Hello all,
with the Fedora 1 kernel 2.4.22-1.2115.nptl I get with hdparm -t
(Timing buffered disk reads) 34 MB/sec. Its very slow for this drive.
With 2.6.0-test9 and -mm3 I get around "62 MB in 3.05 = 20,31". Wow"
Back to ~1998?
UDMA6 is always on. The Abit NF7-S V2 nForce2 board with an siimage
3112a (rev2) raid controller, new BIOS.
Also with the Seagate SATA V ST380023AS I get heavy crashes with
max_kb_per_request when I set it to 128 (all kernel). With 15kb its fine
and stable, but so slow.
The Seagate technical support means in an email to me, that there are no
problems with the SATA seagate drives, its only the driver ... Nice. Is
that really so? And why get other users with new Maxtor or Western
Digital SATA drives (in the same class) much better performance?
Thanks for all the good work,
Marcus
--
from the "Old Europe"
Marcus Hartig wrote:
> Hello all,
>
> with the Fedora 1 kernel 2.4.22-1.2115.nptl I get with hdparm -t
> (Timing buffered disk reads) 34 MB/sec. Its very slow for this drive.
>
> With 2.6.0-test9 and -mm3 I get around "62 MB in 3.05 = 20,31". Wow"
> Back to ~1998?
I have a similar problem: With 2.4.22-ac3 I had 37mb/sec with my Samsung
HD and 49MB/sec with IBM/Hitachi, now with 2.6 (all I tried, including
test9-mm2) I had only 20mb/sec for Samsung and about 39mb/sec for the
IBM. Motherboard is Abit NF7-S Rev2.0, as well, so same situation with
the siimage 1.06 driver. I wanted to run some dd tests as well, but it
is a real performance hit. Playing with readahead or other hdparm
options didn't help either.
Prakash
Prakash K. Cheemplavam wrote:
> I have a similar problem: With 2.4.22-ac3 I had 37mb/sec with my Samsung
> HD and 49MB/sec with IBM/Hitachi, now with 2.6 (all I tried, including
> test9-mm2) I had only 20mb/sec for Samsung and about 39mb/sec for the
> IBM. Motherboard is Abit NF7-S Rev2.0, as well, so same situation with
> the siimage 1.06 driver. I wanted to run some dd tests as well, but it
> is a real performance hit. Playing with readahead or other hdparm
> options didn't help either.
I get a tip from Mark Hahn to set the pci latency to 64. And wow, with
fedora 2.4.22 kernel I get then 41MB/sec with max_k_p_r 128. But,...
after copying big files I get ext3-fs erros, cannot read inode etc and a
bus error. Bumm!
Maybe it runs better with your harddrives.
Marcus
Marcus Hartig wrote:
> Prakash K. Cheemplavam wrote:
>
>> I have a similar problem: With 2.4.22-ac3 I had 37mb/sec with my
>> Samsung HD and 49MB/sec with IBM/Hitachi, now with 2.6 (all I tried,
>> including test9-mm2) I had only 20mb/sec for Samsung and about
>> 39mb/sec for the IBM. Motherboard is Abit NF7-S Rev2.0, as well, so
>> same situation with the siimage 1.06 driver. I wanted to run some dd
>> tests as well, but it is a real performance hit. Playing with
>> readahead or other hdparm options didn't help either.
>
>
> I get a tip from Mark Hahn to set the pci latency to 64. And wow, with
> fedora 2.4.22 kernel I get then 41MB/sec with max_k_p_r 128. But,...
> after copying big files I get ext3-fs erros, cannot read inode etc and a
> bus error. Bumm!
Is there a way to change the latency within Linux? I mean I don't want
to ruin my Windows, as it works w/o problems. Nevertheless I rather make
a backup of my Linux install before messing with that...
Prakash
Prakash K. Cheemplavam wrote:
> Is there a way to change the latency within Linux? I mean I don't want
With setpci and lspci?
> to ruin my Windows, as it works w/o problems. Nevertheless I rather make
> a backup of my Linux install before messing with that...
Thats always good... ;-)
Marcus
--
today until 20:15 CET 2003 Scotland soccer fan
Marcus Hartig wrote:
> Prakash K. Cheemplavam wrote:
>
>> Is there a way to change the latency within Linux? I mean I don't want
>
>
> With setpci and lspci?
Well, I treid setting latenc to 64 even 128, but hd-speed didn't
improve. 23MB/s was max. What else did you set? I just set the siimage
controller's latency higher.
Prakash