This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [rs6000] Print debug log of rs6000 variable_issue invocations
- From: David Edelsohn <dje dot gcc at gmail dot com>
- To: Andrey Belevantsev <abel at ispras dot ru>
- Cc: Alexander Monakov <amonakov at ispras dot ru>, gcc-patches at gcc dot gnu dot org
- Date: Wed, 13 Jan 2010 20:24:36 -0500
- Subject: Re: [rs6000] Print debug log of rs6000 variable_issue invocations
- References: <4B3B7703.1070409@ispras.ru> <1262190835-15573-11-git-send-email-amonakov@ispras.ru> <4B4E0848.9000102@ispras.ru>
2010/1/13 Andrey Belevantsev <abel@ispras.ru>:
> On 30.12.2009 19:33, Alexander Monakov wrote:
>>
>> This patch was useful when debugging scheduler behaviour on powerpc.
>> Certainly not a big deal, but I'd prefer to have it in mainline (it may be
>> useful again in future).
>>
>> 2009-12-30 ?Alexander Monakov<amonakov@ispras.ru>
>>
>> ? ? ? ?* config/rs6000/rs6000.c (rs6000_variable_issue): Rename to...
>> ? ? ? ?(rs6000_variable_issue_1): this. ?Use...
>> ? ? ? ?(rs6000_variable_issue): here. ?Reimplement. ?Print debug info.
The intent is fine, but I am a little concerned about the
implementation: wrapping the entire function for a verbose flag
infrequently used in production mode. I understand the reason given
the multiple return points and multiple callsites.
GCC appears to inline the function, so this patch is okay.
Thanks, David