Hi,

If it helps this can be reproduced in QEMU (ie without any Raspberry Pi hardware).

One just needs to create a root img file with debootstrap[0] (which probably needs to run on Debian), which will contain a kernel and initrd to give to qemu[1]. Once booted one can "ssh -X -p 2222 user@localhost" and start chromium[2] to give a similar backtrace[3].

Kind Regards
Tom

[0] #something along the lines of:
sudo apt-get -y install debootstrap qemu-system
r=/home/thomas/armhf
sudo debootstrap --arch=armhf --include=aptitude,bc,bind9-host,build-essential,curl,dnsutils,fake-hwclock,git,less,locales,netcat-traditional,ntp,ntpdate,pv,screen,sudo,task-ssh-server,tcpdump,telnet,vim-gtk --foreign sid $r http://mirror.bytemark.co.uk/debian
sudo cp /usr/bin/qemu-arm-static $r/usr/bin/qemu-arm-static
sudo chroot $r /debootstrap/debootstrap --second-stage
sudo chroot $r apt-get update
sudo chroot $r apt-get install dbus chromium xbase-clients linux-image-armmp
sudo chroot $r apt-get clean
echo "/dev/vda1 / ext4 defaults 0 0" | sudo tee $r/etc/fstab
echo "armhf" | sudo tee $r/etc/hostname
echo -en "allow-hotplug enp0s1\niface enp0s1 inet dhcp\n" | sudo tee $r/etc/network/interfaces.d/enp0s1
sudo chroot $r passwd
sudo chroot $r adduser user
echo "user ALL=(ALL:ALL) NOPASSWD: ALL" | sudo tee $r/etc/sudoers.d/user
sudo chmod 400 $r/etc/sudoers.d/user
dd if=/dev/zero bs=1M of=$r.img count=4096
sudo losetup -f $r.img
dev=$(sudo losetup | grep $r.img | awk '{print $1}')
echo -en "n\np\n\n\n\nw\n" | sudo fdisk $dev
sudo losetup --detach $dev
sudo rmmod loop
sudo modprobe loop max_part=16
sudo losetup -f $r.img
dev=$(sudo losetup | grep $r.img | awk '{print $1}')
sudo mkfs -t ext4 ${dev}p1
sudo mkdir $r.d
sudo mount ${dev}p1 $r.d
sudo rsync -vax --delete $r/. $r.d/.
sudo umount $r.d
sudo losetup --detach $dev

[1] #something along the lines of:
"qemu-system-arm -M virt -m 1024 -kernel $r/boot/vmlinuz-4.17.0-3-armmp -initrd $r/boot/initrd.img-4.17.0-3-armmp -nographic -append "console=ttyAMA0 root=/dev/vda1" -hda $r.img -net nic -net user,hostfwd=tcp::2222-:22"

[2] #something along the lines of:
echo "deb http://httpredir.debian.org/debian-debug/ sid-debug main" | sudo tee -a /etc/apt/sources.list
sudo apt-get update
sudo apt-get -y install debian-goodies gdb
#sudo apt-get -y install $(find-dbgsym-packages /usr/lib/chromium/chromium)
sudo apt-get -y install chromium-dbgsym libasound2-dbgsym libasyncns0-dbgsym libatk-bridge2.0-0-dbgsym libatk1.0-0-dbgsym libatomic1-dbg libatspi2.0-0-dbgsym libavahi-client3-dbgsym libavahi-common3-dbgsym libavcodec58-dbgsym libavformat58-dbgsym libavutil56-dbgsym libblkid1-dbgsym libbluray2-dbgsym libbsd0-dbgsym libbz2-1.0-dbgsym libcairo-gobject2-dbgsym libcairo2-dbgsym libcap2-dbgsym libchromaprint1-dbgsym libcodec2-0.8.1-dbgsym libcom-err2-dbgsym libcroco3-dbgsym libcups2-dbgsym libdatrie1-dbgsym libdbus-1-3-dbgsym libdrm2-dbgsym libepoxy0-dbgsym libevent-2.1-6-dbgsym libexpat1-dbgsym libffi6-dbg libflac8-dbgsym libfontconfig1-dbgsym libfreetype6-dbgsym libfribidi0-dbgsym libgcrypt20-dbgsym libgdk-pixbuf2.0-0-dbgsym libglib2.0-0-dbgsym libgme0-dbgsym libgmp10-dbgsym libgnutls30-dbgsym libgomp1-dbg libgpg-error0-dbgsym libgraphite2-3-dbgsym libgsm1-dbg libgtk-3-0-dbgsym libharfbuzz0b-dbgsym libhogweed4-dbgsym libice6-dbg libicu60-dbgsym libidn2-0-dbgsym libjpeg62-turbo-dbgsym libkeyutils1-dbgsym libkrb5-dbg liblcms2-2-dbgsym liblz4-1-dbgsym liblzma5-dbgsym libminizip1-dbgsym libmount1-dbgsym libmp3lame0-dbgsym libmpg123-0-dbgsym libnettle6-dbgsym libnspr4-dbgsym libnss3-dbgsym libnuma1-dbgsym libogg-dbg libopenjp2-7-dbgsym libopenmpt0-dbgsym libopus-dbg libp11-kit0-dbgsym libpango-1.0-0-dbgsym libpangocairo-1.0-0-dbgsym libpangoft2-1.0-0-dbgsym libpci3-dbgsym libpcre3-dbg libpixman-1-0-dbgsym libpng16-16-dbgsym libpulse0-dbgsym libre2-4-dbgsym librsvg2-2-dbgsym libselinux1-dbgsym libshine3-dbgsym libsm6-dbg libsnappy1v5-dbgsym libsndfile1-dbgsym libsoxr0-dbgsym libspeex-dbg libssh-gcrypt-4-dbgsym libstdc++6-8-dbg libswresample3-dbgsym libsystemd0-dbgsym libtasn1-6-dbgsym libthai0-dbgsym libtheora-dbg libtwolame0-dbgsym libudev1-dbgsym libunistring2-dbgsym libuuid1-dbgsym libva-drm2-dbgsym libva-x11-2-dbgsym libva2-dbgsym libvdpau1-dbgsym libvorbis0a-dbgsym libvorbisenc2-dbgsym libvorbisfile3-dbgsym libvpx5-dbgsym libwavpack1-dbgsym libwayland-client0-dbgsym libw
ayland-cursor0-dbgsym libwayland-egl1-dbgsym libwebp6-dbgsym libwebpdemux2-dbgsym libwebpmux3-dbgsym libwrap0-dbgsym libx11-6-dbgsym libx11-xcb1-dbgsym libx264-152-dbgsym libx265-160-dbgsym libxau6-dbg libxcb-render0-dbgsym libxcb-shm0-dbgsym libxcb1-dbgsym libxcomposite1-dbg libxcursor1-dbgsym libxdamage1-dbgsym libxdmcp6-dbg libxext6-dbg libxfixes3-dbgsym libxi6-dbgsym libxinerama1-dbg libxkbcommon0-dbgsym libxml2-dbg libxrandr2-dbgsym libxrender1-dbgsym libxslt1.1-dbgsym libxss1-dbg libxtst6-dbgsym libxvidcore4-dbgsym libzvbi0-dbgsym zlib1g-dbg

find / > /dev/null & #generate entropy
cat /proc/sys/kernel/random/entropy_avail
chromium --debug

[3]
user@armhf:~$ /usr/bin/chromium --debug
<SNIP>
(gdb) run
<SNIP>
Thread 1 "chromium" received signal SIGSEGV, Segmentation fault.
0x00b8856c in void std::vector<std::pair<unsigned long long, long long>, std::allocator<std::pair<unsigned long long, long long> > >::_M_insert_aux<std::pair<unsigned long long, long long> >(__gnu_cxx::__normal_iterator<std::pair<unsigned long long, long long>*, std::vector<std::pair<unsigned long long, long long>, std::allocator<std::pair<unsigned long long, long long> > > >, std::pair<unsigned long long, long long>&&) ()
(gdb) bt
#0  0x00b8856c in void std::vector<std::pair<unsigned long long, long long>, std::allocator<std::pair<unsigned long long, long long> > >::_M_insert_aux<std::pair<unsigned long long, long long> >(__gnu_cxx::__normal_iterator<std::pair<unsigned long long, long long>*, std::vector<std::pair<unsigned long long, long long>, std::allocator<std::pair<unsigned long long, long long> > > >, std::pair<unsigned long long, long long>&&) ()
#1  0x01fd7a88 in ukm::internal::UkmEntryBuilderBase::SetMetricInternal(unsigned long long, long long) ()
(gdb) quit
A debugging session is active.

	Inferior 1 [process 4717] will be killed.

Quit anyway? (y or n) y
user@armhf:~$

[?2004l#0  0x5523256c in void std::vector<std::pair<unsigned long long, long long>, std::allocator<std::pair<unsigned long long, long long> > >::_M_insert_aux<std::pair<unsigned long long, long long> >(__gnu_cxx::__normal_iterator<std::pair<unsigned long long, long long>*, std::vector<std::pair<unsigned long long, long long>, std::allocator<std::pair<unsigned long long, long long> > > >, std::pair<unsigned long long, long long>&&) ()
#1  0x56681a88 in ukm::internal::UkmEntryBuilderBase::SetMetricInternal(unsigned long long, long long) ()
#2  0x5676ee46 in ukm::builders::TabManager_WindowMetrics::SetWindowId(long long) ()
#3  0x570c5b98 in (anonymous namespace)::LogWindowMetricsUkmEntry(tab_ranker::WindowFeatures const&) ()
#4  0x570c5c64 in WindowActivityWatcher::BrowserWatcher::MaybeLogWindowMetricsUkmEntry(bool) ()
#5  0x570c5d3e in WindowActivityWatcher::OnBrowserAdded(Browser*) ()
#6  0x570c5dc4 in WindowActivityWatcher::WindowActivityWatcher() ()
#7  0x570c5df6 in WindowActivityWatcher::GetInstance() ()



SetMetricsInternal() is this:

void UkmEntryBuilderBase::SetMetricInternal(uint64_t metric_hash,
                                            int64_t value) {
  entry_->metrics.emplace(metric_hash, value);
}


So why would .emplace() crash?

Either entry_ is corrupt or its members, I would assume.

entry_ is allocated in the constructor here:

UkmEntryBuilderBase::UkmEntryBuilderBase(ukm::SourceId source_id,
                                         uint64_t event_hash)
    : entry_(mojom::UkmEntry::New()) {
  entry_->source_id = source_id;
  entry_->event_hash = event_hash;
}

And reset in Record():

void UkmEntryBuilderBase::Record(UkmRecorder* recorder) {
  if (recorder)
    recorder->AddEntry(std::move(entry_));
  else
    entry_.reset();
}

And logging code looks benign:

void LogWindowMetricsUkmEntry(const WindowFeatures& window_features) {
  // ...
  ukm::builders::TabManager_WindowMetrics entry(ukm::AssignNewSourceId());
  entry.SetWindowId(window_features.window_id.id())
      .SetIsActive(window_features.is_active)
      .SetShowState(window_features.show_state)
      .SetType(window_features.type);

So Record() isn't called before SetWindowId(). So either mojom::UkmEntry::New() is returning some garbage or something is corrupting memory?

Could you debug in GDB to see if entry_ is valid? i.e. if you dump its members, do they look sane?

I can confirm the same bug on two other armhf boards, with plain Debian (i.e. nothing to do with Raspbian or RPI), the software being:

Chromium 68.0.3440.75 on Debian Stretch on kernel 4.9.114
Chromium 68.0.3440.75 on Debian Stretch on kernel 4.14.y

Chromium 67 worked (and still works) fine prior to updating the Debian stable package from chromium from version 67 to version 68.

(All the above is for armhf, I was not able to test arm64 because the debian package for chromium 68 does not even compile from source for arm64 ...).

Does anyone know of any arm-based machines for which chromium 68 does startup without crashing?

An additional test: Was now able to try the 64 bit version of chromium 68 on arm64 under buster (the version of libc on stretch does not support chromium 68), and it does not crash on startup.

In summary, the versions that always crash on startup that I've tested or read about are all 32 bit compiles of chromium 68 for armhf. This happens when running these 32 bit binaries on (at least) chips from broadcom, allwinner, exynos (samsung) and rockchip.

In contrast, 64 bit compiles for arm64 and amd64 never exhibited the bug in my tests.

Perhaps some of the code handling metrics or its input is being a bit sloppy with 32 vs 64 bit words/ints (assuming it is not a compiler bug). 

Hi,

I'm not really a gdb expert, but it seems that it's not behaving as I would expect. I started "chromium --debug" again as normal[0]. I set a breakpoint at UkmEntryBuilderBase[1] and started the program[2] and it hit the breakpoint[3]. However I was not able to list any source or inspect any variables, all I got were errors[4]. I also just noticed that the backtrace didn't contain source file names[5]. I'm not sure if this is my ignorance about GDB, however I repeated the same test on amd64 and found that the backtrace didn't contain filenames and I could not print variables either. So it may be the way Chromium gets compiled.

Kind Regards
Tom

[0]
user@armhf:~$ chromium --debug
# Env:
#     LD_LIBRARY_PATH=
#                PATH=/usr/local/bin:/usr/bin:/bin:/usr/local/games:/usr/games
#            GTK_PATH=
#      CHROMIUM_FLAGS= --show-component-extension-options --ignore-gpu-blacklist --no-default-browser-check --disable-pings --media-router=0 --enable-remote-extensions
/usr/bin/gdb /usr/lib/chromium/chromium -x /tmp/chromiumargs.E4AtqA
GNU gdb (Debian 8.1-4) 8.1
Copyright (C) 2018 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "arm-linux-gnueabihf".
Type "show configuration" for configuration details.
For bug reporting instructions, please see:
<http://www.gnu.org/software/gdb/bugs/>.
Find the GDB manual and other documentation resources online at:
<http://www.gnu.org/software/gdb/documentation/>.
For help, type "help".
Type "apropos word" to search for commands related to "word"...
Reading symbols from /usr/lib/chromium/chromium...Reading symbols from /usr/lib/debug/.build-id/43/f92dc2a09900348561aa51a915a8ec6cf8d852.debug...done.
done.
(gdb) 

[1]
(gdb) break UkmEntryBuilderBase::UkmEntryBuilderBase
Breakpoint 1 at 0x1bd799c
(gdb)

[2]
(gdb) run
Starting program: /usr/lib/chromium/chromium --show-component-extension-options --ignore-gpu-blacklist --no-default-browser-check --disable-pings --media-router=0 --enable-remote-extensions --single-process 
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/arm-linux-gnueabihf/libthread_db.so.1".
[New Thread 0xafe1af70 (LWP 1332)]
[New Thread 0xaf4fef70 (LWP 1337)]
[New Thread 0xaeafef70 (LWP 1338)]
Gtk-Message: 13:26:35.939: Failed to load module "canberra-gtk-module"
Gtk-Message: 13:26:36.028: Failed to load module "canberra-gtk-module"
[New Thread 0xad930f70 (LWP 1339)]
[New Thread 0xad12ff70 (LWP 1340)]
[New Thread 0xaec8cf70 (LWP 1341)]
[New Thread 0xac92ef70 (LWP 1342)]
[1329:1329:0831/132637.498913:ERROR:default_network_context_params.cc(60)] Cannot use V8 Proxy resolver in single process mode.
[New Thread 0xac12df70 (LWP 1343)]
[New Thread 0xab92cf70 (LWP 1344)]
[New Thread 0xab12bf70 (LWP 1345)]
[New Thread 0xaa92af70 (LWP 1346)]
[New Thread 0xaa129f70 (LWP 1347)]
[New Thread 0xa9928f70 (LWP 1348)]
[New Thread 0xa9127f70 (LWP 1349)]
[New Thread 0xa8926f70 (LWP 1350)]
[New Thread 0xa8125f70 (LWP 1351)]
[New Thread 0xa7924f70 (LWP 1352)]
[New Thread 0xa7123f70 (LWP 1353)]
[New Thread 0xa6922f70 (LWP 1354)]
[New Thread 0xa6121f70 (LWP 1355)]
[New Thread 0xa5920f70 (LWP 1356)]
[New Thread 0xa511ff70 (LWP 1357)]
[New Thread 0xa491ef70 (LWP 1358)]
[New Thread 0xa3fbff70 (LWP 1381)]
[New Thread 0xa37bef70 (LWP 1533)]
[New Thread 0xa2fbdf70 (LWP 1534)]
[New Thread 0xa27bcf70 (LWP 1536)]
[New Thread 0xa1e40f70 (LWP 1537)]
[1329:1329:0831/132705.767204:ERROR:default_network_context_params.cc(60)] Cannot use V8 Proxy resolver in single process mode.
[1329:1537:0831/132706.007302:ERROR:object_proxy.cc(616)] Failed to call method: org.freedesktop.Notifications.GetCapabilities: object_path= /org/freedesktop/Notifications: org.freedesktop.DBus.Error.ServiceUnknown: The name org.freedesktop.Notifications was not provided by any .service files
[New Thread 0x9e167f70 (LWP 1538)]
[Thread 0xa2fbdf70 (LWP 1534) exited]
[Thread 0xa3fbff70 (LWP 1381) exited]
[New Thread 0xa2fbdf70 (LWP 1539)]
[Thread 0x9e167f70 (LWP 1538) exited]
[New Thread 0x9e167f70 (LWP 1540)]
[New Thread 0xa3fbff70 (LWP 1541)]
[New Thread 0x9d551f70 (LWP 1542)]
[New Thread 0x9cd50f70 (LWP 1543)]
[New Thread 0x9c54ff70 (LWP 1544)]
[New Thread 0x9bd4ef70 (LWP 1545)]
[New Thread 0x9b54df70 (LWP 1546)]

[3]
Thread 1 "chromium" hit Breakpoint 1, 0x01fd799c in ukm::internal::UkmEntryBuilderBase::UkmEntryBuilderBase(long long, unsigned long long) ()
(gdb) 

[4]
(gdb) list
1	../sysdeps/unix/sysv/linux/arm/libc-do-syscall.S: No such file or directory.
(gdb) p entry_
No symbol "entry_" in current context.
(gdb) p *entry_
No symbol "entry_" in current context.
(gdb) 
(gdb) p *this
No symbol "this" in current context.
(gdb) 
(gdb) info locals
No symbol table info available.
(gdb)

[5]
(gdb) bt
#0  0x01fd799c in ukm::internal::UkmEntryBuilderBase::UkmEntryBuilderBase(long long, unsigned long long) ()
#1  0x020c4d6a in ukm::builders::TabManager_WindowMetrics::TabManager_WindowMetrics(long long) ()
#2  0x02a1bb8e in (anonymous namespace)::LogWindowMetricsUkmEntry(tab_ranker::WindowFeatures const&) ()
#3  0x02a1bc64 in WindowActivityWatcher::BrowserWatcher::MaybeLogWindowMetricsUkmEntry(bool) ()
#4  0x02a1bd3e in WindowActivityWatcher::OnBrowserAdded(Browser*) ()
#5  0x02a1bdc4 in WindowActivityWatcher::WindowActivityWatcher() ()
#6  0x02a1bdf6 in WindowActivityWatcher::GetInstance() ()
#7  0x01a329fc in resource_coordinator::TabActivityWatcher::TabActivityWatcher() ()
#8  0x01a32a32 in resource_coordinator::TabActivityWatcher::GetInstance() ()
#9  0x01907b94 in ChromeBrowserMainParts::PostBrowserStart() ()
#10 0x01910f38 in ChromeBrowserMainParts::PreMainMessageLoopRunImpl() ()
#11 0x019111a2 in ChromeBrowserMainParts::PreMainMessageLoopRun() ()
#12 0x00e28a6e in content::BrowserMainLoop::PreMainMessageLoopRun() ()
#13 0x0104ba0e in content::StartupTaskRunner::RunAllTasksNow() ()
#14 0x00e2fcea in content::BrowserMainLoop::CreateStartupTasks() ()
#15 0x00e30b36 in content::BrowserMainRunnerImpl::Initialize(content::MainFunctionParams const&, std::unique_ptr<content::BrowserProcessSubThread, std::default_delete<content::BrowserProcessSubThread> >) ()
#16 0x00e30d26 in content::BrowserMain(content::MainFunctionParams const&, std::unique_ptr<content::BrowserProcessSubThread, std::default_delete<content::BrowserProcessSubThread> >) ()
#17 0x018e2ab6 in content::RunBrowserProcessMain(content::MainFunctionParams const&, content::ContentMainDelegate*, std::unique_ptr<content::BrowserProcessSubThread, std::default_delete<content::BrowserProcessSubThread> >) ()
#18 0x018e2bb4 in content::ContentMainRunnerImpl::Run() ()
#19 0x018e7ea8 in service_manager::Main(service_manager::MainParams const&) ()
#20 0x018e1ec2 in content::ContentMain(content::ContentMainParams const&) ()
#21 0x008c832c in ChromeMain ()
#22 0xb27ae066 in __libc_start_main (main=0x8b225d <main>, argc=8, argv=0xbefff444, init=<optimized out>, 
    fini=0x3bf8c7d <__libc_csu_fini>, rtld_fini=0xb6fe1761 <_dl_fini>, stack_end=0xbefff444)
    at libc-start.c:310
#23 0x008c81a0 in _start ()
Backtrace stopped: previous frame identical to this frame (corrupt stack?)
(gdb) 

Summary: crash with SEGV on startup persists in version 69 on armhf

Yesterday Debian stretch had an update of the chromium package from version 68 to version 69 (specifically 69.0.3597.81). However, this bug persists on armhf for version 69.

It would be great if the maintainers could take a look and at least mark the bug as confirmed and assign an owner since it is widely experienced. Thanks!

chromium-shell also reliably crashes on startup on debian armh.
This might narrow down what code might be causing the bug.

More specifically, when starting chromium-shell (version 69.0.3497.92-1,deb9u1) under debian armhf, there is a fatal check failure:

"FATAL:v8_initializer.cc(329) Couldn't mmap v8 natives data file, status code is 1"

@asvitk, could you take another look please since you have some familiarity with the v8 code. Thanks!

Seems it is out of scope from TE end as per above all comments , adding TE-NeedsTraige-help label to move this out of our triaging bucket.

Could someone from dev team please take a look into this issue.
Thanks..!

Also seeing this on a SolidRun Hummingboard 2 running Debian Stretch armhf. 67 works fine, upgraded to 69 and segfaults any time I run it. Happy to provide any details needed, all I have copied down is:

Received signal 11 SEGV_MAPERR 000000000000
#0 0x000001c9afdc <unknown>
#1 0x000001c2ac4c <unknown>
#2 0x000001c9b2ce <unknown>
#3 0x000001c9b510 <unknown>
#4 0x0000b2d65fe0 <unknown>
#5 0x000000b394bc <unknown>
#6 0x000001f0c42c <unknown>
#7 0x000002e5a118 <unknown>
#8 0x000002e5a196 <unknown>
#9 0x000002f918c4 <unknown>
#10 0x000002eecf0c <unknown>
#11 0x0000031bdf72 <unknown>
#12 0x0000031bdfe6 <unknown>
#13 0x0000031be584 <unknown>
#14 0x0000031be6a0 <unknown>
#15 0x0000031be958 <unknown>
#16 0x0000031beb1a <unknown>
#17 0x0000031beb60 <unknown>
#18 0x0000031beba0 <unknown>
#19 0x0000031bf002 <unknown>
#20 0x0000031bf1ce <unknown>
#21 0x000002fbd34a <unknown>
#22 0x000003695c08 <unknown>
#23 0x000000c18eec <unknown>
#24 0x000003699af0 <unknown>
#25 0x000001eeffac <unknown>
#26 0x000001ee623e <unknown>
#27 0x000001ee5de2 <unknown>
#28 0x000001c2b10c <unknown>
#29 0x000001c6ec7c <unknown>
#30 0x000001c2b10c <unknown>
#31 0x000001c44574 <unknown>
#32 0x000001c44ca0 <unknown>
#33 0x000001c44ecc <unknown>
#34 0x000001c3cf18 <unknown>
#35 0x000001c5a4a6 <unknown>
#36 0x00000369cdd6 <unknown>
#37 0x000001a27436 <unknown>
#38 0x000001a276e6 <unknown>
#39 0x000001a2beae <unknown>
#40 0x000001a269e0 <unknown>
#41 0x0000038338f6 <unknown>
#42 0x0000038339ce <unknown>
#43 0x000001a2a162 <unknown>
#44 0x0000009bf69e ChromeMain
#45 0x0000b2d574aa __libc_start_main

Hi, have any update?

Hi,

I just upgraded the original Raspberry Pi I encountered this bug on with a "apt-get update; apt-get dist-upgrade" and I have found that the problem has resolved. The version that now works is 70.0.3538.67-2.

I was not able to compile chromium successfully in the end for 32bit arm so was not able to debug further. I was eventually able to cross compile with clang for arm32 but was unable to reproduce, but given the different code version, compile options and compiler it was not a useful test and I didn't get a change to look further. Obviously anything in the 69->70 change could have fixed this issue, however there were also changes in the build process, namely switching from gcc6 to gcc8. I have no idea what has fixed the issue for me.

Kind Regards
Tom

Thanks for the update. Per above, marking as wontfix as there doesn't seem like there's anything left to do for this.