| Message ID | 20220804041121.5672-1-sundeep.kokkonda@gmail.com |
|---|---|
| State | Accepted, archived |
| Commit | ec5de254e40d188cd0cce75568851b0c130533e6 |
| Headers | show |
| Series | [kirkstone] glibc : stable 2.35 branch updates | expand |
Hello, Regression tests are performed. Below is the test summary. > > Summary of test results *before update* of glibc: > 146 FAIL > 4395 PASS > 2 UNRESOLVED > 41 UNSUPPORTED > 12 XFAIL > 6 XPASS > > Summary of test results *after update* of glibc: > 150 FAIL > 4393 PASS > 41 UNSUPPORTED > 12 XFAIL > 6 XPASS > There are new 11 failures after glibc update which all are from ' localedata ' sub-module. I am looking into the failed cases, suspecting that these failures are because of test procedure issue. Thanks, Sundeep K.
Does this fix the build with DEBUG_BUILD? I don't see any commit related to that nor any update in https://sourceware.org/bugzilla/show_bug.cgi?id=29249 On Thu, Aug 4, 2022 at 6:18 AM Sundeep KOKKONDA <sundeep.kokkonda@gmail.com> wrote: > Hello, > > Regression tests are performed. Below is the test summary. > > Summary of test results *before update* of glibc: > 146 FAIL > 4395 PASS > 2 UNRESOLVED > 41 UNSUPPORTED > 12 XFAIL > 6 XPASS > > Summary of test results *after update* of glibc: > 150 FAIL > 4393 PASS > 41 UNSUPPORTED > 12 XFAIL > 6 XPASS > > There are new 11 failures after glibc update which all are from ' > *localedata*' sub-module. I am looking into the failed cases, suspecting > that these failures are because of test procedure issue. > > > > Thanks, > Sundeep K. > > -=-=-=-=-=-=-=-=-=-=-=- > Links: You receive all messages sent to this group. > View/Reply Online (#168836): > https://lists.openembedded.org/g/openembedded-core/message/168836 > Mute This Topic: https://lists.openembedded.org/mt/92807904/3617156 > Group Owner: openembedded-core+owner@lists.openembedded.org > Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [ > Martin.Jansa@gmail.com] > -=-=-=-=-=-=-=-=-=-=-=- > >
Is there a fix available for this issue and was it taken to glibc branches? I checked glibc 2.35 branch now could not see any new commits available. If there is any fix/workaround required then it should be patched in ' meta/recipes-core/glibc/ ' OR if it is a bug in glibc then it should be corrected by glibc. It seems the mentioned issue is still on-going and no fix is available yet.
On Thu, Aug 4, 2022 at 12:20 PM Sundeep KOKKONDA <sundeep.kokkonda@gmail.com> wrote: > Is there a fix available for this issue and was it taken to glibc > branches? I checked glibc 2.35 branch now could not see any new commits > available. > If there is any fix/workaround required then it should be patched in ' > *meta/recipes-core/glibc/*' OR if it is a bug in glibc then it should be > corrected by glibc. It seems the mentioned issue is still on-going and no > fix is available yet. Another thing I am not sure about is if there are fixes which will uncover latent bugs we perhaps need to look into that closely and document it out > -=-=-=-=-=-=-=-=-=-=-=- > Links: You receive all messages sent to this group. > View/Reply Online (#168842): > https://lists.openembedded.org/g/openembedded-core/message/168842 > Mute This Topic: https://lists.openembedded.org/mt/92807904/1997914 > Group Owner: openembedded-core+owner@lists.openembedded.org > Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [ > raj.khem@gmail.com] > -=-=-=-=-=-=-=-=-=-=-=- > >
On Thu, Aug 04, 2022 at 04:20:44AM -0700, Sundeep KOKKONDA wrote: > Is there a fix available for this issue and was it taken to glibc branches? I checked glibc 2.35 branch now could not see any new commits available. > If there is any fix/workaround required then it should be patched in ' meta/recipes-core/glibc/ ' OR if it is a bug in glibc then it should be corrected by glibc. It seems the mentioned issue is still on-going and no fix is available yet. Then maybe we should hold on more "stable" updates for kirkstone, before the DEBUG_BUILD regression from the previous upgrade is fixed? The regression tests are great, but they don't help to detect issues like the DEBUG_BUILD. This change also seems to target kirkstone directly without being tested in master first. I know Khem has a change for 2.36 glibc for master, but maybe merging this update in master first (and resolving the DEBUG_BUILD issue even with that revert I've sent earlier) could be merged before that.
On Thu, Aug 4, 2022 at 1:58 AM Martin Jansa <Martin.Jansa@gmail.com> wrote: > > On Thu, Aug 04, 2022 at 04:20:44AM -0700, Sundeep KOKKONDA wrote: > > Is there a fix available for this issue and was it taken to glibc branches? I checked glibc 2.35 branch now could not see any new commits available. > > If there is any fix/workaround required then it should be patched in ' meta/recipes-core/glibc/ ' OR if it is a bug in glibc then it should be corrected by glibc. It seems the mentioned issue is still on-going and no fix is available yet. > > Then maybe we should hold on more "stable" updates for kirkstone, before > the DEBUG_BUILD regression from the previous upgrade is fixed? The > regression tests are great, but they don't help to detect issues like > the DEBUG_BUILD. > > This change also seems to target kirkstone directly without being tested > in master first. I know Khem has a change for 2.36 glibc for master, but > maybe merging this update in master first (and resolving the DEBUG_BUILD > issue even with that revert I've sent earlier) could be merged before > that. I've been waiting for your DEBUG_BUILD regression fix to hit master so I can take it in kirkstone. The 4.0.3 build is scheduled for Monday, but will likely be held up waiting for a linux-yocto intermittent reproducibility fix. I suspect I should also wait for your patch too before doing the 4.0.3 build. Steve > -=-=-=-=-=-=-=-=-=-=-=- > Links: You receive all messages sent to this group. > View/Reply Online (#168846): https://lists.openembedded.org/g/openembedded-core/message/168846 > Mute This Topic: https://lists.openembedded.org/mt/92807904/3620601 > Group Owner: openembedded-core+owner@lists.openembedded.org > Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [steve@sakoman.com] > -=-=-=-=-=-=-=-=-=-=-=- >
diff --git a/meta/recipes-core/glibc/glibc-version.inc b/meta/recipes-core/glibc/glibc-version.inc index 99017ce1d4..ccb41e5af6 100644 --- a/meta/recipes-core/glibc/glibc-version.inc +++ b/meta/recipes-core/glibc/glibc-version.inc @@ -1,6 +1,6 @@ SRCBRANCH ?= "release/2.35/master" PV = "2.35" -SRCREV_glibc ?= "b6aade18a7e5719c942aa2da6cf3157aca993fa4" +SRCREV_glibc ?= "0e5b239f45992e4b54c6f946ecb0c410afc8bb08" SRCREV_localedef ?= "794da69788cbf9bf57b59a852f9f11307663fa87" GLIBC_GIT_URI ?= "git://sourceware.org/git/glibc.git"
Below commits on glibc-2.35 development branch are updated. glibc: 0e5b239f45 malloc: Simplify implementation of __malloc_assert dc2d843045 Update syscall-names.list for Linux 5.18 d56c7e0965 Apply asm redirections in stdio.h before first use [BZ #27087] c4050b2552 x86: Add missing IS_IN (libc) check to strncmp-sse4_2.S fdf699edea x86: Move mem{p}{mov|cpy}_{chk_}erms to its own file de6f7bb873 x86: Move and slightly improve memset_erms 210e9728ff x86: Add definition for __wmemset_chk AVX2 RTM in ifunc impl list c77cde8172 x86: Put wcs{n}len-sse4.1 in the sse4.1 text section 3eb17048c4 x86: Align entry for memrchr to 64-bytes. dd3b6857ee x86: Add BMI1/BMI2 checks for ISA_V3 check ce32ad91eb x86: Cleanup bounds checking in large memcpy case 863987a6ef x86: Add bounds `x86_non_temporal_threshold` 232b7adb14 x86: Add sse42 implementation to strcmp's ifunc 7f7a728b71 x86: Fix misordered logic for setting `rep_movsb_stop_threshold` 6b4a2ab7e0 x86: Align varshift table to 32-bytes e74385736c x86: ZERO_UPPER_VEC_REGISTERS_RETURN_XTEST expect no transactions b4744d4414 x86: Shrink code size of memchr-evex.S 5321a217cf x86: Shrink code size of memchr-avx2.S c536b318f0 x86: Optimize memrchr-avx2.S f8ba0e0452 x86: Optimize memrchr-evex.S 7001d558a0 x86: Optimize memrchr-sse2.S 54486f520a x86: Add COND_VZEROUPPER that can replace vzeroupper if no `ret` 82560fe43e x86: Create header for VEC classes in x86 strings library b1f23b6128 x86_64: Add strstr function with 512-bit EVEX f9f0fbbf7b x86-64: Ignore r_addend for R_X86_64_GLOB_DAT/R_X86_64_JUMP_SLOT 650bf51c78 x86_64: Implement evex512 version of strlen, strnlen, wcslen and wcsnlen e0cb101d1b x86_64: Remove bzero optimization 51e2d3b53b nptl: Fix ___pthread_unregister_cancel_restore asynchronous restore cd4f43be3d linux: Fix mq_timereceive check for 32 bit fallback code (BZ 29304) Signed-off-by: Sundeep KOKKONDA <sundeep.kokkonda@gmail.com> --- meta/recipes-core/glibc/glibc-version.inc | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)