On Fri, 23 Mar 2012, Jason Mattax wrote:

On 03/23/2012 06:05 AM, Hugh Dickins wrote:

quoted

I'm not surprised that you saw no problem on 2.6.32.27, but I am
very surprised that you see the problem on 2.6.33.1 - I'm wondering
if that's a typo for something else, or a distro kernel which actually
contains changes from later releases?

I can't say why I saw it then, but I got the 2.6.33.1 kernel off of
http://www.kernel.org/pub/linux/kernel/v2.6/ so that I wouldn't have to worry
about distribution changes when reporting the bug here. I just recompiled the
source and verified that it is still affected even with the newest firmware.

Thank you for going to that trouble to narrow it down, thank you for
confirming, and I apologize for shedding doubt on your finding.

It appears that there's something else involved here, which I know nothing
about as yet; perhaps an earlier change in block layer handling of discard.

Although it's just history now, I ought to investigate once my Vertex2
arrives - well, several weeks later, I won't have time immediately.

Please would you clarify one thing.  2.6.33.1 was the earliest kernel
you saw the slowdown on.  I assume from that that you tried 2.6.33 itself,
and it did not show the slowdown.  Or were you testing releases in some
other order? if so please let me know the latest without the slowdown.

Thanks,
Hugh

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>