пт, 12 сент. 2025 г., 09:57 deloptes via tde-users <users@xxxxxxxxxxxxxxxxxx>:
Andrew Randrianasulu via tde-users wrote:
> 32bit firefox just gets killed if grow too big for 32bit address space ;)
>
Did I ever mention, I was using 32bit firefox? No, I did not, because I am
using 64bit.
ok.
> Go 64bit like everyone else, keep they saying ...nah.
And 64bit gets killed too, but it took usually about 10-15min to kick the
oom killer. So it was very unpleasant.
I think it actually doesn't matter in this specific scenario if it is 32 or
64bit. (It actually does, because it utilizes the CPU in full length, but
it is not related to the missing memory, ability to handle and swap)
Then something else is wrong ... I think I see some UI freezes on Android tablet, but I guess it simply mean its dying ... after just a little more than 3 years. grrrr.
Are you using SSD? if yes you may want to check any timeout messages in dmesg, and look for smartctl output (while I never personally averted any hdd failure by looking there ... I use mechanical HDDs so they are audible in their life and death.)
I also use xfs for /home and ext4 for / and /usr
Xfs a bit slow when you update git trees etc, but so far it served me well.
I surely can see firefox crawling out of swap .. on 1gb rammachine. On 6 gb ram laptop and 16 gb ram desktop it behaves ok for me. Well, outside of getting killed after few days of firefox's uptime. But it does not freeze my whole machine. I can freeze everything by filling up artificially enlarged (from default 60% of ram) tmpfs mounted on /dev/shm , and I recall modern distros mount /tmp to tmpfs too, so on /tmp overflow you can get some *p*a*u*s*e ....untill oom killer kicks in. (I think default mostly tuned for servers where admins supposed to watch out their working set, there usually series of tweaks or whole patch sets floated by desktop users, but how effective they are often debated)
____________________________________________________
tde-users mailing list -- users@xxxxxxxxxxxxxxxxxx
To unsubscribe send an email to users-leave@xxxxxxxxxxxxxxxxxx
Web mail archive available at https://mail.trinitydesktop.org/mailman3/hyperkitty/list/users@xxxxxxxxxxxxxxxxxx
____________________________________________________ tde-users mailing list -- users@xxxxxxxxxxxxxxxxxx To unsubscribe send an email to users-leave@xxxxxxxxxxxxxxxxxx Web mail archive available at https://mail.trinitydesktop.org/mailman3/hyperkitty/list/users@xxxxxxxxxxxxxxxxxx
