@@ -335,7 +335,7 @@ def note_constexpr_new_non_replaceable : Note<
def note_constexpr_new_placement : Note<
"this placement new expression is not yet supported in constant
expressions">;
def note_constexpr_placement_new_wrong_type : Note<
- "placement new would change type
@@ -10003,23 +10003,14 @@ bool PointerExprEvaluator::VisitCXXNewExpr(const
CXXNewExpr *E) {
return false;
FunctionDecl *OperatorNew = E->getOperatorNew();
+ QualType AllocType = E->getAllocatedType();
+ QualType TargetType = AllocType;
bool IsNothrow = false;
@@ -994,7 +994,7 @@ namespace placement_new_delete {
constexpr bool bad(int which) {
switch (which) {
case 0:
- delete new (placement_new_arg{}) int; // expected-note {{call to
placement 'operator new'}}
+ delete new (placement_new_arg{}) int; // expected-n
@@ -0,0 +1,78 @@
+// RUN: %clang_cc1 -std=c++2c -verify %s
+
+
+namespace std {
+ using size_t = decltype(sizeof(0));
+}
+
+void *operator new(std::size_t, void *p) { return p; }
+void* operator new[] (std::size_t, void* p) {return p;}
+
+
+consteval int ok() {
+int i;
+n
@@ -0,0 +1,78 @@
+// RUN: %clang_cc1 -std=c++2c -verify %s
+
+
+namespace std {
+ using size_t = decltype(sizeof(0));
+}
+
+void *operator new(std::size_t, void *p) { return p; }
+void* operator new[] (std::size_t, void* p) {return p;}
+
+
+consteval int ok() {
+int i;
+n
@@ -309,7 +309,7 @@ namespace placement_new_delete {
constexpr bool bad(int which) {
switch (which) {
case 0:
- delete new (placement_new_arg{}) int; // ref-note {{call to placement
'operator new'}} \
+ delete new (placement_new_arg{}) int; // ref-note {{th
https://github.com/cor3ntin edited
https://github.com/llvm/llvm-project/pull/104586
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
cor3ntin wrote:
> Thank you for working on this! Given that it's an extension in C++20 mode,
> we're missing diagnostics and test coverage for issuing extension/precompat
> diagnostics.
No longer an extension after our last chat - i just forgot to update the
description
https://github.com/ll
@@ -6691,7 +6691,7 @@ static bool HandleDestructionImpl(EvalInfo &Info,
SourceRange CallRange,
if (Size && Size > Value.getArrayInitializedElts())
expandArray(Value, Value.getArraySize() - 1);
-for (; Size != 0; --Size) {
+for (Size = Value.getArraySize(); S
@@ -245,7 +245,7 @@ namespace std {
namespace PlacementNew {
constexpr int foo() { // both-error {{never produces a constant expression}}
char c[sizeof(int)];
-new (c) int{12}; // ref-note {{call to placement 'operator new'}} \
+new (c) int{12}; // ref-note {{this
@@ -309,7 +309,7 @@ namespace placement_new_delete {
constexpr bool bad(int which) {
switch (which) {
case 0:
- delete new (placement_new_arg{}) int; // ref-note {{call to placement
'operator new'}} \
+ delete new (placement_new_arg{}) int; // ref-note {{th
@@ -0,0 +1,78 @@
+// RUN: %clang_cc1 -std=c++2c -verify %s
+
+
+namespace std {
+ using size_t = decltype(sizeof(0));
+}
+
+void *operator new(std::size_t, void *p) { return p; }
+void* operator new[] (std::size_t, void* p) {return p;}
+
+
+consteval int ok() {
+int i;
+n
@@ -0,0 +1,252 @@
+#ifndef LLVM_CLANG_AST_DYNAMIC_RECURSIVE_AST_VISITOR_H
+#define LLVM_CLANG_AST_DYNAMIC_RECURSIVE_AST_VISITOR_H
+
+#include "clang/AST/Attr.h"
+#include "clang/AST/ExprConcepts.h"
+#include "clang/AST/TypeLoc.h"
+
+namespace clang {
+class ASTContext;
+
+/// Recu
@@ -245,7 +245,7 @@ namespace std {
namespace PlacementNew {
constexpr int foo() { // both-error {{never produces a constant expression}}
char c[sizeof(int)];
-new (c) int{12}; // ref-note {{call to placement 'operator new'}} \
+new (c) int{12}; // ref-note {{this
lenary wrote:
This is also a limitation if you use `@` as a prefix as suggested in the
discourse thread - GCC's `-print-multi-lib` prints out flags with the first
`-` replaced with an `@`, which iirc some build systems also parse (i recall
crosstool-ng, but maybe it's newlib within a crosstoo
@@ -6691,7 +6691,7 @@ static bool HandleDestructionImpl(EvalInfo &Info,
SourceRange CallRange,
if (Size && Size > Value.getArrayInitializedElts())
expandArray(Value, Value.getArraySize() - 1);
-for (; Size != 0; --Size) {
+for (Size = Value.getArraySize(); S
Sirraide wrote:
> I wonder if we can reduce binary size by simply marking RAV as
> `visibility("hidden")`
Hmm, I haven’t thought about that, but I’m not convinced it would help: the
main issue w/ the RAV is that to traverse *every possible AST* that you might
end up handing it, you need an in
https://github.com/cor3ntin updated
https://github.com/llvm/llvm-project/pull/104586
>From 67201d1d2e4d092cf8712b1e1b63f2eceaf8d57c Mon Sep 17 00:00:00 2001
From: Corentin Jabot
Date: Tue, 9 Jul 2024 08:37:18 +0200
Subject: [PATCH 1/4] [Clang] Implement P2747 constexpr placement new
In C++26 a
@@ -1128,15 +1132,22 @@ struct CounterCoverageMappingBuilder
BranchParams = mcdc::BranchParameters{ID, Conds};
// If a condition can fold to true or false, the corresponding branch
- // will be removed. Create a region with both counters hard-coded to
-
@@ -47,7 +47,7 @@
// CHECK: Branch (103:9): [True: 9, False: 1]
// CHECK: switches()
-// CHECK: Branch (113:3): [True: 1, False: 0]
+// CHECK: Branch (113:3): [Folded - Ignored]
evodius96 wrote:
I'm not sure I see why the evaluation of this case (and line 60)
evodius96 wrote:
> I have no authority to request reviewers or commit this patch. So let's @
> someone, @chapuni @ornata @hanickadot
I had some comments in the pending state that I just now noticed :( sorry about
that.
https://github.com/llvm/llvm-project/pull/94137
__
Meinersbur wrote:
I agree with @kparzysz that the examples you are using are generally unsafe.
E.g. with
```c
#pragma omp parallel for collapse(2)
for (int i = 0; i < N; i++) {
arr[i][i] = ...;
for (int j = 0; j < N; j++) {
arr[i][j] = ...;
}
}
```
there shoul
llvm-ci wrote:
LLVM Buildbot has detected a new failure on builder `sanitizer-x86_64-linux`
running on `sanitizer-buildbot2` while building `clang` at step 2 "annotate".
Full details are available at:
https://lab.llvm.org/buildbot/#/builders/66/builds/2982
Here is the relevant piece of the bu
llvm-ci wrote:
LLVM Buildbot has detected a new failure on builder `bolt-x86_64-ubuntu-nfc`
running on `bolt-worker` while building `clang,llvm` at step 9
"test-build-bolt-check-large-bolt".
Full details are available at:
https://lab.llvm.org/buildbot/#/builders/92/builds/4968
Here is the re
https://github.com/andykaylor updated
https://github.com/llvm/llvm-project/pull/104857
>From 59ac754b9a88037e2348f16e5dc7efbd437d65cb Mon Sep 17 00:00:00 2001
From: Andy Kaylor
Date: Mon, 19 Aug 2024 13:41:40 -0700
Subject: [PATCH 1/3] Fix bug with -ffp-contract=fast-honor-pragmas
This fixes a
https://github.com/andykaylor closed
https://github.com/llvm/llvm-project/pull/104857
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Author: Andy Kaylor
Date: 2024-08-21T08:10:26-07:00
New Revision: c0d19a8d01d3945100114256d26cfe833a1c
URL:
https://github.com/llvm/llvm-project/commit/c0d19a8d01d3945100114256d26cfe833a1c
DIFF:
https://github.com/llvm/llvm-project/commit/c0d19a8d01d3945100114256d26cfe833a1c.diff
L
nikic wrote:
Okay, looks like I tracked down all the assertion failures in tests. My plan
would now be to land this in three phases, first to add the new assertions and
implicitTrunc parameter, but with implicitTrunc=true by default. Then to land
fixes (possibly split up further), and finally
lei137 wrote:
> I think the issue could be handled a different (more generic) way, by pulling
> `useFirstFieldIfTransparentUnion` to the caller and taking transparent unions
> into account when emitting LLVM IR for the formal / actual parameters
> somewhere in `CGCall.cpp`, so that ABIInfo imp
https://github.com/MaskRay approved this pull request.
https://github.com/llvm/llvm-project/pull/104742
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
s-barannikov wrote:
> However I feel like it is out of scope for what I am trying to do right now.
> Would this be better as a separate PR to change how this is handled?
Sure
https://github.com/llvm/llvm-project/pull/104816
___
cfe-commits mailing l
https://github.com/MitalAshok edited
https://github.com/llvm/llvm-project/pull/102948
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/matinraayai created
https://github.com/llvm/llvm-project/pull/105541
This PR makes the following changes, addressing issues raised in #104834 and
#98770:
1. MMIWP behaves like the new PM's analysis Pass; It only takes a reference to
an externally created MMI.
2. Removed the
https://github.com/matinraayai converted_to_draft
https://github.com/llvm/llvm-project/pull/105541
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
llvmbot wrote:
@llvm/pr-subscribers-lto
@llvm/pr-subscribers-tools-llvm-exegesis
Author: Matin Raayai (matinraayai)
Changes
This PR makes the following changes, addressing issues raised in #104834 and #98770:
1. MMIWP behaves like the new PM's analysis Pass; It only takes a reference to
a
https://github.com/matinraayai edited
https://github.com/llvm/llvm-project/pull/105541
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
github-actions[bot] wrote:
⚠️ We detected that you are using a GitHub private e-mail address to contribute
to the repo. Please turn off [Keep my email addresses
private](https://github.com/settings/emails) setting in your account. See
[LLVM
Discourse](https://discourse.llvm.org/t/hidden-email
github-actions[bot] wrote:
:warning: C/C++ code formatter, clang-format found issues in your code.
:warning:
You can test this locally with the following command:
``bash
git-clang-format --diff 93eda08babe95188ee41400035abaade79cda7d1
923978640a76a33d77ab2c794953479efa86b73c --e
https://github.com/jurahul created
https://github.com/llvm/llvm-project/pull/105544
- Refactor SetTheory code to use const pointers when possible.
- Use auto for variables initialized using dyn_cast<>.
- Use range based for loops and early continue.
>From b9b2cea49dab03ad2f01b70ffa784be8a5e23b6
@@ -5968,11 +5962,21 @@
ASTNodeImporter::VisitTemplateTypeParmDecl(TemplateTypeParmDecl *D) {
}
if (D->hasDefaultArgument()) {
+// Default argument can be "inherited" when it has a reference to the
+// previous declaration (of the default argument) which is stored
github-actions[bot] wrote:
:warning: C/C++ code formatter, clang-format found issues in your code.
:warning:
You can test this locally with the following command:
``bash
git-clang-format --diff 6816a137985bfa38cda20b9cd4e23c361c3bd0de
b9b2cea49dab03ad2f01b70ffa784be8a5e23b6d --e
https://github.com/jurahul updated
https://github.com/llvm/llvm-project/pull/105544
>From 6ead1e81f891a826b55aa74ef13b132eb854bb47 Mon Sep 17 00:00:00 2001
From: Rahul Joshi
Date: Wed, 21 Aug 2024 05:59:33 -0700
Subject: [PATCH] [NFC][SetTheory] Refactor to use const pointers and range
loops
@@ -5968,11 +5962,21 @@
ASTNodeImporter::VisitTemplateTypeParmDecl(TemplateTypeParmDecl *D) {
}
if (D->hasDefaultArgument()) {
+// Default argument can be "inherited" when it has a reference to the
+// previous declaration (of the default argument) which is stored
@@ -5968,11 +5962,21 @@
ASTNodeImporter::VisitTemplateTypeParmDecl(TemplateTypeParmDecl *D) {
}
if (D->hasDefaultArgument()) {
+// Default argument can be "inherited" when it has a reference to the
+// previous declaration (of the default argument) which is stored
Author: magic-akari
Date: 2024-08-21T18:22:21+02:00
New Revision: 625841c3be4dbaab089c01217726a2906f3a8103
URL:
https://github.com/llvm/llvm-project/commit/625841c3be4dbaab089c01217726a2906f3a8103
DIFF:
https://github.com/llvm/llvm-project/commit/625841c3be4dbaab089c01217726a2906f3a8103.diff
L
https://github.com/HazardyKnusperkeks closed
https://github.com/llvm/llvm-project/pull/100978
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
github-actions[bot] wrote:
@magic-akari Congratulations on having your first Pull Request (PR) merged into
the LLVM Project!
Your changes will be combined with recent changes from other authors, then
tested by our [build bots](https://lab.llvm.org/buildbot/). If there is a
problem with a bu
HazardyKnusperkeks wrote:
How does it look like without the new?
https://github.com/llvm/llvm-project/pull/105168
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/matinraayai updated
https://github.com/llvm/llvm-project/pull/105541
>From 4030c78148a2cb9b584038bd2ce522906b710f7e Mon Sep 17 00:00:00 2001
From: matinraayai <30674652+matinraa...@users.noreply.github.com>
Date: Tue, 20 Aug 2024 14:22:11 -0400
Subject: [PATCH 1/3] Made MMI on
jhuber6 wrote:
```llvm/llvm-project/build/lib/clang/20/include/avx10_2convertintrin.h:29:19:
error: use of undeclared identifier '__builtin_ia32_vcvt2ps2phx128_mask'
29 | return (__m128h)__builtin_ia32_vcvt2ps2phx128_mask(
| ^
```
I'm getting a lot of errors like thi
jhuber6 wrote:
My guess is that you added these as a target dependent builtin, but made the
headers always reference them.
https://github.com/llvm/llvm-project/pull/101600
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/
jhuber6 wrote:
This is all it takes to reproduce this on my machine so I'm going to revert.
x64 Linux, CPU is `znver3`.
```console
echo "#include " | ./bin/clang++ -x c++ -
```
https://github.com/llvm/llvm-project/pull/101600
___
cfe-commits mailing l
@@ -245,7 +245,7 @@ namespace std {
namespace PlacementNew {
constexpr int foo() { // both-error {{never produces a constant expression}}
char c[sizeof(int)];
-new (c) int{12}; // ref-note {{call to placement 'operator new'}} \
+new (c) int{12}; // ref-note {{this
@@ -0,0 +1,115 @@
+// RUN: %clang_cc1 -std=c++2c -verify %s
+
+
+namespace std {
+ using size_t = decltype(sizeof(0));
+}
+
+void *operator new(std::size_t, void *p) { return p; }
+void* operator new[] (std::size_t, void* p) {return p;}
+
+
+consteval int ok() {
+int i;
+
JustinStitt wrote:
@nico
> The test added here was tweaked a bit and then deleted in
> [07a8cba](https://github.com/llvm/llvm-project/commit/07a8cbaf8dc16bebf6e875173d20299d9cc47cc5)
>
> What gives?
>
> Instead of deleting tests, we should revert the PR that adds them and then
> reland the P
@@ -393,22 +393,17 @@ struct CUDADeviceTy : public GenericDeviceTy {
return Plugin::success();
}
- virtual Error callGlobalConstructors(GenericPluginTy &Plugin,
- DeviceImageTy &Image) override {
-// Check for the presense of gl
https://github.com/yonghong-song approved this pull request.
https://github.com/llvm/llvm-project/pull/105417
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/matinraayai updated
https://github.com/llvm/llvm-project/pull/105541
>From 4030c78148a2cb9b584038bd2ce522906b710f7e Mon Sep 17 00:00:00 2001
From: matinraayai <30674652+matinraa...@users.noreply.github.com>
Date: Tue, 20 Aug 2024 14:22:11 -0400
Subject: [PATCH 1/3] Made MMI on
@@ -438,6 +438,36 @@ Moved checkers
Sanitizers
--
+- Added the ``-fsanitize-undefined-ignore-overflow-pattern`` flag which can be
+ used to disable specific overflow-dependent code patterns. The supported
+ patterns are: ``add-overflow-test``, ``negated-unsigned-cons
@@ -293,6 +293,48 @@ To silence reports from unsigned integer overflow, you can
set
``-fsanitize-recover=unsigned-integer-overflow``, is particularly useful for
providing fuzzing signal without blowing up logs.
+Disabling instrumentation for common overflow patterns
+
https://github.com/smanna12 created
https://github.com/llvm/llvm-project/pull/105556
None
>From 3ea17898c7281195fc3def45485cc3528c743ded Mon Sep 17 00:00:00 2001
From: "Manna, Soumi"
Date: Wed, 21 Aug 2024 10:26:42 -0700
Subject: [PATCH] [Clang] Fix null pointer dereference in enum debug info
@@ -293,6 +293,48 @@ To silence reports from unsigned integer overflow, you can
set
``-fsanitize-recover=unsigned-integer-overflow``, is particularly useful for
providing fuzzing signal without blowing up logs.
+Disabling instrumentation for common overflow patterns
+
https://github.com/smanna12 edited
https://github.com/llvm/llvm-project/pull/105556
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -293,6 +293,48 @@ To silence reports from unsigned integer overflow, you can
set
``-fsanitize-recover=unsigned-integer-overflow``, is particularly useful for
providing fuzzing signal without blowing up logs.
+Disabling instrumentation for common overflow patterns
+
https://github.com/vitalybuka edited
https://github.com/llvm/llvm-project/pull/104889
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -438,6 +438,36 @@ Moved checkers
Sanitizers
--
+- Added the ``-fsanitize-undefined-ignore-overflow-pattern`` flag which can be
+ used to disable specific overflow-dependent code patterns. The supported
+ patterns are: ``add-overflow-test``, ``negated-unsigned-cons
@@ -367,6 +367,21 @@ class LangOptionsBase {
PerThread,
};
+ /// Exclude certain code patterns from being instrumented by arithmetic
+ /// overflow sanitizers
+ enum OverflowPatternExclusionKind {
+/// Don't exclude any overflow patterns from sanitizers
+None
@@ -293,6 +293,48 @@ To silence reports from unsigned integer overflow, you can
set
``-fsanitize-recover=unsigned-integer-overflow``, is particularly useful for
providing fuzzing signal without blowing up logs.
+Disabling instrumentation for common overflow patterns
+
https://github.com/vitalybuka edited
https://github.com/llvm/llvm-project/pull/104889
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -293,6 +293,48 @@ To silence reports from unsigned integer overflow, you can
set
``-fsanitize-recover=unsigned-integer-overflow``, is particularly useful for
providing fuzzing signal without blowing up logs.
+Disabling instrumentation for common overflow patterns
+
@@ -293,6 +293,48 @@ To silence reports from unsigned integer overflow, you can
set
``-fsanitize-recover=unsigned-integer-overflow``, is particularly useful for
providing fuzzing signal without blowing up logs.
+Disabling instrumentation for common overflow patterns
+
eddyz87 wrote:
@yonghong-song , @AaronBallman , @nikic , thank you for the review!
https://github.com/llvm/llvm-project/pull/105417
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -293,6 +293,48 @@ To silence reports from unsigned integer overflow, you can
set
``-fsanitize-recover=unsigned-integer-overflow``, is particularly useful for
providing fuzzing signal without blowing up logs.
+Disabling instrumentation for common overflow patterns
+
@@ -0,0 +1,63 @@
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu
-fsanitize=signed-integer-overflow,unsigned-integer-overflow
-fsanitize-undefined-ignore-overflow-pattern=all %s -emit-llvm -o - | FileCheck
%s
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu
-fsanitize
@@ -438,6 +438,36 @@ Moved checkers
Sanitizers
--
+- Added the ``-fsanitize-undefined-ignore-overflow-pattern`` flag which can be
+ used to disable specific overflow-dependent code patterns. The supported
+ patterns are: ``add-overflow-test``, ``negated-unsigned-cons
@@ -293,6 +293,48 @@ To silence reports from unsigned integer overflow, you can
set
``-fsanitize-recover=unsigned-integer-overflow``, is particularly useful for
providing fuzzing signal without blowing up logs.
+Disabling instrumentation for common overflow patterns
+
https://github.com/vitalybuka deleted
https://github.com/llvm/llvm-project/pull/104889
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -293,6 +293,48 @@ To silence reports from unsigned integer overflow, you can
set
``-fsanitize-recover=unsigned-integer-overflow``, is particularly useful for
providing fuzzing signal without blowing up logs.
+Disabling instrumentation for common overflow patterns
+
https://github.com/Harini0924 edited
https://github.com/llvm/llvm-project/pull/102647
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -293,6 +293,48 @@ To silence reports from unsigned integer overflow, you can
set
``-fsanitize-recover=unsigned-integer-overflow``, is particularly useful for
providing fuzzing signal without blowing up logs.
+Disabling instrumentation for common overflow patterns
+
https://github.com/Harini0924 edited
https://github.com/llvm/llvm-project/pull/102647
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -293,6 +293,48 @@ To silence reports from unsigned integer overflow, you can
set
``-fsanitize-recover=unsigned-integer-overflow``, is particularly useful for
providing fuzzing signal without blowing up logs.
+Disabling instrumentation for common overflow patterns
+
Author: Harini0924
Date: 2024-08-21T10:48:32-07:00
New Revision: c975dc1da03d684604ddf787b07b63fb8e903648
URL:
https://github.com/llvm/llvm-project/commit/c975dc1da03d684604ddf787b07b63fb8e903648
DIFF:
https://github.com/llvm/llvm-project/commit/c975dc1da03d684604ddf787b07b63fb8e903648.diff
LO
https://github.com/vitalybuka edited
https://github.com/llvm/llvm-project/pull/104889
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/ilovepi closed
https://github.com/llvm/llvm-project/pull/102647
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -367,6 +367,21 @@ class LangOptionsBase {
PerThread,
};
+ /// Exclude certain code patterns from being instrumented by arithmetic
+ /// overflow sanitizers
+ enum OverflowPatternExclusionKind {
+/// Don't exclude any overflow patterns from sanitizers
+None
https://github.com/vitalybuka edited
https://github.com/llvm/llvm-project/pull/104889
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/vitalybuka edited
https://github.com/llvm/llvm-project/pull/104889
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/vitalybuka edited
https://github.com/llvm/llvm-project/pull/104889
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/vitalybuka edited
https://github.com/llvm/llvm-project/pull/104889
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/vitalybuka edited
https://github.com/llvm/llvm-project/pull/104889
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
kadircet wrote:
```
$ cat a.cc
fooo(BAR(
XXXZ()));
```
```
$ clang-format a.cc
fooo(BAR(
XXXZ()));
```
https://github.com/llvm/llvm-project/pul
@@ -438,6 +438,36 @@ Moved checkers
Sanitizers
--
+- Added the ``-fsanitize-undefined-ignore-overflow-pattern`` flag which can be
+ used to disable specific overflow-dependent code patterns. The supported
+ patterns are: ``add-overflow-test``, ``negated-unsigned-cons
@@ -293,6 +293,48 @@ To silence reports from unsigned integer overflow, you can
set
``-fsanitize-recover=unsigned-integer-overflow``, is particularly useful for
providing fuzzing signal without blowing up logs.
+Disabling instrumentation for common overflow patterns
+
@@ -293,6 +293,48 @@ To silence reports from unsigned integer overflow, you can
set
``-fsanitize-recover=unsigned-integer-overflow``, is particularly useful for
providing fuzzing signal without blowing up logs.
+Disabling instrumentation for common overflow patterns
+
@@ -293,6 +293,48 @@ To silence reports from unsigned integer overflow, you can
set
``-fsanitize-recover=unsigned-integer-overflow``, is particularly useful for
providing fuzzing signal without blowing up logs.
+Disabling instrumentation for common overflow patterns
+
@@ -0,0 +1,63 @@
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu
-fsanitize=signed-integer-overflow,unsigned-integer-overflow
-fsanitize-undefined-ignore-overflow-pattern=all %s -emit-llvm -o - | FileCheck
%s
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu
-fsanitize
https://github.com/JustinStitt edited
https://github.com/llvm/llvm-project/pull/104889
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
hnrklssn wrote:
Regarding there not being a `no-wraps` attribute. What happens with code like
this? Is the attribute lost or casted away during the addition?
```
wrapping_int a = INT_MAX;
a = (int) a + 1;
```
Does it affect converting a number too large to fit in the target type?
```
wrapping_i
aheejin wrote:
> I know it is a little pedantic, but I recently sorted the features
> alphabetically (#90777)... Can you move the new `fp16` entries so that the
> lists are alphabetically sorted in the files?
Sorry nevermind, even if we want to sort it its place is the same... 😅
https://githu
101 - 200 of 328 matches
Mail list logo
