[opensuse-kernel] Re: [opensuse-factory] Different development models of Tumbleweed and Leap
Am Dienstag, 3. November 2020, 12:46:49 CET schrieb Stephan Kulow:
Am 03.11.20 um 09:42 schrieb Hans-Peter Jansen:
Hi,
yesterday, I found myself digging in the guts of a kernel 5.9.3 build, that succeeded with TW and failed with 15.2. This is the second time, a build failed due to stale symlink checking with Leaps (at least).
Given, that all kernel related development is done on TW, this is rather unfortunate. Therefore, I kindly ask to enable the symlink checks, that are enforced with Leaps, for TW as well.
While such issues aren't a big deal, they're a PITA nevertheless.
This has nothing to do with a 'development model', kernel-default.spec sets NO_BRP_STALE_LINK_ERROR=yes
This doesn't explain build differences between Leap and TW, does it? Also, the first example is related to our kernel-firmware package, while the latter affected kernel-source. Cheers, Pete -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
On Tue, 03 Nov 2020 15:48:35 +0100, Hans-Peter Jansen wrote:
Am Dienstag, 3. November 2020, 12:46:49 CET schrieb Stephan Kulow:
Am 03.11.20 um 09:42 schrieb Hans-Peter Jansen:
Hi,
yesterday, I found myself digging in the guts of a kernel 5.9.3 build, that succeeded with TW and failed with 15.2. This is the second time, a build failed due to stale symlink checking with Leaps (at least).
Given, that all kernel related development is done on TW, this is rather unfortunate. Therefore, I kindly ask to enable the symlink checks, that are enforced with Leaps, for TW as well.
While such issues aren't a big deal, they're a PITA nevertheless.
This has nothing to do with a 'development model', kernel-default.spec sets NO_BRP_STALE_LINK_ERROR=yes
This doesn't explain build differences between Leap and TW, does it?
Maybe the term "development model" leads confusion. The reason of different results between TW and Leap is rather whether a specific patch is present or not, so it's irrelevant with "development model", per se. And we have NO_BRP_STALE_LINK_ERROR=yes because of /lib/modules/*/source symlink. It's interesting whether this workaround is still needed, though. AFAIK, it was needed because of the dependency issues in brp check.
Also, the first example is related to our kernel-firmware package, while the latter affected kernel-source.
Yes, those are two different things! Takashi -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
Am Dienstag, 3. November 2020, 16:20:05 CET schrieb Takashi Iwai:
On Tue, 03 Nov 2020 15:48:35 +0100,
Hans-Peter Jansen wrote:
Am Dienstag, 3. November 2020, 12:46:49 CET schrieb Stephan Kulow:
Am 03.11.20 um 09:42 schrieb Hans-Peter Jansen:
Hi,
yesterday, I found myself digging in the guts of a kernel 5.9.3 build, that succeeded with TW and failed with 15.2. This is the second time, a build failed due to stale symlink checking with Leaps (at least).
Given, that all kernel related development is done on TW, this is rather unfortunate. Therefore, I kindly ask to enable the symlink checks, that are enforced with Leaps, for TW as well.
While such issues aren't a big deal, they're a PITA nevertheless.
This has nothing to do with a 'development model', kernel-default.spec sets NO_BRP_STALE_LINK_ERROR=yes
This doesn't explain build differences between Leap and TW, does it?
Maybe the term "development model" leads confusion.
Don't know how to call this, sorry. I renamed the subject to "Different build config" now.
The reason of different results between TW and Leap is rather whether a specific patch is present or not, so it's irrelevant with "development model", per se.
No, the packages I build for TW and Leap are identical.
And we have NO_BRP_STALE_LINK_ERROR=yes because of /lib/modules/*/source symlink. It's interesting whether this workaround is still needed, though. AFAIK, it was needed because of the dependency issues in brp check.
Also, the first example is related to our kernel-firmware package, while the latter affected kernel-source.
Yes, those are two different things!
but the cases, I emphasize here, failed for the same issue (IIRC). Since you might remember the first, here the second. Without: rm -vf linux-%srcversion/tools/testing/selftests/powerpc/copyloops/memcpy_mcsafe_64.S in line 208 of: https://build.opensuse.org/package/view_file/home:frispete:kernel/kernel-sou... The build is going to fail for !TW: [ 2908s] calling /usr/lib/rpm/brp-suse.d/brp-25-symlink [ 2908s] ERROR: link target doesn't exist (neither in build root nor in installed system): [ 2908s] /usr/src/linux-5.9.3-lp152.3-vanilla/tools/testing/selftests/powerpc/copyloops/memcpy_mcsafe_64.S -> /usr/src/linux-5.9.3-lp152.3-vanilla/arch/powerpc/lib/memcpy_mcsafe_64.S [ 2908s] Add the package providing the target to BuildRequires and Requires [ 2909s] INFO: relinking /usr/src/linux-5.9.3-lp152.3-vanilla/tools/testing/selftests/powerpc/primitives/asm/asm-compat.h -> ../../../../../../arch/powerpc/include/asm/asm-compat.h (was ../.././../../../../arch/powerpc/include/asm/asm-compat.h) Again, it's a stale symlink in the package itself. In this case, due to [PATCH 5.9 24/74] x86, powerpc: Rename memcpy_mcsafe() to copy_mc_to_{user, kernel}() applied as part of v5.9.3 fixes: https://lkml.org/lkml/2020/10/31/241 For some reason, it missed to remove a symlink to the deleted file in arch/powerpc/lib/memcpy_mcsafe_64.S More correctly, it was renamed: arch/powerpc/lib/{copy_mc_64,memcpy_mcsafe_64}.S My case is, I saw recurring issues with stale symlinks over time, these I can remember best.. There exist some differences in the build config, and I would like to see the more rigid ones concerning stale symlinks of Leap in TW as well. I hoped, somebody would remember such differences, and could point us to them. Cheers, Pete -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
Am Dienstag, 3. November 2020, 17:29:17 CET schrieb Hans-Peter Jansen:
I hoped, somebody would remember such differences, and could point us to them.
Coolo did, and fixed it: https://github.com/openSUSE/brp-check-suse/pull/36 Just a plain error. Sorry for the churn. Cheers, Pete -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org
participants (2)
-
Hans-Peter Jansen -
Takashi Iwai