PM crashes on large-ish select-all

Users and developers helping users with generic and technical Pale Moon issues on all operating systems.

Moderator: trava90

Forum rules
This board is for technical/general usage questions and troubleshooting for the Pale Moon browser only.
Technical issues and questions not related to the Pale Moon browser should be posted in other boards!
Please keep off-topic and general discussion out of this board, thank you!
User avatar
loxodont
Astronaut
Astronaut
Posts: 725
Joined: 2014-07-26, 23:03
Location: Mare Serenitatis

Re: PM crashes on large-ish select-all

Unread post by loxodont » 2020-02-04, 03:18

I'm using the official PM 28.8.2 for Linux releases on Xubuntu 18.04 x32 and TahrPup 6.0.5 x64 (USB drive)
Important: On a second test to select all 1000 and then 2000 results of that page, no crash happened on TahrPup!
The page took some time to load and scrolling down was very slow first, but after a few seconds scrolling was smooth, even with 2000 results selected. I guess the first crash occurred because "Select-All" had been hit before Pale Moon finished loading the page completely.

The Xubuntu sysinfo and about:buildcofig for xubuntu 18.04 (bionic) x32:

Code: Select all

//System information report, generated by Sysinfo: 04/02/2020 02:50:34
SYSTEM INFORMATION
	Running Ubuntu Linux, the Ubuntu 18.04 (bionic) release.
	GNOME: 3.28.2 (Ubuntu)
	Kernel version: 4.15.0-76-generic (#86-Ubuntu SMP Fri Jan 17 17:25:21 UTC 2020)
	GCC: 7 (i686-linux-gnu)
	Xorg: 1.19.6 (03 June 2019  08:10:35AM) (03 June 2019  08:10:35AM)//

// PM 28.8.2//
about:buildconfig
Build platform
target
i686-pc-linux-gnu
Build tools
Compiler 	Version 	Compiler flags
/opt/rh/devtoolset-3/root/usr/bin/gcc -std=gnu99 	4.9.4 	-Wall -Wempty-body -Wignored-qualifiers -Wpointer-arith -Wsign-compare -Wtype-limits -Wunreachable-code -Wno-error=maybe-uninitialized -Wno-error=deprecated-declarations -Wno-error=array-bounds -fno-strict-aliasing -ffunction-sections -fdata-sections -fno-math-errno -msse2 -mfpmath=sse -pthread -pipe
/opt/rh/devtoolset-3/root/usr/bin/g++ -std=gnu++11 	4.9.4 	-Wall -Wc++11-compat -Wempty-body -Wignored-qualifiers -Woverloaded-virtual -Wpointer-arith -Wsign-compare -Wtype-limits -Wunreachable-code -Wwrite-strings -Wno-invalid-offsetof -Wno-error=maybe-uninitialized -Wno-error=deprecated-declarations -Wno-error=array-bounds -fno-exceptions -fno-strict-aliasing -fno-rtti -ffunction-sections -fdata-sections -fno-exceptions -fno-math-errno -msse2 -mfpmath=sse -pthread -pipe -O2 -msse2 -mfpmath=sse -fomit-frame-pointer
Configure options

--enable-application=palemoon --enable-update-channel=release --disable-debug-symbols --enable-jemalloc --enable-default-toolkit=cairo-gtk2 --enable-av1 MAKE=/usr/bin/gmake --disable-accessibility --disable-eme --disable-necko-wifi --enable-official-branding --enable-official-vendor '--enable-optimize=-O2 -msse2 -mfpmath=sse' --enable-strip --disable-tests --enable-updater --with-pthreads --x-libraries=/usr/lib
And for TahrPup 6.0.5 x64

Code: Select all

about:buildconfig
Build platform
target
x86_64-pc-linux-gnu
Build tools
Compiler 	Version 	Compiler flags
/opt/rh/devtoolset-7/root/usr/bin/gcc -std=gnu99 	7.3.1 	-Wall -Wempty-body -Wignored-qualifiers -Wpointer-arith -Wsign-compare -Wtype-limits -Wunreachable-code -Wno-error=maybe-uninitialized -Wno-error=deprecated-declarations -Wno-error=array-bounds -flifetime-dse=1 -fno-strict-aliasing -ffunction-sections -fdata-sections -fno-math-errno -msse2 -mfpmath=sse -pthread -pipe
/opt/rh/devtoolset-7/root/usr/bin/g++ -std=gnu++11 	7.3.1 	-Wall -Wc++11-compat -Wempty-body -Wignored-qualifiers -Woverloaded-virtual -Wpointer-arith -Wsign-compare -Wtype-limits -Wunreachable-code -Wwrite-strings -Wno-invalid-offsetof -Wc++14-compat -Wc++1z-compat -Wimplicit-fallthrough -Wno-error=maybe-uninitialized -Wno-error=deprecated-declarations -Wno-error=array-bounds -flifetime-dse=1 -fno-exceptions -fno-strict-aliasing -fno-rtti -ffunction-sections -fdata-sections -fno-exceptions -fno-math-errno -msse2 -mfpmath=sse -pthread -pipe -O2 -msse2 -mfpmath=sse -fomit-frame-pointer
Configure options

--enable-application=palemoon --enable-update-channel=release --disable-debug-symbols --enable-jemalloc --enable-default-toolkit=cairo-gtk2 --enable-av1 MAKE=/usr/bin/gmake --disable-accessibility --disable-eme --disable-necko-wifi --enable-official-branding --enable-official-vendor '--enable-optimize=-O2 -msse2 -mfpmath=sse' --enable-strip --disable-tests --enable-updater --with-pthreads --x-libraries=/usr/lib64

arzach

Re: PM crashes on large-ish select-all

Unread post by arzach » 2020-02-04, 03:19

Apologies, but I'm traveling all week and won't be back to my offending machines until sometime Friday. I'll post the about:buildconfig then. I don't think I know how to determine which theme engine is installed.

FWIW, PM in pclinuxos on the laptop I'm traveling with does NOT have the crash problem which seems to suggest it is some kind of configuration issue. The offending machines are pretty plain vanilla pclinuxos. The pclinuxos on the laptop I'm traveling with has been tweaked, though I don't know the details.

arzach

Re: PM crashes on large-ish select-all

Unread post by arzach » 2020-02-04, 03:28

As loxodont observed, I was able to crash PM by select-all before the page's data had completely downloaded. At the time of my original post that was one of the possibilities I had considered and was careful to verify that the crash took place AFTER the entire page was completely downloaded.

arzach

Re: PM crashes on large-ish select-all

Unread post by arzach » 2020-02-04, 03:44

Earlier I said that the laptop I'm traveling with did not suffer the crash problem...but now it does! When I tested it earlier tonight I had 2000 titles selected and was successful in select-all. About 5 minutes later I repeated the experiment and immediately crashed. Subsequent tests found immediate crashes at 1000 and 2000 titles/page on this machine.

About:buildconfig gives this

Code: Select all

about:buildconfig
Build platform
target
x86_64-pc-linux-gnu
Build tools
Compiler 	Version 	Compiler flags
/opt/rh/devtoolset-7/root/usr/bin/gcc -std=gnu99 	7.3.1 	-Wall -Wempty-body -Wignored-qualifiers -Wpointer-arith -Wsign-compare -Wtype-limits -Wunreachable-code -Wno-error=maybe-uninitialized -Wno-error=deprecated-declarations -Wno-error=array-bounds -flifetime-dse=1 -fno-strict-aliasing -fno-math-errno -msse2 -mfpmath=sse -pthread -pipe
/opt/rh/devtoolset-7/root/usr/bin/g++ -std=gnu++11 	7.3.1 	-Wall -Wc++11-compat -Wempty-body -Wignored-qualifiers -Woverloaded-virtual -Wpointer-arith -Wsign-compare -Wtype-limits -Wunreachable-code -Wwrite-strings -Wno-invalid-offsetof -Wc++14-compat -Wc++1z-compat -Wimplicit-fallthrough -Wno-error=maybe-uninitialized -Wno-error=deprecated-declarations -Wno-error=array-bounds -flifetime-dse=1 -fno-exceptions -fno-strict-aliasing -fno-rtti -fno-exceptions -fno-math-errno -msse2 -mfpmath=sse -pthread -pipe -O2 -msse2 -mfpmath=sse -fomit-frame-pointer
Configure options

--enable-application=platform/application/palemoon --with-external-source-dir=/home/mattatobin/development/binoc-central --enable-update-channel=release --disable-debug-symbols --enable-jemalloc --enable-default-toolkit=cairo-gtk2 --enable-av1 MAKE=/opt/rh/devtoolset-7/root/usr/bin/gmake --disable-accessibility --disable-eme --disable-necko-wifi --enable-official-branding --enable-official-vendor '--enable-optimize=-O2 -msse2 -mfpmath=sse' --enable-strip --disable-tests --enable-updater --with-pthreads --x-libraries=/usr/lib64

User avatar
Moonchild
Pale Moon guru
Pale Moon guru
Posts: 35481
Joined: 2011-08-28, 17:27
Location: Motala, SE
Contact:

Re: PM crashes on large-ish select-all

Unread post by Moonchild » 2020-02-04, 09:56

OK so it's an intermittent issue then.
In that case getting a stack trace of the crash would be most helpful. Note: NOT a program execution trace, just a call stack trace of the crash which a debugger should be able to catch.

I'll try again to see if I can entice a crash on Windows.

EDIT: I'm completely unable to make the browser crash no matter what I do on my system. Since HWA was blocked (I'm assuming on all involved systems) I tried with it both off and on, and can't make it crash no matter how often/quickly I make the selection after hitting a load request. So getting a trace or crash dump is the only way forward here to find what's causing this.
"Sometimes, the best way to get what you want is to be a good person." -- Louis Rossmann
"Seek wisdom, not knowledge. Knowledge is of the past; wisdom is of the future." -- Native American proverb
"Linux makes everything difficult." -- Lyceus Anubite

arzach

Re: PM crashes on large-ish select-all

Unread post by arzach » 2020-02-04, 15:57

How can I capture a call stack trace?

A colleague's PM 26.2.0 on a 32-bit installation of Pclinuxos does not crash on 1000 or 2000 titles/page from ArXiv

New Tobin Paradigm

Re: PM crashes on large-ish select-all

Unread post by New Tobin Paradigm » 2020-02-04, 16:12

Are you kidding me?!

User avatar
Moonchild
Pale Moon guru
Pale Moon guru
Posts: 35481
Joined: 2011-08-28, 17:27
Location: Motala, SE
Contact:

Re: PM crashes on large-ish select-all

Unread post by Moonchild » 2020-02-04, 18:18

arzach wrote:
2020-02-04, 15:57
How can I capture a call stack trace?
A Linux guru will have to help you with that.
New Tobin Paradigm wrote:
2020-02-04, 16:12
Are you kidding me?!
To provide some context here for Tobin: Why on earth is your colleague running such an ancient version?
Also: you can't at all compare the two. That's 2 milestones behind where we are now and the browser's internals were vastly different back then. So it's totally irrelevant that that ancient version doesn't happen to crash.
"Sometimes, the best way to get what you want is to be a good person." -- Louis Rossmann
"Seek wisdom, not knowledge. Knowledge is of the past; wisdom is of the future." -- Native American proverb
"Linux makes everything difficult." -- Lyceus Anubite

vannilla
Moon Magic practitioner
Moon Magic practitioner
Posts: 2183
Joined: 2018-05-05, 13:29

Re: PM crashes on large-ish select-all

Unread post by vannilla » 2020-02-04, 18:24

Aside for any stack traces, maybe it could be useful to also provide the version of the X server the crashing system is used.
Just run "X -version" in a terminal and copy the output.
I think it's Linux-specific and it can't happen on Windows due to the different graphics system.

arzach

Re: PM crashes on large-ish select-all

Unread post by arzach » 2020-02-06, 23:05

X -version reports
X.Org X Server 1.20.7

28.8.2 crash after started from command line says
[XXXX@localhost ~]$ palemoon
[11462] ###!!! ABORT: X_ChangeProperty: BadWindow (invalid Window parameter): file /home/PM4Linux/REPO/UXP/toolkit/xre/nsX11ErrorHandler.cpp, line 134
[11462] ###!!! ABORT: X_ChangeProperty: BadWindow (invalid Window parameter): file /home/PM4Linux/REPO/UXP/toolkit/xre/nsX11ErrorHandler.cpp, line 134
Segmentation fault

It's safe to assume that my colleague's antique equipment isn't a matter of personal choice. We are constrained by department equipment budgets, personal finances, and the enormous salaries that administrators pay themselves.

User avatar
Moonchild
Pale Moon guru
Pale Moon guru
Posts: 35481
Joined: 2011-08-28, 17:27
Location: Motala, SE
Contact:

Re: PM crashes on large-ish select-all

Unread post by Moonchild » 2020-02-07, 00:48

arzach wrote:
2020-02-06, 23:05
It's safe to assume that my colleague's antique equipment isn't a matter of personal choice.
I was talking about the browser version, not their equipment.
"Sometimes, the best way to get what you want is to be a good person." -- Louis Rossmann
"Seek wisdom, not knowledge. Knowledge is of the past; wisdom is of the future." -- Native American proverb
"Linux makes everything difficult." -- Lyceus Anubite

Locked