> --------------------------------------------------------------
> ----------
> ==fptools== make all --no-print-directory -r;
> in /home/tetron/hack/lang/haskell/fptools-cvs2/ghc/compiler
> --------------------------------------------------------------
> ----------
> ghc -cpp -fglasgow-exts -Rghc-timing -I. -IcodeGen
> -InativeGen -Iparser
> -iutils:basicTypes:types:hsSyn:prelude:rename:typecheck:deSuga
r:coreSyn:specialise:simplCore:stranal:stgSyn:simplStg:codeGen:absCSyn:main:
profil> ing:parser:usageSP:cprAnalysis:nativeGen
> -recomp -Onot -H45m -K2m -dcore-lint -fno-warn-incomplete-patterns
> -dcore-lint -c rename/ParseIface.hs -o rename/ParseIface.o -osuf o
> make[2]: *** [rename/ParseIface.o] Error 1
> make[1]: *** [all] Interrupt
> make: *** [all] Interrupt
>
> Everything up to here compiles fine, but on this file
> something fucks up.
> Basically it just sits here and *grinds*, in a matter of
> seconds gobbling
> up something like 70 or 80 megabytes of memory (my system
> only has 64!)
This file does need a lot of memory to compile, but not quite that much - I
get by with 48M compiling with 4.02 and using an up-to-date Happy from the
CVS repository.
However, compiling Parser.hs is another story: you'll probably need a
bigger machine to compile that one. Here's the line from one of my
compilations:
ghc-4.02 -cpp -fglasgow-exts -Rghc-timing -I. -IcodeGen -InativeGen -Iparser
-iutils:basicTypes:types:hsSyn:prelude:rename:typecheck:deSugar:coreSyn:spec
ialise:simplCore:stranal:stgSyn:simplStg:codeGen:absCSyn:main:profiling:pars
er:usageSP:cprAnalysis:nativeGen -recomp -O -dcore-lint -H24m -DDEBUG
-optCrts-M128m -Onot -H80m -K2m -dcore-lint -fno-warn-incomplete-patterns
-dcore-lint -c parser/Parser.hs -o parser/Parser.o -osuf o
ghc-4.02: ignoring heap-size-setting option (-H24m)...not the largest seen
<<ghc: 1903045256 bytes, 41 GCs, 11517169/25643824 avg/max bytes residency
(8 samples), 87M in use, 0.01 INIT (0.00 elapsed), 36.32 MUT (39.54
elapsed), 12.66 GC (13.87 elapsed) :ghc>>
Strange that the max. residency was only 25M but it claimed to need 87M (I'd
expect twice the max residency, since GHC has a two-space generational
collector). Oh well, must look into that sometime.
Fiddling around with the garbage collector options may help - take a look in
the User's Guide for details. Don't forget to use -optCrts<option> if
you're sending RTS options to the compiler.
Cheers,
Simon
