On Thu, 10 Feb 2000 [EMAIL PROTECTED] wrote:
> On Thu, Feb 10, 2000 at 08:40:29AM -0700, Rick Macdonald generated a stream
> of 1s and 0s:
> >
> > I added a Maxtor 27GB 7200rpm ATA66 drive to my P200MMX when I upgraded to
> > potatos and kernel 2.2.14.
> >
> > My motherboard (ASUS TX97-E) only supports udma mode 2 (33MB/sec), but I
> > find that this drive and my old WD 4GB (no UDMA) both show the same speed
> > of about 9.4MB/sec with hdparm -t. Even if the new drive is the only drive
> > on the primary IDE interface.
> You can't expect any improvents unless you turn multcount to 16 or to
> whatever your drive supports, and switch I/O into 32bit mode. Just with
> those two turned on, my WD goes from 6 to 12 MB/s without using DMA.
> Also I recommend setting unmaskirq to on, because that frees up your CPU
> when it does disk transfer; the multcount setting saves you the number
> of interrupts per transfer the size of this setting (i.e. 1 interrupt per 16
> blocks instead of 16 interrupts per 16 blocks) that you set it to, so it's
> very important, so is 32 bit transfer mode. The 16 bit mode is really an
> archaic setting
> dating back to early Pentium and 486 machines, this issue really needs
> to be addressed on distribution level, since most people don't bother
> playing with hdparm at all, they're always SLOW. What's ironic, is you
> can configure your kernel to enable DMA, but can't enable 32 bit I/O.
Sheesh! That sounded so good, but as shown below, those settings made no
difference. I tried again with just the one drive on the rimary IDE (no
slave). I tried -X34 too, but it hung the machine. Fortunately, no damage
was done when I hit the reset switch.
Any other ideas? Could I be missing kernel config options, or would hdparm
know that and not set the values?
timshel# hdparm -m16 -c1 -u1 -i -v /dev/hda
/dev/hda:
setting 32-bit I/O support flag to 1
setting multcount to 16
setting unmaskirq to 1 (on)
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 = 3322/255/63, sectors = 53369568, start = 0
Model=Maxtor 92732U8, FwRev=RA530JN0, SerialNo=H8059G4C
Config={ Fixed }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=57
BuffType=3(DualPortCache), BuffSize=2048kB, MaxMultSect=16, MultSect=16
DblWordIO=no, OldPIO=2, DMA=yes, OldDMA=0
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=53369568
tDMA={min:120,rec:120}, DMA modes: mword0 mword1 mword2
IORDY=on/off, tPIO={min:120,w/IORDY:120}, PIO modes: mode3 mode4
UDMA modes: mode0 mode1 *mode2 mode3 mode4
timshel# hdparm -t /dev/hda;hdparm -t /dev/hda
/dev/hda:
Timing buffered disk reads: 64 MB in 6.99 seconds = 9.16 MB/sec
/dev/hda:
Timing buffered disk reads: 64 MB in 6.77 seconds = 9.45 MB/sec
You have mail in /var/spool/mail/rickm
timshel# hdparm -m0 -c0 -u0 -i -v /dev/hda
/dev/hda:
setting 32-bit I/O support flag to 0
setting multcount to 0
setting unmaskirq to 0 (off)
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 = 3322/255/63, sectors = 53369568, start = 0
Model=Maxtor 92732U8, FwRev=RA530JN0, SerialNo=H8059G4C
Config={ Fixed }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=57
BuffType=3(DualPortCache), BuffSize=2048kB, MaxMultSect=16, MultSect=off
DblWordIO=no, OldPIO=2, DMA=yes, OldDMA=0
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=53369568
tDMA={min:120,rec:120}, DMA modes: mword0 mword1 mword2
IORDY=on/off, tPIO={min:120,w/IORDY:120}, PIO modes: mode3 mode4
UDMA modes: mode0 mode1 *mode2 mode3 mode4
timshel# hdparm -t /dev/hda;hdparm -t /dev/hda
/dev/hda:
Timing buffered disk reads: 64 MB in 6.90 seconds = 9.28 MB/sec
/dev/hda:
Timing buffered disk reads: 64 MB in 6.93 seconds = 9.24 MB/sec
...RickM...
