no-branch-likely
Eric Christopher
echristo@redhat.com
Wed Feb 27 13:52:00 GMT 2002
>> > One of our implementation groups is implementing a device using the
>> > sibyte chip. They say that using the branch-likely instruction has a
>> > very large performance penalty and are asking for a later compiler,
>> > which supports a no-branch-likely command line option. I could not
>> > find any compiler
>> version
>> > with that option and so fired back a What gives? message. They
>> > referred me to a message from sibyte/Broadcom and I used that to
>> > track down a message from you in January 2001. Its header is:
>> > 2001-01-24 Mike Stump <mrs@wrs.com>
>>
>> > * config/mips/mips.h (mips_no_branch_likely): Add.
>> > (mips_no_branch_likely_string): Add.
>> > (GENERATE_BRANCHLIKELY): Use mips_no_branch_likely.
>> > (TARGET_OPTIONS): Add -mno-branch-likely. *
>> > config/mips/mips.c (mips_no_branch_likely): Add.
>> > (mips_no_branch_likely_string): Add.
>> > (override_options): Handle -mno-branch-likely.
> Done. "Eric"?
That would be me :) I thought I said _something_ about the patch, but
then again I miss them on occasion.
The question I have is whether or not it would be best just to turn off
branch-likely for the sibyte chip in the sources? However, I can see a
reason for the patch anyhow so if Mike would like to check it in then
that's fine.
-eric
--
I will not use abbrev.
More information about the Gcc-patches
mailing list