Re: [patch] fixed 2.2.1 inode-leakage due bogus design of the free_inodes algorithm [was [Re: [shows

Andrea Arcangeli (andrea@e-mind.com)
Thu, 4 Feb 1999 13:17:20 +0100 (CET)


On Thu, 4 Feb 1999, Oleg Drokin wrote:

> This does not work at all!

Ugh strange works fine here...

> angband:/tmp$ cp -a test test1
> (test is directory with 12 subdirs, with 10k files in each)

Copying the kernel tree is enough as test?

> and machine is dead in few seconds (well, sysrq works, but this is only thing
> that works) When I compile in 2 printks, one in inode.c to show allocated page,
> and one in mm/filemap.c, to show when page is freed, I see a lot of
> pages allocated, then a lot freed and death. sysrq-u does not work.

Could you run `while :; do cat /proc/sys/fs/inode-state; done` in an xterm
before starting the copy? Then could you tell me the last inode-state
before that the machine is lockedup?

Thanks.

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/