oom: create a resource limit for oom_adj
A message that is not in the archive (sent off-list or not yet received): [email protected]
A message that is not in the archive (sent off-list or not yet received): [email protected]
A message that is not in the archive (sent off-list or not yet received): [email protected]
A message that is not in the archive (sent off-list or not yet received): [email protected]
Quoted 13 lines · David Rientjes
Quoted 10 lines · Mandeep Singh Baines
Quoted 3 lines
What is the anticipated use case for this? We know that you want to lower
oom_adjwithoutCAP_SYS_RESOURCE, but what's the expected behavior when an app moves from foreground to background? I assume it's something likeThe focus here is the web browser's tabs. In our case, each is a process. If OOM is going to kill a process, you'd rather it kill the tab you looked at hours ago instead of the one you're looking at now. So you'd like to have a policy where the LRU tab gets killed first. We'd like to use
oom_score_adjas the mechanism to implement an LRU policy like this.
Quoted 5 lines
Hmm, at first glance that seems potentially dangerous if the current tab generates a burt of memory allocations and it ends up killing all other tabs before finally targeting the culprit whereas currently the heuristic should do a good job of finding this problematic tab and killing it instantly.
The original oom_adj design would e.g. adjust all background tabs to seem twice as bad as the current tab, so a ever-growing current tab would only be able to get one or two tabs killed in the worst case:
System is near OOM while the current tab starts using a normal amount of mem and grows beyond limits. After it easily killed the first bg tab by allocating one byte, it needs to grow to twice the normal tab memsize to start the OOM killer again. By then, it's score will be equal to normal background tabs (half size, double score), and killing the second tab will be luck. Killing the third tab should be impossible.
(If you adjust the bg tabs to be four times as killable, you'll get what
you asked for.)
I don't know the current oom_score_adj, it should be able to do something similar? Or should there be a oom_score_mul?
Quoted 9 lines
Quoted 8 lines
Quoted 2 lines
What do you anticipate will be writing to
oom_score_adjwith this patch, the app itself?A process in the browser session will do the adusting. We'd rather not give it
CAP_SYS_RESOURCE. It should only be allowed to changeoom_score_adjup and down within the bounds set by the administrator. Analagous torenice()which we also do using a similar policy.
Quoted 3 lines
So as more and more tabs get used, the least recently used tab gets its
oom_score_adjraised higher and higher until it is reused itself and then it gets reset back to 0 for the current tab?
As far as I understand, a background tab should get a higher score if it rests untouched for some time, and if it gets used again, it will jump to neutral oom_adj.
Quoted 2 lines
Is there a reason you don't want to give the underlying browser session process
CAP_SYS_RESOURCE?
I should not have control over a CAP_SYS_RESOURCE process while being a user, but I should be able to tell the system which process to kill. When I designed the oom_adj, I did not think about one-process-per-tab, so I did not suggest a soft/hard limit.
I think the concept of using an rlimit is the right thing to do, I did not yet look/think about the mapping. Glancing at oom_score_adj does suggest a different range ...
Quoted 11 lines · Bodo Eggert
The original
oom_adjdesign would e.g. adjust all background tabs to seem twice as bad as the current tab, so a ever-growing current tab would only be able to get one or two tabs killed in the worst case:System is near OOM while the current tab starts using a normal amount of mem and grows beyond limits. After it easily killed the first bg tab by allocating one byte, it needs to grow to twice the normal tab memsize to start the OOM killer again. By then, it's score will be equal to normal background tabs (half size, double score), and killing the second tab will be luck. Killing the third tab should be impossible.
Quoted 7 lines
(If you adjust the bg tabs to be four times as killable, you'll get what
you asked for.)I don't know the current
oom_score_adj, it should be able to do something similar? Or should there be aoom_score_mul?
oom_score_adj is done on a linear scale instead of exponential so instead of increasing oom_adj by 1 for each hop to the current tab, you would need to double it.
Quoted 17 lines
Quoted 5 lines
Hmm, at first glance that seems potentially dangerous if the current tab generates a burt of memory allocations and it ends up killing all other tabs before finally targeting the culprit whereas currently the heuristic should do a good job of finding this problematic tab and killing it instantly.
The original
oom_adjdesign would e.g. adjust all background tabs to seem twice as bad as the current tab, so a ever-growing current tab would only be able to get one or two tabs killed in the worst case:System is near OOM while the current tab starts using a normal amount of mem and grows beyond limits. After it easily killed the first bg tab by allocating one byte, it needs to grow to twice the normal tab memsize to start the OOM killer again. By then, it's score will be equal to normal background tabs (half size, double score), and killing the second tab will be luck. Killing the third tab should be impossible.
Quoted 2 lines
(If you adjust the bg tabs to be four times as killable, you'll get what
you asked for.)
Sorry for the delay.
I've sent revert patch to linus. We DON'T want to break any userland seriously. So, we reconsider about this area. personally I hope to reintroduce oom_score_adj later, but It must don't break current userland.
Quoted 4 lines · KOSAKI Motohiro
I've sent revert patch to linus. We DON'T want to break any userland seriously. So, we reconsider about this area. personally I hope to reintroduce
oom_score_adjlater, but It must don't break current userland.
Nobody is going to stall their development waiting for your patches to be merged (especially when you've proposed them four times before and they haven't been), so patches in this area should be against Linus' latest tree.
Quoted 11 lines · KOSAKI Motohiro
Quoted 5 lines
I've sent revert patch to linus. We DON'T want to break any userland seriously. So, we reconsider about this area. personally I hope to reintroduce
oom_score_adjlater, but It must don't break current userland.Nobody is going to stall their development waiting for your patches to be merged (especially when you've proposed them four times before and they haven't been), so patches in this area should be against Linus' latest tree.
Nobody want to accept you break userland.
A message that is not in the archive (sent off-list or not yet received): [email protected]
A message that is not in the archive (sent off-list or not yet received): [email protected]
A message that is not in the archive (sent off-list or not yet received): [email protected]
A message that is not in the archive (sent off-list or not yet received): [email protected]
Re: [PATCH] oom: allow a non-CAP_SYS_RESOURCE proces to oom_score_adj down
Quoted 6 lines · Mandeep Singh Baines
We'd like to be able to
oom_score_adja process up/down as its enters/leaves the foreground. Currently, it is not possible tooom_adjdown withoutCAP_SYS_RESOURCE. This patch allows a task to decrease itsoom_score_adjback to the value that aCAP_SYS_RESOURCEthread set it or its inherited value at fork. Assuming the thread that has forked it hasoom_score_adjof 0, each tab
process
Quoted 3 lines
could decrease it back from 0 upon
activation unless aCAP_SYS_RESOURCEthread elevated it to something higher.