Re: 7200 RPM udma33 same speed as 5400 PIO mode 4
just because the drive has higher rpms does not mean its faster, drive
technology (esp data density) has a major impact on performance, if the
5400rpm drive can store more data in less physical area it will perform
faster then the 7200rpm. ive seen benchmarks that showed more modern 5400
drives outrunning the first couple generations of ibm 7200rpm drives. i
dont know how old your drives are but figure it'd be worth pointing out.
nate
On Thu, 10 Feb 2000, Rick Macdonald wrote:
rickma >On Thu, 10 Feb 2000 dan@cg619985-a.adubn1.nj.home.com wrote:
rickma >
rickma >> On Thu, Feb 10, 2000 at 08:40:29AM -0700, Rick Macdonald generated a stream of 1s and 0s:
rickma >> >
rickma >> > I added a Maxtor 27GB 7200rpm ATA66 drive to my P200MMX when I upgraded to
rickma >> > potatos and kernel 2.2.14.
rickma >> >
rickma >> > My motherboard (ASUS TX97-E) only supports udma mode 2 (33MB/sec), but I
rickma >> > find that this drive and my old WD 4GB (no UDMA) both show the same speed
rickma >> > of about 9.4MB/sec with hdparm -t. Even if the new drive is the only drive
rickma >> > on the primary IDE interface.
rickma >
rickma >> You can't expect any improvents unless you turn multcount to 16 or to
rickma >> whatever your drive supports, and switch I/O into 32bit mode. Just with
rickma >> those two turned on, my WD goes from 6 to 12 MB/s without using DMA.
rickma >> Also I recommend setting unmaskirq to on, because that frees up your CPU
rickma >> when it does disk transfer; the multcount setting saves you the number
rickma >> 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
rickma >> dating back to early Pentium and 486 machines, this issue really needs
rickma >> to be addressed on distribution level, since most people don't bother
rickma >> playing with hdparm at all, they're always SLOW. What's ironic, is you
rickma >> can configure your kernel to enable DMA, but can't enable 32 bit I/O.
rickma >
rickma >Sheesh! That sounded so good, but as shown below, those settings made no
rickma >difference. I tried again with just the one drive on the rimary IDE (no
rickma >slave). I tried -X34 too, but it hung the machine. Fortunately, no damage
rickma >was done when I hit the reset switch.
rickma >
rickma >Any other ideas? Could I be missing kernel config options, or would hdparm
rickma >know that and not set the values?
rickma >
rickma >timshel# hdparm -m16 -c1 -u1 -i -v /dev/hda
rickma >
rickma >/dev/hda:
rickma > setting 32-bit I/O support flag to 1
rickma > setting multcount to 16
rickma > setting unmaskirq to 1 (on)
rickma > multcount = 16 (on)
rickma > I/O support = 1 (32-bit)
rickma > unmaskirq = 1 (on)
rickma > using_dma = 1 (on)
rickma > keepsettings = 0 (off)
rickma > nowerr = 0 (off)
rickma > readonly = 0 (off)
rickma > readahead = 8 (on)
rickma > geometry = 3322/255/63, sectors = 53369568, start = 0
rickma >
rickma > Model=Maxtor 92732U8, FwRev=RA530JN0, SerialNo=H8059G4C
rickma > Config={ Fixed }
rickma > RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=57
rickma > BuffType=3(DualPortCache), BuffSize=2048kB, MaxMultSect=16, MultSect=16
rickma > DblWordIO=no, OldPIO=2, DMA=yes, OldDMA=0
rickma > CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=53369568
rickma > tDMA={min:120,rec:120}, DMA modes: mword0 mword1 mword2
rickma > IORDY=on/off, tPIO={min:120,w/IORDY:120}, PIO modes: mode3 mode4
rickma > UDMA modes: mode0 mode1 *mode2 mode3 mode4
rickma >
rickma >timshel# hdparm -t /dev/hda;hdparm -t /dev/hda
rickma >
rickma >/dev/hda:
rickma > Timing buffered disk reads: 64 MB in 6.99 seconds = 9.16 MB/sec
rickma >
rickma >/dev/hda:
rickma > Timing buffered disk reads: 64 MB in 6.77 seconds = 9.45 MB/sec
rickma >You have mail in /var/spool/mail/rickm
rickma >timshel# hdparm -m0 -c0 -u0 -i -v /dev/hda
rickma >
rickma >/dev/hda:
rickma > setting 32-bit I/O support flag to 0
rickma > setting multcount to 0
rickma > setting unmaskirq to 0 (off)
rickma > multcount = 0 (off)
rickma > I/O support = 0 (default 16-bit)
rickma > unmaskirq = 0 (off)
rickma > using_dma = 1 (on)
rickma > keepsettings = 0 (off)
rickma > nowerr = 0 (off)
rickma > readonly = 0 (off)
rickma > readahead = 8 (on)
rickma > geometry = 3322/255/63, sectors = 53369568, start = 0
rickma >
rickma > Model=Maxtor 92732U8, FwRev=RA530JN0, SerialNo=H8059G4C
rickma > Config={ Fixed }
rickma > RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=57
rickma > BuffType=3(DualPortCache), BuffSize=2048kB, MaxMultSect=16, MultSect=off
rickma > DblWordIO=no, OldPIO=2, DMA=yes, OldDMA=0
rickma > CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=53369568
rickma > tDMA={min:120,rec:120}, DMA modes: mword0 mword1 mword2
rickma > IORDY=on/off, tPIO={min:120,w/IORDY:120}, PIO modes: mode3 mode4
rickma > UDMA modes: mode0 mode1 *mode2 mode3 mode4
rickma >
rickma >timshel# hdparm -t /dev/hda;hdparm -t /dev/hda
rickma >
rickma >/dev/hda:
rickma > Timing buffered disk reads: 64 MB in 6.90 seconds = 9.28 MB/sec
rickma >
rickma >/dev/hda:
rickma > Timing buffered disk reads: 64 MB in 6.93 seconds = 9.24 MB/sec
rickma >
rickma >...RickM...
rickma >
rickma >
rickma >--
rickma >Unsubscribe? mail -s unsubscribe debian-user-request@lists.debian.org < /dev/null
rickma >
----------------------------------------[mailto:aphro@aphroland.org ]--
Vice President Network Operations http://www.firetrail.com/
Firetrail Internet Services Limited http://www.aphroland.org/
Everett, WA 425-348-7336 http://www.linuxpowered.net/
Powered By: http://comedy.aphroland.org/
Debian 2.1 Linux 2.0.36 SMP http://yahoo.aphroland.org/
-----------------------------------------[mailto:aphro@netquest.net ]--
9:39pm up 175 days, 9:54, 2 users, load average: 1.01, 1.03, 1.00
Reply to: