This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re[2]: [Patch Ping] [RFC] Alias export patch
- From: Dmitry Melnik <dm at ispras dot ru>
- To: Andrew MacLeod <amacleod at redhat dot com>
- Cc: Diego Novillo <dnovillo at redhat dot com>, Daniel Berlin <dberlin at dberlin dot org>, <gcc-patches at gcc dot gnu dot org>
- Date: Wed, 10 May 2006 13:22:20 +0400
- Subject: Re[2]: [Patch Ping] [RFC] Alias export patch
- References: <486713706.20051121174009@ispras.ru> <653679614.20051226024922@ispras.ru> <1136267119.20826.36.camel@linux.site> <1434801581.20060222170518@ispras.ru> <444E546C.5030707@redhat.com> <444E6550.5070006@redhat.com> <1146494975.3532.114.camel@pain>
Hello Andrew,
Monday, May 1, 2006, 6:49:34 PM, you wrote:
> At point we'd have more useful information about what the problem really
> is.. is it really the coalescing, or is it something that happens later.
> Your best bet is to copy 'add_coalesce()' in
> tree-ssa-live.c to 'add_coalesce_check_pta()' or some such name. Make
> that routine check the two ssa_names being coalesced just before
> actually coalescing them. If the PTA info is different, return NULL
> indicating they can't be coalesced.
> IN any case, this might help figure out if the coalescing in out-of-ssa
> is the root cause of PTA info not being useful later on in the compiler.
> let me know if there is anything not clear.
Thank you for your explanation!
We are now trying to get some statistics with add_coalesce() and will
report back later with results.
--
Best regards,
Dmitry