On Wed, Mar 19, 2003 at 09:01:57AM +0000, John Bradford wrote:
> Alan - I've cc'ed you in on this, because I think it might point to a
> more subtle disk/controller interaction/bug. Feel free to ignore if
> it's not :-).
Hi Alan, its the Intel ICH4 chipset, I've tried it on 2.4.21-pre5-ac3 too,
to make sure its not something funny in the 2.4.21-pre5-gss kernel, same
problem occurred.
I've attached the lspci -vv output, and the /proc/ide/hda/settings file.
> No, you should get the same results.
On further investigation, I do get valid results on /dev/hdb, but i can't
get /dev/hda (the WDC WD800JB-00CRA1) to give me sane test results. i've disabled
S.M.A.R.T. in the BIOS, done a cold boot, and not poked at the drive at all
with smartctl before running hdparm, and i still get the invalid results:
/dev/hda:
Timing buffer-cache reads: -3008 MB in 0.00 seconds = -inf kB/sec
Timing buffered disk reads: -1504 MB in 0.00 seconds = -inf kB/sec
Hmm.. suspicious results: probably not enough free memory for a proper test.
/dev/hdb:
Timing buffer-cache reads: 128 MB in 0.27 seconds =474.07 MB/sec
Timing buffered disk reads: 64 MB in 1.40 seconds = 45.71 MB/sec
I can't feel a noticable difference in speed for hda though.
leon.
-- in the beginning, was the code.
This archive was generated by hypermail 2b29 : Sun Mar 23 2003 - 22:00:28 EST