Firefox start up with lots of windows tabs make plasma-6.7.90 freeze

Help
,
1

Dear all,

I have found a problem testing plasma 6.7.90. I have a firefox session with 4-5 windows and ~50-60 tabs total. Using 6.7.5 everything works as expected but with the beta version the kwin seems to hang and I have to hard reset.

  • I have tested with firefox-156, 155 and 140.16 both compiled and upstream binary versions with the same results (using gentoo at the moment).
  • An empty profile in firefox do not show the problem but as soon as I include a large set of tabs to load at start up, I can reproduce the behavior.
  • Sound continue working in the background (strawberry), so it seems like this is not a kernel issue but that kwin hang opening this amount of windows.

Thanks in advance

I am using stable Gentoo at the moment (except KDE and firefox). Below I provide kinfo a inxi -zv8

2

cruzki@semiorka ~ $ sudo inxi -zv8
System:
Kernel: 7.2.2-gentoo-dist-bin arch: x86_64 bits: 64 compiler: gcc v: 15.3.0
clocksource: tsc avail: acpi_pm
parameters: BOOT_IMAGE=/vmlinuz-7.2.2-gentoo-dist-bin
root=UUID=77369c12-edc7-4583-aeb8-80028fd334a8 ro rootwait
rootfstype=ext4 mitigations=off noibrs noibpb nopti nospectre_v2
nospectre_v1 l1tf=off nospec_store_bypass_disable no_stf_barrier mds=off
tsx=on tsx_async_abort=off i915.enable_guc=3 zswap.enabled=1
zswap.compressor=zstd mem_sleep_default=deep acpi.ec_no_wakeup=1
UUID=77369c12-edc7-4583-aeb8-80028fd334a8 rootflags=rw,noatime,discard
Desktop: KDE Plasma v: 6.7.5 tk: Qt v: N/A wm: kwin_wayland dm: SDDM
Distro: Gentoo Base System release 2.18
Machine:
Type: Laptop System: Acer product: TravelMate P614-51-G2 v: V1.16
serial: Chassis: type: 10 serial: N/A
Mobo: CML model: Piano_CM v: V1.16 serial:
part-nu: 0000000000000000 uuid: 28b70961-825d-49a1-a3f1-1b37b08d3800
Firmware: UEFI vendor: Insyde v: 1.16 date: 03/11/2021
Battery:
ID-1: BAT0 charge: 39.6 Wh (91.4%) condition: 43.3/59.6 Wh (72.7%)
power: 7.8 W volts: 15.74 min: 15.2 model: Murata KT00404002 AP18L4N
type: Li-ion serial: charging: status: discharging cycles: 140
Memory:
System RAM: total: 16 GiB available: 15.31 GiB used: 11.52 GiB (75.2%)
igpu: 64 MiB
Message: For most reliable report, install dmidecode.
Array-1: capacity: 32 GiB slots: 2 modules: 2 EC: None
max-module-size: 16 GiB note: est.
Device-1: ChannelA-DIMM0 type: DDR4 detail: synchronous size: 8 GiB
speed: 2667 MT/s volts: curr: 1 min: 2 max: 2 width (bits): data: 64
total: 64 manufacturer: SK Hynix part-no: HMAA1GS6CMR6N-VK serial: N/A
Device-2: ChannelB-DIMM0 type: DDR4 detail: synchronous size: 8 GiB
speed: 2667 MT/s volts: curr: 1 min: 1.35 max: 2 width (bits): data: 64
total: 64 manufacturer: Crucial part-no: CT8G4SFRA266.C8FR
serial:
PCI Slots:
Missing: Required tool dmidecode not installed. Check --recommends
CPU:
Info: model: Intel Core i5-10210U bits: 64 type: MT MCP
arch: Comet/Whiskey Lake note: check gen: core 10 level: v3 note: check
built: 2018 process: Intel 14nm family: 6 model-id: 0x8E (142)
stepping: 0xC (12) microcode: 0x100
Topology: cpus: 1x dies: 1 clusters: 4 cores: 4 threads: 8 tpc: 2
smt: enabled cache: L1: 256 KiB desc: d-4x32 KiB; i-4x32 KiB L2: 1024 KiB
desc: 4x256 KiB L3: 6 MiB desc: 1x6 MiB
Speed (MHz): avg: 2082 min/max: 400/4200 scaling: driver: intel_pstate
governor: powersave cores: 1: 2082 2: 2082 3: 2082 4: 2082 5: 2082 6: 2082
7: 2082 8: 2082 bogomips: 33615
Flags: 3dnowprefetch abm acpi adx aes aperfmperf apic arat
arch_capabilities arch_perfmon art avx avx2 bmi1 bmi2 bts clflush
clflushopt cmov constant_tsc cpuid cpuid_fault cx16 cx8 de ds_cpl dtes64
dtherm dts epb ept ept_ad erms est f16c flexpriority flush_l1d fma fpu
fsgsbase fxsr ht hwp hwp_act_window hwp_epp hwp_notify ibpb ibrs
ibrs_enhanced ida intel_pt invpcid lahf_lm lm mca mce md_clear mmx
monitor movbe mpx msr mtrr nonstop_tsc nopl nx pae pat pbe pcid pclmulqdq
pdcm pdpe1gb pebs pge pln pni popcnt pse pse36 pts rdrand rdseed rdtscp
rep_good sdbg sep sgx smap smep ss ssbd sse sse2 sse4_1 sse4_2 ssse3
stibp syscall tm tm2 tpr_shadow tsc tsc_adjust tsc_deadline_timer vme vmx
vnmi vpid x2apic xgetbv1 xsave xsavec xsaveopt xsaves xtopology xtpr
Vulnerabilities:
Type: gather_data_sampling status: Vulnerable
Type: ghostwrite status: Not affected
Type: indirect_target_selection status: Vulnerable
Type: itlb_multihit status: KVM: Vulnerable
Type: l1tf status: Not affected
Type: mds status: Not affected
Type: meltdown status: Not affected
Type: mmio_stale_data status: Vulnerable
Type: old_microcode status: Not affected
Type: reg_file_data_sampling status: Not affected
Type: retbleed status: Vulnerable
Type: spec_rstack_overflow status: Not affected
Type: spec_store_bypass status: Vulnerable
Type: spectre_v1 status: Vulnerable: __user pointer sanitization and
usercopy barriers only; no swapgs barriers
Type: spectre_v2 status: Vulnerable; IBPB: disabled; STIBP: disabled;
PBRSB-eIBRS: Vulnerable; BHI: Vulnerable
Type: srbds status: Vulnerable
Type: tsa status: Not affected
Type: tsx_async_abort status: Not affected
Type: vmscape status: Vulnerable
Graphics:
Device-1: Intel CometLake-U GT2 [UHD Graphics] vendor: Acer Incorporated ALI
driver: i915 v: kernel arch: Gen-9.5 process: Intel 14nm built: 2016-20
ports: active: eDP-1 empty: DP-1,HDMI-A-1 bus-ID: 00:02.0
chip-ID: 8086:9b41 class-ID: 0300
Device-2: Quanta HD User Facing driver: uvcvideo type: USB rev: 2.0
speed: 480 Mb/s lanes: 1 mode: 2.0 bus-ID: 1-7:5 chip-ID: 0408:a061
class-ID: 0e02
Display: unspecified server: X . org v: 1.21.1.24 with: Xwayland v: 24.1.13
compositor: kwin_wayland driver: X: loaded: modesetting
alternate: fbdev,intel,vesa dri: iris gpu: i915 display-ID: :0
note: <missing: xdpyinfo/xrandr>
Monitor-1: eDP-1 model: AU Optronics 0xda91 built: 2020 res: 1920x1080
dpi: 158 gamma: 1.2 chroma: red: x: 0.651 y: 0.329 green: x: 0.290 y: 0.608
blue: x: 0.149 y: 0.055 white: x: 0.314 y: 0.329
size: 309x174mm (12.17x6.85") diag: 355mm (14") ratio: 16:9
modes: 1920x1080
API: OpenGL Message: Unable to show GL data. glxinfo is missing.
Info: Tools: de: kscreen-console,kscreen-doctor x11: xprop
Audio:
Device-1: Intel Comet Lake PCH-LP cAVS vendor: Acer Incorporated ALI
driver: sof-audio-pci-intel-cnl alternate: snd_soc_avs,
snd_sof_pci_intel_cnl, snd_hda_intel bus-ID: 00:1f.3 chip-ID: 8086:02c8
class-ID: 0401
API: ALSA v: k7.2.2-gentoo-dist-bin status: kernel-api
tools: alsactl,alsamixer,amixer
Server-1: PipeWire v: 1.6.8 status: n/a (root, process) with:
1: pipewire-pulse status: active 2: wireplumber status: active 3: pw-jack
type: plugin tools: pactl,pw-cat,pw-cli,wpctl
Network:
Device-1: Intel Comet Lake PCH-LP CNVi WiFi driver: iwlwifi v: kernel
bus-ID: 00:14.3 chip-ID: 8086:02f0 class-ID: 0280
IF: wlan0 state: up mac:
IP v4: type: dynamic noprefixroute scope: global
broadcast:
IP v6: virtual: proto kernel_ll scope: link
Device-2: Intel Ethernet I219-LM vendor: Acer Incorporated ALI
driver: e1000e v: kernel port: N/A bus-ID: 00:1f.6 chip-ID: 8086:15bd
class-ID: 0200
IF: enp0s31f6 state: down mac:
Info: services: iwd,NetworkManager
WAN IP:
Bluetooth:
Device-1: Intel AX201 Bluetooth driver: btusb v: 0.8 type: USB rev: 2.0
speed: 12 Mb/s lanes: 1 mode: 1.1 bus-ID: 1-10:6 chip-ID: 8087:0026
class-ID: e001
Report: rfkill ID: hci0 rfk-id: 0 state: up address: see --recommends
Logical:
Message: No logical block device data found.
RAID:
Message: No RAID data found.
Drives:
Local Storage: total: 715.24 GiB used: 308.95 GiB (43.2%)
ID-1: /dev/mmcblk0 maj-min: 179:0 model: SD256 size: 238.3 GiB block-size:
physical: 512 B logical: 512 B type: USB tech: SSD serial:
scheme: MBR
SMART Message: Unknown smartctl error. Unable to generate data.
ID-2: /dev/nvme0n1 maj-min: 259:0 vendor: Western Digital model: PC SN520
SDAPNUW-512G-1014 size: 476.94 GiB block-size: physical: 512 B
logical: 512 B speed: 15.8 Gb/s lanes: 2 tech: SSD serial:
fw-rev: 20110000 temp: 33.9 C scheme: GPT
SMART: yes health: PASSED on: 1y 68d 20h cycles: 2,891
read-units: 41,366,396 [21.1 TB] written-units: 29,351,012 [15.0 TB]
Message: No optical or floppy data found.
Partition:
ID-1: / raw-size: 20 GiB size: 19.52 GiB (97.59%) used: 14.67 GiB (75.2%)
fs: ext4 block-size: 4096 B dev: /dev/nvme0n1p3 maj-min: 259:3 label: N/A
uuid: 77369c12-edc7-4583-aeb8-80028fd334a8
ID-2: /boot raw-size: 256 MiB size: 252 MiB (98.46%)
used: 153.5 MiB (60.9%) fs: vfat block-size: 512 B dev: /dev/nvme0n1p1
maj-min: 259:1 label: N/A uuid: 7098-F4B4
ID-3: /home raw-size: 452.69 GiB size: 444.51 GiB (98.19%)
used: 78.84 GiB (17.7%) fs: ext4 block-size: 4096 B dev: /dev/nvme0n1p4
maj-min: 259:4 label: N/A uuid: bcaa9642-050a-4077-a1b7-9c6b7fe30479
ID-4: /mnt/musica raw-size: 238.27 GiB size: 238.23 GiB (99.99%)
used: 214.68 GiB (90.1%) fs: exfat block-size: 512 B dev: /dev/mmcblk0p1
maj-min: 179:1 label: N/A uuid: 3434-3733
ID-5: /tmp raw-size: 4 GiB size: 3.92 GiB (98.04%) used: 618.5 MiB (15.4%)
fs: ext4 block-size: 4096 B dev: /dev/zram1 maj-min: 251:1 label: tmp_dir
uuid: a8927ec5-ff8c-4cfa-86d1-79ca43c1493b
ID-6: /var/tmp raw-size: 8 GiB size: 7.84 GiB (97.99%) used: 28 KiB (0.0%)
fs: ext4 block-size: 4096 B dev: /dev/zram2 maj-min: 251:2
label: var_tmp_dir uuid: 97a4600c-a42f-49ae-b781-305af9cc5b8c
Swap:
Kernel: swappiness: 60 (default) cache-pressure: 100 (default) zswap: yes
compressor: zstd max-pool: 20%
ID-1: swap-1 type: partition size: 4 GiB used: 0 KiB (0.0%) priority: -1
dev: /dev/nvme0n1p2 maj-min: 259:2 label: N/A
uuid: b3d3acfa-8a72-417a-83fd-a2b4ed48d41a
ID-2: swap-2 type: zram size: 4 GiB used: 42.4 MiB (1.0%) priority: 16383
comp: zstd avail: lzo-rle,lzo,lz4,lz4hc,deflate,842 dev: /dev/zram0
Unmounted:
Message: No unmounted partitions found.
USB:
Hub-1: 1-0:1 info: hi-speed hub with single TT ports: 12 rev: 2.0
speed: 480 Mb/s (57.2 MiB/s) lanes: 1 mode: 2.0 chip-ID: 1d6b:0002
class-ID: 0900
Device-1: 1-4:27 info: Alcor Micro AU9540 Smartcard Reader
type: smart card driver: usbfs interfaces: 1 rev: 2.0
speed: 12 Mb/s (1.4 MiB/s) lanes: 1 mode: 1.1 power: 50mA
chip-ID: 058f:9540 class-ID: 0b00
Device-2: 1-5:3 info: LighTuning EgisTec EH57A type:
driver: N/A interfaces: 1 rev: 2.0 speed: 480 Mb/s (57.2 MiB/s) lanes: 1
mode: 2.0 power: 100mA chip-ID: 1c7a:057a class-ID: ff00
Device-3: 1-6:4 info: Realtek RTS5129 Card Reader Controller
type: driver: rtsx_usb,rtsx_usb_ms,rtsx_usb_sdmmc
interfaces: 1 rev: 2.0 speed: 480 Mb/s (57.2 MiB/s) lanes: 1 mode: 2.0
power: 500mA chip-ID: 0bda:0129 class-ID: ff00 serial:
Device-4: 1-7:5 info: Quanta HD User Facing type: video driver: uvcvideo
interfaces: 2 rev: 2.0 speed: 480 Mb/s (57.2 MiB/s) lanes: 1 mode: 2.0
power: 500mA chip-ID: 0408:a061 class-ID: 0e02
Device-5: 1-10:6 info: Intel AX201 Bluetooth type: bluetooth driver: btusb
interfaces: 2 rev: 2.0 speed: 12 Mb/s (1.4 MiB/s) lanes: 1 mode: 1.1
power: 100mA chip-ID: 8087:0026 class-ID: e001
Hub-2: 2-0:1 info: super-speed hub ports: 6 rev: 3.1
speed: 10 Gb/s (1.16 GiB/s) lanes: 1 mode: 3.2 gen-2x1 chip-ID: 1d6b:0003
class-ID: 0900
Hub-3: 3-0:1 info: hi-speed hub with single TT ports: 2 rev: 2.0
speed: 480 Mb/s (57.2 MiB/s) lanes: 1 mode: 2.0 chip-ID: 1d6b:0002
class-ID: 0900
Hub-4: 4-0:1 info: super-speed hub ports: 2 rev: 3.1
speed: 10 Gb/s (1.16 GiB/s) lanes: 1 mode: 3.2 gen-2x1 chip-ID: 1d6b:0003
class-ID: 0900
Sensors:
System Temperatures: cpu: 46.0 C pch: 44.0 C mobo: N/A
Fan Speeds (rpm): N/A
Repos:
Packages: 1165 pm: emerge pkgs: 1158 libs: 182 pm: flatpak pkgs: 7
No enabled portage sources in: /etc/portage/repos.conf/eselect-repo.conf
Enabled portage sources in: /etc/portage/repos.conf/gentoo.conf
1: gentoo ~
No enabled portage sources in: /etc/portage/repos.conf/localrepo.conf
Processes:
CPU top: 5 of 274
1: cpu: 100% command: ps pid: 51122 mem: 3.77 MiB (0.0%)
2: cpu: 10.5% command: firefox pid: 47675 mem: 352.2 MiB (2.2%)
3: cpu: 6.4% command: firefox pid: 3168 mem: 1490.0 MiB (9.5%)
4: cpu: 4.2% command: firefox pid: 49176 mem: 475.1 MiB (3.0%)
5: cpu: 2.0% command: firefox pid: 45922 mem: 335.0 MiB (2.1%)
Memory top: 5 of 274
1: mem: 1490.0 MiB (9.5%) command: firefox pid: 3168 cpu: 6.4%
2: mem: 1252.0 MiB (7.9%) command: firefox pid: 3624 cpu: 0.2%
3: mem: 838.6 MiB (5.3%) command: firefox pid: 3701 cpu: 0.4%
4: mem: 823.7 MiB (5.2%) command: firefox pid: 3945 cpu: 0.5%
5: mem: 787.2 MiB (5.0%) command: firefox pid: 3637 cpu: 0.1%
Info:
Processes: 274 Power: uptime: 1d 15h 34m states: freeze,mem,disk
suspend: deep avail: s2idle wakeups: 14 hibernate: suspend avail: platform,
shutdown, reboot, test_resume image: 6.1 GiB
services: org_kde_powerdevil,upowerd Init: SysVinit v: 3.15 rc: OpenRC
v: 0.63.3 runlevel: default default: 3 tool: rc-service
Compilers: gcc: 15.3.0 Shell: Sudo (sudo) v: 1.9.17p2 default: Bash
v: 5.3.15 running-in: konsole inxi: 3.3.41

cruzki@semiorka ~ $ kinfo
Operating System: Gentoo Linux 2.18
KDE Plasma Version: 6.7.5 ← already downgrade to have an usable desktop.
KDE Frameworks Version: 6.30.0
Qt Version: 6.11.2
Kernel Version: 7.2.2-gentoo-dist-bin (64-bit)
Graphics Platform: Wayland
Processors: 8 × Intel® Core™ i5-10210U CPU @ 1.60GHz
Memory: 16 GiB of RAM (15.3 GiB usable)
Graphics Processor: Intel® UHD Graphics

3

Does the OOM killer run? You can check in the system journal for logs.

4

Sorry but I am using openrc so no journal. I have now configure to get the previous dmesg file saved to check if something have crashed. I am recompiling everything at the moment.

5

okay. What I am wondering is if your firefox requires more memory then you have on your system. The OOM killer may well be killing off processes when you run out of memory.

6

This was my first thought but it cannot be for several reasons:

  • It happen with a completely fresh start. Unless it has a pretty hard (and stupidly fast) memory leak, it cannot be. The hang happen instantly.
  • It would be quite strange that the OOM killer consistently kill other process instead of firefox.
  • It only happen with this particular configuration. Other configurations works without an issue.

I any case, I will post the results in a couple of hours. Let’s see if dmegs give any hint of what crashed.

7

Sounds like there may be a site causing the issue. Are the extensions loading when you’re introducing the large set of tabs?

Can you test the same tabs on a live usb stick of your OS?

8

Also see what happens with a different version of Firefox and the same session.
I have far more windows and tabs open with no issues.

Operating System: openSUSE Tumbleweed 20260918
KDE Plasma Version: 6.7.5
KDE Frameworks Version: 6.30.0
Qt Version: 6.11.2
Kernel Version: 7.2.6-1-default (64-bit)
Graphics Platform: X11
Processors: 12 × Intel® Core™ i7-9850H CPU @ 2.60GHz
Memory: 128 GiB of RAM (125.1 GiB usable)
Graphics Processor: Intel® UHD Graphics 630
Manufacturer: HP
Product Name: HP ZBook 17 G6
System Version: SBKPFV3

9

dmesg.old do no show anything usefull.

I do not use any extensión. I have even dissabled the kde integration one. I have reproduced the same issue with an empty firefox profile and with different versions of firefox (stables and actual), both compiled by me and stock ones. I am pretty confident, firefox is not to blame here as it has been working perfectly and in fact, it works perfectly in 6.7.5.

I think the main relevant point is the number of windows opened, not the number of tabs or the sites. Please take into consideration that the first test I did was to open a single firefox window with more tabs than my profile have. Then close it and reopen. The test works. Then I tried to “recreate” the same distribution of “tabs in windows” I use and I am not sure how many windows are needed, but I think more than 3 made the hang (I usually have 5). I have just tested again with “random” sites in windows and not in tabs: Wikipedia, gmail. google calendar, youtube and drive folders. A couple of tabs per windows. When I reach 6 tabs, the same problem arise. If these are to blame, we are in a serious regression here.

The fun part is that this time I was just through out to the virtual terminal. kwin was hang with 200% of the cpu and when I tried to kill it with a 9 signal, the computer hang (!!!).

Now I am wondering, last week I remember that it was mentioned in the last weekly report that a fix for too many one tabs in kwin was solved, no? I could not find it to check if this could be the root of the problem

10

Do you have logs from kwin you can look at for clues?

11

Ubuntu changed its desktop to use lower OOM score to prevent it from being killed under high memory usage.

https://discourse.ubuntu.com/t/improving-ubuntu-desktop-stability-under-memory-pressure/87945?u=d0od

https://bugs.launchpad.net/ubuntu/+source/dbus/+bug/2089800

12

Perhaps not what you are looking for, but on the off chance it is, have you tried separating the tabs into different Firefox Profiles?

I do that by binding different Firefox Profiles to different Plasma Activities:

13

I will try to get this tomorrow as I need to recompile everything [twice as this continue to be my production computer]. Just in case, you want the file that is in this folder, no?

~/.local/share/sddm

Any other thing you can think of?

14

I booted this morning and Plasma killed Firefox with an Out Of Memory warning, while Discover was downloading updates, and I only had 6 tabs open from a few days ago, plus Kwrite and something else small.
It hasn’t done that to me before, so I wonder if it is some transient thing? Because I have 16GB RAM and there’s no way half a dozen open ‘ordinary’ tabs should kill Firefox.

15

You need to monitor the memory use to find data to reason from.
You could run a script that puts the output ps output every few seconds to a log so you know what has been happening to memory.

16

Need to look at the journal logs via cmd or this app.
journalctl -b | grep -i “error|warn|fail|firefox”

17

Note the OP said

18

And more important, the system freeze and I need to do a hard reboot, so whenever I have to look, it has to survive a reboot.

19

Seems like the new version (6.7.91) solve the issue.

Thanks for the help.