From: Richard Biener <rguenther@suse.de>
To: gcc-patches@gcc.gnu.org
Cc: law@redhat.com
Subject: [PATCH] Fix parts of PR61607
Date: Wed, 25 Jun 2014 14:09:00 -0000 [thread overview]
Message-ID: <alpine.LSU.2.11.1406251605030.29270@zhemvz.fhfr.qr> (raw)
This removes prematurely killing loops during jump threading
and moves it to a spot where we know we didn't cancel the
jump thread.
Bootstrapped on x86_64-unknown-linux-gnu, testing in progress.
Ok?
Thanks,
Richard.
2014-06-25 Richard Biener <rguenther@suse.de>
PR tree-optimization/61607
* tree-ssa-threadupdate.c (ssa_redirect_edges): Cancel the
loop if we redirected its latch edge.
(thread_block_1): Do not cancel loops prematurely.
Index: gcc/tree-ssa-threadupdate.c
===================================================================
*** gcc/tree-ssa-threadupdate.c (revision 211978)
--- gcc/tree-ssa-threadupdate.c (working copy)
*************** ssa_redirect_edges (struct redirection_d
*** 764,769 ****
--- 764,777 ----
if ((*path)[1]->type != EDGE_COPY_SRC_JOINER_BLOCK)
EDGE_SUCC (rd->dup_blocks[0], 0)->count += e->count;
+ /* If we redirect a loop latch edge cancel its loop. */
+ if (e->src == e->src->loop_father->latch)
+ {
+ e->src->loop_father->header = NULL;
+ e->src->loop_father->latch = NULL;
+ loops_state_set (LOOPS_NEED_FIXUP);
+ }
+
/* Redirect the incoming edge (possibly to the joiner block) to the
appropriate duplicate block. */
e2 = redirect_edge_and_branch (e, rd->dup_blocks[0]);
*************** thread_block_1 (basic_block bb, bool nol
*** 844,850 ****
edge e, e2;
edge_iterator ei;
ssa_local_info_t local_info;
- struct loop *loop = bb->loop_father;
/* To avoid scanning a linear array for the element we need we instead
use a hash table. For normal code there should be no noticeable
--- 852,857 ----
*************** thread_block_1 (basic_block bb, bool nol
*** 853,884 ****
redirection_data
= new hash_table<struct redirection_data> (EDGE_COUNT (bb->succs));
- /* If we thread the latch of the loop to its exit, the loop ceases to
- exist. Make sure we do not restrict ourselves in order to preserve
- this loop. */
- if (loop->header == bb)
- {
- e = loop_latch_edge (loop);
- vec<jump_thread_edge *> *path = THREAD_PATH (e);
-
- if (path
- && (((*path)[1]->type == EDGE_COPY_SRC_JOINER_BLOCK && joiners)
- || ((*path)[1]->type == EDGE_COPY_SRC_BLOCK && !joiners)))
- {
- for (unsigned int i = 1; i < path->length (); i++)
- {
- edge e2 = (*path)[i]->e;
-
- if (loop_exit_edge_p (loop, e2))
- {
- loop->header = NULL;
- loop->latch = NULL;
- loops_state_set (LOOPS_NEED_FIXUP);
- }
- }
- }
- }
-
/* Record each unique threaded destination into a hash table for
efficient lookups. */
FOR_EACH_EDGE (e, ei, bb->preds)
--- 860,865 ----
next reply other threads:[~2014-06-25 14:09 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-06-25 14:09 Richard Biener [this message]
2014-06-25 18:34 ` Jeff Law
-- strict thread matches above, loose matches on Subject: below --
2014-06-25 14:07 Richard Biener
2014-06-25 18:28 ` Jeff Law
2014-06-26 7:57 ` Richard Biener
2014-06-26 8:33 ` Richard Biener
2014-06-26 9:01 ` Richard Biener
2014-06-26 15:33 ` Jeff Law
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=alpine.LSU.2.11.1406251605030.29270@zhemvz.fhfr.qr \
--to=rguenther@suse.de \
--cc=gcc-patches@gcc.gnu.org \
--cc=law@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).