Bug#1012548: libwebkit2gtk-4.1-0: Offline renderer SIGSEGV on i386

2022-06-17 Thread karogyoker
Package: libwebkit2gtk-4.1-0 Version: 2.36.3-1 Followup-For: Bug #1012548 X-Debbugs-Cc: karogyoker2+deb...@gmail.com Control: tags -1 patch upstream I think not many crashes surfaced because not many people are using real 32 bit CPUs meanwhile visiting websites with Web Audio API. All 32 bit

Bug#1012548: libwebkit2gtk-4.1-0: Offline renderer SIGSEGV on i386

2022-06-14 Thread Alberto Garcia
reopen 1012548 tags 1012548 - unreproducible wontfix thanks On Mon, Jun 13, 2022 at 11:00:03AM +0200, karogyoker wrote: > I think the solution is not to use the code where the ORing happens > with 0x8040 (in case of i386 releases because this combination is > not supported by hardware). For 32

Bug#1012548: libwebkit2gtk-4.1-0: Offline renderer SIGSEGV on i386

2022-06-13 Thread karogyoker
Package: libwebkit2gtk-4.1-0 Version: 2.36.3-1 Followup-For: Bug #1012548 X-Debbugs-Cc: karogyoker2+deb...@gmail.com Dear Maintainer, Please reopen this case because it turned out that it is not a hardware failure at all! I figured out everything about why epiphany is crashing for me, so this

Bug#1012548: libwebkit2gtk-4.1-0: Offline renderer SIGSEGV on i386

2022-06-10 Thread karogyoker
Package: libwebkit2gtk-4.1-0 Followup-For: Bug #1012548 X-Debbugs-Cc: karogyoker2+deb...@gmail.com Dear Maintainer, Based on the stack trace I've found the problematic line on GitHub. It is calling ldmxcsr in an inline asm block. It does something with Web Audio API. I checked the history for

Bug#1012548: libwebkit2gtk-4.1-0: Offline renderer SIGSEGV on i386

2022-06-10 Thread karogyoker
Package: libwebkit2gtk-4.1-0 Version: 2.36.3-1 Followup-For: Bug #1012548 X-Debbugs-Cc: karogyoker2+deb...@gmail.com Thanks. Yes, it was very easy to get the stack trace this way. PID: 1626 (WebKitWebProces) UID: 1000 (user) GID: 1000 (user) Signal: 11

Bug#1012548: libwebkit2gtk-4.1-0: Offline renderer SIGSEGV on i386

2022-06-10 Thread Alberto Garcia
On Fri, Jun 10, 2022 at 10:10:36AM +0200, karogyoker wrote: > I've installed the libwebkit2gtk-4.1-0-dbgsym and > libjavascriptcoregtk-4.1-0-dbgsym packages. > After I attach to the PID exactly how I was doing it before, the > WebKitWebProcess process stops doing anything and the page never loads.

Bug#1012548: libwebkit2gtk-4.1-0: Offline renderer SIGSEGV on i386

2022-06-10 Thread karogyoker
Package: libwebkit2gtk-4.1-0 Version: 2.36.3-1 Followup-For: Bug #1012548 X-Debbugs-Cc: karogyoker2+deb...@gmail.com I've installed the libwebkit2gtk-4.1-0-dbgsym and libjavascriptcoregtk-4.1-0-dbgsym packages. After I attach to the PID exactly how I was doing it before, the WebKitWebProcess

Bug#1012548: libwebkit2gtk-4.1-0: Offline renderer SIGSEGV on i386

2022-06-09 Thread Alberto Garcia
On Thu, Jun 09, 2022 at 01:59:25AM +0200, karogyoker wrote: > The problem is reproducible if I try to open this webpage in epiphany-browser: > https://www.microsoft.com/en-us/software-download/windows10ISO I cannot reproduce the problem with QEMU emulating a pentium 3 CPU. We have had problems

Bug#1012548: libwebkit2gtk-4.1-0: Offline renderer SIGSEGV on i386

2022-06-08 Thread karogyoker
Package: libwebkit2gtk-4.1-0 Version: 2.36.3-1 Severity: normal X-Debbugs-Cc: karogyoker2+deb...@gmail.com Dear Maintainer, The problem is reproducible if I try to open this webpage in epiphany-browser: https://www.microsoft.com/en-us/software-download/windows10ISO The problem is that Gnome Web