Message ID | 20211204071323.2480163-5-alex@linutronix.de |
---|---|
State | Accepted, archived |
Commit | 2eabd3b611f4afa5eafa4490b6b0ef7578b742de |
Headers | show |
Series | [01/26] insane.bbclass: do not hardcode oe-core path in upstream-status check | expand |
On Sat, 2021-12-04 at 08:13 +0100, Alexander Kanavin wrote: > I ran the reproducing sequence on qemux86, and it went fine: > > root@qemux86:~# python3 > Python 3.10.0 (default, Oct 4 2021, 17:55:55) [GCC 11.2.0] on linux > Type "help", "copyright", "credits" or "license" for more information. > > > > import multiprocessing > > > > pool_sema = multiprocessing.BoundedSemaphore(value=1) > > > > pool_sema.acquire() > True > > > > pool_sema.release() > > > > > > Additionally AB testing revealed no issues, and the description isn't entirely > clear about where and how the problem occurs, so I can't submit it upsream > without having that understanding and how to demonstrate the problem. > > Signed-off-by: Alexander Kanavin <alex@linutronix.de> > There is a slightly different explanation here. libpthread was merged into libc so this fix becomes a null-op on glibc. Not sure if musl has a separate pthread but if so we could/should test that? Cheers, Richard
On Mon, Dec 6, 2021 at 3:07 AM Richard Purdie < richard.purdie@linuxfoundation.org> wrote: > On Sat, 2021-12-04 at 08:13 +0100, Alexander Kanavin wrote: > > I ran the reproducing sequence on qemux86, and it went fine: > > > > root@qemux86:~# python3 > > Python 3.10.0 (default, Oct 4 2021, 17:55:55) [GCC 11.2.0] on linux > > Type "help", "copyright", "credits" or "license" for more information. > > > > > import multiprocessing > > > > > pool_sema = multiprocessing.BoundedSemaphore(value=1) > > > > > pool_sema.acquire() > > True > > > > > pool_sema.release() > > > > > > > > > Additionally AB testing revealed no issues, and the description isn't > entirely > > clear about where and how the problem occurs, so I can't submit it > upsream > > without having that understanding and how to demonstrate the problem. > > > > Signed-off-by: Alexander Kanavin <alex@linutronix.de> > > > > There is a slightly different explanation here. libpthread was merged into > libc > so this fix becomes a null-op on glibc. Not sure if musl has a separate > pthread > but if so we could/should test that? Musl has always been a single shared object from day one infact glibc is now moving in that direction so it will work fine on musl > > > Cheers, > > Richard > > > > -=-=-=-=-=-=-=-=-=-=-=- > Links: You receive all messages sent to this group. > View/Reply Online (#159219): > https://lists.openembedded.org/g/openembedded-core/message/159219 > Mute This Topic: https://lists.openembedded.org/mt/87494766/1997914 > Group Owner: openembedded-core+owner@lists.openembedded.org > Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [ > raj.khem@gmail.com] > -=-=-=-=-=-=-=-=-=-=-=- > >
On Mon, 6 Dec 2021 at 17:13, Khem Raj <raj.khem@gmail.com> wrote: > There is a slightly different explanation here. libpthread was merged into > libc > >> so this fix becomes a null-op on glibc. Not sure if musl has a separate >> pthread >> but if so we could/should test that? > > > Musl has always been a single shared object from day one infact glibc is > now moving in that direction so it will work fine on musl > Yep, I just verified this on qemux86 and qemux86-64. Alex
diff --git a/meta/recipes-devtools/python/python3/0001-setup.py-pass-missing-libraries-to-Extension-for-mul.patch b/meta/recipes-devtools/python/python3/0001-setup.py-pass-missing-libraries-to-Extension-for-mul.patch deleted file mode 100644 index 374433c68e..0000000000 --- a/meta/recipes-devtools/python/python3/0001-setup.py-pass-missing-libraries-to-Extension-for-mul.patch +++ /dev/null @@ -1,70 +0,0 @@ -From bad7e6a625436402a01d03021fb9ccd58bc9930f Mon Sep 17 00:00:00 2001 -From: Alexander Kanavin <alex.kanavin@gmail.com> -Date: Thu, 9 Jan 2020 17:44:05 +0100 -Subject: [PATCH] setup.py: pass missing libraries to Extension for - multiprocessing module - -In the following commit: -... -commit e711cafab13efc9c1fe6c5cd75826401445eb585 -Author: Benjamin Peterson <benjamin@python.org> -Date: Wed Jun 11 16:44:04 2008 +0000 - - Merged revisions 64104,64117 via svnmerge from - svn+ssh://pythondev@svn.python.org/python/trunk -... -(see diff in setup.py) -It assigned libraries for multiprocessing module according -the host_platform, but not pass it to Extension. - -In glibc, the following commit caused two definition of -sem_getvalue are different. -https://sourceware.org/git/?p=glibc.git;a=commit;h=042e1521c794a945edc43b5bfa7e69ad70420524 -(see diff in nptl/sem_getvalue.c for detail) -`__new_sem_getvalue' is the latest sem_getvalue@@GLIBC_2.1 -and `__old_sem_getvalue' is to compat the old version -sem_getvalue@GLIBC_2.0. - -To build python for embedded Linux systems: -http://www.yoctoproject.org/docs/2.3.1/yocto-project-qs/yocto-project-qs.html -If not explicitly link to library pthread (-lpthread), it will -load glibc's sem_getvalue randomly at runtime. - -Such as build python on linux x86_64 host and run the python -on linux x86_32 target. If not link library pthread, it caused -multiprocessing bounded semaphore could not work correctly. -... ->>> import multiprocessing ->>> pool_sema = multiprocessing.BoundedSemaphore(value=1) ->>> pool_sema.acquire() -True ->>> pool_sema.release() -Traceback (most recent call last): - File "<stdin>", line 1, in <module> -ValueError: semaphore or lock released too many times -... - -And the semaphore issue also caused multiprocessing.Queue().put() hung. - -Upstream-Status: Pending - -Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com> -Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com> - ---- - setup.py | 2 +- - 1 file changed, 1 insertion(+), 1 deletion(-) - -diff --git a/setup.py b/setup.py -index d92face..f42bcbb 100644 ---- a/setup.py -+++ b/setup.py -@@ -1836,7 +1836,7 @@ class PyBuildExt(build_ext): - if (sysconfig.get_config_var('HAVE_SEM_OPEN') and not - sysconfig.get_config_var('POSIX_SEMAPHORES_NOT_ENABLED')): - multiprocessing_srcs.append('_multiprocessing/semaphore.c') -- self.add(Extension('_multiprocessing', multiprocessing_srcs, -+ self.add(Extension('_multiprocessing', multiprocessing_srcs, libraries=['pthread'], - include_dirs=["Modules/_multiprocessing"])) - - if (not MS_WINDOWS and diff --git a/meta/recipes-devtools/python/python3_3.10.0.bb b/meta/recipes-devtools/python/python3_3.10.0.bb index e3300b6495..24e6c04fe9 100644 --- a/meta/recipes-devtools/python/python3_3.10.0.bb +++ b/meta/recipes-devtools/python/python3_3.10.0.bb @@ -25,7 +25,6 @@ SRC_URI = "http://www.python.org/ftp/python/${PV}/Python-${PV}.tar.xz \ file://0001-Use-FLAG_REF-always-for-interned-strings.patch \ file://0001-test_locale.py-correct-the-test-output-format.patch \ file://0017-setup.py-do-not-report-missing-dependencies-for-disa.patch \ - file://0001-setup.py-pass-missing-libraries-to-Extension-for-mul.patch \ file://0001-Makefile-do-not-compile-.pyc-in-parallel.patch \ file://0020-configure.ac-setup.py-do-not-add-a-curses-include-pa.patch \ file://0001-Skip-failing-tests-due-to-load-variability-on-YP-AB.patch \