C--

[Bulat Ziganshin]

Hello Simon,

Monday, August 08, 2005, 7:17:06 PM, you wrote:

SPJ> Actually using C-- itself as an output medium keeps slipping down the
SPJ> agenda, mainly because no one asks for it!

because you don't advertise it! :)  if you will advertise the
following, preferably with less or more concrete numbers, then we will
know whether we want it or not. just asking "whether you want to have C
or C-- compilation path?" is not enough

what about C-- compiler? you will use the existing version of Quick C--? is
this compiler enough bug-free and generates good enough code?

SPJ> | It should reduce compile time compared to compiling via C, though I
SPJ> | suspect it will still be slower than using GHC's built-in native code
SPJ> | generator.
SPJ> | 
SPJ> | It should improve the speed of compiled programs, because C-- will be
SPJ> | using its own, efficient, calling convention.  For example, this will
SPJ> | mean that some arguments can be passed in registers on x86, which we
SPJ> | can't do currently because of a combination of the lack of registers
SPJ> and
SPJ> | limitations in what we can convince gcc to do.
SPJ> | 
SPJ> | It should improve portability, because there is less work in GHC
SPJ> | required to compile via C-- on a new platform: no mangler to modify,
SPJ> for
SPJ> | example.
SPJ> | 
SPJ> | Eventually, we hope the registerised via-C path would become
SPJ> redundant.
SPJ> | The unregisterised path is still useful for bootstrapping, given that
SPJ> | more platforms have a C compiler than a C-- compiler (if C-- were to
SPJ> | become ubiquitous, the unregisterised path would become irrelevant
SPJ> too).


-- 
Best regards,
 Bulat                            mailto:bulatz at HotPOP.com