs. If we use a non-standard name, everyone has to type it in (and know what it's called) rather than just hitting return. > It'd be good to decide one way or t'other and clear it away. Does anyone else care? > > (This can't be a new problem.) Previously we only generated TAGS, as far as I can tell. I wouldn't object if that continued to be the case :-) Cheers, Simon From davidterei at gmail.com Thu Feb 25 05:57:45 2010 From: davidterei at gmail.com (David Terei) Date: Thu Feb 25 05:28:12 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B864FEE.6070102@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <4B83F293.6020807@gmail.com> <4B85D9D5.5090804@gmail.com> <4B864FEE.6070102@gmail.com> Message-ID: <4B8657A9.8040706@gmail.com> Simon Marlow wrote: > On 25/02/2010 02:00, David Terei wrote: > >>> Do we envisage doing anything complex to the Llvm code inside GHC in the >>> future? I can't think of a reason to. >> >> One thing I would like to investigate at some point is to have the LLVM >> binding use the C++ API of LLVM instead of printing out LLVM Assembly. >> This should potentially speed up the compilation process since it would >> replace the current 4 passes (generate LLVM Assembly, convert to bitcode >> [llvm-as], optimise [opt], compile to native code [llc]). With just one. >> I believe the current design would be easier to extend for this purpose >> then the simplified one you propose. After trying this out I'd be happy >> to simplify the LLVM binding if we stick with the current approach of >> generating LLVM assembly and calling the tools. > > That sounds like a good plan. But in that case wouldn't it make sense to > use the LLVM binding from Hackage? Probably, sorry I didn't specify it but that was the plan I had. My point was more that I think its better to leave the LLVM binding as it is for now since the plan I proposed above seems the better way forward than optimising the current design. Cheers, David From marlowsd at gmail.com Thu Feb 25 06:05:36 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Thu Feb 25 05:36:09 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B8657A9.8040706@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <4B83F293.6020807@gmail.com> <4B85D9D5.5090804@gmail.com> <4B864FEE.6070102@gmail.com> <4B8657A9.8040706@gmail.com> Message-ID: <4B865980.6040900@gmail.com> On 25/02/2010 10:57, David Terei wrote: > Simon Marlow wrote: >> On 25/02/2010 02:00, David Terei wrote: >> >>>> Do we envisage doing anything complex to the Llvm code inside GHC in >>>> the >>>> future? I can't think of a reason to. >>> >>> One thing I would like to investigate at some point is to have the LLVM >>> binding use the C++ API of LLVM instead of printing out LLVM Assembly. >>> This should potentially speed up the compilation process since it would >>> replace the current 4 passes (generate LLVM Assembly, convert to bitcode >>> [llvm-as], optimise [opt], compile to native code [llc]). With just one. >>> I believe the current design would be easier to extend for this purpose >>> then the simplified one you propose. After trying this out I'd be happy >>> to simplify the LLVM binding if we stick with the current approach of >>> generating LLVM assembly and calling the tools. >> >> That sounds like a good plan. But in that case wouldn't it make sense to >> use the LLVM binding from Hackage? > > Probably, sorry I didn't specify it but that was the plan I had. My > point was more that I think its better to leave the LLVM binding as it > is for now since the plan I proposed above seems the better way forward > than optimising the current design. Sure, I buy that. Cheers, Simon From batterseapower at hotmail.com Thu Feb 25 06:09:20 2010 From: batterseapower at hotmail.com (Max Bolingbroke) Date: Thu Feb 25 05:39:43 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <4B7DD35C.2010408@gmail.com> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> Message-ID: <9d4d38821002250309w39727f14jed080d197f2aec70@mail.gmail.com> Hi David, This is great stuff of course! I notice that you have identified some problems with the generated code (e.g. the ESP manipulation stuff you point to at the end of http://groups.google.com/group/llvm-dev/msg/28b99513bcc38e0f). Do you think there are some quick wins here that could be had by writing some optimisation pass plugins for LLVM that tackle the kind of performance bugs commonly seen by code originating from GHC? Is so, this might be a suitable project for a Summer of Code student with some previous knowledge of compiler design. Cheers, Max On 18 February 2010 23:55, David Terei wrote: > Hi all, > > Over the last 6 months I've been working on a new code generator for GHC > which targets the LLVM compiler infrastructure. Most of the work was done as > part of an honours thesis at the University of New South Wales under the > supervision of Manuel Chakravarty. This ended at the start of November but I > have continued to work on the code generator since (although at a much > reduced pace since I'm back at full time work). Its now at a stage where I > feel pretty confident in its correctness and would love to see it merged > into GHC mainline. > > The patch for the llvm back-end can be found here (should apply cleanly to > GHC head): > > http://www.cse.unsw.edu.au/~davidt/downloads/ghc-llvmbackend-full.gz > > This is what I would like/am requesting merged into GHC head. > > The thesis paper which offers a detailed performance evaluation, as well as > the motivation and design of the back-end can be found at: > > http://www.cse.unsw.edu.au/~pls/thesis/davidt-thesis.pdf > > Below I'll quickly detail out the important points though. There are also > instructions on how to get started with the back-end. > > Finally there are also some issues that I think may need to be sorted out > before a merge could be done. They are at the end. > > Performance > ----------- > (All done on linux/x86-32) > > A quick summary of the results are that for the 'nofib' benchmark suite, the > llvm code generator was 3.8% slower than the NCG (the C code generator was > 6.9% slower than the NCG). The DPH project includes a benchmark suite which > I also ran and for this type of code using the llvm back-end shortened the > runtime by an average of 25% compared to the NCG. Also, while not included > in my thesis paper as I ran out of time, I did do some benchmarking with the > 'nobench' benchmark suite. It gave performance ratios for the back-ends of > around: > > NCG : 1.11 > C : 1.05 > LLVM : 1.14 > > > Supported Platforms & 'Correctness' > ----------------------------------- > > Linux x86-32/x86-64 are currently well supported. The back-end can pass the > test suite and build a working version of GHC (bootstrap test). > > Mac OS X 10.5 currently has a rather nasty bug with any dynamic lib calls > (all libffi stuff) [due to the stack not being 16byte aligned when the calls > are made as required by OSX ABI for the curious]. Test suite passes except > for most the ffi tests. > > Other platforms haven't been tested at all. As using the back-end with a > registered build of GHC requires a modified version of LLVM, people wanting > to try it out on those platforms will need to either make the needed changes > to LLVM themselves, or use an unregistered build of GHC which will work with > a vanilla install of LLVM. (A patch for LLVM for x86 is linked to below.) > > Validate > -------- > > I've validated my GHC patch to make sure it won't break anything. This is > just compiling and running GHC normally but with the llvm back-end code > included. It doesn't actually test the llvm code generator, just makes sure > it hasn't broken the NCG or C code generator. > > Linux/x86-32: > > OVERALL SUMMARY for test run started at Do 18. Feb 11:21:48 EST 2010 > 2457 total tests, which gave rise to > 9738 test cases, of which > 0 caused framework failures > 7573 were skipped > > 2088 expected passes > 76 expected failures > 0 unexpected passes > 1 unexpected failures > > Unexpected failures: > user001(normal) > > Linux/x86-64: > > OVERALL SUMMARY for test run started at Thu 18 Feb 15:28:32 EST 2010 > 2458 total tests, which gave rise to > 9739 test cases, of which > 0 caused framework failures > 7574 were skipped > > 2087 expected passes > 77 expected failures > 0 unexpected passes > 1 unexpected failures > > Unexpected failures: > T1969(normal) > > Mac OS X 10.5/x86-32: > > OVERALL SUMMARY for test run started at Thu Feb 18 12:35:49 EST 2010 > 2458 total tests, which gave rise to > 9122 test cases, of which > 0 caused framework failures > 6959 were skipped > > 2085 expected passes > 76 expected failures > 0 unexpected passes > 2 unexpected failures > > Unexpected failures: > T1969(normal) > ffi005(optc) > > All of the test failures fail for me with a unmodified GHC head build as > well as when the llvm patch is included, so the llvm patch isn't introducing > any new failures. > > Installing > ---------- > > http://www.cse.unsw.edu.au/~davidt/downloads/ghc-llvmbackend-full.gz > > Apply the darcs patch linked above to GHC head. This will make some changes > across GHC, with the bulk of the new code ending up in 'compiler/llvmGen'. > > To build GHC you need to add two flags to build.mk, they are: > > GhcWithLlvmCodeGen = YES > GhcEnableTablesNextToCode = NO > > The llvm code generator doesn't support at this time the TABLES_NEXT_TO_CODE > optimisation due to limitations with LLVM. > > You will also need LLVM installed on your computer to use the back-end. > Version 2.6 or SVN trunk is supported. If you want to use the back-end in an > unregistered ghc build, then you can use a vanilla build of LLVM. However if > you want to use a registered ghc build (very likely) then you need to patch > LLVM for this to work. The patch for llvm can be found here: > > http://www.cse.unsw.edu.au/~davidt/downloads/llvm-ghc.patch > > LLVM is very easy to build and install. It can be done as follows: > > $ svn co http://llvm.org/svn/llvm-project/llvm/trunk llvm > $ cd llvm > $ patch -p0 -i ~/llvm-ghc.patch > $ ./configure --enable-optimized # probably also want to set --prefix > $ make > $ make install > > Just make sure this modified version of LLVM is on your path and takes > precedence over any other builds. > > Using > ----- > > Once GHC is built, you can trigger GHC to use the LLVM back-end with the > '-fllvm' flag. There is also a new '-ddump-llvm' which will dump out the > llvm IR code generated. (or use the '-keep-tmp-files' flag). > > 'ghc --info' should also now report that it includes the llvm code > generator. > > Issues > ------ > Issues that might need to be resolved before merging the patch: > > 1. Developed in isolation by 1 person with no Haskell knowledge at first. So > usual issues with that may apply, misused data structures, bad style... ect. > Criticisms of the code are very welcome. There are some specific notes on > what I think may be wrong with the code atm in 'compiler/llvmGen/NOTES'. > > 2. The back-end has a LLVM binding of sorts, this binding is similar in > design to say the Cmm representation used in GHC. It represents the LLVM > Assembly language using a collection of data types and can pretty print it > out correctly. This binding lives in the 'compiler/llvmGen/Llvm' folder. > Should this binding be split out into a separate library? > > 3. As mentioned above, LLVM needs to be patched to work with a registered > build of GHC. If the llvm back-end was merged, how would this be handled? I > would suggest simply carrying the patch with some instructions on how to use > it in the GHC repo. People using GHC head could be expected to grab the LLVM > source code and apply the patch themselves at this stage. > > 4. Finally this email is long. I need to put all this info into a web page > as well. > > Cheers, > David > > _______________________________________________ > Cvs-ghc mailing list > Cvs-ghc@haskell.org > http://www.haskell.org/mailman/listinfo/cvs-ghc > From Malcolm.Wallace at cs.york.ac.uk Thu Feb 25 06:26:31 2010 From: Malcolm.Wallace at cs.york.ac.uk (Malcolm Wallace) Date: Thu Feb 25 05:57:19 2010 Subject: Merge Request: LLVM Code Generator for GHC In-Reply-To: <71FA1768-BC25-40B3-8017-0244439E4B16@cse.unsw.edu.au> References: <4B7CF343.3020608@gmail.com> <4B7DD35C.2010408@gmail.com> <4B8279BE.6000909@gmail.com> <4B82B583.5040103@gmail.com> <4B85568C.6020306@gmail.com> <71FA1768-BC25-40B3-8017-0244439E4B16@cse.unsw.edu.au> Message-ID: <9D645895-146E-4DB6-BD27-A9710CD3E5DC@cs.york.ac.uk> > The most interesting observation seems to be that the results are > largely inconsistent (look at 'atom', for example). Even wheel- > sieve1, where our numbers go in the same direction, varies between > +10% and +22.7%. My numbers are from an older HEAD (as I had nofib > handy there), but I'm still wondering where this difference comes > from. Is it just that different processors are more or less > sensible to TNTC? There is a video of a Google Tech Talk on the topic of how inconsistent and confusing benchmark results arise. http://www.youtube.com/watch?v=DKVRkfXrBpg Well worth investing 50 minutes of your time, if you ever do any kind of performance benchmarking. Warning: you may be shocked at how easy it is for unforeseen and uncontrolled factors to obfuscate your measurements, to the point of indicating a gain instead of a loss, and vice versa. Something as random as library link-order can flip the sign... Regards, Malcolm From marlowsd at gmail.com Thu Feb 25 08:08:42 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Thu Feb 25 07:39:07 2010 Subject: patch applied (ghc): Force encoding to UTF-8 when writing individual .conf files Message-ID: <20100225130841.GA547@abbot.galois.com> Wed Feb 24 07:25:19 PST 2010 Simon Marlow * Force encoding to UTF-8 when writing individual .conf files M ./utils/ghc-pkg/Main.hs -14 +28 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100224152519-12142-96a29065ed03209bb5925f9f1a200d43af5f9708.gz From marlowsd at gmail.com Thu Feb 25 08:11:07 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Thu Feb 25 07:41:30 2010 Subject: patch applied (testsuite): arith012, arith008: use -msse2 on i386 if available Message-ID: <20100225131107.GA902@abbot.galois.com> Wed Feb 24 02:14:34 PST 2010 Simon Marlow * arith012, arith008: use -msse2 on i386 if available M ./tests/ghc-regress/numeric/should_run/all.T -13 +10 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100224101434-12142-f318978358b3de0c6c557c246b26d403a5f3c1a5.gz From ghcbuild at microsoft.com Thu Feb 25 12:34:31 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Thu Feb 25 12:34:32 2010 Subject: [nightly] 25-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100225173431.37809324208@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Thu Feb 25 18:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... failed; relevant barfage is below. GHC Version 6.13.20100224 No successfully built compilers to test; aborting. ------------------------------------------------------------------------ ------------------------------------------------------------------------ The last 30 lines of /playpen/simonmar/nightly/HEAD/logs/i386-unknown-linux-stage1 are ------------------------------------------------------------------------ ------------------------------------------------------------------------ -XCPP -XExistentialQuantification -XDeriveDataTypeable \ -ilibraries/Cabal \ -ilibraries/filepath \ -ilibraries/extensible-exceptions \ -ilibraries/hpc \ -ilibraries/binary/src \ -ilibraries/bin-package-db [ 1 of 27] Compiling Data.Binary.Builder ( libraries/binary/src/Data/Binary/Builder.hs, bootstrapping/Data/Binary/Builder.o ) [ 2 of 27] Compiling Data.Binary.Put ( libraries/binary/src/Data/Binary/Put.hs, bootstrapping/Data/Binary/Put.o ) [ 5 of 27] Compiling Data.Binary.Get ( libraries/binary/src/Data/Binary/Get.hs, bootstrapping/Data/Binary/Get.o ) [ 6 of 27] Compiling Data.Binary ( libraries/binary/src/Data/Binary.hs, bootstrapping/Data/Binary.o ) [ 7 of 27] Compiling Control.Exception.Extensible ( libraries/extensible-exceptions/Control/Exception/Extensible.hs, bootstrapping/Control/Exception/Extensible.o ) Warning: orphan instances: instance Data.Typeable.Typeable [GHC.IOBase.ExitCode] = Control.Exception.Extensible.$f15 [24 of 27] Compiling Distribution.InstalledPackageInfo.Binary ( libraries/bin-package-db/Distribution/InstalledPackageInfo/Binary.hs, bootstrapping/Distribution/InstalledPackageInfo/Binary.o ) [26 of 27] Compiling Version ( utils/ghc-pkg/Version.hs, bootstrapping/Version.o ) [27 of 27] Compiling Main ( utils/ghc-pkg/Main.hs, bootstrapping/Main.o ) utils/ghc-pkg/Main.hs:732:25: Not in scope: `utf8' utils/ghc-pkg/Main.hs:1154:27: Not in scope: `utf8' utils/ghc-pkg/Main.hs:1560:31: Not in scope: type constructor or class `TextEncoding' utils/ghc-pkg/Main.hs:1563:5: Not in scope: `hSetEncoding' gmake[1]: *** [utils/ghc-pkg/dist/build/ghc-pkg] Error 1 gmake: *** [inplace/bin/ghc-stage1] Error 2 Nightly run ended at Thu Feb 25 18:04:04 GMT 2010 From ghcbuild at microsoft.com Thu Feb 25 13:09:34 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Thu Feb 25 13:09:36 2010 Subject: [nightly] 25-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100225180934.492A7324208@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Thu Feb 25 18:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100223 **** Building stage 2 compiler ... failed; relevant barfage is below. **** building testsuite tools ... failed. **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (91 failures) **** running nofib (-O -fvia-C) ... ok. (91 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (91 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. (91 failures) **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Thu Feb 25 18:39:12 GMT 2010 ------------------------------------------------------------------------ ------------------------------------------------------------------------ The last 30 lines of /playpen/simonmar/nightly/STABLE/logs/i386-unknown-linux-stage2 are ------------------------------------------------------------------------ ------------------------------------------------------------------------ ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/ug-book.xml:18: parser error : Entity 'ffi-chap' not defined &ffi-chap; ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/ug-book.xml:19: parser error : Entity 'wrong' not defined &wrong; ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/ug-book.xml:20: parser error : Entity 'utils' not defined &utils; ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/ug-book.xml:21: parser error : Entity 'win32-dll' not defined &win32-dll; ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/ug-book.xml:22: parser error : Entity 'bugs' not defined &bugs; ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/users_guide.xml:10: parser error : Failure to process entity ug-book &ug-book; ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/users_guide.xml:10: parser error : Entity 'ug-book' not defined &ug-book; ^ unable to parse /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/users_guide.xml Error: xsltproc failed gmake[1]: *** [docs/users_guide/users_guide.ps] Error 1 gmake: *** [all] Error 2 real 1m18.538s user 1m15.448s sys 0m2.868s Nightly run ended at Thu Feb 25 18:39:12 GMT 2010 From ghcbuild at microsoft.com Thu Feb 25 13:32:36 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Thu Feb 25 13:32:38 2010 Subject: [nightly] 25-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100225183236.E76CE324441@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Thu Feb 25 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... failed; relevant barfage is below. GHC Version 6.13.20100224 **** building source distribution ... failed. No successfully built compilers to test; aborting. ------------------------------------------------------------------------ ------------------------------------------------------------------------ The last 30 lines of /64playpen/simonmar/nightly/HEAD-cam-04-unx/logs/x86_64-unknown-linux-stage1 are ------------------------------------------------------------------------ ------------------------------------------------------------------------ -XCPP -XExistentialQuantification -XDeriveDataTypeable \ -ilibraries/Cabal \ -ilibraries/filepath \ -ilibraries/extensible-exceptions \ -ilibraries/hpc \ -ilibraries/binary/src \ -ilibraries/bin-package-db [ 1 of 27] Compiling Data.Binary.Builder ( libraries/binary/src/Data/Binary/Builder.hs, bootstrapping/Data/Binary/Builder.o ) [ 2 of 27] Compiling Data.Binary.Put ( libraries/binary/src/Data/Binary/Put.hs, bootstrapping/Data/Binary/Put.o ) [ 5 of 27] Compiling Data.Binary.Get ( libraries/binary/src/Data/Binary/Get.hs, bootstrapping/Data/Binary/Get.o ) [ 6 of 27] Compiling Data.Binary ( libraries/binary/src/Data/Binary.hs, bootstrapping/Data/Binary.o ) [ 7 of 27] Compiling Control.Exception.Extensible ( libraries/extensible-exceptions/Control/Exception/Extensible.hs, bootstrapping/Control/Exception/Extensible.o ) Warning: orphan instances: instance Data.Typeable.Typeable [GHC.IOBase.ExitCode] = Control.Exception.Extensible.$f15 [24 of 27] Compiling Distribution.InstalledPackageInfo.Binary ( libraries/bin-package-db/Distribution/InstalledPackageInfo/Binary.hs, bootstrapping/Distribution/InstalledPackageInfo/Binary.o ) [26 of 27] Compiling Version ( utils/ghc-pkg/Version.hs, bootstrapping/Version.o ) [27 of 27] Compiling Main ( utils/ghc-pkg/Main.hs, bootstrapping/Main.o ) utils/ghc-pkg/Main.hs:732:25: Not in scope: `utf8' utils/ghc-pkg/Main.hs:1154:27: Not in scope: `utf8' utils/ghc-pkg/Main.hs:1560:31: Not in scope: type constructor or class `TextEncoding' utils/ghc-pkg/Main.hs:1563:5: Not in scope: `hSetEncoding' gmake[1]: *** [utils/ghc-pkg/dist/build/ghc-pkg] Error 1 gmake: *** [inplace/bin/ghc-stage1] Error 2 Nightly run ended at Thu Feb 25 19:02:13 GMT 2010 From ghcbuild at microsoft.com Thu Feb 25 13:59:08 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Thu Feb 25 13:59:10 2010 Subject: [nightly] 25-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100225185908.43848324498@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Thu Feb 25 19:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100223 **** Building stage 2 compiler ... failed; relevant barfage is below. **** building source distribution ... ok. **** building testsuite tools ... failed. **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (91 failures) **** running nofib (-O -fvia-C) ... ok. (91 failures) **** running nofib (-O -prof -auto-all) ... ok. (91 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (91 failures) **** running nofib (-fasm) ... ok. (91 failures) **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Thu Feb 25 19:28:46 GMT 2010 ------------------------------------------------------------------------ ------------------------------------------------------------------------ The last 30 lines of /64playpen/simonmar/nightly/STABLE-cam-04-unx/logs/x86_64-unknown-linux-stage2 are ------------------------------------------------------------------------ ------------------------------------------------------------------------ ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/ug-book.xml:18: parser error : Entity 'ffi-chap' not defined &ffi-chap; ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/ug-book.xml:19: parser error : Entity 'wrong' not defined &wrong; ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/ug-book.xml:20: parser error : Entity 'utils' not defined &utils; ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/ug-book.xml:21: parser error : Entity 'win32-dll' not defined &win32-dll; ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/ug-book.xml:22: parser error : Entity 'bugs' not defined &bugs; ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/users_guide.xml:10: parser error : Failure to process entity ug-book &ug-book; ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/users_guide.xml:10: parser error : Entity 'ug-book' not defined &ug-book; ^ unable to parse /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/users_guide.xml Error: xsltproc failed gmake[1]: *** [docs/users_guide/users_guide.ps] Error 1 gmake: *** [all] Error 2 real 0m42.763s user 0m41.252s sys 0m1.526s Nightly run ended at Thu Feb 25 19:28:46 GMT 2010 From ben.lippmeier at anu.edu.au Thu Feb 25 22:33:17 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Thu Feb 25 22:03:40 2010 Subject: patch applied (testsuite): Add DPH dotp test Message-ID: <20100226033317.GA4246@abbot.galois.com> Wed Feb 24 20:26:34 PST 2010 benl@ouroborus.net * Add DPH dotp test A ./tests/ghc-regress/dph/dotp/ A ./tests/ghc-regress/dph/dotp/DotPVect.hs A ./tests/ghc-regress/dph/dotp/Main.hs A ./tests/ghc-regress/dph/dotp/dotp.T A ./tests/ghc-regress/dph/dotp/dotp.stdout View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100225042634-bf82d-132c4e78b9f9514af1b0c3e26cadf603130d3ef9.gz From ben.lippmeier at anu.edu.au Thu Feb 25 22:33:19 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Thu Feb 25 22:03:46 2010 Subject: patch applied (testsuite): Prefix dph tests with 'dph-' to avoid name conflicts Message-ID: <20100226033319.GA4268@abbot.galois.com> Thu Feb 25 17:36:38 PST 2010 benl@ouroborus.net * Prefix dph tests with 'dph-' to avoid name conflicts ./tests/ghc-regress/dph/dotp/dotp.T -> ./tests/ghc-regress/dph/dotp/dph-dotp.T ./tests/ghc-regress/dph/dotp/dotp.stdout -> ./tests/ghc-regress/dph/dotp/dph-dotp.stdout ./tests/ghc-regress/dph/quickhull/quickhull.T -> ./tests/ghc-regress/dph/quickhull/dph-quickhull.T ./tests/ghc-regress/dph/quickhull/quickhull.stdout -> ./tests/ghc-regress/dph/quickhull/dph-quickhull.stdout M ./tests/ghc-regress/dph/dotp/dph-dotp.T -1 +1 M ./tests/ghc-regress/dph/quickhull/dph-quickhull.T -1 +1 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100226013638-bf82d-68613a7e88231467d8c11d7559178ed12167c873.gz From ben.lippmeier at anu.edu.au Thu Feb 25 22:33:20 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Thu Feb 25 22:03:47 2010 Subject: patch applied (testsuite): In dph-dotp, compare with result computed via regular list fns. Message-ID: <20100226033320.GA4290@abbot.galois.com> Thu Feb 25 17:47:00 PST 2010 benl@ouroborus.net * In dph-dotp, compare with result computed via regular list fns. M ./tests/ghc-regress/dph/dotp/Main.hs -3 +11 M ./tests/ghc-regress/dph/dotp/dph-dotp.T -1 +1 M ./tests/ghc-regress/dph/dotp/dph-dotp.stdout +1 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100226014700-bf82d-49fd14faac1567ee1cbb5ab48bd7f37cecd5f7d8.gz From ben.lippmeier at anu.edu.au Thu Feb 25 22:33:23 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Thu Feb 25 22:03:48 2010 Subject: patch applied (testsuite): Add DPH smvm test Message-ID: <20100226033323.GA4334@abbot.galois.com> Thu Feb 25 18:56:30 PST 2010 benl@ouroborus.net * Add DPH smvm test A ./tests/ghc-regress/dph/smvm/ A ./tests/ghc-regress/dph/smvm/Main.hs A ./tests/ghc-regress/dph/smvm/Makefile A ./tests/ghc-regress/dph/smvm/SMVMVect.hs A ./tests/ghc-regress/dph/smvm/dph-smvm.T A ./tests/ghc-regress/dph/smvm/dph-smvm.stdout A ./tests/ghc-regress/dph/smvm/test.dat View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100226025630-bf82d-fbc4a69cfa31bc0a992ed2c9353b42f1aa06e53b.gz From ben.lippmeier at anu.edu.au Thu Feb 25 22:33:22 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Thu Feb 25 22:03:49 2010 Subject: patch applied (testsuite): Add DPH primes test Message-ID: <20100226033321.GA4312@abbot.galois.com> Thu Feb 25 18:00:34 PST 2010 benl@ouroborus.net * Add DPH primes test A ./tests/ghc-regress/dph/primespj/ A ./tests/ghc-regress/dph/primespj/Main.hs A ./tests/ghc-regress/dph/primespj/PrimesVect.hs A ./tests/ghc-regress/dph/primespj/dph-primespj.T A ./tests/ghc-regress/dph/primespj/dph-primespj.stdout View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100226020034-bf82d-19f8ce2f4a1c504fc87b3aa1ee6ed23af50040bb.gz From ben.lippmeier at anu.edu.au Thu Feb 25 22:37:02 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Thu Feb 25 22:07:24 2010 Subject: patch applied (testsuite): Add missing Makefiles Message-ID: <20100226033702.GA4766@abbot.galois.com> Thu Feb 25 19:36:07 PST 2010 benl@ouroborus.net * Add missing Makefiles A ./tests/ghc-regress/dph/dotp/Makefile A ./tests/ghc-regress/dph/primespj/Makefile View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100226033607-bf82d-105f4d259b13c2ebe0cf6b66e8d49aaa7b19b811.gz From ben.lippmeier at anu.edu.au Thu Feb 25 23:58:28 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Thu Feb 25 23:28:50 2010 Subject: patch applied (testsuite): Add DPH solution for 1st Euler problem Message-ID: <20100226045828.GA7344@abbot.galois.com> Thu Feb 25 20:51:05 PST 2010 benl@ouroborus.net * Add DPH solution for 1st Euler problem A ./tests/ghc-regress/dph/sumnats/ A ./tests/ghc-regress/dph/sumnats/Main.hs A ./tests/ghc-regress/dph/sumnats/Makefile A ./tests/ghc-regress/dph/sumnats/SumNatsVect.hs A ./tests/ghc-regress/dph/sumnats/dph-sumnats.T A ./tests/ghc-regress/dph/sumnats/dph-sumnats.stdout View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100226045105-bf82d-4abd289812f601c2a6ad7365c2b30d6e93e566ba.gz From marlowsd at gmail.com Fri Feb 26 04:36:43 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Fri Feb 26 04:07:17 2010 Subject: patch applied (testsuite): Add DPH smvm test In-Reply-To: <20100226033323.GA4334@abbot.galois.com> References: <20100226033323.GA4334@abbot.galois.com> Message-ID: <4B87962B.7050007@gmail.com> On 26/02/2010 03:33, Ben Lippmeier wrote: > Thu Feb 25 18:56:30 PST 2010 benl@ouroborus.net > * Add DPH smvm test > > A ./tests/ghc-regress/dph/smvm/ > A ./tests/ghc-regress/dph/smvm/Main.hs > A ./tests/ghc-regress/dph/smvm/Makefile > A ./tests/ghc-regress/dph/smvm/SMVMVect.hs > A ./tests/ghc-regress/dph/smvm/dph-smvm.T > A ./tests/ghc-regress/dph/smvm/dph-smvm.stdout > A ./tests/ghc-regress/dph/smvm/test.dat I presume this isn't supposed to happen? :) (on x86-64/Linux) =====> dph-smvm(normal) 2 of 2464 [0, 0, 0] cd ./dph/smvm && '/64playpen/simonmar/validate/bindisttest/install dir/bin/ghc' -fforce-recomp -dcore-lint -dcmm-lint -no-user-package-conf -dno-debug-output --make -o dph-smvm Main -Odph -package dph-par -package dph-prim-par >dph-smvm.comp.stderr 2>&1 cd ./dph/smvm && ./dph-smvm dph-smvm.run.stdout 2>dph-smvm.run.stderr Wrong exit code (expected 0 , actual 1 ) Stdout: Stderr: dph-smvm: out of memory (requested 790275031040 bytes) From marlowsd at gmail.com Fri Feb 26 05:14:08 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Fri Feb 26 04:44:29 2010 Subject: patch applied (ghc): Fix crash when using printf format specifiers in traceEvent (#3874) Message-ID: <20100226101408.GA18133@abbot.galois.com> Fri Feb 26 01:32:15 PST 2010 Simon Marlow * Fix crash when using printf format specifiers in traceEvent (#3874) M ./rts/Trace.c -3 +11 M ./rts/eventlog/EventLog.c -2 +2 M ./rts/eventlog/EventLog.h -1 +1 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100226093215-12142-99a3d07da0b0f864b06731c184d6c545f39e0e07.gz From marlowsd at gmail.com Fri Feb 26 05:14:06 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Fri Feb 26 04:44:30 2010 Subject: patch applied (ghc): hack to make the docs build again in a lndir build tree (see comments) Message-ID: <20100226101406.GA18113@abbot.galois.com> Thu Feb 25 05:16:16 PST 2010 Simon Marlow * hack to make the docs build again in a lndir build tree (see comments) M ./docs/users_guide/ghc.mk +14 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100225131616-12142-aa4bb9a172d6beb7e2a5686cde2b7af11dc1b715.gz From marlowsd at gmail.com Fri Feb 26 05:16:33 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Fri Feb 26 04:46:53 2010 Subject: patch applied (testsuite): add test for GHC.Exts.traceEvent (#3874) Message-ID: <20100226101633.GA18486@abbot.galois.com> Fri Feb 26 02:13:21 PST 2010 Simon Marlow * add test for GHC.Exts.traceEvent (#3874) M ./tests/ghc-regress/rts/all.T +3 A ./tests/ghc-regress/rts/traceEvent.hs View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100226101321-12142-5a4384479a043e5e42855d892e970c6efae4493d.gz From marlowsd at gmail.com Fri Feb 26 08:12:50 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Fri Feb 26 07:43:11 2010 Subject: patch applied (ghc): Tweak the tracing flags slightly, and clean up error handling and diagnostics Message-ID: <20100226131250.GA25448@abbot.galois.com> Fri Feb 26 03:06:08 PST 2010 Simon Marlow * Tweak the tracing flags slightly, and clean up error handling and diagnostics Tracing flags are now: -l[flags] Log events in binary format to the file .eventlog -v[flags] Log events to stderr where [flags] can contain: s scheduler events t add time stamps (only useful with -v) and there are more helpful error messages when using flags that are only available in particular variants of the RTS: the flag -v requires the program to be built with -debug the flag -Ds requires the program to be built with -debug the flag -p requires the program to be built with -prof the flag -N requires the program to be built with -threaded the flag -v requires the program to be built with -debug Also, I fixed the crash reported in #3874, with +RTS -ls -v. M ./rts/RtsFlags.c -64 +69 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100226110608-12142-d5e16ed31e0050c5df4dfa8c2ecb22e2e2b84a40.gz From marlowsd at gmail.com Fri Feb 26 08:12:51 2010 From: marlowsd at gmail.com (Simon Marlow) Date: Fri Feb 26 07:43:15 2010 Subject: patch applied (ghc): update the docs to reflect changes in the tracing flags Message-ID: <20100226131251.GA25469@abbot.galois.com> Fri Feb 26 03:12:05 PST 2010 Simon Marlow * update the docs to reflect changes in the tracing flags M ./docs/users_guide/runtime_control.xml -4 +11 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100226111205-12142-f9d51d34cf759876726eb15f0b3a287e65c28902.gz From ghcbuild at microsoft.com Fri Feb 26 12:34:35 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Fri Feb 26 12:34:37 2010 Subject: [nightly] 26-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100226173435.CFB27324214@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Fri Feb 26 18:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... failed; relevant barfage is below. GHC Version 6.13.20100226 No successfully built compilers to test; aborting. ------------------------------------------------------------------------ ------------------------------------------------------------------------ The last 30 lines of /playpen/simonmar/nightly/HEAD/logs/i386-unknown-linux-stage1 are ------------------------------------------------------------------------ ------------------------------------------------------------------------ -XCPP -XExistentialQuantification -XDeriveDataTypeable \ -ilibraries/Cabal \ -ilibraries/filepath \ -ilibraries/extensible-exceptions \ -ilibraries/hpc \ -ilibraries/binary/src \ -ilibraries/bin-package-db [ 1 of 27] Compiling Data.Binary.Builder ( libraries/binary/src/Data/Binary/Builder.hs, bootstrapping/Data/Binary/Builder.o ) [ 2 of 27] Compiling Data.Binary.Put ( libraries/binary/src/Data/Binary/Put.hs, bootstrapping/Data/Binary/Put.o ) [ 5 of 27] Compiling Data.Binary.Get ( libraries/binary/src/Data/Binary/Get.hs, bootstrapping/Data/Binary/Get.o ) [ 6 of 27] Compiling Data.Binary ( libraries/binary/src/Data/Binary.hs, bootstrapping/Data/Binary.o ) [ 7 of 27] Compiling Control.Exception.Extensible ( libraries/extensible-exceptions/Control/Exception/Extensible.hs, bootstrapping/Control/Exception/Extensible.o ) Warning: orphan instances: instance Data.Typeable.Typeable [GHC.IOBase.ExitCode] = Control.Exception.Extensible.$f15 [24 of 27] Compiling Distribution.InstalledPackageInfo.Binary ( libraries/bin-package-db/Distribution/InstalledPackageInfo/Binary.hs, bootstrapping/Distribution/InstalledPackageInfo/Binary.o ) [26 of 27] Compiling Version ( utils/ghc-pkg/Version.hs, bootstrapping/Version.o ) [27 of 27] Compiling Main ( utils/ghc-pkg/Main.hs, bootstrapping/Main.o ) utils/ghc-pkg/Main.hs:732:25: Not in scope: `utf8' utils/ghc-pkg/Main.hs:1154:27: Not in scope: `utf8' utils/ghc-pkg/Main.hs:1560:31: Not in scope: type constructor or class `TextEncoding' utils/ghc-pkg/Main.hs:1563:5: Not in scope: `hSetEncoding' gmake[1]: *** [utils/ghc-pkg/dist/build/ghc-pkg] Error 1 gmake: *** [inplace/bin/ghc-stage1] Error 2 Nightly run ended at Fri Feb 26 18:04:15 GMT 2010 From ghcbuild at microsoft.com Fri Feb 26 13:09:55 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Fri Feb 26 13:09:58 2010 Subject: [nightly] 26-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100226180955.AB19032421B@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Fri Feb 26 18:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100223 **** Building stage 2 compiler ... failed; relevant barfage is below. **** building testsuite tools ... failed. **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (91 failures) **** running nofib (-O -fvia-C) ... ok. (91 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (91 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. (91 failures) **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Fri Feb 26 18:39:36 GMT 2010 ------------------------------------------------------------------------ ------------------------------------------------------------------------ The last 30 lines of /playpen/simonmar/nightly/STABLE/logs/i386-unknown-linux-stage2 are ------------------------------------------------------------------------ ------------------------------------------------------------------------ ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/ug-book.xml:18: parser error : Entity 'ffi-chap' not defined &ffi-chap; ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/ug-book.xml:19: parser error : Entity 'wrong' not defined &wrong; ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/ug-book.xml:20: parser error : Entity 'utils' not defined &utils; ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/ug-book.xml:21: parser error : Entity 'win32-dll' not defined &win32-dll; ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/ug-book.xml:22: parser error : Entity 'bugs' not defined &bugs; ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/users_guide.xml:10: parser error : Failure to process entity ug-book &ug-book; ^ /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/users_guide.xml:10: parser error : Entity 'ug-book' not defined &ug-book; ^ unable to parse /playpen/simonmar/nightly/STABLE/fptools/docs/users_guide/users_guide.xml Error: xsltproc failed gmake[1]: *** [docs/users_guide/users_guide.ps] Error 1 gmake: *** [all] Error 2 real 1m19.406s user 1m16.339s sys 0m2.762s Nightly run ended at Fri Feb 26 18:39:36 GMT 2010 From ghcbuild at microsoft.com Fri Feb 26 13:32:35 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Fri Feb 26 13:32:37 2010 Subject: [nightly] 26-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100226183235.D1C9B3244B5@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Fri Feb 26 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... failed; relevant barfage is below. GHC Version 6.13.20100226 **** building source distribution ... failed. No successfully built compilers to test; aborting. ------------------------------------------------------------------------ ------------------------------------------------------------------------ The last 30 lines of /64playpen/simonmar/nightly/HEAD-cam-04-unx/logs/x86_64-unknown-linux-stage1 are ------------------------------------------------------------------------ ------------------------------------------------------------------------ -XCPP -XExistentialQuantification -XDeriveDataTypeable \ -ilibraries/Cabal \ -ilibraries/filepath \ -ilibraries/extensible-exceptions \ -ilibraries/hpc \ -ilibraries/binary/src \ -ilibraries/bin-package-db [ 1 of 27] Compiling Data.Binary.Builder ( libraries/binary/src/Data/Binary/Builder.hs, bootstrapping/Data/Binary/Builder.o ) [ 2 of 27] Compiling Data.Binary.Put ( libraries/binary/src/Data/Binary/Put.hs, bootstrapping/Data/Binary/Put.o ) [ 5 of 27] Compiling Data.Binary.Get ( libraries/binary/src/Data/Binary/Get.hs, bootstrapping/Data/Binary/Get.o ) [ 6 of 27] Compiling Data.Binary ( libraries/binary/src/Data/Binary.hs, bootstrapping/Data/Binary.o ) [ 7 of 27] Compiling Control.Exception.Extensible ( libraries/extensible-exceptions/Control/Exception/Extensible.hs, bootstrapping/Control/Exception/Extensible.o ) Warning: orphan instances: instance Data.Typeable.Typeable [GHC.IOBase.ExitCode] = Control.Exception.Extensible.$f15 [24 of 27] Compiling Distribution.InstalledPackageInfo.Binary ( libraries/bin-package-db/Distribution/InstalledPackageInfo/Binary.hs, bootstrapping/Distribution/InstalledPackageInfo/Binary.o ) [26 of 27] Compiling Version ( utils/ghc-pkg/Version.hs, bootstrapping/Version.o ) [27 of 27] Compiling Main ( utils/ghc-pkg/Main.hs, bootstrapping/Main.o ) utils/ghc-pkg/Main.hs:732:25: Not in scope: `utf8' utils/ghc-pkg/Main.hs:1154:27: Not in scope: `utf8' utils/ghc-pkg/Main.hs:1560:31: Not in scope: type constructor or class `TextEncoding' utils/ghc-pkg/Main.hs:1563:5: Not in scope: `hSetEncoding' gmake[1]: *** [utils/ghc-pkg/dist/build/ghc-pkg] Error 1 gmake: *** [inplace/bin/ghc-stage1] Error 2 Nightly run ended at Fri Feb 26 19:02:15 GMT 2010 From ghcbuild at microsoft.com Fri Feb 26 13:58:59 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Fri Feb 26 13:59:00 2010 Subject: [nightly] 26-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100226185859.31C5A324348@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Fri Feb 26 19:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100223 **** Building stage 2 compiler ... failed; relevant barfage is below. **** building source distribution ... ok. **** building testsuite tools ... failed. **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (91 failures) **** running nofib (-O -fvia-C) ... ok. (91 failures) **** running nofib (-O -prof -auto-all) ... ok. (91 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (91 failures) **** running nofib (-fasm) ... ok. (91 failures) **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Fri Feb 26 19:28:40 GMT 2010 ------------------------------------------------------------------------ ------------------------------------------------------------------------ The last 30 lines of /64playpen/simonmar/nightly/STABLE-cam-04-unx/logs/x86_64-unknown-linux-stage2 are ------------------------------------------------------------------------ ------------------------------------------------------------------------ ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/ug-book.xml:18: parser error : Entity 'ffi-chap' not defined &ffi-chap; ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/ug-book.xml:19: parser error : Entity 'wrong' not defined &wrong; ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/ug-book.xml:20: parser error : Entity 'utils' not defined &utils; ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/ug-book.xml:21: parser error : Entity 'win32-dll' not defined &win32-dll; ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/ug-book.xml:22: parser error : Entity 'bugs' not defined &bugs; ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/users_guide.xml:10: parser error : Failure to process entity ug-book &ug-book; ^ /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/users_guide.xml:10: parser error : Entity 'ug-book' not defined &ug-book; ^ unable to parse /64playpen/simonmar/nightly/STABLE-cam-04-unx/fptools/docs/users_guide/users_guide.xml Error: xsltproc failed gmake[1]: *** [docs/users_guide/users_guide.ps] Error 1 gmake: *** [all] Error 2 real 0m42.615s user 0m41.162s sys 0m1.459s Nightly run ended at Fri Feb 26 19:28:40 GMT 2010 From igloo at earth.li Fri Feb 26 18:06:51 2010 From: igloo at earth.li (Ian Lynagh) Date: Fri Feb 26 17:37:10 2010 Subject: patch applied (ghc): Fix the build with GHC < 6.12 Message-ID: <20100226230650.GA22295@abbot.galois.com> Fri Feb 26 14:39:31 PST 2010 Ian Lynagh * Fix the build with GHC < 6.12 M ./utils/ghc-pkg/Main.hs -5 +7 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100226223931-3fd76-b2e5b07b4ce814f86fcd9558715f01466044ccf5.gz From bit.bucket at galois.com Sat Feb 27 08:17:02 2010 From: bit.bucket at galois.com (Builder) Date: Sat Feb 27 07:47:23 2010 Subject: igloo, build 4 Message-ID: igloo, build 4 Build succeeded Details: http://darcs.haskell.org/ghcBuilder/builders/igloo/4.html darcs checkout | Success create mk/build.mk | Success make darcs-all executable | Success get subrepos | Success repo versions | Success setting version date | Success booting | Success configuring | Success compiling | Success making bindist | Success testing bindist | Success testing | Success Build succeeded Details: http://darcs.haskell.org/ghcBuilder/builders/igloo/4.html From igloo at earth.li Sat Feb 27 10:01:52 2010 From: igloo at earth.li (Ian Lynagh) Date: Sat Feb 27 09:32:10 2010 Subject: patch applied (ghc-6.12/ghc): Force encoding to UTF-8 when writing individual .conf files Message-ID: <20100227150151.GA21637@abbot.galois.com> Wed Feb 24 07:25:19 PST 2010 Simon Marlow * Force encoding to UTF-8 when writing individual .conf files M ./utils/ghc-pkg/Main.hs -14 +28 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100224152519-12142-96a29065ed03209bb5925f9f1a200d43af5f9708.gz From igloo at earth.li Sat Feb 27 10:01:54 2010 From: igloo at earth.li (Ian Lynagh) Date: Sat Feb 27 09:32:12 2010 Subject: patch applied (ghc-6.12/ghc): Fix the build with GHC < 6.12 Message-ID: <20100227150154.GA21659@abbot.galois.com> Fri Feb 26 14:39:31 PST 2010 Ian Lynagh * Fix the build with GHC < 6.12 M ./utils/ghc-pkg/Main.hs -5 +7 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100226223931-3fd76-b2e5b07b4ce814f86fcd9558715f01466044ccf5.gz From igloo at earth.li Sat Feb 27 10:01:55 2010 From: igloo at earth.li (Ian Lynagh) Date: Sat Feb 27 09:32:15 2010 Subject: patch applied (ghc-6.12/ghc): Fix crash when using printf format specifiers in traceEvent (#3874) Message-ID: <20100227150155.GA21681@abbot.galois.com> Fri Feb 26 01:32:15 PST 2010 Simon Marlow * Fix crash when using printf format specifiers in traceEvent (#3874) M ./rts/Trace.c -3 +11 M ./rts/eventlog/EventLog.c -2 +2 M ./rts/eventlog/EventLog.h -1 +1 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100226093215-12142-99a3d07da0b0f864b06731c184d6c545f39e0e07.gz From igloo at earth.li Sat Feb 27 10:01:57 2010 From: igloo at earth.li (Ian Lynagh) Date: Sat Feb 27 09:32:16 2010 Subject: patch applied (ghc-6.12/ghc): hack to make the docs build again in a lndir build tree (see comments) Message-ID: <20100227150157.GA21701@abbot.galois.com> Thu Feb 25 05:16:16 PST 2010 Simon Marlow * hack to make the docs build again in a lndir build tree (see comments) M ./docs/users_guide/ghc.mk +14 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100225131616-12142-aa4bb9a172d6beb7e2a5686cde2b7af11dc1b715.gz From igloo at earth.li Sat Feb 27 10:01:58 2010 From: igloo at earth.li (Ian Lynagh) Date: Sat Feb 27 09:32:21 2010 Subject: patch applied (ghc-6.12/ghc): update the docs to reflect changes in the tracing flags Message-ID: <20100227150158.GA21721@abbot.galois.com> Fri Feb 26 03:12:05 PST 2010 Simon Marlow * update the docs to reflect changes in the tracing flags M ./docs/users_guide/runtime_control.xml -4 +11 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100226111205-12142-f9d51d34cf759876726eb15f0b3a287e65c28902.gz From igloo at earth.li Sat Feb 27 13:14:51 2010 From: igloo at earth.li (Ian Lynagh) Date: Sat Feb 27 12:45:09 2010 Subject: patch applied (ghc): Fix trac #2578 Message-ID: <20100227181451.GA28207@abbot.galois.com> Sat Feb 27 09:39:51 PST 2010 Ian Lynagh * Fix trac #2578 We define empty datatypes as not being enumerations, which means the empty blocks aren't generated. M ./compiler/iface/BuildTyCl.lhs -1 +7 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100227173951-3fd76-df1c33074de137c5e6d4674ed22b0e973e1d6262.gz From igloo at earth.li Sat Feb 27 13:18:12 2010 From: igloo at earth.li (Ian Lynagh) Date: Sat Feb 27 12:48:28 2010 Subject: patch applied (testsuite): Add a test for #2578 Message-ID: <20100227181812.GA28536@abbot.galois.com> Sat Feb 27 09:24:29 PST 2010 Ian Lynagh * Add a test for #2578 A ./tests/ghc-regress/codeGen/should_compile/2578.hs M ./tests/ghc-regress/codeGen/should_compile/Makefile +4 M ./tests/ghc-regress/codeGen/should_compile/all.T +1 View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100227172429-3fd76-d40e2edbe8da4e5352469c50db43ef4fb8239e0c.gz From cvs-ghc at haskell.org Sat Feb 27 14:11:00 2010 From: cvs-ghc at haskell.org (Builder) Date: Sat Feb 27 13:41:20 2010 Subject: igloo, build 5 Message-ID: igloo, build 5 Build succeeded Details: http://darcs.haskell.org/ghcBuilder/builders/igloo/5.html darcs checkout | Success create mk/build.mk | Success make darcs-all executable | Success get subrepos | Success repo versions | Success setting version date | Success booting | Success configuring | Success compiling | Success making bindist | Success testing bindist | Success testing | Success Build succeeded Details: http://darcs.haskell.org/ghcBuilder/builders/igloo/5.html From igloo at earth.li Sat Feb 27 14:13:06 2010 From: igloo at earth.li (Ian Lynagh) Date: Sat Feb 27 13:43:22 2010 Subject: patch applied (ghc-6.12/ghc): Fix trac #2578 Message-ID: <20100227191305.GA30662@abbot.galois.com> Sat Feb 27 09:39:51 PST 2010 Ian Lynagh * Fix trac #2578 We define empty datatypes as not being enumerations, which means the empty blocks aren't generated. M ./compiler/iface/BuildTyCl.lhs -1 +7 View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100227173951-3fd76-df1c33074de137c5e6d4674ed22b0e973e1d6262.gz From igloo at earth.li Sat Feb 27 14:16:13 2010 From: igloo at earth.li (Ian Lynagh) Date: Sat Feb 27 13:46:29 2010 Subject: patch applied (ghc-6.12/testsuite): Add a test for #2578 Message-ID: <20100227191613.GA30920@abbot.galois.com> Sat Feb 27 09:24:29 PST 2010 Ian Lynagh * Add a test for #2578 A ./tests/ghc-regress/codeGen/should_compile/2578.hs M ./tests/ghc-regress/codeGen/should_compile/Makefile +4 M ./tests/ghc-regress/codeGen/should_compile/all.T +1 View patch online: http://darcs.haskell.org/ghc-6.12/testsuite/_darcs/patches/20100227172429-3fd76-d40e2edbe8da4e5352469c50db43ef4fb8239e0c.gz From igloo at earth.li Sat Feb 27 15:56:21 2010 From: igloo at earth.li (Ian Lynagh) Date: Sat Feb 27 15:26:40 2010 Subject: patch applied (ghc): Have separate rules to generate tags and TAGS Message-ID: <20100227205620.GA1856@abbot.galois.com> Sat Feb 27 12:25:17 PST 2010 Ian Lynagh * Have separate rules to generate tags and TAGS This works around the two filenames being the same on case-insensitive file systems. M ./ghc.mk +3 M ./rules/tags-package.mk -1 +5 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100227202517-3fd76-49ba8952380e3a7f37dc57df42869430884559f5.gz From igloo at earth.li Sat Feb 27 15:57:46 2010 From: igloo at earth.li (Ian Lynagh) Date: Sat Feb 27 15:28:01 2010 Subject: patch applied (ghc): Make "make tags" work in the new build system In-Reply-To: <20100225092459.GA8088@doc.ic.ac.uk> References: <20100221172858.GA15637@abbot.galois.com> <59543203684B2244980D7E4057D5FBC10B08F069@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224140511.GA2994@matrix.chaos.earth.li> <59543203684B2244980D7E4057D5FBC10B08F3BB@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224154540.GB5117@matrix.chaos.earth.li> <6147FF6B-2779-4066-BC88-F0584B4FC27D@cs.york.ac.uk> <59543203684B2244980D7E4057D5FBC10B08F4B3@DB3EX14MBXC312.europe.corp.microsoft.com> <20100224185636.GA28070@matrix.chaos.earth.li> <59543203684B2244980D7E4057D5FBC10B08FA68@DB3EX14MBXC312.europe.corp.microsoft.com> <20100225092459.GA8088@doc.ic.ac.uk> Message-ID: <20100227205746.GA20140@matrix.chaos.earth.li> On Thu, Feb 25, 2010 at 09:24:59AM +0000, Tristan Allwood wrote: > +1 for having separate 'make tags' and 'make TAGS'. Done. Thanks Ian From igloo at earth.li Sat Feb 27 16:33:21 2010 From: igloo at earth.li (Ian Lynagh) Date: Sat Feb 27 16:03:38 2010 Subject: patch applied (ghc): Add a LICENSE file to libraries/bin-package-db Message-ID: <20100227213321.GA3262@abbot.galois.com> Sat Feb 27 12:56:26 PST 2010 Ian Lynagh * Add a LICENSE file to libraries/bin-package-db Silences a (harmless) error from tar when making bindists A ./libraries/bin-package-db/LICENSE View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100227205626-3fd76-95b1760c04f95a684b581d6260718df8f88215e3.gz From igloo at earth.li Sat Feb 27 17:03:44 2010 From: igloo at earth.li (Ian Lynagh) Date: Sat Feb 27 16:34:00 2010 Subject: patch applied (ghc-6.12/ghc): Add a LICENSE file to libraries/bin-package-db Message-ID: <20100227220344.GA4661@abbot.galois.com> Sat Feb 27 12:56:26 PST 2010 Ian Lynagh * Add a LICENSE file to libraries/bin-package-db Silences a (harmless) error from tar when making bindists A ./libraries/bin-package-db/LICENSE View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100227205626-3fd76-95b1760c04f95a684b581d6260718df8f88215e3.gz From igloo at earth.li Sat Feb 27 17:03:46 2010 From: igloo at earth.li (Ian Lynagh) Date: Sat Feb 27 16:34:03 2010 Subject: patch applied (ghc-6.12/ghc): Update directory tarball Message-ID: <20100227220345.GA4681@abbot.galois.com> Sat Feb 27 13:31:47 PST 2010 Ian Lynagh * Update directory tarball Includes: Wed Feb 3 07:09:57 GMT 2010 benl@cse.unsw.edu.au * On Solaris, _FILE_OFFSET_BITS must be defined before including sys/stat.h Fri Feb 26 01:31:44 PST 2010 Simon Marlow * Fix #3878: various directory tests were wrong on Unix systems R ./libraries/tarballs/directory-1.0.1.0.tar.gz A ./libraries/tarballs/directory-1.0.1.1.tar.gz View patch online: http://darcs.haskell.org/ghc-6.12/ghc/_darcs/patches/20100227213147-3fd76-94eb9e2aca3aec304f148a658c626ea71ddfbf73.gz From ghcbuild at microsoft.com Sat Feb 27 19:52:47 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sat Feb 27 19:52:48 2010 Subject: [nightly] 27-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100228005247.0A1F6324120@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Sat Feb 27 19:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100226 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Sun Feb 28 01:22:32 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sat Feb 27 21:20:40 GMT 2010 2369 total tests, which gave rise to 13116 test cases, of which 0 caused framework failures 2760 were skipped 10031 expected passes 315 expected failures 0 unexpected passes 10 unexpected failures Unexpected failures: 2592(profc,profasm) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci,threaded2) drvfail006(normal) drvfail008(normal) outofmem2(normal) ---------------------------------------------------- Nightly run ended at Sun Feb 28 01:22:32 GMT 2010 From ghcbuild at microsoft.com Sat Feb 27 19:56:30 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sat Feb 27 19:56:32 2010 Subject: [nightly] 27-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100228005630.E700B324295@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Sat Feb 27 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100226 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... failed. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Sun Feb 28 01:26:15 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sat Feb 27 21:26:13 GMT 2010 2464 total tests, which gave rise to 13631 test cases, of which 0 caused framework failures 2838 were skipped 10423 expected passes 359 expected failures 0 unexpected passes 11 unexpected failures Unexpected failures: T1969(normal) concprog001(ghci) dph-smvm(normal,threaded1,threaded2) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) traceEvent(profc,profasm,profthreaded) ---------------------------------------------------- Nightly run ended at Sun Feb 28 01:26:15 GMT 2010 From ghcbuild at microsoft.com Sat Feb 27 23:42:43 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sat Feb 27 23:42:45 2010 Subject: [nightly] 27-Feb-2010 build of HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100228044243.6326432416B@www.haskell.org> Build description = HEAD on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/HEAD Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-02-unx Nightly build started on cam-02-unx at Sat Feb 27 18:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100226 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. (1 failures) **** running nofib (-O -fvia-C) ... ok. (2 failures) **** running nofib (-O -prof -auto-all) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-fasm) ... ok. (1 failures) **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Sun Feb 28 05:12:28 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sat Feb 27 22:20:20 GMT 2010 2464 total tests, which gave rise to 13631 test cases, of which 0 caused framework failures 2845 were skipped 10413 expected passes 353 expected failures 0 unexpected passes 20 unexpected failures Unexpected failures: concprog001(ghci) ffi002(profthreaded) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) joao-circular(profc) traceEvent(profc,profasm,profthreaded) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Sun Feb 28 05:12:28 GMT 2010 From ghcbuild at microsoft.com Sun Feb 28 00:01:52 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sun Feb 28 00:01:54 2010 Subject: [nightly] 27-Feb-2010 build of STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Message-ID: <20100228050152.4DD043242B1@www.haskell.org> Build description = STABLE on i386-unknown-linux (cam-02-unx.europe.corp.microsoft.com) Build location = /playpen/simonmar/nightly/STABLE Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-02-unx Nightly build started on cam-02-unx at Sat Feb 27 18:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.12.1.20100226 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O -fasm) ... ok. (1 failures) **** running nofib (-O -fvia-C) ... ok. (1 failures) **** running nofib (-O -prof -auto-all -fasm) ... ok. (2 failures) **** running nofib (-O -prof -auto-all -fvia-C) ... ok. (1 failures) **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/stable/logs Dists are at http://www.haskell.org/ghc/dist/stable/dist Docs are at http://www.haskell.org/ghc/dist/stable/docs All done! Nightly build finished successfully at Sun Feb 28 05:31:38 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sat Feb 27 21:58:40 GMT 2010 2369 total tests, which gave rise to 13116 test cases, of which 0 caused framework failures 2775 were skipped 9998 expected passes 317 expected failures 0 unexpected passes 26 unexpected failures Unexpected failures: 2592(profc,profasm) CPUTime001(threaded2) T1969(normal) annrun01(dyn) barton-mangler-bug(profc) break024(ghci) concprog001(ghci) drvfail006(normal) drvfail008(normal) ffi002(threaded1) joao-circular(profc) outofmem2(normal) signals002(threaded2) signals004(threaded1) user001(normal,optc,hpc,optasm,profc,profasm,ghci,threaded1,threaded2,dyn,profthreaded) ---------------------------------------------------- Nightly run ended at Sun Feb 28 05:31:38 GMT 2010 From ghcbuild at microsoft.com Sun Feb 28 13:55:13 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sun Feb 28 13:55:14 2010 Subject: [nightly] 28-Feb-2010 build of STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100228185513.A32B8324132@www.haskell.org> Build description = STABLE on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/STABLE-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-STABLE-cam-04-unx Nightly build started on cam-04-unx at Sun Feb 28 19:10:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... failed; relevant barfage is below. GHC Version 6.12.1.20100227 **** building source distribution ... ok. No successfully built compilers to test; aborting. ------------------------------------------------------------------------ ------------------------------------------------------------------------ The last 30 lines of /64playpen/simonmar/nightly/STABLE-cam-04-unx/logs/x86_64-unknown-linux-stage1 are ------------------------------------------------------------------------ ------------------------------------------------------------------------ Linking utils/ghc-cabal/dist/build/tmp/ghc-cabal ... touch utils/ghc-cabal/dist/build/tmp/ghc-cabal "inplace/bin/mkdirhier" inplace/bin/ "cp" utils/ghc-cabal/dist/build/tmp/ghc-cabal inplace/bin/ghc-cabal "rm" -f utils/ghc-pkg/Version.hs echo "module Version where" >> utils/ghc-pkg/Version.hs echo "version, targetOS, targetARCH :: String" >> utils/ghc-pkg/Version.hs echo "version = \"6.12.1.20100227\"" >> utils/ghc-pkg/Version.hs echo "targetOS = \"linux\"" >> utils/ghc-pkg/Version.hs echo "targetARCH = \"x86_64\"" >> utils/ghc-pkg/Version.hs "inplace/bin/mkdirhier" bootstrapping "inplace/bin/mkdirhier" utils/ghc-pkg/dist/build "ghc-6.8.3" -H32m -O --make utils/ghc-pkg/Main.hs -o utils/ghc-pkg/dist/build/ghc-pkg \ -Wall -fno-warn-unused-imports \ -DCABAL_VERSION=1,8,0,2 \ -DBOOTSTRAPPING \ -odir bootstrapping \ -hidir bootstrapping \ -iutils/ghc-pkg \ -XCPP -XExistentialQuantification -XDeriveDataTypeable \ -ilibraries/Cabal \ -ilibraries/filepath \ -ilibraries/extensible-exceptions \ -ilibraries/hpc \ -ilibraries/ghc-binary/src \ -ilibraries/bin-package-db libraries/filepath/System/FilePath.hs:1:13: cannot parse LANGUAGE pragma gmake[1]: *** [utils/ghc-pkg/dist/build/ghc-pkg] Error 1 gmake: *** [inplace/bin/ghc-stage1] Error 2 Nightly run ended at Sun Feb 28 19:25:00 GMT 2010 From igloo at earth.li Sun Feb 28 15:22:17 2010 From: igloo at earth.li (Ian Lynagh) Date: Sun Feb 28 14:52:32 2010 Subject: patch applied (ghc): Whitespace only in docs/users_guide/packages.xml Message-ID: <20100228202216.GA17390@abbot.galois.com> Sun Feb 28 10:29:45 PST 2010 Ian Lynagh * Whitespace only in docs/users_guide/packages.xml M ./docs/users_guide/packages.xml -681 +681 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100228182945-3fd76-ab5300f93998efde69322abdc37d1c55df0182ca.gz From igloo at earth.li Sun Feb 28 15:22:18 2010 From: igloo at earth.li (Ian Lynagh) Date: Sun Feb 28 14:52:35 2010 Subject: patch applied (ghc): Add -no-auto-link-packages docs to the user guide; fixes trac #3759 Message-ID: <20100228202218.GA17410@abbot.galois.com> Sun Feb 28 11:18:37 PST 2010 Ian Lynagh * Add -no-auto-link-packages docs to the user guide; fixes trac #3759 M ./docs/users_guide/flags.xml +6 M ./docs/users_guide/packages.xml +11 View patch online: http://darcs.haskell.org/ghc/_darcs/patches/20100228191837-3fd76-c2a47d3c6b595cc51b039fbbf4e977a3ff0161a2.gz From ghcbuild at microsoft.com Sun Feb 28 19:30:44 2010 From: ghcbuild at microsoft.com (GHC Build Reports) Date: Sun Feb 28 19:30:45 2010 Subject: [nightly] 28-Feb-2010 build of HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Message-ID: <20100301003044.8CA6F3241AE@www.haskell.org> Build description = HEAD on x86_64-unknown-linux (cam-04-unx.europe.corp.microsoft.com) Build location = /64playpen/simonmar/nightly/HEAD-cam-04-unx Build config file = /home/simonmar/nightly/site/msrc/conf-HEAD-cam-04-unx Nightly build started on cam-04-unx at Sun Feb 28 19:00:01 GMT 2010. **** checking out new source tree ... warning: . already present; omitting ok. **** Building stage 1 compiler ... ok. GHC Version 6.13.20100227 **** Building stage 2 compiler ... ok. **** Building stage 3 compiler ... ok. **** building source distribution ... failed. **** building testsuite tools ... ok. **** running tests ... ok (summary below). **** building compiler binary distribution ... ok. **** uploading binary distribution ... ok. **** running nofib (-O) ... ok. **** running nofib (-O -fvia-C) ... ok. **** running nofib (-O -prof -auto-all) ... ok. **** running nofib (-O -prof -auto-all -fasm) ... ok. **** running nofib (-fasm) ... ok. **** publishing logs ... ok. Logs are at http://www.haskell.org/ghc/dist/current/logs Dists are at http://www.haskell.org/ghc/dist/current/dist Docs are at http://www.haskell.org/ghc/dist/current/docs All done! Nightly build finished successfully at Mon Mar 1 01:00:32 GMT 2010 ------------- GHC Test summary --------------------- OVERALL SUMMARY for test run started at Sun Feb 28 21:18:54 GMT 2010 2465 total tests, which gave rise to 13632 test cases, of which 0 caused framework failures 2838 were skipped 10424 expected passes 359 expected failures 0 unexpected passes 11 unexpected failures Unexpected failures: T1969(normal) concprog001(ghci) dph-smvm(normal,threaded1,threaded2) hpc_markup_multi_001(normal) hpc_markup_multi_002(normal) hpc_markup_multi_003(normal) traceEvent(profc,profasm,profthreaded) ---------------------------------------------------- Nightly run ended at Mon Mar 1 01:00:32 GMT 2010 From ben.lippmeier at anu.edu.au Sun Feb 28 23:14:10 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Sun Feb 28 22:44:25 2010 Subject: patch applied (testsuite): Add DPH smvm test In-Reply-To: <4B87962B.7050007@gmail.com> References: <20100226033323.GA4334@abbot.galois.com> <4B87962B.7050007@gmail.com> Message-ID: <903ACE2B-9886-4A19-91B3-90745F925807@anu.edu.au> :-P That'll be the bugzors. The test file format is probably word-size dependent -- I'll fix it. Thanks, Ben. On 26/02/2010, at 20:36 , Simon Marlow wrote: > On 26/02/2010 03:33, Ben Lippmeier wrote: >> Thu Feb 25 18:56:30 PST 2010 benl@ouroborus.net >> * Add DPH smvm test >> >> A ./tests/ghc-regress/dph/smvm/ >> A ./tests/ghc-regress/dph/smvm/Main.hs >> A ./tests/ghc-regress/dph/smvm/Makefile >> A ./tests/ghc-regress/dph/smvm/SMVMVect.hs >> A ./tests/ghc-regress/dph/smvm/dph-smvm.T >> A ./tests/ghc-regress/dph/smvm/dph-smvm.stdout >> A ./tests/ghc-regress/dph/smvm/test.dat > > I presume this isn't supposed to happen? :) (on x86-64/Linux) > > =====> dph-smvm(normal) 2 of 2464 [0, 0, 0] > cd ./dph/smvm && '/64playpen/simonmar/validate/bindisttest/install dir/bin/ghc' -fforce-recomp -dcore-lint -dcmm-lint -no-user-package-conf -dno-debug-output --make -o dph-smvm Main -Odph -package dph-par -package dph-prim-par >dph-smvm.comp.stderr 2>&1 > cd ./dph/smvm && ./dph-smvm dph-smvm.run.stdout 2>dph-smvm.run.stderr > Wrong exit code (expected 0 , actual 1 ) > Stdout: > > Stderr: > dph-smvm: out of memory (requested 790275031040 bytes) > > From ben.lippmeier at anu.edu.au Sun Feb 28 23:30:31 2010 From: ben.lippmeier at anu.edu.au (Ben Lippmeier) Date: Sun Feb 28 23:00:43 2010 Subject: patch applied (testsuite): Add DPH solution for 108th Euler problem Message-ID: <20100301043030.GA4021@abbot.galois.com> Sun Feb 28 20:16:57 PST 2010 benl@ouroborus.net * Add DPH solution for 108th Euler problem ./tests/ghc-regress/dph/broken-diophantine -> ./tests/ghc-regress/dph/diophantine A ./tests/ghc-regress/dph/broken-diophantine/ A ./tests/ghc-regress/dph/diophantine/DiophantineVect.hs A ./tests/ghc-regress/dph/diophantine/Main.hs A ./tests/ghc-regress/dph/diophantine/Makefile A ./tests/ghc-regress/dph/diophantine/dph-diophantine.T A ./tests/ghc-regress/dph/diophantine/dph-diophantine.stdout View patch online: http://darcs.haskell.org/testsuite/_darcs/patches/20100301041657-bf82d-f79ffe35ddf6b67cbcbbd15828a5ffed85d571a8.gz