From: Richard Biener <rguenther@suse.de>
To: gcc-patches@gcc.gnu.org
Subject: [PATCH][2/2] Fix PR77768
Date: Wed, 28 Sep 2016 11:42:00 -0000 [thread overview]
Message-ID: <alpine.LSU.2.11.1609281334480.26629@t29.fhfr.qr> (raw)
The following patch makes VN not choke on code that stores to readonly
memory it knows the constant value of. I took the liberty to clean up
the surrounding code a bit as well.
Bootstrap / regtest in progress on x86_64-unknown-linux-gnu.
Richard.
2016-09-28 Richard Biener <rguenther@suse.de>
PR tree-optimization/77768
* tree-ssa-sccvn.c (visit_reference_op_store): Properly deal
with stores to a place we know has a constant value.
Index: gcc/tree-ssa-sccvn.c
===================================================================
--- gcc/tree-ssa-sccvn.c (revision 240566)
+++ gcc/tree-ssa-sccvn.c (working copy)
@@ -3575,7 +3575,7 @@ visit_reference_op_store (tree lhs, tree
{
bool changed = false;
vn_reference_t vnresult = NULL;
- tree result, assign;
+ tree assign;
bool resultsame = false;
tree vuse = gimple_vuse (stmt);
tree vdef = gimple_vdef (stmt);
@@ -3599,9 +3599,11 @@ visit_reference_op_store (tree lhs, tree
Otherwise, the vdefs for the store are used when inserting into
the table, since the store generates a new memory state. */
- result = vn_reference_lookup (lhs, vuse, VN_NOWALK, &vnresult, false);
- if (result)
+ vn_reference_lookup (lhs, vuse, VN_NOWALK, &vnresult, false);
+ if (vnresult
+ && vnresult->result)
{
+ tree result = vnresult->result;
if (TREE_CODE (result) == SSA_NAME)
result = SSA_VAL (result);
resultsame = expressions_equal_p (result, op);
@@ -3616,22 +3618,21 @@ visit_reference_op_store (tree lhs, tree
}
}
- if ((!result || !resultsame)
+ if (!resultsame)
+ {
/* Only perform the following when being called from PRE
which embeds tail merging. */
- && default_vn_walk_kind == VN_WALK)
- {
- assign = build2 (MODIFY_EXPR, TREE_TYPE (lhs), lhs, op);
- vn_reference_lookup (assign, vuse, VN_NOWALK, &vnresult, false);
- if (vnresult)
+ if (default_vn_walk_kind == VN_WALK)
{
- VN_INFO (vdef)->use_processed = true;
- return set_ssa_val_to (vdef, vnresult->result_vdef);
+ assign = build2 (MODIFY_EXPR, TREE_TYPE (lhs), lhs, op);
+ vn_reference_lookup (assign, vuse, VN_NOWALK, &vnresult, false);
+ if (vnresult)
+ {
+ VN_INFO (vdef)->use_processed = true;
+ return set_ssa_val_to (vdef, vnresult->result_vdef);
+ }
}
- }
- if (!result || !resultsame)
- {
if (dump_file && (dump_flags & TDF_DETAILS))
{
fprintf (dump_file, "No store match\n");
@@ -3644,9 +3645,7 @@ visit_reference_op_store (tree lhs, tree
/* Have to set value numbers before insert, since insert is
going to valueize the references in-place. */
if (vdef)
- {
- changed |= set_ssa_val_to (vdef, vdef);
- }
+ changed |= set_ssa_val_to (vdef, vdef);
/* Do not insert structure copies into the tables. */
if (is_gimple_min_invariant (op)
reply other threads:[~2016-09-28 11:36 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
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.2.11.1609281334480.26629@t29.fhfr.qr \
--to=rguenther@suse.de \
--cc=gcc-patches@gcc.gnu.org \
/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).