Re: [PATCH] fix spurious OOM kills

[Martin MOKREJŠ]

Hi,
 I just tested 2.6.7 and it works comparably to 2.4.28.
swpd value reported by vmstat is actually 1172736,
actually inspecting the previous results from 2.4.28
show also this number. But "swapon -s" reports 1172732.
I'm puzzled. vmstat comes from gentoo procps-3.2.3-r1.


 Anyway, plain 2.6.7 kills only the application asking for
so much memory and logs via syslog:
Out of Memory: Killed process 58888 (RNAsubopt)

 It's a lot better compared to what we have in 2.6.10-rc2,
from my user's view.

 I cannot easily reverse the tbtc patch from 2.6.10-rc2 tree,
but applying the tbtc patch over 2.6.7 gives me same behaviour
as on plain 2.6.7 - so only the RNAsubopt application get's
killed.

 The problem must have been introduced between 2.6.7
and 2.6.10-rc2 but is not directly related to tbtc patch in it's
original form:
http://marc.theaimsgroup.com/?l=linux-kernel&m=109122597407401&w=2

Hope this helps.
Martin
 
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/