Re: [PATCH] oom killer (Core)

[Thomas Gleixner <tglx@linutronix.de>]

On Fri, 2004-12-03 at 03:28 +0100, Andrea Arcangeli wrote:
> On Fri, Dec 03, 2004 at 12:34:59AM +0100, Andrea Arcangeli wrote:
> > I'll add to my last patch the removal of the PF_MEMDIE check in oom_kill
> > plus I'll fix the remaining race with PF_EXITING/DEAD, and I'll add a
> > cond_resched. Then you can try again with my simple way (w/ and w/o
> > PREEMPT ;).
> 
> Ok, I expect this patch to fix the problem completely. 
> <SNIP>
> With this thing, I doubt any wrong task will ever be killed again...

You're right. oom-kill() did not do anything wrong. See log below

This is w/o PREEMPT. Is it neccecary to verify w/ PREEMPT too ?

If it would have booted it still would have killed sshd instead of the
application which was forking a lot of childs.

tglx

Dentry cache hash table entries: 32768 (order: 5, 131072 bytes)
Inode-cache hash table entries: 16384 (order: 4, 65536 bytes)
Memory: 126476k/131060k available (1690k kernel code, 4044k reserved,
732k data)Checking if this processor honours the WP bit even in
supervisor mode... Ok.
Mount-cache hash table entries: 512 (order: 0, 4096 bytes)
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 128K
Intel machine check architecture supported.
Intel machine check reporting enabled on CPU#0.
CPU: Intel Celeron (Mendocino) stepping 00
Enabling fast FPU save and restore... done.
Checking 'hlt' instruction... OK.

END OF LOG