Re: slows to a crawl when hard drive is in use
I've also been running my iBook with interrupts unmasked for many months
without problems.
Is it possible to get UDMA66 working on the iBook? My version of the kernel
configures the drive for UDMA33. I did try forcing it to UDMA66 using hdparm,
and it utterly toasted the filesystem ;( I doubt it makes a major difference
to the performance, but it'd be nice to get it working.
Ben -- is it just accurate timings for the chipset that are required to get
UDMA66 working? Do you need access to the iBook hardware or would it be
possible to gather them with the assistance of an iBook owner?
Will
On Tue, May 21, 2002 at 12:22:42PM -0600, Will Aoki wrote:
>On Tue, May 21, 2002 at 06:15:40PM +0200, David N. Welton wrote:
>> Egidio Corsini <egidio.corsini@fastweb.it> writes:
>>
>> > Have you tryed to tune some parameters with hdparm?
>> > Mine looks like that
>> > /dev/hda:
>> > multcount = 0 (off)
>> > I/O support = 0 (default 16-bit)
>> > unmaskirq = 0 (off)
>> > using_dma = 1 (on)
>> > keepsettings = 0 (off)
>> > nowerr = 0 (off)
>> > readonly = 0 (off)
>> > readahead = 8 (on)
>> > geometry = 38760/16/63, sectors = 39070080, start = 0
>> > busstate = 1 (on)
>> >
>> > And they're still not optimized, but it works well.
>> > Try at least hdparm -d 1 /dev/hd? in order to enable dma
>>
>> These are the default settings, afaik. What I (and the previous
>> poster, I suspect) are curious about is 'unmaskirq', I guess, or
>> anything else that will make the system more usable when the hard
>> drive is at work.
>
>IRQ unmasking on my ibook hasn't made me lose any data yet (that I'm
>aware of...):
>
>/dev/hda:
> multcount = 16 (on)
> I/O support = 1 (32-bit)
> unmaskirq = 1 (on)
> using_dma = 1 (on)
> keepsettings = 0 (off)
> nowerr = 0 (off)
> readonly = 0 (off)
> readahead = 8 (on)
> geometry = 38760/16/63, sectors = 39070080, start = 0
> busstate = 1 (on)
>
>However, the system is only marginally more useful when there's heavy
>disk I/O. I run the same settings on a G4 desktop (unknown model) that
>acts as a mail server, and it sometimes bogs down quite badly for a few
>seconds at a time when there's heavy IMAP activity (although now that
>I've ditched wu-imap, it dosen't thrash as badly as before). If the mail
>server's settings caused data loss, I'd probably have heard about it by
>now from angry users.
>
>--
>William Aoki waoki@umnh.utah.edu /"\ ASCII Ribbon Campaign
>B1FB C169 C7A6 238B 280B <- key change \ / No HTML in mail or news!
>99AF A093 29AE 0AE1 9734 prev. expired X
> / \
>
>
>--
>To UNSUBSCRIBE, email to debian-powerpc-request@lists.debian.org
>with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
>
_________________________________________________________________________
William R Sowerbutts will@sowerbutts.com
Coder / Guru / Nrrrd http://sowerbutts.com
main(){char*s=">#=0> ^#X@#@^7=",c=0,m;for(;c<15;c++)for
(m=-1;m<7;putchar(m++/6&c%3/2?10:s[c]-31&1<<m?42:32));}
--
To UNSUBSCRIBE, email to debian-powerpc-request@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Reply to: