This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug target/50135] Loop optimization.


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50135

Uros Bizjak <ubizjak at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |SUSPENDED
   Last reconfirmed|                            |2011-08-21
     Ever Confirmed|0                           |1

--- Comment #2 from Uros Bizjak <ubizjak at gmail dot com> 2011-08-21 10:25:31 UTC ---
Don't worry about loop insn, it is *SLOW* on every half-decent recent
processor, see instruction_tables.pdf at [1]. OTOH, the insn has only short
jump displacement, so you are out of luck for every loop larger than 128 bytes.

So, until x86 implements something like powerpc bdnz insn, there will be no
doloop_* patterns.

[1] http://agner.org/optimize/


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]