Messages in this thread | | | From | KOSAKI Motohiro <> | Subject | Re: [resend][PATCH 4/4] oom: don't ignore rss in nascent mm | Date | Thu, 25 Nov 2010 20:06:51 +0900 (JST) |
| |
> > > Stupid question. > > > > > > Can't we just account these allocations in the old -mm temporary? > > > > > > IOW. Please look at the "patch" below. It is of course incomplete > > > and wrong (to the point inc_mm_counter() is not safe without > > > SPLIT_RSS_COUNTING), and copy_strings/flush_old_exec are not the > > > best places to play with mm-counters, just to explain what I mean. > > > > > > It is very simple. copy_strings() increments MM_ANONPAGES every > > > time we add a new page into bprm->vma. This makes this memory > > > visible to select_bad_process(). > > > > > > When exec changes ->mm (or if it fails), we change MM_ANONPAGES > > > counter back. > > > > > > Most probably I missed something, but what do you think? > > > > Because, If the pages of argv is swapping out when processing execve, > > This accouing doesn't work. > > Why? > > If copy_strings() inserts the new page into bprm->vma and then > this page is swapped out, inc_mm_counter(current->mm, MM_ANONPAGES) > becomes incorrect, yes. And we can't turn it into MM_SWAPENTS. > > But does this really matter? oom_badness() counts MM_ANONPAGES + > MM_SWAPENTS, and result is the same.
Ah, I got it. I did too strongly get stucked correct accounting. but you mean it's not must.
Okey, I'll tackle this one at this weekend hopefully.
> > Is this enough explanation? Please don't hesitate say "no". If people > > don't like my approach, I don't hesitate change my thinking. > > Well, certainly I can't say no ;) > > But it would be nice to find a more simple fix (if it can work, > of course). > > > And. I need a simple solution for the older kernels.
Alright. It is certinally considerable one.
| |