<br><br><div class="gmail_quote">On Wed, Sep 14, 2011 at 13:28, Bruno Wolff III <span dir="ltr"><<a href="mailto:bruno@wolff.to">bruno@wolff.to</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<div class="im">On Wed, Sep 14, 2011 at 13:46:40 -0700,<br>
Adam Williamson <<a href="mailto:awilliam@redhat.com">awilliam@redhat.com</a>> wrote:<br>
><br>
> They weren't; see my numbers in the related bug<br>
> (<a href="https://bugzilla.redhat.com/show_bug.cgi?id=735268" target="_blank">https://bugzilla.redhat.com/show_bug.cgi?id=735268</a> ). The debug<br>
> overhead appears to have gotten much heavier in 3.1.<br>
<br>
</div>I added myself to that bug, though I wasn't really seeing that much of<br>
a slow down in graphics. (I have an rv280 which might not have the issue<br>
that later radeon cards were having.) My issue seems to be related to<br>
disk I/O. I am not sure if it is heavier cpu use in journalling, raid or<br>
encryption or if the actual I/O is slower. But it seems that I/O heavy<br>
activities such as yum updates and kernel builds are taking several times<br>
longer than they were a few weeks ago (in rawhide).<br>
<div><div></div><div class="h5"><br></div></div></blockquote><div><br></div><div>I'm using the latest rawhide kernel. During times of heavy I/O the system becomes very unresponsive. The cursor will stop moving long enough to count to between 5 and 10. I know it happens with heavy network I/O so it might be the particular driver in my case. Not sure if the slow-down is true for other types of I/O. This has been true for all of the recent rawhide kernels.</div>
<div><br></div><div>darrell</div></div><br>