[PATCH] .gitlab-ci.d/windows: Do not run the qtests in the msys2-32bit job

[PATCH] .gitlab-ci.d/windows: Do not run the qtests in the msys2-32bit job

[Thomas Huth]

The qtests are not stable in the msys2-32bit job yet - especially
the test-hmp and the qom-test are failing randomly. Until this is
fixed, let's better disable the qtests here again to avoid failing
CI tests.

Signed-off-by: Thomas Huth <thuth@redhat.com>
---
 .gitlab-ci.d/windows.yml | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/.gitlab-ci.d/windows.yml b/.gitlab-ci.d/windows.yml
index 22f794e537..a1d5790580 100644
--- a/.gitlab-ci.d/windows.yml
+++ b/.gitlab-ci.d/windows.yml
@@ -116,4 +116,5 @@ msys2-32bit:
   - ..\msys64\usr\bin\bash -lc '../configure --target-list=ppc64-softmmu
         --disable-opengl'
   - ..\msys64\usr\bin\bash -lc 'make'
-  - ..\msys64\usr\bin\bash -lc 'make check || { cat meson-logs/testlog.txt; exit 1; } ;'
+  - ..\msys64\usr\bin\bash -lc 'make check MTESTARGS=\"--no-suite qtest\" ||
+                                { cat meson-logs/testlog.txt; exit 1; }'
-- 
2.31.1

Re: [PATCH] .gitlab-ci.d/windows: Do not run the qtests in the msys2-32bit job

[Philippe Mathieu-Daudé]

On 5/1/23 21:48, Thomas Huth wrote:
> The qtests are not stable in the msys2-32bit job yet - especially
> the test-hmp and the qom-test are failing randomly. Until this is
> fixed,

Who is gonna look after this?

I'm not against this patch, but I'm afraid this config starts to
bitrot more. That said, maybe it is time to deprecate the 32-bit
hosts?

> let's better disable the qtests here again to avoid failing
> CI tests.
> 
> Signed-off-by: Thomas Huth <thuth@redhat.com>
> ---
>   .gitlab-ci.d/windows.yml | 3 ++-
>   1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/.gitlab-ci.d/windows.yml b/.gitlab-ci.d/windows.yml
> index 22f794e537..a1d5790580 100644
> --- a/.gitlab-ci.d/windows.yml
> +++ b/.gitlab-ci.d/windows.yml
> @@ -116,4 +116,5 @@ msys2-32bit:
>     - ..\msys64\usr\bin\bash -lc '../configure --target-list=ppc64-softmmu
>           --disable-opengl'
>     - ..\msys64\usr\bin\bash -lc 'make'
> -  - ..\msys64\usr\bin\bash -lc 'make check || { cat meson-logs/testlog.txt; exit 1; } ;'
> +  - ..\msys64\usr\bin\bash -lc 'make check MTESTARGS=\"--no-suite qtest\" ||
> +                                { cat meson-logs/testlog.txt; exit 1; }'

Re: [PATCH] .gitlab-ci.d/windows: Do not run the qtests in the msys2-32bit job

[Thomas Huth]

On 05/01/2023 22.42, Philippe Mathieu-Daudé wrote:
> On 5/1/23 21:48, Thomas Huth wrote:
>> The qtests are not stable in the msys2-32bit job yet - especially
>> the test-hmp and the qom-test are failing randomly. Until this is
>> fixed,
> 
> Who is gonna look after this?

It certainly has to be someone who's got a proper Windows installation. I've 
now tried to debug the failures for two days via the gitlab-CI jobs, and 
that just does not work. The turnaround times are way to long, and I really 
cannot waste all my limited CI minutes for such problems.

> I'm not against this patch, but I'm afraid this config starts to
> bitrot more.

It's not really a big step backward - the qtests have just been enabled 
there 3 weeks ago (see commit a35e2ee929741fd), so this is just restoring 
the state from the time before the qtests have been enabled for Windows.

 > That said, maybe it is time to deprecate the 32-bit
 > hosts?

Certainly fine for me, but that's up to the Windows folks to decide. Maybe 
you could just suggest a patch to start the discussion?

  Thomas

Re: [PATCH] .gitlab-ci.d/windows: Do not run the qtests in the msys2-32bit job

[Stefan Weil via]

Am 06.01.23 um 08:49 schrieb Thomas Huth:

> On 05/01/2023 22.42, Philippe Mathieu-Daudé wrote:
>
> > That said, maybe it is time to deprecate the 32-bit
> > hosts?
>
> Certainly fine for me, but that's up to the Windows folks to decide. 
> Maybe you could just suggest a patch to start the discussion?
>
>  Thomas


Download numbers from yesterday for my latest Windows installers:

qemu-w32-setup-20221230.exe - 243

qemu-w64-setup-20221230.exe - 6540

On Wednesday the ratio was 288 : 3516.

As expected the 64-bit variant is used much more often, but it looks 
like there is still a certain desire for the 32-bit variant.

Stefan

Re: [PATCH] .gitlab-ci.d/windows: Do not run the qtests in the msys2-32bit job

[Thomas Huth]

On 06/01/2023 09.15, Stefan Weil wrote:
> Am 06.01.23 um 08:49 schrieb Thomas Huth:
> 
>> On 05/01/2023 22.42, Philippe Mathieu-Daudé wrote:
>>
>> > That said, maybe it is time to deprecate the 32-bit
>> > hosts?
>>
>> Certainly fine for me, but that's up to the Windows folks to decide. Maybe 
>> you could just suggest a patch to start the discussion?
>>
>>  Thomas
> 
> 
> Download numbers from yesterday for my latest Windows installers:
> 
> qemu-w32-setup-20221230.exe - 243
> 
> qemu-w64-setup-20221230.exe - 6540
> 
> On Wednesday the ratio was 288 : 3516.
> 
> As expected the 64-bit variant is used much more often, but it looks like 
> there is still a certain desire for the 32-bit variant.

OK, thanks. Could you maybe also check the browser types in the logs? ... 
I'm wondering whether a big part of those w32 downloads were just automatic 
web crawlers?

  Thomas

Re: [PATCH] .gitlab-ci.d/windows: Do not run the qtests in the msys2-32bit job

[Philippe Mathieu-Daudé]

On 6/1/23 09:19, Thomas Huth wrote:
> On 06/01/2023 09.15, Stefan Weil wrote:
>> Am 06.01.23 um 08:49 schrieb Thomas Huth:
>>
>>> On 05/01/2023 22.42, Philippe Mathieu-Daudé wrote:
>>>
>>> > That said, maybe it is time to deprecate the 32-bit
>>> > hosts?
>>>
>>> Certainly fine for me, but that's up to the Windows folks to decide. 
>>> Maybe you could just suggest a patch to start the discussion?
>>>
>>>  Thomas
>>
>>
>> Download numbers from yesterday for my latest Windows installers:
>>
>> qemu-w32-setup-20221230.exe - 243
>>
>> qemu-w64-setup-20221230.exe - 6540
>>
>> On Wednesday the ratio was 288 : 3516.
>>
>> As expected the 64-bit variant is used much more often, but it looks 
>> like there is still a certain desire for the 32-bit variant.
> 
> OK, thanks. Could you maybe also check the browser types in the logs? 
> ... I'm wondering whether a big part of those w32 downloads were just 
> automatic web crawlers?

Or people downloading both variants "just in case" but only install the
64-bit one ;)

Re: [PATCH] .gitlab-ci.d/windows: Do not run the qtests in the msys2-32bit job

[Stefan Weil via]

Am 06.01.23 um 09:19 schrieb Thomas Huth:

> On 06/01/2023 09.15, Stefan Weil wrote:
>>
>> Download numbers from yesterday for my latest Windows installers:
>>
>> qemu-w32-setup-20221230.exe - 243
>>
>> qemu-w64-setup-20221230.exe - 6540
>>
>> On Wednesday the ratio was 288 : 3516.
>>
>> As expected the 64-bit variant is used much more often, but it looks 
>> like there is still a certain desire for the 32-bit variant.
>
> OK, thanks. Could you maybe also check the browser types in the logs? 
> ... I'm wondering whether a big part of those w32 downloads were just 
> automatic web crawlers?
>
>  Thomas


I now checked all downloads of the latests installers since 2022-12-30.

qemu-w32-setup-20221230.exe – 509 different IP addresses
qemu-w64-setup-20221230.exe - 5471 different IP addresses

339 unique IP addresses are common for 32- and 64-bit, either crawlers 
or people who simply got both variants. So there remain 170 IP addresses 
which only downloaded the 32-bit variant in the last week.

I see 437 different strings for the browser type, but surprisingly none 
of them looks like a crawler.

Stefan