Author: Nico Weber
Date: 2025-01-07T09:23:50-05:00
New Revision: ab5133bbc62af4686f305a3c7d85f74b9f5b949f
URL:
https://github.com/llvm/llvm-project/commit/ab5133bbc62af4686f305a3c7d85f74b9f5b949f
DIFF:
https://github.com/llvm/llvm-project/commit/ab5133bbc62af4686f305a3c7d85f74b9f5b949f.diff
LO
nico wrote:
Also fails here: http://45.33.8.238/linux/156731/step_6.txt
http://45.33.8.238/macm1/98544/step_6.txt
https://github.com/llvm/llvm-project/pull/120507
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/m
nico wrote:
Reverted in ab5133bbc62af4686f305a3c7d85f74b9f5b949f for now.
https://github.com/llvm/llvm-project/pull/120507
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/nico closed https://github.com/llvm/llvm-project/pull/120787
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Author: Nico Weber
Date: 2025-01-02T09:13:40-05:00
New Revision: a4deb809be8f5ec3adec3626e9d700f6168d0e9f
URL:
https://github.com/llvm/llvm-project/commit/a4deb809be8f5ec3adec3626e9d700f6168d0e9f
DIFF:
https://github.com/llvm/llvm-project/commit/a4deb809be8f5ec3adec3626e9d700f6168d0e9f.diff
LO
nico wrote:
Here's a command you can use to reproduce this:
```
cmake -GNinja -DCMAKE_BUILD_TYPE=Release -DLLVM_ENABLE_ASSERTIONS=ON
-DLLVM_ENABLE_PROJECTS='clang;lld;compiler-rt' -DLLVM_APPEND_VC_REV=NO
-DLLVM_TARGETS_TO_BUILD='X86' -DLLVM_ENABLE_PER_TARGET_RUNTIME_DIR=OFF
../llvm-project/ll
nico wrote:
I did remove the contents of out/gn/lib/clang/20/lib/linux, but
`./lib/clang/20/lib/linux/libclang_rt.builtins-x86_64.a` just reappeared on the
next build.
I'm guessing this might break tests in cmake builds that use
`-DLLVM_ENABLE_PER_TARGET_RUNTIME_DIR=OFF` (?)
http://45.33.8.2
https://github.com/nico created https://github.com/llvm/llvm-project/pull/120787
Also move the -fno-wrapv option definition next to the -fwrapv one while here.
>From 48b974d50a6b5d91f094663e446b94910a22f097 Mon Sep 17 00:00:00 2001
From: Nico Weber
Date: Fri, 20 Dec 2024 14:25:20 -0500
Subject:
nico wrote:
@alanzhao1 has been trying to reproduce the problem, I think. I'm not sure what
the current status is.
https://github.com/llvm/llvm-project/pull/119071
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/
nico wrote:
IIRC there are two possible compiler-rt directory layouts. @MaskRay worked on
moving many platforms from one to another. I think the newer layout doesn't use
platform suffixes.
We shouldn't change where clang looks for things just for my bot. Thanks for
figuring it out; I can dele
nico wrote:
Aha! Yes, that bot does do incremental builds. (As do many devs.)
https://github.com/llvm/llvm-project/pull/118192
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
nico wrote:
I'll try to find a Linux box tomorrow to get more details.
https://logs.chromium.org/logs/chromium/buildbucket/cr-buildbucket/8728518597999528929/+/u/package_clang/stdout?format=raw
is the full build output. If you ctrl-f for "cmake -G", you should find all
cmake vars that that bot
nico wrote:
I reverted this in 1464b8ec8a675fd11dc7280db1c56aac03771b0a for now.
https://github.com/llvm/llvm-project/pull/119071
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Author: Nico Weber
Date: 2024-12-15T14:04:56-05:00
New Revision: 1464b8ec8a675fd11dc7280db1c56aac03771b0a
URL:
https://github.com/llvm/llvm-project/commit/1464b8ec8a675fd11dc7280db1c56aac03771b0a
DIFF:
https://github.com/llvm/llvm-project/commit/1464b8ec8a675fd11dc7280db1c56aac03771b0a.diff
LO
Author: Nico Weber
Date: 2024-12-15T14:04:55-05:00
New Revision: 50046221b8e913ec6506eb96ce4c0cd267a5cc99
URL:
https://github.com/llvm/llvm-project/commit/50046221b8e913ec6506eb96ce4c0cd267a5cc99
DIFF:
https://github.com/llvm/llvm-project/commit/50046221b8e913ec6506eb96ce4c0cd267a5cc99.diff
LO
nico wrote:
This broke Linux/dn_expand.cpp on our bots, see
https://issues.chromium.org/issues/384188036
https://github.com/llvm/llvm-project/pull/119071
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/li
nico wrote:
If all other bots are happy, then I'm doing something wrong probably. If you
don't mind reverting, it'd be nice since then I can catch other, actual
problems, with the bot, but it's not a requirement.
https://github.com/llvm/llvm-project/pull/118192
nico wrote:
This broke check-clang on my linux box:
http://45.33.8.238/linux/155432/step_6.txt
That bot uses the GN build – are there any prerequisite changes elsewhere that
I need to port to it?
https://github.com/llvm/llvm-project/pull/118192
___
nico wrote:
Looks like this breaks tests: http://45.33.8.238/linux/153081/step_6.txt
Please take a look and revert for now if it takes a while to fix.
https://github.com/llvm/llvm-project/pull/113881
___
cfe-commits mailing list
cfe-commits@lists.llvm
@@ -0,0 +1,111 @@
+// REQUIRES: amdgpu-registered-target
+// RUN: %clang -xhip --offload-arch=gfx1030 --offload-host-only -pedantic
-nogpuinc -nogpulib -nobuiltininc -fsyntax-only -Xclang -verify %s
+// RUN: %clang -xhip --offload-arch=gfx1030 --offload-device-only -pedantic
-no
@@ -0,0 +1,111 @@
+// REQUIRES: amdgpu-registered-target
+// RUN: %clang -xhip --offload-arch=gfx1030 --offload-host-only -pedantic
-nogpuinc -nogpulib -nobuiltininc -fsyntax-only -Xclang -verify %s
+// RUN: %clang -xhip --offload-arch=gfx1030 --offload-device-only -pedantic
-no
@@ -0,0 +1,218 @@
+//===-- CachedConstAccessorsLattice.h ---*- C++
-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Ap
nico wrote:
This seems to break building of clang-tools-extra binaries:
http://45.33.8.238/win/95986/step_3.txt
Please take a look and revert for now if it takes a while to fix.
https://github.com/llvm/llvm-project/pull/112640
___
cfe-commits mailing
@@ -0,0 +1,218 @@
+//===-- CachedConstAccessorsLattice.h ---*- C++
-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Ap
nico wrote:
I think this'll fix the build, thanks!
However, the root cause is probably that the `AST_MATCHER_P_OVERLOAD`
instantiation in
clang/lib/Analysis/FlowSensitive/Models/UncheckedOptionalAccessModel.cpp is in
an unnamed namespace that's inside namespace `clang::dataflow`, right? Shoul
nico wrote:
Looks like this breaks check-clang: http://45.33.8.238/linux/149333/step_6.txt
Please take a look and revert for now if it takes a while to fix.
https://github.com/llvm/llvm-project/pull/110198
___
cfe-commits mailing list
cfe-commits@list
nico wrote:
Thanks for merging the fix! At least on one of my bots, it helped and it's
green again :)
https://github.com/llvm/llvm-project/pull/100128
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listi
nico wrote:
The test is broken on trunk. It won't get more broken if you merge the fix. So
please merge, and if it helps, great.
https://github.com/llvm/llvm-project/pull/100128
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llv
nico wrote:
Looks like this breaks tests on Mac: http://45.33.8.238/macm1/92471/step_6.txt
Please take a look and revert for now if it takes a while to fix.
https://github.com/llvm/llvm-project/pull/76838
___
cfe-commits mailing list
cfe-commits@lists
@@ -181,16 +181,16 @@ llvm::Expected>
getFoldingRanges(ParsedAST &AST) {
// Related issue: https://github.com/clangd/clangd/issues/310
llvm::Expected>
getFoldingRanges(const std::string &Code, bool LineFoldingOnly) {
nico wrote:
Should we revert 70914aa631561
nico wrote:
@thurstond ^
https://github.com/llvm/llvm-project/pull/106588
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
nico wrote:
Looks like the _revert_ broke check-clang:
http://45.33.8.238/linux/148045/step_6.txt
Please take a look, and uh, mark that test as UNSUPPORTED if it takes a while
to fix, I suppose.
https://github.com/llvm/llvm-project/pull/106588
___
c
nico wrote:
This breaks tests on windows: http://45.33.8.238/win/94083/step_6.txt
Please take a look and revert for now if it takes a while to fix.
https://github.com/llvm/llvm-project/pull/97369
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://github.com/nico closed https://github.com/llvm/llvm-project/pull/108452
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/nico approved this pull request.
https://github.com/llvm/llvm-project/pull/108452
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
nico wrote:
(To be clear, I'm fine with landing this.)
> non conforming two phase lookup by default would also be a concern.
It is! There were quite a few discussions around this. Aha, see #42377.
We usually say the goal is to have a very standards compliant impl, while also
being able to par
Author: Nico Weber
Date: 2024-09-06T09:17:45-04:00
New Revision: c64dac2e6c39f0f7f1c676857e7d34c764b4d632
URL:
https://github.com/llvm/llvm-project/commit/c64dac2e6c39f0f7f1c676857e7d34c764b4d632
DIFF:
https://github.com/llvm/llvm-project/commit/c64dac2e6c39f0f7f1c676857e7d34c764b4d632.diff
LO
nico wrote:
For this patch here, I don't know if gcc even has a `fms-volatile`. If not,
this here seems less surprising to me than the strict aliasing patch.
https://github.com/llvm/llvm-project/pull/107509
___
cfe-commits mailing list
cfe-commits@lis
nico wrote:
I don't have a very strong opinion. I find it surprising in that the clang
driver is supposed to be gcc-compatible, and having to use `clang
-fstrict-aliasing` (but only when targeting Windows) to get the default
behavior looks a bit surprising.
(I agree it's a better default, but
nico wrote:
I made that change in 5cf3677a7bcdf5a9e603c054bd40c1282db984a9.
https://github.com/llvm/llvm-project/pull/98736
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Author: Nico Weber
Date: 2024-09-06T08:27:41-04:00
New Revision: 5cf3677a7bcdf5a9e603c054bd40c1282db984a9
URL:
https://github.com/llvm/llvm-project/commit/5cf3677a7bcdf5a9e603c054bd40c1282db984a9
DIFF:
https://github.com/llvm/llvm-project/commit/5cf3677a7bcdf5a9e603c054bd40c1282db984a9.diff
LO
nico wrote:
…passing `Wno-msvc-not-found` as suggested above would also make the test go.
But `-c` seems generally nicer here anyways?
https://github.com/llvm/llvm-project/pull/98736
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://list
nico wrote:
> Somehow `clang.exe -Ofast -O2 -### -Werror
> C:\src\llvm-project\clang\test\Driver\Ofast.c 2>&1` is not producing any
> output for you, despite `-###` being present. I'm not sure why, but all
> windows buildbots seem to be happy with this patch. Can you investigate
> further?
F
nico wrote:
The test added here was tweaked a bit and then deleted in
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 PR with working tests.
https://github.c
https://github.com/nico approved this pull request.
Makes sense, thanks!
https://github.com/llvm/llvm-project/pull/101973
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
nico wrote:
That sounds believable.
We did update our compiler just an hour or so ago, so it's possible it will
work for us now.
However, it suggests that the assumption `defined(__ARM_FEATURE_GCS_DEFAULT)`
=> arm_acle.h has `__chkfeat()` isn't valid in this snippet in
libunwind/src/cet_unwi
nico wrote:
Looks like this breaks building on Android:
https://ci.chromium.org/ui/p/chromium/builders/try/android-arm64-rel/680348/overview
https://github.com/llvm/llvm-project/pull/99335
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https:
nico wrote:
This seems to break tests on my Windows box:
http://45.33.8.238/win/91548/step_6.txt
https://github.com/llvm/llvm-project/pull/98736
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cf
nico wrote:
Independent of the other issues, it looks like this also re-introduces #98878
https://github.com/llvm/llvm-project/pull/98547
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commit
https://github.com/nico edited https://github.com/llvm/llvm-project/pull/66462
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/nico edited https://github.com/llvm/llvm-project/pull/66462
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -161,6 +163,7 @@ clang_target_link_libraries(clangDaemon
clangAST
clangASTMatchers
clangBasic
+ clangDependencyScanning
nico wrote:
This dependency makes clangd depend (transitively) on clangCodeGen, which it
didn't depend on previously.
This makes
nico wrote:
In addition to the test failures, this apparently also breaks building lldb on
windows and Mac:
http://45.33.8.238/macm1/88231/step_3.txt
http://45.33.8.238/win/91235/step_3.txt
Please check if that's something real or if my bots are being silly before
relanding.
https://github.co
@@ -654,6 +654,9 @@ Improvements to Clang's diagnostics
- Clang now shows implicit deduction guides when diagnosing overload
resolution failure. #GH92393.
+- Clong no longer emits a no previous prototype warning for Win32 entry points
under ``-Wmissing-prototypes``.
---
Author: Nico Weber
Date: 2024-06-28T13:42:54+02:00
New Revision: 69c99ad7e84b075bbafc541a2e4397e18975391d
URL:
https://github.com/llvm/llvm-project/commit/69c99ad7e84b075bbafc541a2e4397e18975391d
DIFF:
https://github.com/llvm/llvm-project/commit/69c99ad7e84b075bbafc541a2e4397e18975391d.diff
LO
Author: Nico Weber
Date: 2024-06-28T13:38:14+02:00
New Revision: 7878d9c0363528c44effe55aae2843fbabb6dd0e
URL:
https://github.com/llvm/llvm-project/commit/7878d9c0363528c44effe55aae2843fbabb6dd0e
DIFF:
https://github.com/llvm/llvm-project/commit/7878d9c0363528c44effe55aae2843fbabb6dd0e.diff
LO
nico wrote:
(I now ported ade28a77ed17760bf2fde57c6571b69489b0bac0 to the GN build in
3ba7599842be. Again, apologies for missing this – the GN build shouldn't affect
other people. But I'm kind of glad it did since I found several GN-unrelated
issues both here and on #93928. Feel free to do wit
@@ -127,16 +133,86 @@ std::string getFormatString() {
// GetMainExecutable (since some platforms don't support taking the
// address of main, and some platforms can't implement GetMainExecutable
// without being given the address of a function in the main executable).
-std::str
@@ -0,0 +1,21 @@
+#include "Calculator.h"
+#include
nico wrote:
Tests must be freestanding and cannot include system headers.
https://github.com/llvm/llvm-project/pull/93928
___
cfe-commits mailing list
cfe-commits@li
@@ -182,23 +258,9 @@ Example usage for a project using a compile commands
database:
{"index.js", "index_json.js"}};
if (Format == "html") {
-void *MainAddr = (void *)(intptr_t)GetExecutablePath;
-std::string ClangDocPath = GetExecutablePath(argv[0], MainAddr);
@@ -0,0 +1,358 @@
+// RUN: rm -rf %t && mkdir -p %t/docs %t/build
+// RUN: sed 's|$test_dir|%/S|g' %S/Inputs/basic-project/database_template.json
> %t/build/compile_commands.json
+// RUN: clang-doc --format=html --output=%t/docs --executor=all-TUs
%t/build/compile_commands.json
@@ -0,0 +1,2 @@
+// RUN: clang-doc --format=html --executor=standalone %s -output=%t/docs |
FileCheck %s
+// CHECK: Using default asset: {{.*}}{{[\/]}}share{{[\/]}}clang
nico wrote:
A test that only tests for logspam doesn't seem very useful. I would've
expecte
@@ -127,16 +133,86 @@ std::string getFormatString() {
// GetMainExecutable (since some platforms don't support taking the
// address of main, and some platforms can't implement GetMainExecutable
// without being given the address of a function in the main executable).
-std::str
@@ -182,23 +258,9 @@ Example usage for a project using a compile commands
database:
{"index.js", "index_json.js"}};
if (Format == "html") {
-void *MainAddr = (void *)(intptr_t)GetExecutablePath;
-std::string ClangDocPath = GetExecutablePath(argv[0], MainAddr);
nico wrote:
Ah, I see now, the failing test is also very new (70ec8419dd7), and it depends
on ade28a77ed177 which the GN build doesn't yet have. But my bot points at this
change here for breaking that fairly new end-to-end test. I think it's probably
because absence of the css file wasn't an e
nico wrote:
Sorry, I'm confused, how is this build-system dependent? The patch doesn't
change any cmake files, and it changes the behavior of an existing test.
I'm not saying it can't be build-system dependent, but it isn't clear to me why
it is. Do you have a few more words on what's happenin
nico wrote:
Looks Like this might break tests: http://45.33.8.238/linux/141118/step_7.txt
Please take a look and revert for now if it takes a while to fix.
https://github.com/llvm/llvm-project/pull/94717
___
cfe-commits mailing list
cfe-commits@lists.
https://github.com/nico closed https://github.com/llvm/llvm-project/pull/95546
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
nico wrote:
Merging to green up tests on a bot. Happy to address post-commit comments in a
follow-up 🙂
https://github.com/llvm/llvm-project/pull/95546
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listi
https://github.com/nico created https://github.com/llvm/llvm-project/pull/95546
At least on my Windows machine, these two tests fail due to not being able to
look up `??3@YAXPEAX_K@Z` (which is
`void __cdecl operator delete(void *, unsigned __int64)` in demangled) after
130e93cc26ca. Since they
https://github.com/nico approved this pull request.
https://github.com/llvm/llvm-project/pull/95406
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
nico wrote:
Oh, you have that in the release notes already, even better.
https://github.com/llvm/llvm-project/pull/95406
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
nico wrote:
Can you add something like "use /O2 /clang:-O2 to restore the previous
behavior" to the commit message, on case someone prefers that?
Otherwise LG.
https://github.com/llvm/llvm-project/pull/95406
___
cfe-commits mailing list
cfe-commits@l
nico wrote:
This causes #95451.
https://github.com/llvm/llvm-project/pull/90373
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Author: Nico Weber
Date: 2024-06-11T14:00:24-04:00
New Revision: 9b4f8acf9dd1cd517f923c6de8274eed80879f6c
URL:
https://github.com/llvm/llvm-project/commit/9b4f8acf9dd1cd517f923c6de8274eed80879f6c
DIFF:
https://github.com/llvm/llvm-project/commit/9b4f8acf9dd1cd517f923c6de8274eed80879f6c.diff
LO
https://github.com/nico closed https://github.com/llvm/llvm-project/pull/94762
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/nico updated https://github.com/llvm/llvm-project/pull/94762
>From 2cbc9f7e066066ffb04480be6bd7e19855086b80 Mon Sep 17 00:00:00 2001
From: Nico Weber
Date: Fri, 7 Jun 2024 11:17:29 -0400
Subject: [PATCH 1/2] [clang] Add fixit for using declaration with a
(qualified) namespace
https://github.com/nico created https://github.com/llvm/llvm-project/pull/94762
For `using std::literals`, we now output:
error: using declaration cannot refer to a namespace
4 | using std::literals;
| ~^
note: did you mean 'using namespace'?
4 |
nico wrote:
Any chance to put these behind a subgroup? This now fires in a bunch of places
where it didn't before.
https://github.com/llvm/llvm-project/pull/91699
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/m
nico wrote:
Thanks for the quick revert!
Is the failure due to a conflict with another commit that landed?
https://github.com/llvm/llvm-project/pull/88266
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/l
https://github.com/nico closed https://github.com/llvm/llvm-project/pull/89842
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
nico wrote:
If your project disables warning for system headers (why even do that?), I'm
guessing there's going to be a bunch of other issues you'll run into. But this
change here is fine, it makes cpuid.h consisten with all the other headers in
clang/lib/Headers :)
https://github.com/llvm/ll
nico wrote:
> Again, clangDriver does file probing in various places.
This is the only place where we probe link-time inputs for compile-time
decisions as far as I know.
https://github.com/llvm/llvm-project/pull/89775
___
cfe-commits mailing list
cfe
nico wrote:
Anyway, for this specific issue it sounds like the change here was
unintentional. As suggested by mstorsjo at
https://github.com/llvm/llvm-project/pull/87866#issuecomment-2073231116, I
think it'd be good to revert these changes, and then make a mindful decision
about this when rel
nico wrote:
> My recollection of the past discussions is that we want users to switch to
> the new hierarchy.
FWIW this doesn't match my understanding. phosek added the new hierarchy to
Fuchsia, maskray put in some work to enable it elsewhere, but it's a huge
headache, and we've forcibly turn
nico wrote:
> I would suggest we revert this - and at least collect all the observed side
> effects and declare them before considering relanding it.
That sounds good to me. Do you have a list of PRs to revert?
> ... then we do still get the old name embedded.
Sure. The motivation on our sid
https://github.com/nico closed https://github.com/llvm/llvm-project/pull/89642
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
nico wrote:
I now did build clang at ccdebbae4d77 and ccdebbae4d77^. ccdebbae4d77 has
`/DEFAULTLIB:clang_rt.profile.lib` in its output, ccdebbae4d77^ has
/DEFAULTLIB:clang_rt.profile-x86_64.lib`. So definitely due to this change.
It sounds like you're saying that's not intentional?
(It's kind
nico wrote:
Here's a minimal repro:
```
% cat empty.c
% foo/clang --driver-mode=cl empty.c --target=x86_64-pc-windows
-fprofile-instr-generate -c /Fa && rg DEFAULTLIB empty.asm
23: .ascii " /DEFAULTLIB:libcmt.lib"
24: .ascii " /DEFAULTLIB:oldnames.lib"
25: .ascii " /DEFAULTLIB:c
nico wrote:
Good call-out, thanks. I meant #87866, which is new.
I also forgot that llvm-project likes to do squash commits and put the new
commit message in the old commit and `git push -f`'d. But the only thing that
changed in the new commit is the commit message text.
https://github.com/ll
https://github.com/nico edited https://github.com/llvm/llvm-project/pull/89642
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/nico updated https://github.com/llvm/llvm-project/pull/89642
>From 39bd2d2ece1f55c3c77e1f376913846fb604be27 Mon Sep 17 00:00:00 2001
From: Nico Weber
Date: Mon, 22 Apr 2024 14:05:04 -0400
Subject: [PATCH] clang/win: Add a flag to disable default-linking of
compiler-rt librari
nico wrote:
This is a behavior change: In distributed build environments, neither lib file
exists at compile time. Previously, this would result in the "old" style, now
(together with #81037) it results in the "new" style (which we disable
everywhere since it causes all kinds of issues – from
@@ -656,19 +656,29 @@ std::string ToolChain::getCompilerRT(const ArgList &Args,
StringRef Component,
// Check for runtime files in the new layout without the architecture first.
std::string CRTBasename =
buildCompilerRTBasename(Args, Component, Type, /*AddArch=*/fals
https://github.com/nico edited https://github.com/llvm/llvm-project/pull/89642
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/nico edited https://github.com/llvm/llvm-project/pull/89642
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/nico created https://github.com/llvm/llvm-project/pull/89642
…ries
For ASan, users already manually have to pass in the path to the lib, and for
other libraries they have to pass in the path to the libpath.
With LLVM's unreliable name of the lib (due to
LLVM_ENABLE_PER_TARGE
@@ -656,19 +656,29 @@ std::string ToolChain::getCompilerRT(const ArgList &Args,
StringRef Component,
// Check for runtime files in the new layout without the architecture first.
std::string CRTBasename =
buildCompilerRTBasename(Args, Component, Type, /*AddArch=*/fals
@@ -1,5 +1,5 @@
-// RUN: %clang -S -fno-discard-value-names -emit-llvm -o - %s | FileCheck %s
-// Issue no: 41441
+// RUN: %clang --target=x86_64-pc-linux -S -fno-discard-value-names -emit-llvm
-o - %s | FileCheck %s
+
#include
nico wrote:
Thanks!
https://git
nico wrote:
Yes, #83124 broke it in the meantime 😅
https://github.com/llvm/llvm-project/pull/88101
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
1 - 100 of 1112 matches
Mail list logo
