summaryrefslogtreecommitdiff
path: root/src/cmd/cc/pgen.c
AgeCommit message (Collapse)AuthorFilesLines
2011-02-14Imported Upstream version 2011-02-01.1upstream/2011-02-01.1Ondřej Surý1-2/+1
2011-01-17Imported Upstream version 2011.01.12upstream/2011.01.12Ondřej Surý1-4/+4
2010-03-04cc: disallow ... argument unless NOSPLIT is set.Russ Cox1-0/+1
check that NOSPLIT functions don't use too much stack. correct some missing NOSPLITs in the runtime library. Fixes bug reported in https://groups.google.com/group/golang-nuts/t/efff68b73941eccf R=ken2 CC=golang-dev http://codereview.appspot.com/236041
2009-11-06with rsc: argsize rounding to 8 smashes stack data on 32bit machines.Kai Backman1-1/+5
R=rsc http://go/go-review/1024011
2009-07-13fix bug in codegen when we modified move instruction insteadKai Backman1-2/+1
of text. R=rsc APPROVED=rsc DELTA=3 (1 added, 2 deleted, 0 changed) OCL=31575 CL=31575
2009-07-13compilers were inconsistent aboutRuss Cox1-7/+7
whether no register argument was REGARG == 0 or REGARG < 0. use REGARG < 0 because arm needs 0 for R0. R=ken OCL=31562 CL=31566
2009-07-07move 6c/pgen.c, 6c/pswt.c into ccRuss Cox1-0/+591
and make 5c, 8c use them. centralizes reachability analysis and switch generation. now 8c doesn't have spurious warnings in pkg/runtime. R=ken OCL=31266 CL=31266
2009-04-10bug in stack size used inKen Thompson1-550/+0
extending segmented stack R=r OCL=27319 CL=27319
2008-08-03make 6a, 6c, 6g, 6l, libmach_amd64 build on 64-bit gcc.Russ Cox1-2/+2
these guys really really want long to be 32-bits, so ,s/long/int32/ (and then manual fixup). still passes all tests. (i started out looking for just those longs that needed to be int32 instead, and it was just too hard to track them down one by one.) the longs were rare enough that i don't think it will cause integration problems. R=ken OCL=13787 CL=13789
2008-06-04Add compiler source to new directory structureRob Pike1-0/+550
SVN=121164