View previous topic :: View next topic |
Author |
Message |
scape n00b
Joined: 22 Oct 2002 Posts: 16 Location: Stuttgart, Germany
|
Posted: Thu Feb 27, 2003 11:36 pm Post subject: hd dma problems |
|
|
Update:
it seems it actually was a kernel issue. switching from gentoo-sources to ac-sources actually solved the problem.
so you folks out there with a ss51g (anyone?), better stick with alan cox's kernels...
this is driving me nuts. no matter what i do, my harddrives perform poorly. i've been looking around the forums quite a while now, but no solution came up.
first some information about my setup:
Code: |
root@atair root # hdparm -I /dev/hda
/dev/hda:
ATA device, with non-removable media
powers-up in standby; SET FEATURES subcmd spins-up.
Model Number: IC35L120AVVA07-0
Serial Number: VNC602A6C97LTG
Firmware Revision: VA6OA52A
Standards:
Used: ATA/ATAPI-5 T13 1321D revision 1
Supported: 5 4 3 2 & some of 6
Configuration:
Logical max current
cylinders 16383 16383
heads 16 16
sectors/track 63 63
--
CHS current addressable sectors: 16514064
LBA user addressable sectors: 241254720
device size with M = 1024*1024: 117800 MBytes
device size with M = 1000*1000: 123522 MBytes (123 GB)
Capabilities:
LBA, IORDY(can be disabled)
bytes avail on r/w long: 52 Queue depth: 32
Standby timer values: spec'd by Standard, no device specific minimum
R/W multiple sector transfer: Max = 16 Current = 0
Advanced power management level: unknown setting (0x0000)
Recommended acoustic management value: 128, current value: 254
DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 *udma5
Cycle time: min=120ns recommended=120ns
PIO: pio0 pio1 pio2 pio3 pio4
Cycle time: no flow control=240ns IORDY flow control=120ns
Commands/features:
Enabled Supported:
* NOP cmd
* READ BUFFER cmd
* WRITE BUFFER cmd
* Host Protected Area feature set
Release interrupt
* Look-ahead
* Write cache
* Power Management feature set
Security Mode feature set
* SMART feature set
* Mandatory FLUSH CACHE command
* Device Configuration Overlay feature set
Automatic Acoustic Management feature set
SET MAX security extension
Address Offset Reserved Area Boot
SET FEATURES subcommand required to spinup after power up
Power-Up In Standby feature set
Advanced Power Management feature set
* READ/WRITE DMA QUEUED
Security:
Master password revision code = 65534
supported
not enabled
not locked
not frozen
not expired: security count
not supported: enhanced erase
72min for SECURITY ERASE UNIT.
HW reset results:
CBLID- above Vih
Device num = 0 determined by the jumper
Checksum: correct
|
and
Code: |
root@atair root # hdparm -I /dev/hdb
/dev/hdb:
ATA device, with non-removable media
Model Number: Maxtor 96147H6
Serial Number: V604A0EC
Firmware Revision: ZAH814Y0
Standards:
Used: ATA/ATAPI-6 T13 1410D revision 0
Supported: 6 5 4 3
Configuration:
Logical max current
cylinders 16383 16383
heads 16 16
sectors/track 63 63
--
CHS current addressable sectors: 16514064
LBA user addressable sectors: 120064896
device size with M = 1024*1024: 58625 MBytes
device size with M = 1000*1000: 61473 MBytes (61 GB)
Capabilities:
LBA, IORDY(can be disabled)
bytes avail on r/w long: 57 Queue depth: 1
Standby timer values: spec'd by Standard, no device specific minimum
R/W multiple sector transfer: Max = 16 Current = 0
Advanced power management level: unknown setting (0x0000)
Recommended acoustic management value: 192, current value: 192
DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 *udma5
Cycle time: min=120ns recommended=120ns
PIO: pio0 pio1 pio2 pio3 pio4
Cycle time: no flow control=120ns IORDY flow control=120ns
Commands/features:
Enabled Supported:
* NOP cmd
* READ BUFFER cmd
* WRITE BUFFER cmd
* Host Protected Area feature set
* Look-ahead
* Write cache
* Power Management feature set
* SMART feature set
* Automatic Acoustic Management feature set
SET MAX security extension
Advanced Power Management feature set
* DOWNLOAD MICROCODE cmd
HW reset results:
CBLID- above Vih
Device num = 1 determined by the jumper
Checksum: correct
|
I don't have one of these 80-whatsoever-cables, but nonetheless they should go faster than this:
Code: |
root@atair root # hdparm -Tt /dev/hda
/dev/hda:
Timing buffer-cache reads: 128 MB in 0.46 seconds =277.66 MB/sec
Timing buffered disk reads: 64 MB in 28.41 seconds = 2.25 MB/sec
|
same applies to /dev/hdb. i've been trying around with different kernel configurations - nothing. tried adding /etc/init.d/hdparm - no effect either.
i keep playing around with hdparm without success, the strange thing though is this:
Code: |
root@atair root # hdparm -d1 -u1 /dev/hdb
/dev/hdb:
setting unmaskirq to 1 (on)
setting using_dma to 1 (on)
unmaskirq = 1 (on)
using_dma = 1 (on)
root@atair root # hdparm -d /dev/hdb
/dev/hdb:
using_dma = 0 (off)
|
oh yeah, btw: got a ss51g (german) with shuttle mobo fs51 (SiS651/SiS962L chipset)
so, any suggestions? i'm really desperate _________________ a slave begins by demanding justice and ends by wanting to wear a crown
Last edited by scape on Mon Mar 03, 2003 4:41 pm; edited 1 time in total |
|
Back to top |
|
|
Malakin Veteran
Joined: 14 Apr 2002 Posts: 1692 Location: Victoria BC Canada
|
Posted: Fri Feb 28, 2003 1:12 am Post subject: |
|
|
Well I suppose this is obvious but I can tell you your dma isn't working ;) This is about what performance you get without dma.
I looked around and couldn't really find any answers.
What kernel are you using? does dmesg say anything interesting about your ide? |
|
Back to top |
|
|
scape n00b
Joined: 22 Oct 2002 Posts: 16 Location: Stuttgart, Germany
|
Posted: Fri Feb 28, 2003 2:11 am Post subject: |
|
|
Malakin wrote: | Well I suppose this is obvious but I can tell you your dma isn't working ;) This is about what performance you get without dma.
|
yeah, i was thinking of that, too ;)
Malakin wrote: |
I looked around and couldn't really find any answers.
What kernel are you using? does dmesg say anything interesting about your ide? |
Code: | root@atair root # uname -a
Linux atair 2.4.20-gentoo-r1 #4 Fri Feb 28 00:05:32 CET 2003 i686 Intel(R) Pentium(R) 4 CPU 2.00GHz GenuineIntel GNU/Linux
|
and the important (i guess) part from dmesg:
Code: | Uniform Multi-Platform E-IDE driver Revision: 6.31
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
SIS5513: IDE controller on PCI bus 00 dev 15
PCI: Found IRQ 11 for device 00:02.5
SIS5513: chipset revision 0
SIS5513: not 100% native mode: will probe irqs later
SiS5513
ide0: BM-DMA at 0x4000-0x4007, BIOS settings: hda:DMA, hdb:DMA
ide1: BM-DMA at 0x4008-0x400f, BIOS settings: hdc:DMA, hdd:pio
hda: IC35L120AVVA07-0, ATA DISK drive
hdb: Maxtor 96147H6, ATA DISK drive
hdc: HL-DT-ST RW/DVD GCC-4320B, ATAPI CD/DVD-ROM drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide1 at 0x170-0x177,0x376 on irq 15
hda: 241254720 sectors (123522 MB) w/1863KiB Cache, CHS=15017/255/63
hdb: 120064896 sectors (61473 MB) w/2048KiB Cache, CHS=7473/255/63
hdc: ATAPI 40X DVD-ROM CD-R/RW drive, 2048kB Cache
|
_________________ a slave begins by demanding justice and ends by wanting to wear a crown |
|
Back to top |
|
|
modal Apprentice
Joined: 02 Oct 2002 Posts: 277
|
Posted: Fri Feb 28, 2003 5:30 am Post subject: |
|
|
First, make sure that your IDE controller is supported and compiled in your kernel (either as modules or not)...for me, I have an nforce ATA100 controller card, which is only supported in certain patches (not in gentoo-sources, or vanilla-sources)
Second, run Code: | hdparm -Tt /dev/hdx | and look at your performance. Your cached setting (the first one) should be good, while your second one should be really poor.
Third, your ata controller supports udma5, which if you forced it with an -X option, would be:
udma5 : "-X 69"
the others are this:
udma2 : "-X 66"
udma4 : "-X 68"
udma5 : "-X 69"
udma6 : "-X 70"
Forth, make sure when you run hdparm to you -d1 to enable ultradma...
These are my hdparm settings:
Code: | hdparm -d1 -X69 -c1 A1 -m16 -u1 -a64 |
My performance went from 3MB/s to 30+MB/s
Hope this helps. |
|
Back to top |
|
|
scape n00b
Joined: 22 Oct 2002 Posts: 16 Location: Stuttgart, Germany
|
Posted: Fri Feb 28, 2003 11:17 am Post subject: |
|
|
nope, already tried all of this stuff. as i said, i'm browsing the forums for quite a while now. all this stuff came up, but with no or little effect. for example if i set all this hdparm stuff + run the tests (-Tt) in one single command, the performance seems quite ok. afterwards its just shitty again, so i reallly dont know what to do anymore _________________ a slave begins by demanding justice and ends by wanting to wear a crown |
|
Back to top |
|
|
Malakin Veteran
Joined: 14 Apr 2002 Posts: 1692 Location: Victoria BC Canada
|
Posted: Fri Feb 28, 2003 7:20 pm Post subject: |
|
|
Quote: | for example if i set all this hdparm stuff + run the tests (-Tt) in one single command, the performance seems quite ok. afterwards its just shitty again, so i reallly dont know what to do anymore |
So your performance will speed up for a while after enabling dma but will go back to being slow? This is without a reboot? How long will it remain fast for?
If ide errors are generated I believe it jumps down speeds until it stops gettings errors so a bad cable could cause this.
You could try a different kernel but I have a feeling it's not the kernel. |
|
Back to top |
|
|
scape n00b
Joined: 22 Oct 2002 Posts: 16 Location: Stuttgart, Germany
|
Posted: Fri Feb 28, 2003 9:40 pm Post subject: |
|
|
yep, without reboot. most of the time it stays fast just during the test, but once in a while it even remains for 2-3 secs.
i don't believe it's the kernel either, as i tried many different configurations.
i'll try switching cables, but running win2000 i get normal performance. that's why i think it isn't the cable.... _________________ a slave begins by demanding justice and ends by wanting to wear a crown |
|
Back to top |
|
|
scape n00b
Joined: 22 Oct 2002 Posts: 16 Location: Stuttgart, Germany
|
Posted: Sat Mar 01, 2003 5:51 pm Post subject: |
|
|
as far as i can tell it isn't the cable. i tried another one but that didn't fix it... _________________ a slave begins by demanding justice and ends by wanting to wear a crown |
|
Back to top |
|
|
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
|