patch applied (ghc): SMP bugfix: reload capability from
task-> cap after scheduleDoGC
simonmarhaskell at gmail.com
Thu Feb 23 09:32:43 EST 2006
> On 2/22/06, Simon Marlow <simonmarhaskell at gmail.com> wrote:
>>David Himmelstrup wrote:
>>>Sat Feb 11 19:47:36 PST 2006 wolfgang.thaller at gmx.net
>>> * SMP bugfix: reload capability from task->cap after scheduleDoGC
>>> M ./ghc/rts/Schedule.c -2 +14
>>Evidently commit mailings aren't working for everyone - sometimes they
>>get misattributed (like this one).
>>If you see an error message when pushing, please could you send it to me
>>so I can investigate, thanks.
> Some time ago, I unpulled an erroneous patch (it was a tag so I
> couldn't just undo it). The unpull caused .current-hash to be invalid
> and thus no commit mails has been sent for a while. I've just now
> fixed it and mailed all the missing messages.
BTW, it's not a good idea to unpull from the main repo, because other
people might already have pulled your patch - in this case, Simon PJ had
pulled your tag and got confused when darcs later asked whether it
should push it back. I should probably disable unpull/obliterate in the
More information about the Cvs-ghc