[GHC] #6042: GHC is bloated
GHC
cvs-ghc at haskell.org
Mon Aug 13 17:00:01 CEST 2012
#6042: GHC is bloated
---------------------------------+------------------------------------------
Reporter: simonmar | Owner: pcapriotti
Type: bug | Status: new
Priority: highest | Milestone: 7.6.1
Component: Compiler | Version: 7.4.1
Keywords: | Os: Unknown/Multiple
Architecture: Unknown/Multiple | Failure: None/Unknown
Difficulty: Unknown | Testcase:
Blockedby: | Blocking:
Related: |
---------------------------------+------------------------------------------
Comment(by simonpj):
Paolo writes: I've spent some time looking at #6042, but with not much
success. Here's how the GHC executable size progressed with the past few
releases:
{{{
7.0 31228882
7.2 36553006
7.4 43645088
7.6 45291472
head 45443324
}}}
Here are listings showing module sizes for 7.0, 7.2 and 7.4. CmmParse has
increased dramatically between 7.0 and 7.2, but then it went back down in
later releases. There doesn't seem anything else sticking out there, it
seems that the growth is spread mostly uniformly across modules.
{{{
module 7.0 7.2 7.4
HsExpr.o 1314060 1372352 1432064
DynFlags.o 908576 1044420 1144280
Parser.o 844896 862364 745108
HsDecls.o 712864 748692 838848
ZipDataflow.o 572676 - -
Lexer.o 550864 810460 599104
DsMeta.o 532032 543612 526708
CmmParse.o 511732 2548668 492312
HsBinds.o 486876 487872 453996
PrimOp.o 337680 494072 547644
}}}
Not sure if this is helpful, but here's a summary of the differences in
section sizes between 7.0 and 7.2:
{{{
text +2536464 14.3 %
rodata +1494840 151.5 %
data + 190880 17.9 %
}}}
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/6042#comment:14>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler
More information about the Glasgow-haskell-bugs
mailing list