bug-gettext
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: GetText 0.21 and Asan testing


From: Bruno Haible
Subject: Re: GetText 0.21 and Asan testing
Date: Fri, 18 Oct 2024 20:22:46 +0200

Hi,

Jeffrey Walton wrote in
<https://lists.gnu.org/archive/html/bug-gettext/2021-01/msg00020.html>:
> It looks like GetText 0.21 is experiencing some test failures under Asan:
> 
>     PASS: msgfilter-1
>     FAIL: msgfilter-2
>     PASS: msgfilter-3
>     FAIL: msgfilter-4
>     PASS: msgfilter-5
>     ...
>     PASS: msginit-2
>     FAIL: msginit-3
>     FAIL: msginit-4
>     PASS: msgmerge-1
>     ...
>     SKIP: msgunfmt-csharp-1
>     FAIL: msgunfmt-java-1
>     PASS: msgunfmt-properties-1
>     ...
>     SKIP: lang-python-2
>     FAIL: lang-java
>     SKIP: lang-csharp
> 
> I think this is the first time I tested 0.21 under Asan. If it's
> already been reported, then accept my apologies.

It's all fine when I use gcc-11.4 with ASAN on the current gettext
sources, both with and without --disable-shared.

Unlike you, I don't put the -fsanitize=address option into CFLAGS
and CXXFLAGS and LDFLAGS, but into CC and CXX:
CC="gcc -fsanitize=address"; CXX="g++ -fsanitize=address"; export CC CXX
Maybe that is what caused the error in your testing:
  ASan runtime does not come first in initial library list; you should either 
link runtime to your application or manually preload it with LD_PRELOAD.

Thanks for the testing. By now, I use -fsanitize=address routinely, as
it allows me to catch memory corruption bugs early in the development.

Bruno






reply via email to

[Prev in Thread] Current Thread [Next in Thread]