This is the mail archive of the
mailing list for the GCC project.
Re: [PING][PATCH]: Allow CFG with empty BBs in selective scheduler
- From: Vladimir Makarov <vmakarov at redhat dot com>
- To: Alexander Monakov <amonakov at ispras dot ru>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Fri, 24 Apr 2009 13:51:45 -0400
- Subject: Re: [PING][PATCH]: Allow CFG with empty BBs in selective scheduler
- References: <20081031105239.GC21606@endeed2.ispras.ru> <20090125233259.GH3780@endeed2.ispras.ru>
Alexander Monakov wrote:
Hi,Thanks, Alexander. The patch is ok to commit it into the trunk.
I'm not sure if this may count as a regression, as empty basic blocks are
rarely expected at the start of second scheduling pass (the testcase enables
selective scheduling at -Os to reproduce). OK for stage 1?
Please, only merge two changelog entries for sel-sched-ir.c into one:
* sel-sched-ir.c (maybe_tidy_empty_bb): Do not attempt to delete a
block if there are complex incoming edges.
(sel_merge_blocks): Remove useless assert.
(sel_redirect_edge_and_branch): Check that edge was redirected.
* sel-sched-ir.h (_eligible_successor_edge_p): Remove assert.
(sel_find_rgns): Delete declaration.
The patch would be more readable if you used -up in diff. Please, remember about this next time.