Messages in this thread | | | From | jmcmullan@linuxcar ... | Subject | Re: OOM in 2.2.14 | Date | 14 Jul 2000 19:37:10 GMT |
| |
Rik van Riel <riel@conectiva.com.br> wrote: > As long as there is enough _freeable_ memory around we must not > kill anything with OOM. In fact, my OOM code doesn't kill anything > as long as do_try_to_free_pages is able to free the pages it needs.
For in-kernel OOM, this makes sense. For a pro-active user-land version, we would start warning at a low-water mark, and killing at some high water mark.
> It is a very common situation that systems have no free memory but > tons of easily _freeable_ memory, like cache or pages which are > duplicated in both swap and memory. In that case it is just not > acceptable to start killing processes.
Cache is not included in my VM measurement. It's just pages currently allocated to processes.
> Also, how is your "90%" measure applicable to machines without > swap?
90% of RAM is used.
(In case I haven't been clear, when I say ``X% of VM'' I mean:
X% = (pages used by processes in MEM + pages used by processes in SWAP) ------------------------------------------------------------------ * 100 (total pages of MEM + total pages of SWAP)
Clear as mud?
-- Jason McMullan, Senior Linux Consultant, Linuxcare, Inc. 412.422.8077 tel, 412.656.3519 cell, 415.701.0792 fax jmcmullan@linuxcare.com, http://www.linuxcare.com/ Linuxcare. Support for the revolution.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |