https://github.com/michael-jabbour-sonarsource commented:
LGTM. Thank you.
https://github.com/llvm/llvm-project/pull/151259
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
michael-jabbour-sonarsource wrote:
@sbc100 Yes, this looks like the same failure addressed in #151259 to me
https://github.com/llvm/llvm-project/pull/150430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/
@@ -0,0 +1,98 @@
+// REQUIRES: staticanalyzer
+//
+// RUN: rm -rf %t
+// RUN: mkdir -p %t
+// RUN: split-file --leading-lines %s %t
+//
+// Prepare the BMIs.
+// RUN: %clang_cc1 -std=c++20 -triple x86_64-unknown-linux-gnu
-emit-module-interface -o %t/mod_a-part1.pcm %t/mod_a-part
https://github.com/michael-jabbour-sonarsource commented:
LGTM. Thanks for fixing this.
I left two minor comments. Feel free to ignore if they are not relevant.
https://github.com/llvm/llvm-project/pull/151259
___
cfe-commits mailing list
cfe-commits@
@@ -0,0 +1,98 @@
+// REQUIRES: staticanalyzer
+//
+// RUN: rm -rf %t
+// RUN: mkdir -p %t
+// RUN: split-file --leading-lines %s %t
+//
+// Prepare the BMIs.
+// RUN: %clang_cc1 -std=c++20 -triple x86_64-unknown-linux-gnu
-emit-module-interface -o %t/mod_a-part1.pcm %t/mod_a-part
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/151259
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -8,10 +10,7 @@
// RUN: %clang_cc1 -std=c++20 -triple x86_64-unknown-linux-gnu
-emit-module-interface -o %t/mod_a.pcm %t/mod_a.cppm
-fmodule-file=mod_a:part2=%t/mod_a-part2.pcm
-fmodule-file=mod_a:part1=%t/mod_a-part1.pcm
// RUN: %clang_cc1 -std=c++20 -triple x86_64-unknown
michael-jabbour-sonarsource wrote:
Thank you @hahnjo very much for the hint. I had wrongly assumed this is coming
from the commit metadata.
I just changed the setting. Let me know if it still doesn't work.
Thanks again to you all...
https://github.com/llvm/llvm-project/pull/150430
___
michael-jabbour-sonarsource wrote:
I am a bit confused where this comes from. The correct email address is also
shown on the commit:
https://api.github.com/repos/llvm/llvm-project/commits/c3cf089fb721a52993c42d0fe18547685d8badb6
https://github.com/llvm/llvm-project/pull/150430
michael-jabbour-sonarsource wrote:
I am a bit confused where this comes from. The correct email address is also
shown on the commit:
https://api.github.com/repos/llvm/llvm-project/commits/c3cf089fb721a52993c42d0fe18547685d8badb6
https://github.com/llvm/llvm-project/pull/150430
michael-jabbour-sonarsource wrote:
Hi @ChuanqiXu9, could you please clarify the request to "update the mail"? I've
double-checked that the commit author is currently set to my email, and I can't
see any github autogenerated email addresses on the commit metdata.
I want to make sure I understan
michael-jabbour-sonarsource wrote:
Thank you all for the reviews. I don’t have commit access, so I’ll need someone
to merge this for me eventually. I’ll be away for the next week and won’t be
able to respond to any regressions or follow-ups during that time.
If you’re comfortable merging it as
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/150430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/150430
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/michael-jabbour-sonarsource updated
https://github.com/llvm/llvm-project/pull/150430
>From c3cf089fb721a52993c42d0fe18547685d8badb6 Mon Sep 17 00:00:00 2001
From: Michael Jabbour
Date: Thu, 24 Jul 2025 15:53:08 +0200
Subject: [PATCH] Fix crash while lazy-loading template spec
https://github.com/michael-jabbour-sonarsource updated
https://github.com/llvm/llvm-project/pull/150430
>From 3e49b5c4f7c572f09e39c7c26dfbf6f1c58351ef Mon Sep 17 00:00:00 2001
From: Michael Jabbour
Date: Thu, 24 Jul 2025 15:53:08 +0200
Subject: [PATCH] Fix crash while lazy-loading template spec
michael-jabbour-sonarsource wrote:
Hello,
We have encountered crashes after this patch, with stack traces that look like
this:
```
PLEASE submit a bug report to https://github.com/llvm/llvm-project/issues/ and
include the crash backtrace, preprocessed source, and associated run script.
Stack
https://github.com/michael-jabbour-sonarsource created
https://github.com/llvm/llvm-project/pull/150430
## Problem
This is a regression that was observed in Clang 20.
The lazy-loading mechanism for template specializations introduced in #119333
can currently load additional nodes when called
michael-jabbour-sonarsource wrote:
Hello,
I think that the following cases used to be valid reports on clang 19, but they
are not reported in clang 20 after this patch. Is this intended?
```cpp
namespace classEnums {
int globalVar = 0;
class ClassWithEnum {
enum Enum {
globalVar // No -
michael-jabbour-sonarsource wrote:
Hi @tristanlabelle and thanks for your response,
> 1. Code pervasively access files through canonicalized paths rather than
> using canonicalized paths only for identity comparisons. The two usages are
> often conflated and can't be teased apart easily.
I un
michael-jabbour-sonarsource wrote:
Gentle ping :smile:
I would also appreciate any opinions on my previous comment, maybe this can
help us align on a solution that satisfies all users...
https://github.com/llvm/llvm-project/pull/139739
___
cfe-commit
michael-jabbour-sonarsource wrote:
Thank you very much @compnerd and @tristanlabelle for the information. I am
writing my thoughts below:
> Everything is being built in S:, there is no reason that the path should be
> "canonicalised" to %UserProfile%\SourceCache\llvm-project\llvm\... instead o
michael-jabbour-sonarsource wrote:
Hello @compnerd, and thank you for sharing the information and for the prompt
response.
> Without this change, a file on a substituted drive will have two separate
> paths.
I am not sure I understand this part. For me, a file on a substituted drive (or
unde
michael-jabbour-sonarsource wrote:
Thank you all for looking into this.
Note that I don't have commit access, and I would appreciate it if someone
merges this if/when it is ready to merge!
https://github.com/llvm/llvm-project/pull/114240
___
cfe-comm
https://github.com/michael-jabbour-sonarsource updated
https://github.com/llvm/llvm-project/pull/114240
>From d5c710e5a16f77b5a44bb842d09674a8b40dbd0d Mon Sep 17 00:00:00 2001
From: Michael Jabbour
Date: Fri, 7 Mar 2025 09:32:54 +0100
Subject: [PATCH] [clang] Fix ASTWriter crash after merging n
https://github.com/michael-jabbour-sonarsource updated
https://github.com/llvm/llvm-project/pull/114240
>From 9eede4531ad27ce1d77df492076d5fbcd237fd3f Mon Sep 17 00:00:00 2001
From: Michael Jabbour
Date: Fri, 7 Mar 2025 09:32:54 +0100
Subject: [PATCH] [clang] Fix ASTWriter crash after merging n
michael-jabbour-sonarsource wrote:
Hello,
This seems to cause a noticeable increase in memory consumption on some
examples. See
https://github.com/llvm/llvm-project/issues/129808#issuecomment-2700015065.
https://github.com/llvm/llvm-project/pull/87824
__
https://github.com/michael-jabbour-sonarsource updated
https://github.com/llvm/llvm-project/pull/114240
>From cc3cf25da67c9f8b9edabb318c6011cad9bd2f58 Mon Sep 17 00:00:00 2001
From: Michael Jabbour
Date: Tue, 29 Oct 2024 11:16:09 +0100
Subject: [PATCH 1/9] [NFC] Factor out RetireNodesFromMerged
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -2551,18 +2551,7 @@ void Sema::MergeTypedefNameDecl(Scope *S,
TypedefNameDecl *New,
// Make the old tag definition visible.
makeMergedDefinitionVisible(Hidden);
- // If this was an unscoped enumeration, yank all of its enumerators
- // out of the scop
https://github.com/michael-jabbour-sonarsource updated
https://github.com/llvm/llvm-project/pull/114240
>From cc3cf25da67c9f8b9edabb318c6011cad9bd2f58 Mon Sep 17 00:00:00 2001
From: Michael Jabbour
Date: Tue, 29 Oct 2024 11:16:09 +0100
Subject: [PATCH 1/8] [NFC] Factor out RetireNodesFromMerged
https://github.com/michael-jabbour-sonarsource updated
https://github.com/llvm/llvm-project/pull/114240
>From cc3cf25da67c9f8b9edabb318c6011cad9bd2f58 Mon Sep 17 00:00:00 2001
From: Michael Jabbour
Date: Tue, 29 Oct 2024 11:16:09 +0100
Subject: [PATCH 1/8] [NFC] Factor out RetireNodesFromMerged
https://github.com/michael-jabbour-sonarsource updated
https://github.com/llvm/llvm-project/pull/114240
>From cc3cf25da67c9f8b9edabb318c6011cad9bd2f58 Mon Sep 17 00:00:00 2001
From: Michael Jabbour
Date: Tue, 29 Oct 2024 11:16:09 +0100
Subject: [PATCH 1/8] [NFC] Factor out RetireNodesFromMerged
michael-jabbour-sonarsource wrote:
Gentle ping :smile:
@vsapsai, @vgvassilev, I think I have responded to your inquiries and added a
few related questions. Could you have a look?
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailin
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -2551,18 +2551,7 @@ void Sema::MergeTypedefNameDecl(Scope *S,
TypedefNameDecl *New,
// Make the old tag definition visible.
makeMergedDefinitionVisible(Hidden);
- // If this was an unscoped enumeration, yank all of its enumerators
- // out of the scop
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -2639,6 +2628,19 @@ void Sema::MergeTypedefNameDecl(Scope *S,
TypedefNameDecl *New,
notePreviousDefinition(Old, New->getLocation());
}
+void Sema::CleanupMergedEnum(Scope *S, Decl *New) {
michael-jabbour-sonarsource wrote:
> Is MergeTypedefNameDecl not
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -2551,18 +2551,7 @@ void Sema::MergeTypedefNameDecl(Scope *S,
TypedefNameDecl *New,
// Make the old tag definition visible.
makeMergedDefinitionVisible(Hidden);
- // If this was an unscoped enumeration, yank all of its enumerators
- // out of the scop
michael-jabbour-sonarsource wrote:
@vsapsai Thank you for sharing the previous patch. This is really interesting.
> For the record, the previous work I've abandoned is
> https://reviews.llvm.org/D114833 Doesn't seem particularly relevant to this
> change to be honest.
As far as I can see, it
michael-jabbour-sonarsource wrote:
@vsapsai Thanks again for the feedback :pray:
> I've realized that the test doesn't have to be Objective-C, the failure is
> reproducible in C. Curiously, we aren't hitting the assertion in
> C++/Objective-C++ but I don't know why.
I dug a bit into this tod
https://github.com/michael-jabbour-sonarsource updated
https://github.com/llvm/llvm-project/pull/114240
>From cc3cf25da67c9f8b9edabb318c6011cad9bd2f58 Mon Sep 17 00:00:00 2001
From: Michael Jabbour
Date: Tue, 29 Oct 2024 11:16:09 +0100
Subject: [PATCH 1/8] [NFC] Factor out RetireNodesFromMerged
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
michael-jabbour-sonarsource wrote:
@vsapsai Thank you very much for the input. :pray:
> Can you trigger the crash without -ast-dump-all?
> Can you use in the test %clang_cc1 instead of %clang?
I addressed both of these in f6f0887d22172c9db4f602b7f272630a97a708bd. I added
a separate invocatio
@@ -2551,18 +2551,7 @@ void Sema::MergeTypedefNameDecl(Scope *S,
TypedefNameDecl *New,
// Make the old tag definition visible.
makeMergedDefinitionVisible(Hidden);
- // If this was an unscoped enumeration, yank all of its enumerators
- // out of the scop
@@ -2551,18 +2551,7 @@ void Sema::MergeTypedefNameDecl(Scope *S,
TypedefNameDecl *New,
// Make the old tag definition visible.
makeMergedDefinitionVisible(Hidden);
- // If this was an unscoped enumeration, yank all of its enumerators
- // out of the scop
@@ -4034,6 +4034,12 @@ class Sema final : public SemaBase {
void MergeTypedefNameDecl(Scope *S, TypedefNameDecl *New,
LookupResult &OldDecls);
+ /// RetireNodesFromMergedDecl - We have just merged the decl 'New' by making
+ /// another definitio
@@ -2551,18 +2551,7 @@ void Sema::MergeTypedefNameDecl(Scope *S,
TypedefNameDecl *New,
// Make the old tag definition visible.
makeMergedDefinitionVisible(Hidden);
- // If this was an unscoped enumeration, yank all of its enumerators
- // out of the scop
https://github.com/michael-jabbour-sonarsource updated
https://github.com/llvm/llvm-project/pull/114240
>From cc3cf25da67c9f8b9edabb318c6011cad9bd2f58 Mon Sep 17 00:00:00 2001
From: Michael Jabbour
Date: Tue, 29 Oct 2024 11:16:09 +0100
Subject: [PATCH 1/7] [NFC] Factor out RetireNodesFromMerged
https://github.com/michael-jabbour-sonarsource updated
https://github.com/llvm/llvm-project/pull/114240
>From cc3cf25da67c9f8b9edabb318c6011cad9bd2f58 Mon Sep 17 00:00:00 2001
From: Michael Jabbour
Date: Tue, 29 Oct 2024 11:16:09 +0100
Subject: [PATCH 1/6] [NFC] Factor out RetireNodesFromMerged
michael-jabbour-sonarsource wrote:
Gentle ping, and I wish you all a happy new year. :smile:
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-c
michael-jabbour-sonarsource wrote:
Thanks for the update @vsapsai. Sounds good. Let me know if there are any
changes that can make reviewing the PR easier!
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@lists
michael-jabbour-sonarsource wrote:
Gentle ping :smile:
I would love to hear your feedback, and if any changes are needed, I am happy
to address them promptly.
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@l
michael-jabbour-sonarsource wrote:
Gentle ping :smile:
The aim of the PR is to fix an ASTWriter crash that currently happens when
serializing merged enums. Such a case was encountered while parsing Objective-C
code that uses the XCode 16 SDK. If any of the changes look risky, I am happy
to re
michael-jabbour-sonarsource wrote:
Ping?
Note that this PR fixes a crash in ASTWriter on valid Objective-C modules code;
The initial PR title didn't accurately reflect that.
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing lis
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/michael-jabbour-sonarsource edited
https://github.com/llvm/llvm-project/pull/114240
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/michael-jabbour-sonarsource created
https://github.com/llvm/llvm-project/pull/114240
Clang already has a mechanism to cleanup enumerators for typedefs to anonymous
enums. So the following example code used to be handled correctly while
merging, and ASTWriter behaves as expec
68 matches
Mail list logo
