# Re: Failing tests on s390x and aarch64 architectures

3 messages from 2019-11-26 to 2019-11-26. Participants: Marketa Calabkova, Carlo Arenas, Andreas Schwab.
Thread: https://gitlist.dev/t/52343

## Marketa Calabkova, 2019-11-26 14:51

Subject: Re: Failing tests on s390x and aarch64 architectures
Message-ID: <94b6452d-9888-a129-78f9-38eb0f03eea3@suse.cz>
URL: https://gitlist.dev/e/94b6452d-9888-a129-78f9-38eb0f03eea3%40suse.cz
In-Reply-To: <2d7f373f-5cf5-a956-53d7-1e67ba157e9d@suse.cz>

```
Hi,

just bumping this issue up. I found out that the fail occurs from the
very beginning, so from commit [1]. Hope this helps.

Marketa
[1]
https://github.com/git/git/commit/a73f91774cbfb6f31bca328ebf200498fe92d97a

On 15/11/2019 14:01, Marketa Calabkova wrote:
> Hi all,
>
> the test "PCRE v2: grep non-ASCII from invalid UTF-8 data with -i" from
> file t7812-grep-icase-non-ascii.sh fails on the architectures mentioned
> above (for more distributions). See the attached buildlogs. We are using
> git version 2.24.0 from [0].
>
> I think it could be caused by commit [1], I am not sure at all. It is
> not caused directly by new docbook5, we had it before 2.24.0 and the
> tests were passing.
>
> Could you please look at that? Thank you in advance.
>
> Best regards,
> Marketa Calabkova
>
> [0] https://build.opensuse.org/package/show/openSUSE:Factory/git
> [1]
> https://github.com/git/git/commit/e0ff2d4c7ec338e30ea5e0340cda7f5fe8a187dc



```

## Carlo Arenas, 2019-11-26 19:09

Subject: Re: Failing tests on s390x and aarch64 architectures
Message-ID: <CAPUEsph-sP79ov0DoTj9vfhFeD_dt=FpPWgk78syX+qKeVBT3Q@mail.gmail.com>
URL: https://gitlist.dev/e/CAPUEsph-sP79ov0DoTj9vfhFeD_dt%3DFpPWgk78syX%2BqKeVBT3Q%40mail.gmail.com
In-Reply-To: <94b6452d-9888-a129-78f9-38eb0f03eea3@suse.cz>

```
is this failing only in factory? couldn't find any failing logs going
through the links in the original email but while I don't have
OpenSUSE with me right now, I hadn't been able to reproduce the
problem with an aarch64 chromebook and its emulated Linux runtime
(based on Debian 9)

my question about factory is because I'd seen reports of other
problems with it and pcre2 because the system library was compiled
with the experimental selinux aware allocator (--enable-jit-sealloc)
by mistake that might be also the source for this issue.

Carlo

```

## Andreas Schwab, 2019-11-26 21:24

Subject: Re: Failing tests on s390x and aarch64 architectures
Message-ID: <87ftial4a9.fsf@igel.home>
URL: https://gitlist.dev/e/87ftial4a9.fsf%40igel.home
In-Reply-To: <CAPUEsph-sP79ov0DoTj9vfhFeD_dt=FpPWgk78syX+qKeVBT3Q@mail.gmail.com>

```
On Nov 27 2019, Carlo Arenas wrote:

> is this failing only in factory? couldn't find any failing logs going
> through the links in the original email but while I don't have
> OpenSUSE with me right now, I hadn't been able to reproduce the
> problem with an aarch64 chromebook and its emulated Linux runtime
> (based on Debian 9)

The test only fails if pcre2 was configured without JIT.

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 7578 EB47 D4E5 4D69 2510  2552 DF73 E780 A9DA AEC1
"And now for something completely different."

```
