Discussion:
optimum disk performance
Craig Main
2003-10-30 14:17:45 UTC
Permalink
Hi all,

I have been looking at the speeds of the disks in some of the threads
lately. I noticed that mine is a lot slower than most. I have attached
some info. Is my disk running as optimally as it can or is there some
more tweaking I can do to improve performance?

laptop root # hdparm -tT /dev/hda

/dev/hda:
Timing buffer-cache reads: 644 MB in 2.00 seconds = 322.00 MB/sec
Timing buffered disk reads: 22 MB in 3.25 seconds = 6.77 MB/sec

laptop root # hdparm /dev/hda

/dev/hda:
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 0 (off)
using_dma = 0 (off)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 2432/255/63, sectors = 39070080, start = 0

laptop root # hdparm -i /dev/hda

/dev/hda:

Model=HITACHI_DK23CA-20, FwRev=00H1A0J1, SerialNo=12N68B
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=36477, SectSize=579, ECCbytes=4
BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=39070080
IORDY=yes, tPIO={min:400,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: sdma0 sdma1 sdma2 mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 *udma2 udma3 udma4 udma5
AdvancedPM=yes: mode=0x80 (128) WriteCache=enabled
Drive conforms to: ATA/ATAPI-5 T13 1321D revision 3:

* signifies the current active mode

What other hdparm flags can I _safely_ use, I don't want to trash my
disk.

Thanks and Regards
Craig

--
Craig Main
RHCE
ST Solutions
Cell: 082-323-4670
Home Tel: 011-760-1900
AIM: SATuxman; ICQ: 75815570
http://www.stsolutions.co.za
http://www.linuxsa.com


____________________________
I can't believe I ate the whole thing. -- Homer Simpson The Front
Mark Knecht
2003-10-30 14:35:40 UTC
Permalink
Post by Craig Main
laptop root # hdparm /dev/hda
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 0 (off)
using_dma = 0 (off)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 2432/255/63, sectors = 39070080, start = 0
See if you can get 'using_dma' and possibly 'unmaskirq' turned on for better
performance.

- Mark



--
gentoo-***@gentoo.org mailing list
Hall Stevenson
2003-10-30 14:38:25 UTC
Permalink
Post by Craig Main
using_dma = 0 (off)
What other hdparm flags can I _safely_ use, I don't want to trash my
disk.
The fact that DMA is *off* immediately jumped out at me. This can/should be
on with any or most modern HDs and/or controllers.

Hall


--
gentoo-***@gentoo.org mailing list
Craig Main
2003-10-30 14:50:31 UTC
Permalink
Post by Hall Stevenson
Post by Craig Main
using_dma = 0 (off)
What other hdparm flags can I _safely_ use, I don't want to trash my
disk.
The fact that DMA is *off* immediately jumped out at me. This can/should be
on with any or most modern HDs and/or controllers.
I get this when trying to enable dma, any ideas?

laptop root # hdparm -d1 /dev/hda

/dev/hda:
setting using_dma to 1 (on)
HDIO_SET_DMA failed: Operation not permitted
using_dma = 0 (off)
Post by Hall Stevenson
Hall
--
--
Craig Main
RHCE
ST Solutions
Cell: 082-323-4670
Home Tel: 011-760-1900
AIM: SATuxman; ICQ: 75815570
http://www.stsolutions.co.za
http://www.linuxsa.com


____________________________
There are no great men, only great challenges that ordinary men are
forced by circumstances to meet. -- Admiral William Halsey
Hall Stevenson
2003-10-30 15:13:22 UTC
Permalink
Post by Hall Stevenson
Post by Hall Stevenson
Post by Craig Main
using_dma = 0 (off)
What other hdparm flags can I _safely_ use, I don't want to trash my
disk.
The fact that DMA is *off* immediately jumped out at me. This
can/should be
Post by Hall Stevenson
on with any or most modern HDs and/or controllers.
I get this when trying to enable dma, any ideas?
laptop root # hdparm -d1 /dev/hda
setting using_dma to 1 (on)
HDIO_SET_DMA failed: Operation not permitted
using_dma = 0 (off)
My idea ?? Your hardware doesn't support DMA.

Hall


--
gentoo-***@gentoo.org mailing list
Mike Wojcikiewicz
2003-10-30 22:21:50 UTC
Permalink
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Post by Craig Main
Post by Hall Stevenson
Post by Craig Main
using_dma = 0 (off)
What other hdparm flags can I _safely_ use, I don't want to trash my
disk.
The fact that DMA is *off* immediately jumped out at me. This can/should
be on with any or most modern HDs and/or controllers.
I get this when trying to enable dma, any ideas?
laptop root # hdparm -d1 /dev/hda
setting using_dma to 1 (on)
HDIO_SET_DMA failed: Operation not permitted
using_dma = 0 (off)
I used to get that after a kernel upgrade.. after some digging around it turns
out i didnt have the right IDE chipset compiled into my kernel.. check yours
and make sure youre using the right one and that its compiled IN to the
kernel (as opposed to a module)
- --
- --mike
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQE/oY8CzK7WDkEewTARAqy/AKCswEa/TsW+yXRZAS+4NZrQ6Yq3egCeMSCF
xQkTg9HNcg7P7w8UUFhcjso=
=+KtC
-----END PGP SIGNATURE-----

--
gentoo-***@gentoo.org mailing list
flacycads
2003-10-30 17:56:32 UTC
Permalink
Post by Mike Wojcikiewicz
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Post by Craig Main
Post by Hall Stevenson
Post by Craig Main
using_dma = 0 (off)
What other hdparm flags can I _safely_ use, I don't want to trash my
disk.
The fact that DMA is *off* immediately jumped out at me. This
can/should be on with any or most modern HDs and/or controllers.
I get this when trying to enable dma, any ideas?
laptop root # hdparm -d1 /dev/hda
setting using_dma to 1 (on)
HDIO_SET_DMA failed: Operation not permitted
using_dma = 0 (off)
I used to get that after a kernel upgrade.. after some digging around it
turns out i didnt have the right IDE chipset compiled into my kernel..
check yours and make sure youre using the right one and that its compiled
IN to the kernel (as opposed to a module)
- --
- --mike
----------------------------------------------------
Hmmm. I just looked at one of my 2.4.22 kernel configs. Make sure all these
items are supported in the kernel, in addition to your chipset, as Mike says.
The "# forced" line I'm not sure of, but I don't have it set, and DMA works
with my 2.4 kernels. If you have these all set "y" already, it must be the
hard drive itself.

Robert Crawford

#
# IDE, ATA and ATAPI Block devices
#
CONFIG_BLK_DEV_IDE=y
CONFIG_BLK_DEV_IDEDISK=y
CONFIG_IDEDISK_MULTI_MODE=y
CONFIG_BLK_DEV_IDECD=y
CONFIG_BLK_DEV_CMD640=y
CONFIG_BLK_DEV_IDEPCI=y
CONFIG_IDEPCI_SHARE_IRQ=y
CONFIG_BLK_DEV_IDEDMA_PCI=y
# CONFIG_BLK_DEV_IDEDMA_FORCED is not set
CONFIG_IDEDMA_PCI_AUTO=y
CONFIG_BLK_DEV_IDEDMA=y
CONFIG_IDEDMA_AUTO=y
CONFIG_BLK_DEV_IDE_MODES=y




















--
gentoo-***@gentoo.org mailing list
James H.Cloos Jr.
2003-11-08 20:18:27 UTC
Permalink
Hall> The fact that DMA is *off* immediately jumped out at me. This
Hall> can/should be on with any or most modern HDs and/or controllers.

The using_dma flag is not relevant to udma modes, based on old
postings on lkml.

Since the -i info on that drive said *udma2 using_dma does not need
to be bothered with.

-JimC




--
gentoo-***@gentoo.org mailing list
Hall Stevenson
2003-11-08 15:30:44 UTC
Permalink
Post by James H.Cloos Jr.
Hall> The fact that DMA is *off* immediately jumped out at me. This
Hall> can/should be on with any or most modern HDs and/or controllers.
The using_dma flag is not relevant to udma modes, based on old
postings on lkml.
Since the -i info on that drive said *udma2 using_dma does not need
to be bothered with.
I honestly noticed that discrepency, but then thought that it meant the
drive could or should be in 'udma2' mode, but because of a configuration
error, it wasn't. Oh well..

Hall


--
gentoo-***@gentoo.org mailing list
James H.Cloos Jr.
2003-11-08 20:43:49 UTC
Permalink
Hall> I honestly noticed that discrepency, but then thought that it
Hall> meant the drive could or should be in 'udma2' mode, but because
Hall> of a configuration error, it wasn't. Oh well..

The discrepency is a result of all of the cruft in the ata spec from
past versions. With luck sata will leave all of that in the past. :)

Anyway, I do clearly remember a post from Anrdre saying to ignore the
hdparm output and only worry about what is printk()ed. Ie look in
the dmesg(8) output to see what the drives are using.

-JimC




--
gentoo-***@gentoo.org mailing list

keanu
2003-10-30 14:48:43 UTC
Permalink
Post by Craig Main
Model=HITACHI_DK23CA-20, FwRev=00H1A0J1, SerialNo=12N68B
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=36477, SectSize=579, ECCbytes=4
BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=39070080
IORDY=yes, tPIO={min:400,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: sdma0 sdma1 sdma2 mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 *udma2 udma3 udma4 udma5
AdvancedPM=yes: mode=0x80 (128) WriteCache=enabled
* signifies the current active mode
What other hdparm flags can I _safely_ use, I don't want to trash my
disk.
try
-X70 -d1 -c3


--
gentoo-***@gentoo.org mailing list
Joao Seabra
2003-10-30 14:57:27 UTC
Permalink
Post by Craig Main
Hi all,
I have been looking at the speeds of the disks in some of the threads
lately. I noticed that mine is a lot slower than most. I have attached
some info. Is my disk running as optimally as it can or is there some
more tweaking I can do to improve performance?
laptop root # hdparm -tT /dev/hda
Model=HITACHI_DK23CA-20, FwRev=00H1A0J1, SerialNo=12N68B
This HD sucks thats why you cant get more speed.Anyway try using -d1.
Take a look at hdparm man page because you will find a nice description
for all flags,including the dangerous ones...
Be careful

Kind Regards,

Joao Seabra

--
gentoo-***@gentoo.org mailing list
Redeeman
2003-10-30 15:34:47 UTC
Permalink
i would like to know if i could improve too, here is data for my disks
(3 disks):

hda:
***@redeeman redeeman # hdparm -tT /dev/hda

/dev/hda:
Timing buffer-cache reads: 808 MB in 2.01 seconds = 401.99 MB/sec
Timing buffered disk reads: 102 MB in 3.05 seconds = 33.44 MB/sec

***@redeeman redeeman # hdparm /dev/hda

/dev/hda:
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 7474/255/63, sectors = 120069936, start = 0

***@redeeman redeeman # hdparm -i /dev/hda

/dev/hda:

Model=Maxtor 4D060H3, FwRev=DAH017K0, SerialNo=D3H7MPTE
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=120069936
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
UDMA modes: udma0 udma1 udma2
AdvancedPM=yes: disabled (255) WriteCache=enabled
Drive conforms to: ATA/ATAPI-6 T13 1410D revision 0:

* signifies the current active mode

#################################################

hdc:

***@redeeman redeeman # hdparm -tT /dev/hdc

/dev/hdc:
Timing buffer-cache reads: 812 MB in 2.01 seconds = 403.98 MB/sec
Timing buffered disk reads: 142 MB in 3.00 seconds = 47.33 MB/sec

***@redeeman redeeman # hdparm /dev/hdc

/dev/hdc:
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 14946/255/63, sectors = 240121728, start = 0

***@redeeman redeeman # hdparm -i /dev/hdc

/dev/hdc:

Model=Maxtor 6Y120L0, FwRev=YAR41VW0, SerialNo=Y4091JNE
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=240121728
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
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 udma6
AdvancedPM=yes: disabled (255) WriteCache=enabled
Drive conforms to: (null):

* signifies the current active mode

#################################################
hdd:

***@redeeman redeeman # hdparm -tT /dev/hdd

/dev/hdd:
Timing buffer-cache reads: 792 MB in 2.00 seconds = 396.00 MB/sec
Timing buffered disk reads: 154 MB in 3.03 seconds = 50.83 MB/sec
***@redeeman redeeman # hdparm /dev/hdd

/dev/hdd:
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 19929/255/63, sectors = 320173056, start = 0

***@redeeman redeeman # hdparm -i /dev/hdd

/dev/hdd:

Model=Maxtor 6Y160P0, FwRev=YAR41VW0, SerialNo=Y60D80ME
Config={ Fixed }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=57
BuffType=DualPortCache, BuffSize=7936kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=268435455
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
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 udma6
AdvancedPM=yes: disabled (255) WriteCache=enabled
Drive conforms to: (null):

* signifies the current active mode
Post by Craig Main
Hi all,
I have been looking at the speeds of the disks in some of the threads
lately. I noticed that mine is a lot slower than most. I have attached
some info. Is my disk running as optimally as it can or is there some
more tweaking I can do to improve performance?
laptop root # hdparm -tT /dev/hda
Timing buffer-cache reads: 644 MB in 2.00 seconds = 322.00 MB/sec
Timing buffered disk reads: 22 MB in 3.25 seconds = 6.77 MB/sec
laptop root # hdparm /dev/hda
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 0 (off)
using_dma = 0 (off)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 2432/255/63, sectors = 39070080, start = 0
laptop root # hdparm -i /dev/hda
Model=HITACHI_DK23CA-20, FwRev=00H1A0J1, SerialNo=12N68B
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=36477, SectSize=579, ECCbytes=4
BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=39070080
IORDY=yes, tPIO={min:400,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: sdma0 sdma1 sdma2 mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 *udma2 udma3 udma4 udma5
AdvancedPM=yes: mode=0x80 (128) WriteCache=enabled
* signifies the current active mode
What other hdparm flags can I _safely_ use, I don't want to trash my
disk.
Thanks and Regards
Craig
--
Craig Main
RHCE
ST Solutions
Cell: 082-323-4670
Home Tel: 011-760-1900
AIM: SATuxman; ICQ: 75815570
http://www.stsolutions.co.za
http://www.linuxsa.com
____________________________
I can't believe I ate the whole thing. -- Homer Simpson The Front
--
Regards, Redeeman
() ascii ribbon campaign - against html e-mail
/\ - against microsoft attachments


--
gentoo-***@gentoo.org mailing list
Javier Villavicencio
2003-10-30 15:51:52 UTC
Permalink
Just add -a255 (for 2.4 kernels) or -a256(for 2.6 kernels) to increase the
"readahead" of the disk, this is one of the "safe" parameters of hdparm.
Post by Redeeman
i would like to know if i could improve too, here is data for my disks
Timing buffer-cache reads: 808 MB in 2.01 seconds = 401.99 MB/sec
Timing buffered disk reads: 102 MB in 3.05 seconds = 33.44 MB/sec
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 7474/255/63, sectors = 120069936, start = 0
Model=Maxtor 4D060H3, FwRev=DAH017K0, SerialNo=D3H7MPTE
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=120069936
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
UDMA modes: udma0 udma1 udma2
AdvancedPM=yes: disabled (255) WriteCache=enabled
* signifies the current active mode
#################################################
Timing buffer-cache reads: 812 MB in 2.01 seconds = 403.98 MB/sec
Timing buffered disk reads: 142 MB in 3.00 seconds = 47.33 MB/sec
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 14946/255/63, sectors = 240121728, start = 0
Model=Maxtor 6Y120L0, FwRev=YAR41VW0, SerialNo=Y4091JNE
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=240121728
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
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 udma6
AdvancedPM=yes: disabled (255) WriteCache=enabled
* signifies the current active mode
#################################################
Timing buffer-cache reads: 792 MB in 2.00 seconds = 396.00 MB/sec
Timing buffered disk reads: 154 MB in 3.03 seconds = 50.83 MB/sec
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 19929/255/63, sectors = 320173056, start = 0
Model=Maxtor 6Y160P0, FwRev=YAR41VW0, SerialNo=Y60D80ME
Config={ Fixed }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=57
BuffType=DualPortCache, BuffSize=7936kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=268435455
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
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5 udma6
AdvancedPM=yes: disabled (255) WriteCache=enabled
* signifies the current active mode
Post by Craig Main
Hi all,
I have been looking at the speeds of the disks in some of the threads
lately. I noticed that mine is a lot slower than most. I have attached
some info. Is my disk running as optimally as it can or is there some
more tweaking I can do to improve performance?
laptop root # hdparm -tT /dev/hda
Timing buffer-cache reads: 644 MB in 2.00 seconds = 322.00 MB/sec
Timing buffered disk reads: 22 MB in 3.25 seconds = 6.77 MB/sec
laptop root # hdparm /dev/hda
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 0 (off)
using_dma = 0 (off)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 2432/255/63, sectors = 39070080, start = 0
laptop root # hdparm -i /dev/hda
Model=HITACHI_DK23CA-20, FwRev=00H1A0J1, SerialNo=12N68B
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=36477, SectSize=579, ECCbytes=4
BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=39070080
IORDY=yes, tPIO={min:400,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: sdma0 sdma1 sdma2 mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 *udma2 udma3 udma4 udma5
AdvancedPM=yes: mode=0x80 (128) WriteCache=enabled
* signifies the current active mode
What other hdparm flags can I _safely_ use, I don't want to trash my
disk.
Thanks and Regards
Craig
--
Craig Main
RHCE
ST Solutions
Cell: 082-323-4670
Home Tel: 011-760-1900
AIM: SATuxman; ICQ: 75815570
http://www.stsolutions.co.za
http://www.linuxsa.com
____________________________
I can't believe I ate the whole thing. -- Homer Simpson The Front
--
gentoo-***@gentoo.org mailing list
Hemmann, Volker Armin
2003-10-30 18:13:22 UTC
Permalink
Hi,

hdparm -d1 -c1 -u1 should be save.
The udma level is set by the kernel and should not manipulated by the user. (I
do not have to touch hdparm at all, do you have the correct kernel settings?)

Glück Auf
Volker


--
gentoo-***@gentoo.org mailing list
eise
2003-11-01 20:17:15 UTC
Permalink
On Thu, 30 Oct 2003 19:13:22 +0100
"Hemmann, Volker Armin" <***@heim9.tu-clausthal.de> wrote:

Hi Volker,
Post by Hemmann, Volker Armin
hdparm -d1 -c1 -u1 should be save.
Are you sure that -u1 is safe? I've read the hdparm man page,
but I (obviously) didn't understand it in full - is -u1 safe for
_all_ chipsets now (got some sort of CMD640 chipset...) ?
If it is, what can I expect from it concerning speedup ?

Greetings, Dennis

--
gentoo-***@gentoo.org mailing list
Hemmann, Volker Armin
2003-11-01 21:57:53 UTC
Permalink
Post by eise
On Thu, 30 Oct 2003 19:13:22 +0100
Hi Volker,
Post by Hemmann, Volker Armin
hdparm -d1 -c1 -u1 should be save.
Are you sure that -u1 is safe? I've read the hdparm man page,
but I (obviously) didn't understand it in full - is -u1 safe for
_all_ chipsets now (got some sort of CMD640 chipset...) ?
If it is, what can I expect from it concerning speedup ?
as man hdparm says:
-u Get/set interrupt-unmask flag for the drive. A setting of 1
permits the driver to unmask other interrupts during processing
of a disk interrupt, which greatly improves Linux's responsive-
ness and eliminates "serial port overrun" errors.

the chipsets mentioned there are from the stone age of ide-controllers. If you
have one of them, you do not need to worry about dma etc ;o) And man hdparm
further says, that the kernels 2.0.13 and later incorporate an apropriate bug
fix. If you do not believe man, don't use it.

I am using this options since ages without probs but with other chipsets. (SiS
735, 746, hpt366 , via kt133 (I do not know the southbridge anymore, AFAIR
an udma 66 type.))

Glück Auf
Volker
--
Conclusions
In a straight-up fight, the Empire squashes the Federation like a bug. Even
with its numerical advantage removed, the Empire would still squash the
Federation like a bug. Accept it. -Michael Wong


--
gentoo-***@gentoo.org mailing list
Sean Johnson
2003-11-02 02:27:51 UTC
Permalink
For what it's worth, I use the -u1 flag on both my DVD-ROM and my
DVD-RW. Why? Well, it was driving me nuts that anytime I would copy
large amounts of data from the DVD-ROM or burn a DVD-R that the data
rate on my modem went to almost zero. Turns out that since the COM port
interrupts aren't considered as "important" as the IDE interrupts, that
the modem was simply getting no "interrupt" time. Or at least this is
how I understood what I could find via google. Unmasking the irq on both
devices completely fixed the problem. Occassionally when I switch
kernels, I think about checking to see if the setting is still required,
but I haven't tried it recently.

Just my 2c.

Sean
Post by Hemmann, Volker Armin
Post by eise
On Thu, 30 Oct 2003 19:13:22 +0100
Hi Volker,
Post by Hemmann, Volker Armin
hdparm -d1 -c1 -u1 should be save.
Are you sure that -u1 is safe? I've read the hdparm man page,
but I (obviously) didn't understand it in full - is -u1 safe for
_all_ chipsets now (got some sort of CMD640 chipset...) ?
If it is, what can I expect from it concerning speedup ?
-u Get/set interrupt-unmask flag for the drive. A setting of 1
permits the driver to unmask other interrupts during processing
of a disk interrupt, which greatly improves Linux's responsive-
ness and eliminates "serial port overrun" errors.
the chipsets mentioned there are from the stone age of ide-controllers. If you
have one of them, you do not need to worry about dma etc ;o) And man hdparm
further says, that the kernels 2.0.13 and later incorporate an apropriate bug
fix. If you do not believe man, don't use it.
I am using this options since ages without probs but with other chipsets. (SiS
735, 746, hpt366 , via kt133 (I do not know the southbridge anymore, AFAIR
an udma 66 type.))
Glück Auf
Volker
--
gentoo-***@gentoo.org mailing list
Alan
2003-10-30 18:49:32 UTC
Permalink
Post by Craig Main
Hi all,
[..]

Figured I'd chime in here with my own question. I have software RAID0
set up on my / partition across hda and hdb, should the throughput of
the raid on /dev/md0 be basically double the throughput of the
individual disk?

IE:

phoenix warez # hdparm -tT /dev/hda /dev/hdb /dev/md0

/dev/hda:
Timing buffer-cache reads: 1652 MB in 2.00 seconds = 826.00 MB/sec
Timing buffered disk reads: 84 MB in 3.03 seconds = 27.72 MB/sec

/dev/hdb:
Timing buffer-cache reads: 1664 MB in 2.00 seconds = 832.00 MB/sec
Timing buffered disk reads: 94 MB in 3.03 seconds = 31.02 MB/sec

/dev/md0:
Timing buffer-cache reads: 1652 MB in 2.00 seconds = 826.00 MB/sec
Timing buffered disk reads: 136 MB in 3.03 seconds = 44.88 MB/sec


Shouldn't /dev/md0 be a bit faster, basically 50-60MB/sec? My drives
and raid config look like the following, any suggestions on getting
100MB/sec like another poster here with RAID0 would be appreciated :)

I'd guess it's either overhead or I don't have the proper 133 cables on
my drives (pretty sure I do though) or something like that. This is on
an A7N8X-DX board with IDE hard drives, agpgart with nvidia support
built into the kernel, dma on by default in the kernel.

phoenix warez # cat /etc/raidtab
raiddev /dev/md0
raid-level 0
nr-raid-disks 2
nr-spare-disks 0
persistent-superblock 1
chunk-size 32
device /dev/hdb1
raid-disk 0
device /dev/hda6
raid-disk 1

phoenix warez # hdparm -i /dev/hda /dev/hdb

/dev/hda:

Model=Maxtor 54098H8, FwRev=DAC10SC0, SerialNo=K80JJDFC
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=80041248
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
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5
AdvancedPM=yes: disabled (255) WriteCache=enabled
Drive conforms to: ATA/ATAPI-6 T13 1410D revision 0:

* signifies the current active mode


/dev/hdb:

Model=Maxtor 52049H3, FwRev=JAC61HU0, SerialNo=F3H0HWFC
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=40021632
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
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5
AdvancedPM=yes: disabled (255) WriteCache=enabled
Drive conforms to: ATA/ATAPI-6 T13 1410D revision 0:

* signifies the current active mode


phoenix warez # hdparm /dev/hda /dev/hdb

/dev/hda:
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 1 (on)
readonly = 0 (off)
readahead = 255 (on)
geometry = 4982/255/63, sectors = 80041248, start = 0

/dev/hdb:
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 1 (on)
readonly = 0 (off)
readahead = 255 (on)
geometry = 39704/16/63, sectors = 40021632, start = 0

Any hints appreciated.

alan
--
Alan <***@ufies.org> - http://arcterex.net
--------------------------------------------------------------------
"There are only 3 real sports: bull-fighting, car racing and mountain
climbing. All the others are mere games." -- Hemingway
HvR
2003-10-30 20:16:57 UTC
Permalink
see below...
Post by Alan
Post by Craig Main
Hi all,
[..]
Figured I'd chime in here with my own question. I have software RAID0
set up on my / partition across hda and hdb, should the throughput of
the raid on /dev/md0 be basically double the throughput of the
individual disk?
phoenix warez # hdparm -tT /dev/hda /dev/hdb /dev/md0
Timing buffer-cache reads: 1652 MB in 2.00 seconds = 826.00 MB/sec
Timing buffered disk reads: 84 MB in 3.03 seconds = 27.72 MB/sec
Timing buffer-cache reads: 1664 MB in 2.00 seconds = 832.00 MB/sec
Timing buffered disk reads: 94 MB in 3.03 seconds = 31.02 MB/sec
Timing buffer-cache reads: 1652 MB in 2.00 seconds = 826.00 MB/sec
Timing buffered disk reads: 136 MB in 3.03 seconds = 44.88 MB/sec
with mdo striped over hda and hdb the best you can do is 27+31 for
speed, but then only if each disk is on its own IDE controller (so not
hanging of the same cable), if you want 100mb then you have to buy more
disks and stripe them, or go to scsi
Post by Alan
Shouldn't /dev/md0 be a bit faster, basically 50-60MB/sec? My drives
and raid config look like the following, any suggestions on getting
100MB/sec like another poster here with RAID0 would be appreciated :)
I'd guess it's either overhead or I don't have the proper 133 cables on
my drives (pretty sure I do though) or something like that. This is on
an A7N8X-DX board with IDE hard drives, agpgart with nvidia support
built into the kernel, dma on by default in the kernel.
phoenix warez # cat /etc/raidtab
raiddev /dev/md0
raid-level 0
nr-raid-disks 2
nr-spare-disks 0
persistent-superblock 1
chunk-size 32
device /dev/hdb1
raid-disk 0
device /dev/hda6
raid-disk 1
phoenix warez # hdparm -i /dev/hda /dev/hdb
Model=Maxtor 54098H8, FwRev=DAC10SC0, SerialNo=K80JJDFC
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=80041248
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
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5
AdvancedPM=yes: disabled (255) WriteCache=enabled
* signifies the current active mode
Model=Maxtor 52049H3, FwRev=JAC61HU0, SerialNo=F3H0HWFC
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=40021632
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
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5
AdvancedPM=yes: disabled (255) WriteCache=enabled
* signifies the current active mode
phoenix warez # hdparm /dev/hda /dev/hdb
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 1 (on)
readonly = 0 (off)
readahead = 255 (on)
geometry = 4982/255/63, sectors = 80041248, start = 0
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 1 (on)
using_dma = 1 (on)
keepsettings = 1 (on)
readonly = 0 (off)
readahead = 255 (on)
geometry = 39704/16/63, sectors = 40021632, start = 0
Any hints appreciated.
alan
Alan
2003-10-30 20:36:34 UTC
Permalink
Post by HvR
Post by Alan
Timing buffer-cache reads: 1652 MB in 2.00 seconds = 826.00 MB/sec
Timing buffered disk reads: 84 MB in 3.03 seconds = 27.72 MB/sec
Timing buffer-cache reads: 1664 MB in 2.00 seconds = 832.00 MB/sec
Timing buffered disk reads: 94 MB in 3.03 seconds = 31.02 MB/sec
Timing buffer-cache reads: 1652 MB in 2.00 seconds = 826.00 MB/sec
Timing buffered disk reads: 136 MB in 3.03 seconds = 44.88 MB/sec
with mdo striped over hda and hdb the best you can do is 27+31 for
speed, but then only if each disk is on its own IDE controller (so not
hanging of the same cable), if you want 100mb then you have to buy more
disks and stripe them, or go to scsi
Right, of course, my bad. I had it set up with hda and hdc some time
ago, but in some hardware shuffling I must have moved things around
without thinking. I'll reset things tonight.

Thanks for the pointer!

alan
--
Alan <***@ufies.org> - http://arcterex.net
--------------------------------------------------------------------
"There are only 3 real sports: bull-fighting, car racing and mountain
climbing. All the others are mere games." -- Hemingway

--
gentoo-***@gentoo.org mailing list
Michael Niethen
2003-10-31 01:51:18 UTC
Permalink
Post by HvR
with mdo striped over hda and hdb the best you can do is 27+31 for
speed, but then only if each disk is on its own IDE controller (so not
hanging of the same cable), if you want 100mb then you have to buy more
disks and stripe them, or go to scsi
I would rather expect an increase of 50% in practice with a 2-disk stripeset, in this case : (27 + 31) / 2 * 1.5
i.e. a transfer rate of approx. 43.5 MB/sec.

Having the discs on the same cable or not didn't make any difference for me. No bottlenecks since we don't exceed the bandwidth of each IDE channel (133 MB/sec, i.e. 66 MB/sec per single disk).
--
Michael
Linux: Because rebooting is for adding new hardware.
--------------------------------------
Gentoo Linux 1.4 (Intel) Kernel 2.4.20 [SMP,ACPI,RAID] 12281.44 BogoMIPS

--
gentoo-***@gentoo.org mailing list
Michael Niethen
2003-10-31 01:21:22 UTC
Permalink
Post by Alan
Shouldn't /dev/md0 be a bit faster, basically 50-60MB/sec? My drives
and raid config look like the following, any suggestions on getting
100MB/sec like another poster here with RAID0 would be appreciated :)
Well, I'm the other poster I think :-)
I just would like to remind the 100 MB/sec are obtained with 4 disks, not 2.
It's not possible with 2 disks I guess, unless you use very expensive ones, like 10,000 or 15,000 rpm SCSI disks.
Post by Alan
I'd guess it's either overhead or I don't have the proper 133 cables on
my drives (pretty sure I do though) or something like that. This is on
an A7N8X-DX board with IDE hard drives, agpgart with nvidia support
built into the kernel, dma on by default in the kernel.
No need for ATA/133 cables if you use ATA/100 disks.
Post by Alan
phoenix warez # cat /etc/raidtab
raiddev /dev/md0
raid-level 0
nr-raid-disks 2
nr-spare-disks 0
persistent-superblock 1
chunk-size 32
device /dev/hdb1
raid-disk 0
device /dev/hda6
raid-disk 1
As mentioned before, I have "nr-raid-disks 4".
You could give a try for "chunk-size 64", this worked best for my box.
--
Michael
Linux: Because rebooting is for adding new hardware.
--------------------------------------
Gentoo Linux 1.4 (Intel) Kernel 2.4.20 [SMP,ACPI,RAID] 12281.44 BogoMIPS

--
gentoo-***@gentoo.org mailing list
HvR
2003-10-30 20:06:24 UTC
Permalink
here is my set:


/dev/hda:
Timing buffer-cache reads: 592 MB in 2.00 seconds = 296.00 MB/sec
Timing buffered disk reads: 80 MB in 3.02 seconds = 26.49 MB/sec

/dev/hda:
multcount = 16 (on)
IO_support = 3 (32-bit w/sync)
unmaskirq = 0 (off)
using_dma = 1 (on)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 16 (on)
geometry = 9729/255/63, sectors = 156301488, start = 0

/dev/hda:

Model=IC25N080ATMR04-0, FwRev=MO4OAD0A, SerialNo=MRG401K4C0MEKC
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
BuffType=DualPortCache, BuffSize=7884kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=156301488
IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5
AdvancedPM=yes: mode=0x80 (128) WriteCache=enabled
Drive conforms to: ATA/ATAPI-6 T13 1410D revision 3a:

* signifies the current active mode

Command i use to set this:

hdparm -d1 -a16 -c3 -A1 -m16 /dev/hda
Post by Craig Main
Hi all,
I have been looking at the speeds of the disks in some of the threads
lately. I noticed that mine is a lot slower than most. I have attached
some info. Is my disk running as optimally as it can or is there some
more tweaking I can do to improve performance?
laptop root # hdparm -tT /dev/hda
Timing buffer-cache reads: 644 MB in 2.00 seconds = 322.00 MB/sec
Timing buffered disk reads: 22 MB in 3.25 seconds = 6.77 MB/sec
laptop root # hdparm /dev/hda
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 0 (off)
using_dma = 0 (off)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 2432/255/63, sectors = 39070080, start = 0
laptop root # hdparm -i /dev/hda
Model=HITACHI_DK23CA-20, FwRev=00H1A0J1, SerialNo=12N68B
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=36477, SectSize=579, ECCbytes=4
BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=39070080
IORDY=yes, tPIO={min:400,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: sdma0 sdma1 sdma2 mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 *udma2 udma3 udma4 udma5
AdvancedPM=yes: mode=0x80 (128) WriteCache=enabled
* signifies the current active mode
What other hdparm flags can I _safely_ use, I don't want to trash my
disk.
Thanks and Regards
Craig
--
Craig Main
RHCE
ST Solutions
Cell: 082-323-4670
Home Tel: 011-760-1900
AIM: SATuxman; ICQ: 75815570
http://www.stsolutions.co.za
http://www.linuxsa.com
____________________________
I can't believe I ate the whole thing. -- Homer Simpson The Front
g***@lentner.com
2003-10-30 20:19:05 UTC
Permalink
Does forcing UDMA5 using -X69 help any? I tested both ways but didn't notice

any difference.

Does anyone know of a shell script that will test all the optimal settings? I

build a lot of systems and it's a drag always testing different settings. A
script that could probe and test a few sets of settings and output the best
possible settings would be really cool!
Post by HvR
Timing buffer-cache reads: 592 MB in 2.00 seconds = 296.00 MB/sec
Timing buffered disk reads: 80 MB in 3.02 seconds = 26.49 MB/sec
multcount = 16 (on)
IO_support = 3 (32-bit w/sync)
unmaskirq = 0 (off)
using_dma = 1 (on)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 16 (on)
geometry = 9729/255/63, sectors = 156301488, start = 0
Model=IC25N080ATMR04-0, FwRev=MO4OAD0A, SerialNo=MRG401K4C0MEKC
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
BuffType=DualPortCache, BuffSize=7884kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=156301488
IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5
AdvancedPM=yes: mode=0x80 (128) WriteCache=enabled
* signifies the current active mode
hdparm -d1 -a16 -c3 -A1 -m16 /dev/hda
Post by Craig Main
Hi all,
I have been looking at the speeds of the disks in some of the threads
lately. I noticed that mine is a lot slower than most. I have attached
some info. Is my disk running as optimally as it can or is there some
more tweaking I can do to improve performance?
laptop root # hdparm -tT /dev/hda
Timing buffer-cache reads: 644 MB in 2.00 seconds = 322.00 MB/sec
Timing buffered disk reads: 22 MB in 3.25 seconds = 6.77 MB/sec
laptop root # hdparm /dev/hda
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 0 (off)
using_dma = 0 (off)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 2432/255/63, sectors = 39070080, start = 0
laptop root # hdparm -i /dev/hda
Model=HITACHI_DK23CA-20, FwRev=00H1A0J1, SerialNo=12N68B
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=36477, SectSize=579, ECCbytes=4
BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=39070080
IORDY=yes, tPIO={min:400,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: sdma0 sdma1 sdma2 mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 *udma2 udma3 udma4 udma5
AdvancedPM=yes: mode=0x80 (128) WriteCache=enabled
* signifies the current active mode
What other hdparm flags can I _safely_ use, I don't want to trash my
disk.
Thanks and Regards
Craig
--
Craig Main
RHCE
ST Solutions
Cell: 082-323-4670
Home Tel: 011-760-1900
AIM: SATuxman; ICQ: 75815570
http://www.stsolutions.co.za
http://www.linuxsa.com
____________________________
I can't believe I ate the whole thing. -- Homer Simpson The Front
--
gentoo-***@gentoo.org mailing list
William Kenworthy
2003-10-31 11:47:42 UTC
Permalink
Mandrake had something like that. However all I remember is it took
ages and settled on something that tested slower than when I just set to
spec.

BillK
Post by g***@lentner.com
Does forcing UDMA5 using -X69 help any? I tested both ways but didn't notice
any difference.
Does anyone know of a shell script that will test all the optimal settings? I
build a lot of systems and it's a drag always testing different settings. A
script that could probe and test a few sets of settings and output the best
possible settings would be really cool!
Post by HvR
Timing buffer-cache reads: 592 MB in 2.00 seconds = 296.00 MB/sec
Timing buffered disk reads: 80 MB in 3.02 seconds = 26.49 MB/sec
multcount = 16 (on)
IO_support = 3 (32-bit w/sync)
unmaskirq = 0 (off)
using_dma = 1 (on)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 16 (on)
geometry = 9729/255/63, sectors = 156301488, start = 0
Model=IC25N080ATMR04-0, FwRev=MO4OAD0A, SerialNo=MRG401K4C0MEKC
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
BuffType=DualPortCache, BuffSize=7884kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=156301488
IORDY=on/off, tPIO={min:240,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 udma2 udma3 udma4 *udma5
AdvancedPM=yes: mode=0x80 (128) WriteCache=enabled
* signifies the current active mode
hdparm -d1 -a16 -c3 -A1 -m16 /dev/hda
Post by Craig Main
Hi all,
I have been looking at the speeds of the disks in some of the threads
lately. I noticed that mine is a lot slower than most. I have attached
some info. Is my disk running as optimally as it can or is there some
more tweaking I can do to improve performance?
laptop root # hdparm -tT /dev/hda
Timing buffer-cache reads: 644 MB in 2.00 seconds = 322.00 MB/sec
Timing buffered disk reads: 22 MB in 3.25 seconds = 6.77 MB/sec
laptop root # hdparm /dev/hda
multcount = 16 (on)
IO_support = 1 (32-bit)
unmaskirq = 0 (off)
using_dma = 0 (off)
keepsettings = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 2432/255/63, sectors = 39070080, start = 0
laptop root # hdparm -i /dev/hda
Model=HITACHI_DK23CA-20, FwRev=00H1A0J1, SerialNo=12N68B
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=36477, SectSize=579, ECCbytes=4
BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=39070080
IORDY=yes, tPIO={min:400,w/IORDY:120}, tDMA={min:120,rec:120}
PIO modes: pio0 pio1 pio2 pio3 pio4
DMA modes: sdma0 sdma1 sdma2 mdma0 mdma1 mdma2
UDMA modes: udma0 udma1 *udma2 udma3 udma4 udma5
AdvancedPM=yes: mode=0x80 (128) WriteCache=enabled
* signifies the current active mode
What other hdparm flags can I _safely_ use, I don't want to trash my
disk.
Thanks and Regards
Craig
--
Craig Main
RHCE
ST Solutions
Cell: 082-323-4670
Home Tel: 011-760-1900
AIM: SATuxman; ICQ: 75815570
http://www.stsolutions.co.za
http://www.linuxsa.com
____________________________
I can't believe I ate the whole thing. -- Homer Simpson The Front
--
--
William Kenworthy <***@iinet.net.au>


--
gentoo-***@gentoo.org mailing list
Continue reading on narkive:
Loading...