|Summary:||app-text/ghostscript-gpl-9.54.0 - Fatal error: cant create obj/extract_alloc.o: No such file or directory (maybe parallel build issue ?)|
|Product:||Gentoo Linux||Reporter:||Toralf Förster <toralf>|
|Component:||Current packages||Assignee:||Codec Project <codec>|
|Severity:||normal||CC:||ionen, jamiegetty1, juippis, mva, printing, sam|
|Package list:||Runtime testing required:||---|
build.log + emerge --info.txt
Description Toralf Förster 2021-07-03 06:10:10 UTC
make -f Makefile so-subtarget BUILDDIRPREFIX=so Assembler messages: Fatal error: can't create obj/extract_alloc.o: No such file or directory make: *** [devices/extract.mak:5: obj/extract_alloc.o] Error 2 make: *** Waiting for unfinished jobs.... make: Entering directory '/var/tmp/portage/app-text/ghostscript-gpl-9.54.0/work/ghostscript-9.54.0' make -f Makefile GS_DOT_O= REALMAIN_OBJ= GS_XE=./sobin/libgs.so.9.54 GPCL_XE=./sobin/libno_gpcl6.so.9.54 GXPS_XE=./sobin/libno_gxps.so.9.54 GPDL_XE=./sobin/libno_gpdl.so.9.54 DISPLAY_DEV=./soobj/display.dev BUILDDIRPREFIX=so GENOPT='' LDFLAGS='-Wl,-O1 -Wl,--as-needed -Wl,--defsym=__gentoo_check_ldflags__=0 '\ ------------------------------------------------------------------- This is an unstable amd64 chroot image at a tinderbox (==build bot) name: 17.1-j3-20210702-091244 ------------------------------------------------------------------- gcc-config -l:  x86_64-pc-linux-gnu-11.1.0 * /usr/lib/llvm/12 12.0.0 Python 3.9.6 Available Ruby profiles:  ruby26 (with Rubygems) * Available Rust versions:  rust-bin-1.53.0 * The following VMs are available for generation-2: *) AdoptOpenJDK 8.292_p10 [openjdk-bin-8] Available Java Virtual Machines:  openjdk-bin-8 system-vm The Glorious Glasgow Haskell Compilation System, version 8.10.4 HEAD of ::gentoo commit 2eb48bc48857694cad1cf010eb3ecc8f332ee8f5 Author: Repository mirror & CI <email@example.com> Date: Fri Jul 2 22:36:27 2021 +0000 2021-07-02 22:36:25 UTC emerge -qpvO app-text/ghostscript-gpl [ebuild N ] app-text/ghostscript-gpl-9.54.0 USE="jpeg2k (unicode) -X -cups -dbus -gtk -static-libs" L10N="-de (-ja) -ko -zh-CN -zh-TW"
Comment 1 Toralf Förster 2021-07-03 06:10:11 UTC
Created attachment 721159 [details] emerge-info.txt
Comment 2 Toralf Förster 2021-07-03 06:10:12 UTC
Created attachment 721163 [details] app-text:ghostscript-gpl-9.54.0:20210703-005141.log.bz2
Comment 3 Toralf Förster 2021-07-03 06:10:14 UTC
Created attachment 721165 [details] emerge-history.txt
Comment 5 Toralf Förster 2021-07-03 06:10:16 UTC
Created attachment 721169 [details] etc.portage.tar.bz2
Comment 8 Sam James 2021-08-28 18:10:46 UTC
*** Bug 810295 has been marked as a duplicate of this bug. ***
Comment 9 Thomas Deutschmann (RETIRED) 2021-08-29 23:39:29 UTC
I cannot reproduce. Are you able to reproduce all the time? I wonder if this is a parallel build issue. Can you try with MAKEOPTS=-j1?
Comment 10 Toralf Förster 2021-08-30 20:24:45 UTC
(In reply to Thomas Deutschmann from comment #9) > I cannot reproduce. Are you able to reproduce all the time? I wonder if this > is a parallel build issue. Can you try with MAKEOPTS=-j1? It cannot be reproduced and builds fine here at various images with -j4. Will change the title.
Comment 11 Toralf Förster 2021-09-04 07:49:07 UTC
happened here again at ~/img/17.1_desktop_plasma_systemd-j4-20210903-200003 - so pbly a parallel build issue
Comment 12 Sam James 2021-09-13 00:50:20 UTC
*** Bug 809929 has been marked as a duplicate of this bug. ***
Comment 13 Joonas Niilola 2021-09-13 11:19:10 UTC
^ It seemed random to me too. I've managed to build 3 binpkgs with different USE flags so I haven't triggered this in a while.
Comment 14 Toralf Förster 2021-09-13 11:21:00 UTC
(In reply to Joonas Niilola from comment #13) > ^ It seemed random to me too. I've managed to build 3 binpkgs with different > USE flags so I haven't triggered this in a while. here it happened again yesterday at an image
Comment 15 Sam James 2021-09-15 06:03:01 UTC
I'm inclined to just force -j1 for now if nobody has any ideas and given a new release is coming up shortly. Could somebody report it upstream who can reproduce this?
Comment 16 Thomas Deutschmann (RETIRED) 2021-09-16 11:58:16 UTC
I hit this today by myself during world upgrade on my Gentoo notebook. Still unable to force it in dev environment.
Comment 17 Sam James 2021-09-18 04:20:39 UTC
(In reply to Thomas Deutschmann from comment #16) > I hit this today by myself during world upgrade on my Gentoo notebook. Still > unable to force it in dev environment. Do you think you could try the 9.55.0 RC (not in tree)?
Comment 18 Toralf Förster 2021-09-18 09:48:15 UTC
It is repeating but not 100% reproducible even at the same image. SO I do bet that the load of the system plays an important role. IMO a lack in the make process but hard to tell for me, where.
Comment 19 Rolf Eike Beer 2021-09-20 15:17:08 UTC
I have hit this twice on hppa during my tatt runs: USE='X -cups dbus -gtk jpeg2k l10n_de l10n_ko l10n_zh-CN -l10n_zh-TW -static-libs' failed for =app-text/ghostscript-gpl-9.54.0-r1 USE='X -cups dbus gtk -jpeg2k -l10n_de -l10n_ko -l10n_zh-CN l10n_zh-TW static-libs' failed for =app-text/ghostscript-gpl-9.54.0-r1
Comment 20 Sam James 2021-09-21 19:00:38 UTC
Speaking to upstream about this (they were extremely helpful). Got a candidate patch: https://git.ghostscript.com/?p=ghostpdl.git;a=commitdiff;h=4874418fd031617212336b8b55f8fcba3b9cfb68#patch1. Going to add it given it seems right and we want to move forward with stabling for the security bug. Will drop it if it causes issues.
Comment 21 Larry the Git Cow 2021-09-21 19:02:30 UTC
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=b7805f02f359138edf4384f2b7df317bfc27e3cb commit b7805f02f359138edf4384f2b7df317bfc27e3cb Author: Sam James <firstname.lastname@example.org> AuthorDate: 2021-09-21 19:01:50 +0000 Commit: Sam James <email@example.com> CommitDate: 2021-09-21 19:02:14 +0000 app-text/ghostscript-gpl: add parallel build patch to 9.54.0-r1 Bug: https://bugs.gentoo.org/800125 Signed-off-by: Sam James <firstname.lastname@example.org> .../ghostscript-gpl-9.54.0-makefile-deps.patch | 110 +++++++++++++++++++++ .../ghostscript-gpl-9.54.0-r1.ebuild | 2 + 2 files changed, 112 insertions(+)
Comment 22 Joonas Niilola 2021-09-22 16:15:02 UTC
Don't think I've ever been able to reproduce when emerging ghostscript-gpl alone, but when it's a dep amongst other deps being pulled and installed, then it historically has failed. Tried to break the installation few times today, tried under stress, pulling as a dep few times, straight up build it 20 times with pkg-testing-tools. But couldn't reproduce anymore.
Comment 23 Ionen Wolkens 2021-09-25 02:47:26 UTC
Created attachment 741176 [details] build.log + emerge --info.txt I've ran into this three times despite patch, and so did a forum user Three cases were on extract_docx_template.o, and one on extract_astring.o  https://forums.gentoo.org/viewtopic-t-1142930.html
Comment 24 Larry the Git Cow 2021-09-25 03:05:35 UTC
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=2cb90af63cbe0372a74eaee522915bf75fe63641 commit 2cb90af63cbe0372a74eaee522915bf75fe63641 Author: Sam James <email@example.com> AuthorDate: 2021-09-25 03:04:40 +0000 Commit: Sam James <firstname.lastname@example.org> CommitDate: 2021-09-25 03:04:40 +0000 app-text/ghostscript-gpl: force -j1 for now in 9.54.0 Pending investigation upstream but it's not obvious to users what to do. Especially important given the vulnerability fixed that users are able to build the new one. Bug: https://bugs.gentoo.org/800125 Signed-off-by: Sam James <email@example.com> app-text/ghostscript-gpl/ghostscript-gpl-9.54.0-r1.ebuild | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
Comment 25 Sam James 2021-09-26 03:38:51 UTC
*** Bug 814917 has been marked as a duplicate of this bug. ***
Comment 26 Larry the Git Cow 2021-09-28 02:50:18 UTC
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=655a62e570746ba83e8fbadd977d58ae608984ba commit 655a62e570746ba83e8fbadd977d58ae608984ba Author: Sam James <firstname.lastname@example.org> AuthorDate: 2021-09-28 02:49:56 +0000 Commit: Sam James <email@example.com> CommitDate: 2021-09-28 02:49:56 +0000 app-text/ghostscript-gpl: add more upstream parallel build patches to 94.0-r1 Bug: https://bugs.gentoo.org/800125 Signed-off-by: Sam James <firstname.lastname@example.org> .../ghostscript-gpl-9.54.0-makefile-deps-2.patch | 96 ++++++++++++++++++++++ .../ghostscript-gpl-9.54.0-makefile-deps.patch | 2 + .../ghostscript-gpl-9.54.0-r1.ebuild | 7 +- 3 files changed, 101 insertions(+), 4 deletions(-)
Comment 27 Ionen Wolkens 2021-09-28 06:34:16 UTC
I gave it the same stress test as last time for an extended period, and I can't get it to fail anymore. Fairly confident it's fixed. My thanks to upstream for for the patches.