Re: [OE-core] [PATCH 4/8] debug_build.inc: collect debug build tuning configuration

[Hongxu Jia <hongxu.jia@windriver.com>]

[-- Attachment #1: Type: text/plain, Size: 9379 bytes --]

On 9/19/25 11:08, hongxu via lists.openembedded.org wrote:
> On 9/17/25 19:24, Mathieu Dubois-Briand via lists.openembedded.org wrote:
>> CAUTION: This email comes from a non Wind River email account!
>> Do not click links or open attachments unless you recognize the sender and know the content is safe.
>>
>> On Tue Sep 16, 2025 at 6:09 AM CEST, hongxu via lists.openembedded.org wrote:
>>> The modern compilers and code seem to require extra steps to avoid DEBUG errors,
>>> Move debug tuning configuration from recipes to an include file to address these
>>> errors.
>>>
>>> Introduce separate variables to avoid the :remove operations, such DEBUG_OPTLEVEL,
>>> BUILD_OPTLEVEL and OECMAKE_WEBKIT_NO_INLINE_HINTS
>>>
>>> Suggested-by: Peter Kjellerstedt<peter.kjellerstedt@axis.com>
>>> Signed-off-by: Hongxu Jia<hongxu.jia@windriver.com>
>>> ---
>> Just a note about this patches: I had perf reproducibility issues [1][2]
>> after taking this series and I believe this is linked with this patch. I
>> suspect this is not directly related with it, but more with the issue we
>> have had with perf reproducibility last week.
>>
>> So I dropped this patch for now, but I will take it a bit later this
>> week and try to go further. I still wanted to provide some feedback
>> here. Also, if anyone believes this issue IS related with this series,
>> please share your thoughts.
>>
>> [1]https://autobuilder.yoctoproject.org/valkyrie/#/builders/37/builds/2505
>> [2]http://valkyrie.yocto.io/pub/repro-fail/oe-reproducible-20250916-0ss72vce/

I am digging in the issue, and found the issue only happen on world 
build of reproducible, have different debuginfo data.

I found the commit [1] to fix reproducibility issue for perf, but it 
seems the issue still existed,

...

     perf: Fix reproducibility issue

     Perf's build process generates two copies of its internal headers,
     one at tools/lib/perf/include/internal and one at 
libperf/include/internal
     with files like xyarray.h. Although the files are identical, the 
binaries including
     them would see them at different paths and have different debuginfo 
data.

     To avoid this, build libperf.a first, which will ensure the libperf
     headers directory exists. This is used in preference to the other by
     the build process and results in consistent binaries.

...

[1] 
https://git.openembedded.org/openembedded-core/commit/?id=0a886fcacaab0fbce1306c0f99d482f940a8f705

oe-selftest -r reproducible.ReproducibleTests.test_reproducible_builds

$ recipe-sysroot-native/usr/bin/dwarfsrcfiles 
sources/perf-1.0/libperf/core.o
/usr/src/debug/perf/1.0/tools/lib/perf/core.c
...
         /usr/src/debug/perf/1.0/tools/lib/perf/include/perf/core.h
         /usr/src/debug/perf/1.0/tools/lib/perf/include/internal/lib.h
...

$ recipe-sysroot-native/usr/bin/dwarfsrcfiles 
sources/perf-1.0/libperf/core.o
/usr/src/debug/perf/1.0/tools/lib/perf/core.c
...
     /usr/src/debug/perf/1.0/libperf/include/perf/core.h
     /usr/src/debug/perf/1.0/libperf/include/internal/lib.h
...


If specify perf of reproducible for oe selftest, the issue was not found

$ echo 'OEQA_REPRODUCIBLE_TEST_TARGET = "perf"' >> conf/local.conf

$ oe-selftest -r reproducible.ReproducibleTests.test_reproducible_builds

$ recipe-sysroot-native/usr/bin/dwarfsrcfiles 
sources/perf-1.0/libperf/core.o
/usr/src/debug/perf/1.0/tools/lib/perf/core.c
...
     /usr/src/debug/perf/1.0/libperf/include/perf/core.h
     /usr/src/debug/perf/1.0/libperf/include/internal/lib.h
...

$ recipe-sysroot-native/usr/bin/dwarfsrcfiles 
sources/perf-1.0/libperf/core.o
/usr/src/debug/perf/1.0/tools/lib/perf/core.c
...
     /usr/src/debug/perf/1.0/libperf/include/perf/core.h
     /usr/src/debug/perf/1.0/libperf/include/internal/lib.h
...

//Hongxu

> Hi
>
> I clone the branch of you on contrib
>
> https://git.openembedded.org/openembedded-core-contrib/log/?h=mathieu/master-next-success
>
> https://git.openembedded.org/bitbake-contrib/log/?h=mathieu/master-next-success
>
> And do oe-selftest without the debug build patches, the issue we have 
> had with perf reproducibility still exist
>
> $ oe-selftest -r reproducible.ReproducibleTests.test_reproducible_builds
>
> ...
>
> 2025-09-19 05:05:31,099 - oe-selftest - INFO - 
> ======================================================================
> 2025-09-19 05:05:31,099 - oe-selftest - INFO - FAIL: 
> test_reproducible_builds (reproducible.ReproducibleTests)
> 2025-09-19 05:05:31,099 - oe-selftest - INFO - 
> ----------------------------------------------------------------------
> 2025-09-19 05:05:31,099 - oe-selftest - INFO - Traceback (most recent 
> call last):
>   File 
> "/buildarea5/hjia/contrib/openembedded-core-contrib/meta/lib/oeqa/selftest/cases/reproducible.py", 
> line 406, in test_reproducible_builds
>     self.fail('\n'.join(fails))
> AssertionError: Bitbake reproducibleA failure
> Bitbake reproducibleB-extended failure
> The following deb packages are different and not in exclusion list:
> /buildarea5/hjia/contrib/openembedded-core-contrib/build-st/reproducibleB-extended/tmp/deploy/deb/./qemux86_64/perf-dbg_6.16.7-r0_amd64.deb
> /buildarea5/hjia/contrib/openembedded-core-contrib/build-st/reproducibleB-extended/tmp/deploy/deb/./qemux86_64/perf-python_6.16.7-r0_amd64.deb
> /buildarea5/hjia/contrib/openembedded-core-contrib/build-st/reproducibleB-extended/tmp/deploy/deb/./qemux86_64/perf-src_6.16.7-r0_amd64.deb
> /buildarea5/hjia/contrib/openembedded-core-contrib/build-st/reproducibleB-extended/tmp/deploy/deb/./qemux86_64/perf_6.16.7-r0_amd64.deb
> /buildarea5/hjia/contrib/openembedded-core-contrib/build-st/reproducibleB-extended/tmp/deploy/deb/./x86-64-v3/apt-dbg_3.0.3-r0_amd64.deb
> /buildarea5/hjia/contrib/openembedded-core-contrib/build-st/reproducibleB-extended/tmp/deploy/deb/./x86-64-v3/apt_3.0.3-r0_amd64.deb
> The following ipk packages are different and not in exclusion list:
> /buildarea5/hjia/contrib/openembedded-core-contrib/build-st/reproducibleB-extended/tmp/deploy/ipk/./qemux86_64/perf-dbg_6.16.7-r0_qemux86_64.ipk
> /buildarea5/hjia/contrib/openembedded-core-contrib/build-st/reproducibleB-extended/tmp/deploy/ipk/./qemux86_64/perf-python_6.16.7-r0_qemux86_64.ipk
> /buildarea5/hjia/contrib/openembedded-core-contrib/build-st/reproducibleB-extended/tmp/deploy/ipk/./qemux86_64/perf-src_6.16.7-r0_qemux86_64.ipk
> /buildarea5/hjia/contrib/openembedded-core-contrib/build-st/reproducibleB-extended/tmp/deploy/ipk/./qemux86_64/perf_6.16.7-r0_qemux86_64.ipk
> /buildarea5/hjia/contrib/openembedded-core-contrib/build-st/reproducibleB-extended/tmp/deploy/ipk/./x86-64-v3/apt-dbg_3.0.3-r0_x86-64-v3.ipk
> /buildarea5/hjia/contrib/openembedded-core-contrib/build-st/reproducibleB-extended/tmp/deploy/ipk/./x86-64-v3/apt_3.0.3-r0_x86-64-v3.ipk
> The following rpm packages are different and not in exclusion list:
> /buildarea5/hjia/contrib/openembedded-core-contrib/build-st/reproducibleB-extended/tmp/deploy/rpm/./qemux86_64/perf-6.16.7-r0.qemux86_64.rpm
> /buildarea5/hjia/contrib/openembedded-core-contrib/build-st/reproducibleB-extended/tmp/deploy/rpm/./qemux86_64/perf-dbg-6.16.7-r0.qemux86_64.rpm
> /buildarea5/hjia/contrib/openembedded-core-contrib/build-st/reproducibleB-extended/tmp/deploy/rpm/./qemux86_64/perf-python-6.16.7-r0.qemux86_64.rpm
> /buildarea5/hjia/contrib/openembedded-core-contrib/build-st/reproducibleB-extended/tmp/deploy/rpm/./qemux86_64/perf-src-6.16.7-r0.qemux86_64.rpm
> /buildarea5/hjia/contrib/openembedded-core-contrib/build-st/reproducibleB-extended/tmp/deploy/rpm/./x86_64_v3/apt-3.0.3-r0.x86_64_v3.rpm
> /buildarea5/hjia/contrib/openembedded-core-contrib/build-st/reproducibleB-extended/tmp/deploy/rpm/./x86_64_v3/apt-dbg-3.0.3-r0.x86_64_v3.rpm
>
> 2025-09-19 05:05:31,099 - oe-selftest - INFO - 
> ----------------------------------------------------------------------
> 2025-09-19 05:05:31,099 - oe-selftest - INFO - Ran 1 test in 32633.939s
> 2025-09-19 05:05:31,099 - oe-selftest - INFO - FAILED
> 2025-09-19 05:05:31,099 - oe-selftest - INFO -  (failures=1)
> 2025-09-19 05:05:34,691 - oe-selftest - INFO - RESULTS:
> 2025-09-19 05:05:34,691 - oe-selftest - INFO - RESULTS - 
> reproducible.ReproducibleTests.test_reproducible_builds: FAILED 
> (32633.79s)
> 2025-09-19 05:05:34,736 - oe-selftest - INFO - SUMMARY:
> 2025-09-19 05:05:34,736 - oe-selftest - INFO - oe-selftest () - Ran 1 
> test in 32633.939s
> 2025-09-19 05:05:34,736 - oe-selftest - INFO - oe-selftest - FAIL - 
> Required tests failed (successes=0, skipped=0, failures=1, errors=0)
>
> ...
>
> //Hongxu
>
>> Thanks,
>> Mathieu
>>
>> --
>> Mathieu Dubois-Briand, Bootlin
>> Embedded Linux and Kernel engineering
>> https://bootlin.com
>>
>>
>
>
> -=-=-=-=-=-=-=-=-=-=-=-
> Links: You receive all messages sent to this group.
> View/Reply Online (#223719):https://lists.openembedded.org/g/openembedded-core/message/223719
> Mute This Topic:https://lists.openembedded.org/mt/115268059/3617049
> Group Owner:openembedded-core+owner@lists.openembedded.org
> Unsubscribe:https://lists.openembedded.org/g/openembedded-core/unsub  [hongxu.jia@eng.windriver.com]
> -=-=-=-=-=-=-=-=-=-=-=-
>

[-- Attachment #2: Type: text/html, Size: 13455 bytes --]