Hello folks.
Some of you may remember my trouble with SI0680-based ULTRADMA ATA
controller card. Well, the problem was obvivously the faulty card.
After replacing it works fine.
I don't know what magic Soeren has put in the SI driver, but unlike
Windows it never crashed, never hang, it even tried
It seems Buckie wrote:
> Hello folks.
> Some of you may remember my trouble with SI0680-based ULTRADMA ATA
> controller card. Well, the problem was obvivously the faulty card.
> After replacing it works fine.
>
> I don't know what magic Soeren has put in the SI driver, but unlike
> Windows it neve
John-Mark Gurney wrote:
> Ruben de Groot wrote this message on Fri, Aug 01, 2003 at 10:15 +0200:
> > On Fri, Aug 01, 2003 at 04:33:08AM +0200, mh typed:
> > The following comparison is probably bogus, but can anybody explain the
> > huge difference?
>
> It's called micro optimization. Linux feels
> The transfer rates are usually given for the outside of the disk I think.
Average transfer rates is what's usually given ([inside+outside]/2) - at least
that's what manufacturers say. However, as they usually seem too high, probably
the data rate for the first few thousand sectors are given.
01
Ruben de Groot wrote this message on Fri, Aug 01, 2003 at 10:15 +0200:
> On Fri, Aug 01, 2003 at 04:33:08AM +0200, mh typed:
> The following comparison is probably bogus, but can anybody explain the
> huge difference?
It's called micro optimization. Linux feels the need to special case
/dev/zero
On Fri, Aug 01, 2003 at 04:33:08AM +0200, mh typed:
> Soeren Schmidt wrote:
>
> >It seems Buckie wrote:
> >
> >
> >>I ran a dd:
> >>dd if=/dev/ad1 of=/dev/zero ibs=8192
> >>
> >>OH, you need to output to /dev/null NOT /dev/zero :)
> >>
> >>
> It doesn't matter as far as cdevsw entry for zer
On Thu, Jul 31, 2003 at 06:33:28PM -0600, [EMAIL PROTECTED] wrote:
>
> > > Maybe not, but they do give a transferspeed from medium range and that
> > > is what can be expected.
> >
> > Hmm, I guess not everyone does that. We have some seagates here at work we
> > were wondering about because they
Soeren Schmidt wrote:
It seems Buckie wrote:
I ran a dd:
dd if=/dev/ad1 of=/dev/zero ibs=8192
OH, you need to output to /dev/null NOT /dev/zero :)
It doesn't matter as far as cdevsw entry for zero-device write leads
to null_write() in /usr/src/sys/dev/null.c
I was testing the memory t
> > Maybe not, but they do give a transferspeed from medium range and that
> > is what can be expected.
>
> Hmm, I guess not everyone does that. We have some seagates here at work we
> were wondering about because they seemed too slow, and we couldn't find
> anything aside from what we already kne
Wow. But still, can it be that something bottlenecks somewhere so that
I can't get these loverly speeds? Can I measure pci performance as
well? Would be cool to understand what doesn't allow this system to
use the 133 to its max...
Z
>> > Guess I just didn't look hard enough or in the right place
> > Guess I just didn't look hard enough or in the right place. I only spent a
> > minute or 2 on it, but yeah ATA drives don't tell you that sort of thing,
> > they say stuff like "ATA133 for a max transfer rate of 133 MB/sec *" then
> > at the bottom the * says something like "133 MB/sec burst ra
It seems Kenneth Culver wrote:
> >
> Guess I just didn't look hard enough or in the right place. I only spent a
> minute or 2 on it, but yeah ATA drives don't tell you that sort of thing,
> they say stuff like "ATA133 for a max transfer rate of 133 MB/sec *" then
> at the bottom the * says somethin
> http://www.maxtor.com/en/products/scsi/atlas_10k_family/atlas_10k_iv/index.htm
> "maximum sustained data transfer rate up to 72MB/sec."
>
> http://www.seagate.com/cda/products/discsales/enterprise/family/0,1086,530,00.html
> Lists not only sustained transfer rate, but tells you the center and
> e
In the last episode (Jul 31), Kenneth Culver said:
> > What I find fascinating is that Maxtor's site never actually tells you
> > the true throughput of that disk anywhere.
> > http://www.maxtor.com/en/products/ata/desktop/diamondmax_plus_9/
>
> Almost none of the hard disk manufacturers do. In fa
> Maybe not, but they do give a transferspeed from medium range and that
> is what can be expected.
Hmm, I guess not everyone does that. We have some seagates here at work we
were wondering about because they seemed too slow, and we couldn't find
anything aside from what we already knew... the tra
It seems Kenneth Culver wrote:
> > What I find fascinating is that Maxtor's site never actually tells you
> > the true throughput of that disk anywhere.
> > http://www.maxtor.com/en/products/ata/desktop/diamondmax_plus_9/
>
> Almost none of the hard disk manufacturers do. In fact I've never seen
>
> What I find fascinating is that Maxtor's site never actually tells you
> the true throughput of that disk anywhere.
> http://www.maxtor.com/en/products/ata/desktop/diamondmax_plus_9/
Almost none of the hard disk manufacturers do. In fact I've never seen
"true throughput" numbers from ANY manufac
In the last episode (Jul 31), Buckie said:
> Soeren,
>
> Heh, QNX in first RTP incarnations only had /dev/zero and I sort of
> get used to it.
> But nevermind. I've already tried large bsizes (1m and more) and
> /dev/null too. But I can't really get past the 16Mb/s barrier...
What I find fascinat
Soeren,
Heh, QNX in first RTP incarnations only had /dev/zero and I sort of
get used to it.
But nevermind. I've already tried large bsizes (1m and more) and
/dev/null too. But I can't really get past the 16Mb/s barrier...
Does machine speed really should affect the hard drive performance so
much?
It seems Buckie wrote:
> I ran a dd:
> dd if=/dev/ad1 of=/dev/zero ibs=8192
> ...and cancelled it after some amount of time:
> 2321+0 records in
> 37136+0 records out
> 19013632 bytes transferred in 27.876118 secs (682076 bytes/sec)
>
> I don't use static2 ata device numbering hence Maxtor is assi
Hello again. Sorry I just forgot to provide at least a version number.
Stupid me. It's FreeBSD 5.1 RELEASE and here's a dmesg output.
Also, it's not that it 'doesn't work', it works, but I see no
difference in speed after changing transfer modes (from WDMA2 to
UDMA6)
Copyright (c) 1992-2003 The Fr
quot; <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Thursday, July 31, 2003 12:50 PM
Subject: Ultra ATA card doesn't seem to provide Ultra speeds.
> Hello hackers!
>
> I hope someone can help me with this since you know the internals of
> FreeBSD much more than me. Hop
It seems Buckie wrote:
> Hello hackers!
>
> I hope someone can help me with this since you know the internals of
> FreeBSD much more than me. Hope it doesn't take much time.
>
> The sort of problem I'm having is this: after installing a new Ultra
> ATA card (it's based on Silicon Image 0680 chip
Hello hackers!
I hope someone can help me with this since you know the internals of
FreeBSD much more than me. Hope it doesn't take much time.
The sort of problem I'm having is this: after installing a new Ultra
ATA card (it's based on Silicon Image 0680 chip by the way) although
the disks start
24 matches
Mail list logo
