Dog slow IDE

Dog slow IDE

[Tony Hoyle]

For some reason the my IDE is running extremely slow (which accounts for 
why this box feels so sluggish).

hdparm gives:

/dev/hda:

  Model=Maxtor 53073U6, FwRev=DA6207V0, SerialNo=K607RFNC
  Config={ Fixed }
  RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=57
  BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=16
  CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=60030432
  IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
  PIO modes: pio0 pio1 pio2 pio3 pio4
  DMA modes: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 *udma4
  AdvancedPM=yes: disabled (255) WriteCache=enabled
  Drive Supports : ATA/ATAPI-4 T13 1153D revision 17 : ATA-1 ATA-2 ATA-3 
ATA-4 ATA-5

/dev/hdb:

  Model=MAXTOR 4K040H2, FwRev=A08.1500, SerialNo=572124112782
  Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
  RawCHS=16383/16/63, TrkSize=32256, SectSize=21298, ECCbytes=4
  BuffType=DualPortCache, BuffSize=2000kB, MaxMultSect=16, MultSect=16
  CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=78198750
  IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
  PIO modes: pio0 pio1 pio2 pio3 pio4
  DMA modes: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 *udma5
  AdvancedPM=no WriteCache=enabled
  Drive Supports : ATA/ATAPI-5 T13 1321D revision 1 : ATA-1 ATA-2 ATA-3 
ATA-4 ATA-5

Both drives claim to be using UDMA if I read the above correctly.

However:

/dev/hda:
  Timing buffered disk reads:  64 MB in 16.27 seconds =  3.93 MB/sec

/dev/hdb:
  Timing buffered disk reads:  64 MB in 32.99 seconds =  1.94 MB/sec

1.94MB/sec is *way* to slow for a UDMA5 hard disk surely?

Tony

Re: Dog slow IDE

[Vojtech Pavlik]

Hi!

What does haparm /dev/hda (without the -i) say? Is it using DMA at all?
It may be set up for UDMA4 or UDMA5 but still run PIO only ... btw what
chipset is this?

Vojtech

On Mon, Mar 11, 2002 at 01:40:59PM +0000, Tony Hoyle wrote:
> For some reason the my IDE is running extremely slow (which accounts for 
> why this box feels so sluggish).
> 
> hdparm gives:
> 
> /dev/hda:
> 
>   Model=Maxtor 53073U6, FwRev=DA6207V0, SerialNo=K607RFNC
>   Config={ Fixed }
>   RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=57
>   BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=16
>   CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=60030432
>   IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
>   PIO modes: pio0 pio1 pio2 pio3 pio4
>   DMA modes: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 *udma4
>   AdvancedPM=yes: disabled (255) WriteCache=enabled
>   Drive Supports : ATA/ATAPI-4 T13 1153D revision 17 : ATA-1 ATA-2 ATA-3 
> ATA-4 ATA-5
> 
> /dev/hdb:
> 
>   Model=MAXTOR 4K040H2, FwRev=A08.1500, SerialNo=572124112782
>   Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
>   RawCHS=16383/16/63, TrkSize=32256, SectSize=21298, ECCbytes=4
>   BuffType=DualPortCache, BuffSize=2000kB, MaxMultSect=16, MultSect=16
>   CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=78198750
>   IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
>   PIO modes: pio0 pio1 pio2 pio3 pio4
>   DMA modes: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 *udma5
>   AdvancedPM=no WriteCache=enabled
>   Drive Supports : ATA/ATAPI-5 T13 1321D revision 1 : ATA-1 ATA-2 ATA-3 
> ATA-4 ATA-5
> 
> Both drives claim to be using UDMA if I read the above correctly.
> 
> However:
> 
> /dev/hda:
>   Timing buffered disk reads:  64 MB in 16.27 seconds =  3.93 MB/sec
> 
> /dev/hdb:
>   Timing buffered disk reads:  64 MB in 32.99 seconds =  1.94 MB/sec
> 
> 1.94MB/sec is *way* to slow for a UDMA5 hard disk surely?
> 
> Tony
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

-- 
Vojtech Pavlik
SuSE Labs

Re: Dog slow IDE

[Tony Hoyle]

Vojtech Pavlik wrote:
> Hi!
> 
> What does haparm /dev/hda (without the -i) say? Is it using DMA at all?
> It may be set up for UDMA4 or UDMA5 but still run PIO only ... btw what
> chipset is this?
> 
Judging by the pci ids database it's either an AMD Unknown or AMD Opus 
chipset.

hdparm says DMA is enabled (it's enabled by a script a bootup AFAIK).

Installing smartd seems to have helped a lot - it's jumped to 28MB/32MB 
which is a hell of a lot closer to what I was expecting (not sure if 
ATA133 should give 133MB a second or whether that's just a theoretical 
limit).

Not sure why enabling smart would help that much (I'm still working my 
way around the BIOS settings and the way to enable it was far from 
obvious - you have to do a manual config of the hard drive first).

Tony

-- 
"Wipe Info uses hexadecimal values to wipe files. This provides more 
security than wiping with decimal values." -- Norton SystemWorks 2002 Manual

tmh@nothing-on.tv 
http://www.nothing-on.tv

Re: Dog slow IDE

[Mark Cooke]

Hi Tony,

Just a me too earlier today with 2.4.19-pre2-ac4, on a dual celeron
440BX-based system.

Im my case, I have 3 hdds in a raid-5, hda, hdb, and hdc.

hda and hdb were both performing as expected 18-22MB/sec.  hdc was 
down at the 1.8-2.2MB/sec - even after trying to use hdparm to reset
the drive transfer rates (-p, -d1 -X66, -d0 -X12 (to try pio) etc).

None of the attempts using hdparm altered the hdc transfer rate in any
noticable fashion.  I powered down the machine, and on reboot, the hdc
transfer rate returned to the expected range 18-22MB/sec), and the
raid performance went up to it's usual level from the 5MB/sec it was
showing.

I couldn't see any reason for the odd behaviour, and I had put it down
to 'just one of those things' as the 3 or 4 test reboots since haven't
produced the strangely low-transfer rate again.

Cheers,

Mark

On Mon, 11 Mar 2002, Tony Hoyle wrote:

> For some reason the my IDE is running extremely slow (which accounts for 
> why this box feels so sluggish).

<snip>

> However:
> 
> /dev/hda:
>   Timing buffered disk reads:  64 MB in 16.27 seconds =  3.93 MB/sec
> 
> /dev/hdb:
>   Timing buffered disk reads:  64 MB in 32.99 seconds =  1.94 MB/sec
> 
> 1.94MB/sec is *way* to slow for a UDMA5 hard disk surely?

Yes.  On another system, I pull 35-40MB/sec off each individual
barracuda IV drive, and 50+ from the combined raid array (I assume PCI 
bandwidth limited, as twiddling BIOS PCI settings boosted this from 41 
to 50).

Regards,

Mark

-- 
+-------------------------------------------------------------------------+
Mark Cooke                  The views expressed above are mine and are not
Systems Programmer          necessarily representative of university policy
University Of Birmingham    URL: http://www.sr.bham.ac.uk/~mpc/
+-------------------------------------------------------------------------+

Re: Dog slow IDE

[Adam Keys]

On March 11, 2002 07:40, Tony Hoyle wrote:
> For some reason the my IDE is running extremely slow (which accounts for
> why this box feels so sluggish).

Are they dog slow after a *cold* start?  My solution to slow Maxtor drives 
was to fix the CPU fans in my machine and put the main drive in a bay cooler. 
 The drive wasn't spewing bad data, so it wasn't the drive controller 
overheating.  That leads me to believe that the motor (or the motor control 
DSP) was overheating and degrading performance on that route.  Trying that 
will save you lots of hours toying with hdparm like I did :).

Good luck..its a sucky problem but once you get it fixed your machine will 
seem like new!
-- 
akk~

Re: Dog slow IDE

[Tony Hoyle]

Adam Keys wrote:

> Are they dog slow after a *cold* start?  My solution to slow Maxtor drives 
> was to fix the CPU fans in my machine and put the main drive in a bay cooler. 
>
Heck, the machine won't even boot after a cold start....  depending on 
the ambient temperature it can take anything from a few seconds to a few 
minutes to 'warm up'.  I think this will be my last Asus board for a 
while...

Tony

-- 
"Wipe Info uses hexadecimal values to wipe files. This provides more 
security than wiping with decimal values." -- Norton SystemWorks 2002 Manual

tmh@nothing-on.tv 
http://www.nothing-on.tv