public inbox for gcc-cvs@sourceware.org
help / color / mirror / Atom feed
* [gcc r14-8910] gimple-low: Fix up handling of volatile automatic vars in assume attribute [PR110754]
@ 2024-02-10 10:28 Jakub Jelinek
0 siblings, 0 replies; only message in thread
From: Jakub Jelinek @ 2024-02-10 10:28 UTC (permalink / raw)
To: gcc-cvs
https://gcc.gnu.org/g:39920447f876128ff7942a9cd931021800865894
commit r14-8910-g39920447f876128ff7942a9cd931021800865894
Author: Jakub Jelinek <jakub@redhat.com>
Date: Sat Feb 10 11:28:00 2024 +0100
gimple-low: Fix up handling of volatile automatic vars in assume attribute [PR110754]
As the following testcases show, the gimple-low outlining of assume
magic functions handled volatile automatic vars (including
parameters/results) like non-volatile ones except it copied volatile
to the new PARM_DECL, which has the undesirable effect that a load
from the volatile var is passed to IFN_ASSUME and so there is a
side-effect there even when side-effects of the assume attribute
shouldn't be evaluated.
The following patch fixes that by passing address of the volatile
variables/parameters/results instead and doing loads or stores from it
or to it where it was originally accessed in the assume attribute
expression.
2024-02-10 Jakub Jelinek <jakub@redhat.com>
PR middle-end/110754
* gimple-low.cc (assumption_copy_decl): For TREE_THIS_VOLATILE
decls create PARM_DECL with pointer to original type, set
TREE_READONLY and keep TREE_THIS_VOLATILE, TREE_ADDRESSABLE,
DECL_NOT_GIMPLE_REG_P and DECL_BY_REFERENCE cleared.
(adjust_assumption_stmt_op): For remapped TREE_THIS_VOLATILE decls
wrap PARM_DECL into a simple TREE_THIS_NO_TRAP MEM_REF.
(lower_assumption): For TREE_THIS_VOLATILE vars pass ADDR_EXPR
of the var as argument.
* gcc.dg/attr-assume-6.c: New test.
* g++.dg/cpp23/attr-assume12.C: New test.
Diff:
---
gcc/gimple-low.cc | 27 ++++++++++++++++++++++-----
gcc/testsuite/g++.dg/cpp23/attr-assume12.C | 14 ++++++++++++++
gcc/testsuite/gcc.dg/attr-assume-6.c | 14 ++++++++++++++
3 files changed, 50 insertions(+), 5 deletions(-)
diff --git a/gcc/gimple-low.cc b/gcc/gimple-low.cc
index e19fc2cce9d4..e03719887050 100644
--- a/gcc/gimple-low.cc
+++ b/gcc/gimple-low.cc
@@ -374,15 +374,22 @@ assumption_copy_decl (tree decl, copy_body_data *id)
gcc_assert (VAR_P (decl)
|| TREE_CODE (decl) == PARM_DECL
|| TREE_CODE (decl) == RESULT_DECL);
+ if (TREE_THIS_VOLATILE (decl))
+ type = build_pointer_type (type);
tree copy = build_decl (DECL_SOURCE_LOCATION (decl),
PARM_DECL, DECL_NAME (decl), type);
if (DECL_PT_UID_SET_P (decl))
SET_DECL_PT_UID (copy, DECL_PT_UID (decl));
- TREE_ADDRESSABLE (copy) = TREE_ADDRESSABLE (decl);
- TREE_READONLY (copy) = TREE_READONLY (decl);
- TREE_THIS_VOLATILE (copy) = TREE_THIS_VOLATILE (decl);
- DECL_NOT_GIMPLE_REG_P (copy) = DECL_NOT_GIMPLE_REG_P (decl);
- DECL_BY_REFERENCE (copy) = DECL_BY_REFERENCE (decl);
+ TREE_THIS_VOLATILE (copy) = 0;
+ if (TREE_THIS_VOLATILE (decl))
+ TREE_READONLY (copy) = 1;
+ else
+ {
+ TREE_ADDRESSABLE (copy) = TREE_ADDRESSABLE (decl);
+ TREE_READONLY (copy) = TREE_READONLY (decl);
+ DECL_NOT_GIMPLE_REG_P (copy) = DECL_NOT_GIMPLE_REG_P (decl);
+ DECL_BY_REFERENCE (copy) = DECL_BY_REFERENCE (decl);
+ }
DECL_ARG_TYPE (copy) = type;
((lower_assumption_data *) id)->decls.safe_push (decl);
return copy_decl_for_dup_finish (id, decl, copy);
@@ -466,6 +473,11 @@ adjust_assumption_stmt_op (tree *tp, int *, void *datap)
case PARM_DECL:
case RESULT_DECL:
*tp = remap_decl (t, &data->id);
+ if (TREE_THIS_VOLATILE (t) && *tp != t)
+ {
+ *tp = build_simple_mem_ref (*tp);
+ TREE_THIS_NOTRAP (*tp) = 1;
+ }
break;
default:
break;
@@ -600,6 +612,11 @@ lower_assumption (gimple_stmt_iterator *gsi, struct lower_data *data)
/* Remaining arguments will be the variables/parameters
mentioned in the condition. */
vargs[i - sz] = lad.decls[i - 1];
+ if (TREE_THIS_VOLATILE (lad.decls[i - 1]))
+ {
+ TREE_ADDRESSABLE (lad.decls[i - 1]) = 1;
+ vargs[i - sz] = build_fold_addr_expr (lad.decls[i - 1]);
+ }
/* If they have gimple types, we might need to regimplify
them to make the IFN_ASSUME call valid. */
if (is_gimple_reg_type (TREE_TYPE (vargs[i - sz]))
diff --git a/gcc/testsuite/g++.dg/cpp23/attr-assume12.C b/gcc/testsuite/g++.dg/cpp23/attr-assume12.C
new file mode 100644
index 000000000000..c955cd1d1e83
--- /dev/null
+++ b/gcc/testsuite/g++.dg/cpp23/attr-assume12.C
@@ -0,0 +1,14 @@
+// PR middle-end/110754
+// { dg-do compile { target c++11 } }
+// { dg-options "-O2 -fdump-tree-optimized" }
+// { dg-final { scan-tree-dump-times "a ={v} x" 1 "optimized" } }
+// { dg-final { scan-tree-dump-not "={v} a" "optimized" } }
+// { dg-final { scan-tree-dump-times "return 0;" 1 "optimized" } }
+
+int
+foo (int x)
+{
+ volatile int a = x;
+ [[gnu::assume (x == (a & 0))]];
+ return x;
+}
diff --git a/gcc/testsuite/gcc.dg/attr-assume-6.c b/gcc/testsuite/gcc.dg/attr-assume-6.c
new file mode 100644
index 000000000000..346845f1600f
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/attr-assume-6.c
@@ -0,0 +1,14 @@
+/* PR middle-end/110754 */
+/* { dg-do compile } */
+/* { dg-options "-std=c23 -O2 -fdump-tree-optimized" } */
+/* { dg-final { scan-tree-dump-times "a ={v} x" 1 "optimized" } } */
+/* { dg-final { scan-tree-dump-not "={v} a" "optimized" } } */
+/* { dg-final { scan-tree-dump-times "return 0;" 1 "optimized" } } */
+
+int
+foo (int x)
+{
+ volatile int a = x;
+ [[gnu::assume (x == (a & 0))]];
+ return x;
+}
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2024-02-10 10:28 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2024-02-10 10:28 [gcc r14-8910] gimple-low: Fix up handling of volatile automatic vars in assume attribute [PR110754] Jakub Jelinek
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).