------- Comment #5 from rob1weld at aol dot com 2007-05-23 06:29 ------- The patch for 4.1.1 was created by simply using the GEM revision 4.1.0 patch on the release version of the GCC 4.1.1 source. All "hunks" succeeded with minimal offsets.
The patch for 4.2.1 was created by simply using the GEM revision 4.1.0 patch on the 124902 version of the GCC 4.2.1 source. Most "hunks" succeeded with small offsets, some with larger offsets. The were rejects in: gcc/calls.c (1 hunk), gcc/c-common.c (1 hunk) and gcc/Makefile.in (2 hunks). All the rejects were repaired. The patch for 4.3.0 was created by (not so) simply using the GEM revision 4.1.0 patch on the 124940 version of the GCC 4.3.0 source and referring to the 4.2.1 source. Most "hunks" succeeded with moderate (under 200 max) offsets, some went together easily. The were rejects in: gcc/c-decl.c, gcc/gimplify.c, gcc/calls.c, gcc/cfgexpand.c, gcc/c-common.c and gcc/Makefile.in . I did my best to try and apply the 4.3.0 fixes, I am more familiar with 4.2.0. All the rejects were repaired, I hope they are OK. Search for "/* Added " to see a couple of spots were the patch should be fine were it is _OR_ needs to go one line down. It would be great if someone with more 4.3.0 experience could check them out for us and post any fixes. Thanks and enjoy. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32051
