Message ID | CAO2gOZWkp6S05ukHFhjO1CTh6svTrkraX5UO21ygB-sLn_Gheg@mail.gmail.com |
---|---|
State | New |
Headers | show |
> This patch replaces expand_location_to_spelling_point with > expand_location so that macros will not be expanded. I'm confused. Doesn't this just undo the fix you made a month ago? How does this not break discriminator assignment? -cary
This patch does not break the regression test. And because the final assembly is emitted with "expand_location" instead of "expand_location_to_spelling_point", thus when we check same_line_p, it should also reflect this. Dehao On Mon, Jul 1, 2013 at 4:28 PM, Cary Coutant <ccoutant@google.com> wrote: >> This patch replaces expand_location_to_spelling_point with >> expand_location so that macros will not be expanded. > > I'm confused. Doesn't this just undo the fix you made a month ago? How > does this not break discriminator assignment? > > -cary
>>> This patch replaces expand_location_to_spelling_point with >>> expand_location so that macros will not be expanded. OK for google/gcc-4_8. -cary
Index: gcc/tree-cfg.c =================================================================== --- gcc/tree-cfg.c (revision 200584) +++ gcc/tree-cfg.c (working copy) @@ -732,8 +732,8 @@ same_line_p (location_t locus1, location_t locus2) if (locus1 == locus2) return true; - from = expand_location_to_spelling_point (locus1); - to = expand_location_to_spelling_point (locus2); + from = expand_location (locus1); + to = expand_location (locus2); if (from.line != to.line) return false;