[ltp] SuSE 8.2 kernel patches
Valient Gough
linux-thinkpad@linux-thinkpad.org
24 Jun 2003 13:50:39 -0700
--=-lKube9A/DCFNbTIc2lPh
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Hmm, I think maybe you've misread. I said that what the benchmarks
don't show is just how much more responsive the system is during disk
intensive tasks..
The patch contains just the minimal amount of stuff needed to enable DMA
support. It doesn't contain any of the other IDE changes from the
2.4.21-* series.
You can check that DMA is enabled using hdparm. Also, turning on
unmaskirq helps responsiveness a lot:
> sudo hdparm /dev/hda
/dev/hda:
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 5168/240/63, sectors = 78140160, start = 0
Try hdparm -d 1 -u 1 /dev/hda , to enable DMA and the unmaskirq flag.
regards,
Val
On Tue, 2003-06-24 at 13:12, Steve Krulewitz wrote:
> I noticed on your web page you mentioned in your 2003-05-11 update
> that even with the IDE files from 2.4.21-rc2, the system was still
> sluggish during disk intensive tasks. I noticed that on my t40p last
> night as well -- I'm using the 2.4.21-laptopkernel and I believe DMA
> is enabled on my disk. Do your new patches address this problem? Or
> is it just a fact of life?
>
> cheers,
> -steve
--=-lKube9A/DCFNbTIc2lPh
Content-Type: text/html; charset=utf-8
Content-Transfer-Encoding: 7bit
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 TRANSITIONAL//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; CHARSET=UTF-8">
<META NAME="GENERATOR" CONTENT="GtkHTML/3.0.5">
</HEAD>
<BODY>
<BR>
Hmm, I think maybe you've misread. I said that what the benchmarks don't show is just how much more responsive the system is during disk intensive tasks..<BR>
<BR>
The patch contains just the minimal amount of stuff needed to enable DMA support. It doesn't contain any of the other IDE changes from the 2.4.21-* series.<BR>
<BR>
You can check that DMA is enabled using hdparm. Also, turning on unmaskirq helps responsiveness a lot:<BR>
<BR>
<BLOCKQUOTE>
<PRE>> sudo hdparm /dev/hda
/dev/hda:
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 5168/240/63, sectors = 78140160, start = 0</PRE>
</BLOCKQUOTE>
<BR>
Try hdparm -d 1 -u 1 /dev/hda , to enable DMA and the unmaskirq flag.<BR>
<BR>
regards,<BR>
Val<BR>
<BR>
<BR>
On Tue, 2003-06-24 at 13:12, Steve Krulewitz wrote:
<BLOCKQUOTE TYPE=CITE>
<FONT COLOR="#68151e"><I>I noticed on your web page you mentioned in your 2003-05-11 update that even with the IDE files from 2.4.21-rc2, the system was still sluggish during disk intensive tasks. I noticed that on my t40p last night as well -- I'm using the 2.4.21-laptopkernel and I believe DMA is enabled on my disk. Do your new patches address this problem? Or is it just a fact of life?<BR>
<BR>
cheers,<BR>
-steve</I></FONT>
</BLOCKQUOTE>
</BODY>
</HTML>
--=-lKube9A/DCFNbTIc2lPh--