>>> On Fri, 16 Sep 2005 10:40:30 -0500, Dave Kleikamp
>>> <[EMAIL PROTECTED]> said:

[ ... crash on mounting JFS partition with '-o noatime' ... ]

>> Spoke too soon, it happened again, and I got a backtrace (I
>> haven't applied the patch you just sent me yet). It does not
>> look directly JFS related.

shaggy> No, I googled and found a similar oops reported on a -ck
shaggy> kernel with reiserfs: [ ... ] I haven't read the whole
shaggy> thread, so I don't know what's known about it yet.

Looks a bit different as to where it arises. However it looks
indeed like it is somewhere else.

[ ... ]

shaggy> Try my patch and avoid anticipatory, and I think that
shaggy> you may be okay.

Well, the patch and staying clear of 'anticipatory' seem to be
doing fine, so far so good. I have had some other problems with
'ancipatory', so I am suspicious of it too. For interactive use
I think that 'cfq' is better anyhow, so that's my default.

Unfortunately I am trying to do some simple file system driver
performance tests, and 'anticipatory' is the kernel default
elevator, and thus presumably the one that matters to most
people.

I am considering doing the test with 'noop', but that's not fair
either, as it seems some file systems are designed assuming some
sort of elevator sorting does happen.



-------------------------------------------------------
SF.Net email is sponsored by:
Tame your development challenges with Apache's Geronimo App Server. Download
it for free - -and be entered to win a 42" plasma tv or your very own
Sony(tm)PSP.  Click here to play: http://sourceforge.net/geronimo.php
_______________________________________________
Jfs-discussion mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/jfs-discussion

Reply via email to