Michael137 wrote:
> We are seeing test failures on `lldb-api :: lang/cpp/stl/TestSTL.py` after
> this patch is landed.
>
> ```
> Error when building test subject.
>
> Build Command:
> /b/s/w/ir/x/w/cipd/bin/make
> VPATH=/b/s/w/ir/x/w/llvm-llvm-project/lldb/test/API/lang/cpp/stl -C
> /b/s/w/i
@@ -1,5 +1,9 @@
CXX_SOURCES := main.cpp
-USE_SYSTEM_STDLIB := 1
+ifneq ($(OS),Darwin)
+USE_LIBSTDCPP := 1
zeroomega wrote:
This will break every test under `lldb/test/API/lang/cpp/stl` on non darwin
machines if GNU libstdc++ is not installed.
I don't thi
zeroomega wrote:
We are seeing test failures on `lldb-api :: lang/cpp/stl/TestSTL.py` after this
patch is landed.
```
Error when building test subject.
Build Command:
/b/s/w/ir/x/w/cipd/bin/make
VPATH=/b/s/w/ir/x/w/llvm-llvm-project/lldb/test/API/lang/cpp/stl -C
/b/s/w/ir/x/w/llvm_build/lldb-
https://github.com/slydiman closed
https://github.com/llvm/llvm-project/pull/112357
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
slydiman wrote:
> Can we change the Darwin case to `USE_LIBCPP` though? The less reliance on
> the system stdlib the better i think.
I cannot test it on Darwin, so I would leave that to the Darwin experts.
> Also lets check that the PR CI actually does run the test (aka doesn't
> trigger the
https://github.com/Michael137 approved this pull request.
Doing both seems fine, thanks!
Can we change the Darwin case to `USE_LIBCPP` though? The less reliance on the
system stdlib the better i think.
Also lets check that the PR CI actually does run the test (aka doesn't trigger
the skip add
https://github.com/slydiman updated
https://github.com/llvm/llvm-project/pull/112357
>From 29326dedbc1c786e699fb30655d4105c867d1b84 Mon Sep 17 00:00:00 2001
From: Dmitry Vasilyev
Date: Tue, 15 Oct 2024 17:20:13 +0400
Subject: [PATCH 1/2] [lldb][test] Fix TestStdCXXDisassembly test
The path #98
https://github.com/slydiman edited
https://github.com/llvm/llvm-project/pull/112357
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
slydiman wrote:
@Michael137
Thanks. I have updated the patch with both solutions to be sure.
https://github.com/llvm/llvm-project/pull/112357
___
lldb-commits mailing list
lldb-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/lld
https://github.com/slydiman updated
https://github.com/llvm/llvm-project/pull/112357
>From 29326dedbc1c786e699fb30655d4105c867d1b84 Mon Sep 17 00:00:00 2001
From: Dmitry Vasilyev
Date: Tue, 15 Oct 2024 17:20:13 +0400
Subject: [PATCH 1/2] [lldb][test] Fix TestStdCXXDisassembly test
The path #98
Michael137 wrote:
> > maybe we could just skip the test if we can't locate the module?
>
> Can you suggest how to do that (locate the module)?
Instead of,
```
self.expect(
lib_stdcxx, "Libraray StdC++ is located", exe=False, substr
slydiman wrote:
> maybe we could just skip the test if we can't locate the module?
Can you suggest how to do that (locate the module)?
> Won't setting USE_LIBSTDCPP be an issue on Darwin?
I can't test on Darwin. Maybe the following?
```
ifneq ($(OS),Darwin)
USE_LIBSTDCPP := 1
else
USE_S
Michael137 wrote:
So the problem is that because we statically link we fail to find a
`libstdc++.so`? Won't setting `USE_LIBSTDCPP` be an issue on Darwin?
Generally, why is this test trying to do disassembly on every STL symbol? Can't
really tell if this was actually STL specific from the git
llvmbot wrote:
@llvm/pr-subscribers-lldb
Author: Dmitry Vasilyev (slydiman)
Changes
The patch #98694 was not enough. This test is still failed on the
buildbot https://lab.llvm.org/staging/#/builders/195/builds/4438 Use
`USE_LIBSTDCPP := 1` instead.
---
Full diff: https://github.com/llvm
https://github.com/slydiman created
https://github.com/llvm/llvm-project/pull/112357
The patch #98694 was not enough. This test is still failed on the buildbot
https://lab.llvm.org/staging/#/builders/195/builds/4438 Use `USE_LIBSTDCPP :=
1` instead.
>From 29326dedbc1c786e699fb30655d4105c867d1
15 matches
Mail list logo
