linux-kernel - Re: PATA-disk named sda

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:	Sat, 07 Jul 2007 23:40:33 +0200
From:	Rene Herman <rene.herman@...il.com>
To:	Christoph Pleger <Christoph.Pleger@...uni-dortmund.de>
CC:	linux-kernel@...r.kernel.org
Subject: Re: PATA-disk named sda

On 07/06/2007 02:30 PM, Christoph Pleger wrote:

> And what about hdparm (setting 32bit I/O and multi-sector mode)? Suren 
> wrote that 32bit I/O makes no sense when using DMA. Maybe that's right, 
> but it does not correspond with my experiences. At least, I have the 
> "feeling" that my IDE disks work much faster since I enabled 32bit 
> support (DMA already was on before).

hdparm -t /dev/hda (or /dev/sda -- it works for the SD interface as well) is 
a quick test of a drive's sequential read speed.

I have, at the time, noticed at least on older controllers/drives (Intel 430 
generation chipsets with things like 8G UDMA33 disks) that I could reliably 
increase the result with something like 1MB/s (to a total of 6 to 8, so it 
wasn't insignificant) by enabling 32-bit I/O. Had also understood that it 
shouldn't make a difference with DMA, but just went "oh well" and stuck a 
"hdparm -c1" in my bootup scripts.

(if anyone tries; note that hdparm -a can have a large effect on that result 
as well on some setups -- on machines where it does, -a 1024 usually gives 
me best results)

Rene.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/