2.2.7_andrea2.bz2 [Re: cache killer memory death test - 2.2.6 vs

Andrea Arcangeli (andrea@e-mind.com)
Thu, 29 Apr 1999 21:21:36 +0200 (CEST)


On Thu, 29 Apr 1999, Andrea Arcangeli wrote:

>I also released a 2.2.7_andrea1.bz2 but don't try it!!! because since I
>merged 2.2.7 my VM tree is unstable under high swap load. I don't think

I fixed the bug. I forgot one of my lru_refile_unmapped_cache() in
unuse_pte ;). So when delete_from_swap_cache() was running, it was doing a
list_del(page->lru) it had page->lru pointing to the old ->lru
not-valid-anymore values...

Since I was in swapfile.c I also implemented a smarter cluster logic for
the swap allocation.

Now everything seems rock solid also after a swapoff -a.

So I would ask if you could try out 2.2.7_andrea2.bz2 and see if it's
responsive as 2.2.6_andrea2. I think it should be equally repsonsive but
quite faster. Let me know. And thanks for the great feedback!! ;))

ftp://e-mind.com/pub/andrea/kernel/2.2.7_andrea2.bz2

Andrea Arcangeli

-
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/