[opensuse] sig 7 bus error
Hi again, all -- This time I do have a real problem. I shut down my laptop davidtg@wench:~> uname -a ; cat /etc/os-release Linux wench 4.4.140-62-default #1 SMP Tue Jul 17 12:14:34 UTC 2018 (951943d) x86_64 x86_64 x86_64 GNU/Linux NAME="openSUSE Leap" VERSION="42.3" ID=opensuse ID_LIKE="suse" VERSION_ID="42.3" PRETTY_NAME="openSUSE Leap 42.3" ANSI_COLOR="0;32" CPE_NAME="cpe:/o:opensuse:leap:42.3" BUG_REPORT_URL="https://bugs.opensuse.org" HOME_URL="https://www.opensuse.org/" to bring it home, and when I brought it back up I suddenly couldn't run Chrome (Chromium); it pops up its window but then immediately dies. Here's what it throws: davidtg@wench:~> /usr/bin/chromium --disable-gpu --user-data-dir=/mnt/data/home/davidtg/.chromium/profiles/gift [5387:5387:1004/222017.561228:ERROR:gpu_process_transport_factory.cc(1016)] Lost UI shared context. libva info: VA-API version 0.39.4 libva info: va_getDriverName() returns 0 libva info: Trying to open /usr/lib64/dri/i965_drv_video.so libva info: va_openDriver() returns -1 [5471:5471:1004/222017.875318:ERROR:vaapi_wrapper.cc(323)] vaInitialize failed: unknown libva error Received signal 7 BUS_ADRERR 7f4faef5c1bc #0 0x556ce6714fae <unknown> #1 0x556ce41f78a0 <unknown> #2 0x556ce6715367 <unknown> #3 0x7f4fc5203c10 <unknown> #4 0x556ce6a23f9a <unknown> #5 0x556ce6a2423c <unknown> #6 0x556ce6a1eb51 <unknown> #7 0x556ce6a1ed17 <unknown> #8 0x556ce6a1f6da <unknown> #9 0x556ce69e2275 <unknown> #10 0x556ce67846ec <unknown> #11 0x556ce6784859 <unknown> #12 0x556ce6785423 <unknown> #13 0x556ce7d28569 <unknown> #14 0x556ce7d419d0 <unknown> #15 0x556ce7d41f6c <unknown> #16 0x556ce9b89991 <unknown> #17 0x556ce668e448 <unknown> #18 0x556ce66e5afd <unknown> #19 0x556ce6721ff2 <unknown> #20 0x556ce66e769e <unknown> #21 0x556ce67373a4 <unknown> #22 0x556ce6737904 <unknown> #23 0x556ce6722621 <unknown> #24 0x7f4fc51fb724 start_thread #25 0x7f4fbc645e8d __clone r8: 00007f4f9269ccc8 r9: 0000556ceb968721 r10: 0000556ce6a231f0 r11: 0000000000000014 r12: 000008e930b9c928 r13: 00007f4f9269cd80 r14: 00007f4f9269cec4 r15: 0000556ceb90ef38 di: 000008e930b9c928 si: 0000000000000002 bp: 00007f4f9269ccb0 bx: 000008e930b9c928 dx: 00007f4faef5a600 ax: 00007f4faef5c1bc cx: 0000556ceb968721 sp: 00007f4f9269cc50 ip: 0000556ce6a23f9a efl: 0000000000010202 cgf: 0000000000000033 erf: 0000000000000004 trp: 000000000000000e msk: 0000000000000000 cr2: 00007f4faef5c1bc [end of stack trace] Calling _exit(1). Core file will not be generated. Firefox gives me a much shorter error with similar disappearing-window results: davidtg@wench:~> /usr/bin/firefox -p gift ExceptionHandler::GenerateDump cloned child 6233 ExceptionHandler::SendContinueSignalToChild sent continue signal to child ExceptionHandler::WaitForContinueSignal waiting for continue signal... Bus error (core dumped) I've actually bounced the box a few times with no better luck. I know that I tried something else that failed, but I don't recall just what it was. Dolphin works, though, and shotcut will view a video file, so it isn't simply all graphics stuff. Um, what did I miss here?!? TIA & HANN :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Hello, On Fri, 04 Oct 2019, David T-G wrote:
This time I do have a real problem. I shut down my laptop [..] davidtg@wench:~> /usr/bin/chromium --disable-gpu --user-data-dir=/mnt/data/home/davidtg/.chromium/profiles/gift [5387:5387:1004/222017.561228:ERROR:gpu_process_transport_factory.cc(1016)] Lost UI shared context. libva info: VA-API version 0.39.4 libva info: va_getDriverName() returns 0 libva info: Trying to open /usr/lib64/dri/i965_drv_video.so libva info: va_openDriver() returns -1 [5471:5471:1004/222017.875318:ERROR:vaapi_wrapper.cc(323)] vaInitialize failed: unknown libva error
Hm. Despite your --disable-gpu it tries to init libva... Anyway:
Received signal 7 BUS_ADRERR 7f4faef5c1bc [..] davidtg@wench:~> /usr/bin/firefox -p gift [..] Bus error (core dumped)
My first bet is on faulty RAM. But try to roll back the latest updates concerning the kernel or your GPU (Mesa, radeon/fglrx, libva, etc. pp.). And/or run memtest... Anyway, SIGBUS usually is either a faulty driver or failing hardware... HTH, -dnh -- "Oh, NT is reliable. You can count on it to keel over under just any circumstance." -- Rik Steenwinkel -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
David, et al -- ...and then David Haller said... % % Hello, Good morning again! % % On Fri, 04 Oct 2019, David T-G wrote: % >This time I do have a real problem. I shut down my laptop % [..] % > davidtg@wench:~> /usr/bin/chromium --disable-gpu --user-data-dir=/mnt/data/home/davidtg/.chromium/profiles/gift % > [5387:5387:1004/222017.561228:ERROR:gpu_process_transport_factory.cc(1016)] Lost UI shared context. % > libva info: VA-API version 0.39.4 % > libva info: va_getDriverName() returns 0 % > libva info: Trying to open /usr/lib64/dri/i965_drv_video.so % > libva info: va_openDriver() returns -1 % > [5471:5471:1004/222017.875318:ERROR:vaapi_wrapper.cc(323)] vaInitialize failed: unknown libva error % % Hm. Despite your --disable-gpu it tries to init libva... Anyway: I tripped over that after getting the s3tc error earlier, but that wasn't enough of a fix. % % > Received signal 7 BUS_ADRERR 7f4faef5c1bc % [..] % > davidtg@wench:~> /usr/bin/firefox -p gift % [..] % > Bus error (core dumped) % % My first bet is on faulty RAM. But try to roll back the latest updates Ewwwwwwww... OK; I'll check that. % concerning the kernel or your GPU (Mesa, radeon/fglrx, libva, etc. pp.). *sigh* That means chasing down what updates I *have* had, which is always a pain. % % And/or run memtest... % % Anyway, SIGBUS usually is either a faulty driver or failing hardware... Thanks. MEMTEST is a good place to start, too. % % HTH, % -dnh HAND again :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Hello, On Sat, 05 Oct 2019, David T-G wrote:
% My first bet is on faulty RAM. But try to roll back the latest updates
Ewwwwwwww... OK; I'll check that.
% concerning the kernel or your GPU (Mesa, radeon/fglrx, libva, etc. pp.).
*sigh* That means chasing down what updates I *have* had, which is always a pain.
Just have a look at $ rpm -qa --last | less and roll back the last updates (concerning kernel and or video i.e. mesa, radeon/fglrx, nvidia, libdrm etc.pp.)... Considering how long memtest should run to be of any significance (and in the meantime, you can't use the box), and how long it'd take to roll back on some packages, I'd first try the latter. Or you might just try reencoding some video (~30min+ in duration), not sure about x264, but the mpeg4 stuff in mencoder seems to be quite a good memtest, and one which you can have running while doing whatever besides, it'll just crash _that mencoder process_ at some point, usually. Good parameters are mencoder -oac copy -ovc lavc -lavcopts \ vcodec=mpeg4:trell:vqmin=1:vqmax=12:mbd=2:cmp=3:vqscale=3:vbitrate=24000000:vb_strategy=1:b_sensitivity=4:threads=1 \ -o foo.avi input.whatever (or use '-nosound' instead of '-oac copy') Run that, and if it's bad RAM, mencoder should crash the sooner, the more instances or loaded the RAM is, i.e. run it once on each (real) core you have ;) At least that was the call to mencoder that _reproducibly_ triggered errors (and mencoder aborting/crashing) on my box with faulty RAM. Albeit, that was with IIRC just 2GiB RAM and 2 cores... The upside is, you can just keep that running while doing whatever you do, and if it crashes, esp. on MPEG2/DVD input, repeatedly, it's likely bad RAM ;) Oh, and you can even "nice" the mencoder instances, so you can almost work on as usual. Ok, you should NOT do important stuff, the box might crash hard after all, ISTR I had that a couple of times... HTH, -dnh PS: I once had bad RAM. memtest did not detect anything wrong in one run (not enough, I know, but it was some hours). But I could crash mencoder reproducibly. Either with a SIGBUS or a SEGFAULT. Esp. when running two instances in parallel (on two cores). Got new RAM (on warranty) and all was fine after that. -- "`Ford, you're turning into a penguin. Stop it.'" -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
David, et al -- ...and then David Haller said... % % Hello, % % On Sat, 05 Oct 2019, David T-G wrote: % >% My first bet is on faulty RAM. But try to roll back the latest updates % > % >Ewwwwwwww... OK; I'll check that. % > % >% concerning the kernel or your GPU (Mesa, radeon/fglrx, libva, etc. pp.). % > % >*sigh* That means chasing down what updates I *have* had, which is % >always a pain. % % Just have a look at % % $ rpm -qa --last | less % % and roll back the last updates (concerning kernel and or video i.e. % mesa, radeon/fglrx, nvidia, libdrm etc.pp.)... Thanks! I'll check that when I get back on. % % Considering how long memtest should run to be of any significance (and % in the meantime, you can't use the box), and how long it'd take to % roll back on some packages, I'd first try the latter. [snip] I actually had some trouble grabbing a good memtest for YUMI to install, so I just let the Dell BIOS run its memory checks for about 20 minutes. All came back clean, so as I reboot any moment now I'll check software as you suggest. Thanks again & HANW :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On Sat, 5 Oct 2019 13:38:10 +0200 David Haller <dnh@opensuse.org> wrote:
Hello,
On Sat, 05 Oct 2019, David T-G wrote:
% My first bet is on faulty RAM. But try to roll back the latest updates
Ewwwwwwww... OK; I'll check that.
% concerning the kernel or your GPU (Mesa, radeon/fglrx, libva, etc. pp.).
*sigh* That means chasing down what updates I *have* had, which is always a pain.
Just have a look at
$ rpm -qa --last | less
and roll back the last updates (concerning kernel and or video i.e. mesa, radeon/fglrx, nvidia, libdrm etc.pp.)...
Considering how long memtest should run to be of any significance (and in the meantime, you can't use the box), and how long it'd take to roll back on some packages, I'd first try the latter.
Or you might just try reencoding some video (~30min+ in duration), not sure about x264, but the mpeg4 stuff in mencoder seems to be quite a good memtest, and one which you can have running while doing whatever besides, it'll just crash _that mencoder process_ at some point, usually. Good parameters are
mencoder -oac copy -ovc lavc -lavcopts \ vcodec=mpeg4:trell:vqmin=1:vqmax=12:mbd=2:cmp=3:vqscale=3:vbitrate=24000000:vb_strategy=1:b_sensitivity=4:threads=1 \ -o foo.avi input.whatever
(or use '-nosound' instead of '-oac copy')
Run that, and if it's bad RAM, mencoder should crash the sooner, the more instances or loaded the RAM is, i.e. run it once on each (real) core you have ;) At least that was the call to mencoder that _reproducibly_ triggered errors (and mencoder aborting/crashing) on my box with faulty RAM. Albeit, that was with IIRC just 2GiB RAM and 2 cores... The upside is, you can just keep that running while doing whatever you do, and if it crashes, esp. on MPEG2/DVD input, repeatedly, it's likely bad RAM ;)
Oh, and you can even "nice" the mencoder instances, so you can almost work on as usual. Ok, you should NOT do important stuff, the box might crash hard after all, ISTR I had that a couple of times...
HTH, -dnh
PS: I once had bad RAM. memtest did not detect anything wrong in one run (not enough, I know, but it was some hours). But I could crash mencoder reproducibly. Either with a SIGBUS or a SEGFAULT. Esp. when running two instances in parallel (on two cores). Got new RAM (on warranty) and all was fine after that.
Given that he's running 42.3, have there been any updates recently? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 10/05/2019 02:44 PM, Dave Howorth wrote:
Given that he's running 42.3, have there been any updates recently?
Nothing that would provide commonality between firefox and chromium that I can think of. I still have 42.3 on this box, and some repos still update, (Mozilla, etc..), but I haven't noticed any issues. -- David C. Rankin, J.D.,P.E. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
David*, et al -- ...and then David Haller said... % % Just have a look at % % $ rpm -qa --last | less Thanks again! I do see some libraries, but don't know how much they matter: davidtg@wench:~> rpm -qa --last | head -15 mkvtoolnix-35.0.0-1.1.x86_64 Mon Jul 15 21:38:24 2019 vlc-qt-2.2.6-5.1.x86_64 Mon Jul 15 21:38:23 2019 vlc-2.2.6-5.1.x86_64 Mon Jul 15 21:38:23 2019 vlc-noX-2.2.6-5.1.x86_64 Mon Jul 15 21:38:22 2019 libvlc5-2.2.6-5.1.x86_64 Mon Jul 15 21:38:21 2019 libmatroska6-1.5.0-1.3.x86_64 Mon Jul 15 21:38:21 2019 libvlccore8-2.2.6-5.1.x86_64 Mon Jul 15 21:38:20 2019 libpugixml1-1.9-3.3.x86_64 Mon Jul 15 21:38:20 2019 libfmt5-5.3.0-2.3.x86_64 Mon Jul 15 21:38:20 2019 libboost_filesystem1_69_0-1.69.0-1.5.x86_64 Mon Jul 15 21:38:20 2019 libstdc++6-9.1.1+r272147-1.1.x86_64 Mon Jul 15 21:38:19 2019 boost-license1_69_0-1.69.0-1.5.noarch Mon Jul 15 21:38:19 2019 gpg-pubkey-4b6baa9c-5c733f48 Wed Mar 20 22:14:05 2019 gpg-pubkey-324e6311-5b999b6d Fri Sep 14 10:03:38 2018 fuse-exfat-1.2.7-3.1.x86_64 Thu Sep 6 07:18:03 2018 % % and roll back the last updates (concerning kernel and or video i.e. % mesa, radeon/fglrx, nvidia, libdrm etc.pp.)... [snip] Maybe libmatroska? And none of those are terribly recent, but I'm trying to recall when I last booted and it might well not have been for a few weeks, which makes them likely culprits. Thanks again & HAND :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
On 10/04/2019 09:43 PM, David Haller wrote:
My first bet is on faulty RAM. But try to roll back the latest updates concerning the kernel or your GPU (Mesa, radeon/fglrx, libva, etc. pp.).
+1, I'd reseat the RAM and make sure all your fans are working. Depending on the age, you may be looking at a capacitor issue. If it is a desktop, you can look at the capacitors, if any look like a "puffy beer can" or a "soda you left in the freezer too long", you've got the likely culprit. -- David C. Rankin, J.D.,P.E. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
David & David, et al -- ...and then David C. Rankin said... % % On 10/04/2019 09:43 PM, David Haller wrote: % > My first bet is on faulty RAM. But try to roll back the latest updates % > concerning the kernel or your GPU (Mesa, radeon/fglrx, libva, etc. pp.). % % +1, % % I'd reseat the RAM and make sure all your fans are working. Depending on the Done (and swapped, even) and done. All seems good so far, except that the error stays the same. % age, you may be looking at a capacitor issue. If it is a desktop, you can look % at the capacitors, if any look like a "puffy beer can" or a "soda you left in % the freezer too long", you've got the likely culprit. Didn't see any of those in the brief trip I took around the laptop (Dell E6540) mobo, but it was *really* brief. I'll check software and repos next, or maybe even just pop last year's pre-SSD drive back in and see if it is happy (if I can wait long enough for it to boot ;-) % % -- % David C. Rankin, J.D.,P.E. Thanks again & stay tuned :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
Hi, all -- ...and then David T-G said... % ... % davidtg@wench:~> /usr/bin/chromium --disable-gpu --user-data-dir=/mnt/data/home/davidtg/.chromium/profiles/gift ... % #25 0x7f4fbc645e8d __clone % r8: 00007f4f9269ccc8 r9: 0000556ceb968721 r10: 0000556ce6a231f0 r11: 0000000000000014 % r12: 000008e930b9c928 r13: 00007f4f9269cd80 r14: 00007f4f9269cec4 r15: 0000556ceb90ef38 % di: 000008e930b9c928 si: 0000000000000002 bp: 00007f4f9269ccb0 bx: 000008e930b9c928 % dx: 00007f4faef5a600 ax: 00007f4faef5c1bc cx: 0000556ceb968721 sp: 00007f4f9269cc50 % ip: 0000556ce6a23f9a efl: 0000000000010202 cgf: 0000000000000033 erf: 0000000000000004 % trp: 000000000000000e msk: 0000000000000000 cr2: 00007f4faef5c1bc % [end of stack trace] % Calling _exit(1). Core file will not be generated. % % Firefox gives me a much shorter error with similar disappearing-window results: % % davidtg@wench:~> /usr/bin/firefox -p gift % ExceptionHandler::GenerateDump cloned child 6233 % ExceptionHandler::SendContinueSignalToChild sent continue signal to child % ExceptionHandler::WaitForContinueSignal waiting for continue signal... % Bus error (core dumped) [snip] This one fixed itself the other day, by the way... I popped out and reseated both RAM modules with no improvement, ran memory diagnostics with no signs of error, and was just as stuck until I bounced it a few more times and suddenly Chrome was fine again. Then I was busy for the past few days and haven't been able to follow up until now when something else fun decides to happen. Urgh. So, no, I don't know what went wrong here nor how I fixed it :-/ :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
ARRRRGH! ...and then David T-G said... % ... % with no signs of error, and was just as stuck until I bounced it a few % more times and suddenly Chrome was fine again. Then I was busy for the % past few days and haven't been able to follow up until now when something [snip] AND NOW CHROME IS BROKEN AGAIN!!! *sigh* :-D -- David T-G See http://justpickone.org/davidtg/email/ See http://justpickone.org/davidtg/tofu.txt -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org
participants (4)
-
Dave Howorth -
David C. Rankin -
David Haller -
David T-G