Re: [gentoo-user] Firefox crashes on some www-pages on a newer Gentoo system

[Mick <michaelkintzios@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 3005 bytes --]

On Friday, 27 July 2018 19:49:45 BST gevisz wrote:
> 2018-07-27 18:29 GMT+03:00 Mick <michaelkintzios@gmail.com>:
> > On Friday, 27 July 2018 14:00:43 BST gevisz wrote:
> >> 2018-07-27 15:44 GMT+03:00 gevisz <gevisz@gmail.com>:
> >> > Updating world without global -pam, -polkit and -consolekit use flags
> >> > in /etc/portage/make.conf have not changed the behaviour of Fireforx
> >> > described above: it is still cannot display the page
> >> > https://www.privat24.ua/ and shows "Gah. Your tab just crashed."
> >> > message.
> >> > 
> >> > So, I will probably revert changes tomorrow.
> >> > 
> >> > Playing with strace has been put on todo list.
> >> 
> >> Last lines of the log file after runnig the following command
> >> 
> >> $ strace -o strace.log qupzilla https://www.privat24.ua/
> >> QupZilla: 0 extensions loaded
> >> [15707:15725:0727/155144.939329:ERROR:nss_util.cc(808)] After loading
> >> Root Certs, loaded==false: NSS error code: -8018
> >> [15707:15719:0727/155145.900259:ERROR:nss_ocsp.cc(591)] No
> >> URLRequestContext for NSS HTTP handler. host: ocsp.digicert.com
> > 
> > Mozilla is using libnss to check the validity of the website certificate
> > with the OCSP of the CA issuer of the certificate (digicert.com).  For
> > some reason it seems the connection is interrupted.  See lines 605 to 621
> > here:
> > 
> > https://code.woboq.org/qt5/qtwebengine/src/3rdparty/chromium/net/cert_net/
> > nss_ocsp.cc.html
> > 
> > The connection may be interrupted because the browser/tab crashes (see
> > below).> 
> >> r300 FP: Compiler Error:
> >> /var/tmp/portage/media-libs/mesa-17.3.9/work/mesa-17.3.9/src/gallium/driv
> >> ers /r300/compiler/r300_fragprog_emit.c::translate_rgb_opcode():
> >> translate_rgb_opcode: Unknown opcode DDY
> >> Using a dummy shader instead.
> >> QupZilla: Crashed :( Saving backtrace in
> >> /home/gevis/.config/qupzilla/crashlog ...
> >> Backtrace successfully saved in
> >> /home/gevis/.config/qupzilla/crashlog/Crash-2018-07-27T15:51:49.txt
> >> zsh: abort      strace -o strace.log qupzilla https://www.privat24.ua/
> >> 
> >> are the following:
> >> 
> >> clock_gettime(CLOCK_MONOTONIC, {tv_sec=1167, tv_nsec=68524671}) = 0
> >> ioctl(8, DRM_IOCTL_RADEON_GEM_MMAP, 0x7ffd31773280) = 0
> >> mmap(NULL, 6868992, PROT_READ|PROT_WRITE, MAP_SHARED, 8, 0x107616000)
> >> = 0x7f22f5650000
> >> +++ killed by SIGABRT +++
> > 
> > This looks like a radeon video driver problem.  You could go into a loop
> > of
> > rebuilding xorg, mesa, dev-libs/nss, what-ever and see if things improve,
> > or you could wait for/keyword later versions of these packages.
> 
> Thank you for the tip. In this case, I should first try reconfigure the
> kernel.
> > Alternatively, you could try reporting this to the application devs and
> > see
> > what they say.
> 
> Do you mean developpers of FF/QZ?

Yes, although unless this is a known bug they may point you to the devs of the 
radeon drivers.  I've been around this loop a couple of times (sigh ...)
-- 
Regards,
Mick

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 833 bytes --]