Il 27/02/2014 23:26, Aaron Ballman ha scritto: > On Thu, Feb 27, 2014 at 5:17 PM, Abramo Bagnara > <[email protected]> wrote: >> $ cat z.c >> void __attribute__((noinline,noreturn)) >> f(void); >> $ _clang -cc1 -ast-dump z.c >> TranslationUnitDecl 0x6630730 <<invalid sloc>> >> |-TypedefDecl 0x6630c30 <<invalid sloc>> __int128_t '__int128' >> |-TypedefDecl 0x6630c90 <<invalid sloc>> __uint128_t 'unsigned __int128' >> |-TypedefDecl 0x6630fe0 <<invalid sloc>> __builtin_va_list >> '__va_list_tag [1]' >> `-FunctionDecl 0x6631120 <z.c:1:1, line:2:7> f 'void (void) >> __attribute__((noreturn))' >> |-NoInlineAttr 0x66311c0 <line:1:21> >> `-NoInlineAttr 0x6631200 <col:21> >> $ clang -cc1 -ast-print z.c >> void f() __attribute__((noinline)) __attribute__((noinline)); >> >> Can you confirm that the doubled attribute noinline should be considered >> a bug? >> >> Should I submit a bugzilla issue? > > Yes, that is definitely a bug -- please CC me on the bugzilla issue, > if you don't mind filing a report.
http://llvm.org/bugs/show_bug.cgi?id=19000 -- Abramo Bagnara BUGSENG srl - http://bugseng.com mailto:[email protected] _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
