[PATCH] valgrind: Disable drd/tests/bar_bad ptest

[PATCH] valgrind: Disable drd/tests/bar_bad ptest

[Randy MacLeod]

This is failing intermettently in the YP autobuilder
so disable it until we can fix the root cause.

[YOCTO #14311]

Signed-off-by: Randy MacLeod <Randy.MacLeod@windriver.com>
---
 meta/recipes-devtools/valgrind/valgrind/remove-for-all | 1 +
 1 file changed, 1 insertion(+)

diff --git a/meta/recipes-devtools/valgrind/valgrind/remove-for-all b/meta/recipes-devtools/valgrind/valgrind/remove-for-all
index 5d1e7ed102..e26088f3b1 100644
--- a/meta/recipes-devtools/valgrind/valgrind/remove-for-all
+++ b/meta/recipes-devtools/valgrind/valgrind/remove-for-all
@@ -3,6 +3,7 @@ gdbserver_tests/hginfo
 memcheck/tests/supp_unknown
 helgrind/tests/tls_threads
 helgrind/tests/pth_mempcpy_false_races
+drd/tests/bar_bad
 drd/tests/bar_bad_xml
 drd/tests/pth_barrier_thr_cr
 drd/tests/std_thread2
-- 
2.34.1

Re: [OE-core] [PATCH] valgrind: Disable drd/tests/bar_bad ptest

[Alexander Kanavin]

Ok, I'll bite. Should we just drop valgrind from ptests? It's by far,
the most brittle, finicky ptest we have, and doesn't seem to expose
actual issues, instead just giving everyone a never ending stream of
intermittent failures or tests that won't work on anything except x86
configured in a particular way.

Alex

On Thu, 9 Mar 2023 at 22:23, Randy MacLeod <randy.macleod@windriver.com> wrote:
>
> This is failing intermettently in the YP autobuilder
> so disable it until we can fix the root cause.
>
> [YOCTO #14311]
>
> Signed-off-by: Randy MacLeod <Randy.MacLeod@windriver.com>
> ---
>  meta/recipes-devtools/valgrind/valgrind/remove-for-all | 1 +
>  1 file changed, 1 insertion(+)
>
> diff --git a/meta/recipes-devtools/valgrind/valgrind/remove-for-all b/meta/recipes-devtools/valgrind/valgrind/remove-for-all
> index 5d1e7ed102..e26088f3b1 100644
> --- a/meta/recipes-devtools/valgrind/valgrind/remove-for-all
> +++ b/meta/recipes-devtools/valgrind/valgrind/remove-for-all
> @@ -3,6 +3,7 @@ gdbserver_tests/hginfo
>  memcheck/tests/supp_unknown
>  helgrind/tests/tls_threads
>  helgrind/tests/pth_mempcpy_false_races
> +drd/tests/bar_bad
>  drd/tests/bar_bad_xml
>  drd/tests/pth_barrier_thr_cr
>  drd/tests/std_thread2
> --
> 2.34.1
>
>
> -=-=-=-=-=-=-=-=-=-=-=-
> Links: You receive all messages sent to this group.
> View/Reply Online (#178271): https://lists.openembedded.org/g/openembedded-core/message/178271
> Mute This Topic: https://lists.openembedded.org/mt/97506606/1686489
> Group Owner: openembedded-core+owner@lists.openembedded.org
> Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [alex.kanavin@gmail.com]
> -=-=-=-=-=-=-=-=-=-=-=-
>

Re: [OE-core] [PATCH] valgrind: Disable drd/tests/bar_bad ptest

[Randy MacLeod]

[-- Attachment #1: Type: text/plain, Size: 3687 bytes --]

On 2023-03-10 05:04, Alexander Kanavin wrote:
> Ok, I'll bite. Should we just drop valgrind from ptests? It's by far,
> the most brittle, finicky ptest we have, and doesn't seem to expose
> actual issues, instead just giving everyone a never ending stream of
> intermittent failures or tests that won't work on anything except x86
> configured in a particular way.


Hi Alex,

It's a bit of a challenge but I'd like to keep the valgrind ptests around
since most actually pass and we'd likely never know if the tool was broken
otherwise.

Non-x86 used to be a problem but not so much in hardknott. (1)

We do run > 600 tests and we have a mechanism to skip the
10-20 that are a problem:

❯ wc -l meta/recipes-devtools/valgrind/valgrind/remove-for-a*
  13 meta/recipes-devtools/valgrind/valgrind/remove-for-aarch64
  11 meta/recipes-devtools/valgrind/valgrind/remove-for-all

../Randy


1)

commit cbeb9418c43ec834868aa65b774dc09e983d26d9
Author: Qiu, Zheng <Zheng.Qiu@windriver.com>
Date:   Fri Nov 4 14:47:05 2022

     valgrind: remove most hidden tests for arm64

     An earlier version of valgrind fixed the defunct processes bug, so 
those
     tests that were skipped specifically for arm can pass now in master,
     kirkstone, honister, hardknott, and dunfell.

     Detailed test result with remove-for-aarch64 skipped on qemuarm64:

         Commit           Pass   Fail    Skip
         master           624    9       21
         kirkstone        618    10      20
         honister         616    10      19
         hardknott        609    13      18
         dunfell          598    16      17
         zeus             Out of memory: Killed (with many defunct 
processes)

     There are now only 12 skipped by remove-for-aarch64 because 9 fail on
     qemuarm64 and 3 more fail on raspberry pi. These are tracked by:
https://bugzilla.yoctoproject.org/show_bug.cgi?id=14960

     Signed-off-by: Zheng Qiu <zheng.qiu@windriver.com>

>
> Alex
>
> On Thu, 9 Mar 2023 at 22:23, Randy MacLeod<randy.macleod@windriver.com>  wrote:
>> This is failing intermettently in the YP autobuilder
>> so disable it until we can fix the root cause.
>>
>> [YOCTO #14311]
>>
>> Signed-off-by: Randy MacLeod<Randy.MacLeod@windriver.com>
>> ---
>>   meta/recipes-devtools/valgrind/valgrind/remove-for-all | 1 +
>>   1 file changed, 1 insertion(+)
>>
>> diff --git a/meta/recipes-devtools/valgrind/valgrind/remove-for-all b/meta/recipes-devtools/valgrind/valgrind/remove-for-all
>> index 5d1e7ed102..e26088f3b1 100644
>> --- a/meta/recipes-devtools/valgrind/valgrind/remove-for-all
>> +++ b/meta/recipes-devtools/valgrind/valgrind/remove-for-all
>> @@ -3,6 +3,7 @@ gdbserver_tests/hginfo
>>   memcheck/tests/supp_unknown
>>   helgrind/tests/tls_threads
>>   helgrind/tests/pth_mempcpy_false_races
>> +drd/tests/bar_bad
>>   drd/tests/bar_bad_xml
>>   drd/tests/pth_barrier_thr_cr
>>   drd/tests/std_thread2
>> --
>> 2.34.1
>>
>>
>> -=-=-=-=-=-=-=-=-=-=-=-
>> Links: You receive all messages sent to this group.
>> View/Reply Online (#178271):https://lists.openembedded.org/g/openembedded-core/message/178271
>> Mute This Topic:https://lists.openembedded.org/mt/97506606/1686489
>> Group Owner:openembedded-core+owner@lists.openembedded.org
>> Unsubscribe:https://lists.openembedded.org/g/openembedded-core/unsub  [alex.kanavin@gmail.com]
>> -=-=-=-=-=-=-=-=-=-=-=-
>>

-- 
# Randy MacLeod
# Wind River Linux

[-- Attachment #2: Type: text/html, Size: 6595 bytes --]