sata, low HDD performance

marcin7427 · 2008-04-11 17:12:36

Hi guys.

I've got a problem with my Archlinux: I can't force my ASUS A6r (http://uk.asus.com/products.aspx?modelm … =24&l3=134) to use DMA - hdparm -Tt /dev/sda returned:

Timing cached reads:   318 MB in  2.00 seconds = 158.63 MB/sec
Timing buffered disk reads:   26 MB in  4.00 seconds =   6.51 MB/sec

Just in case, hdparm -I /dev/sda

ATA device, with non-removable media
    Model Number:       HTS541060G9AT00                         
    Serial Number:      MPB3RAXGGXLATM
    Firmware Revision:  MB3VA60A
Standards:
    Used: ATA/ATAPI-6 T13 1410D revision 3a 
    Supported: 6 5 4 
Configuration:
    Logical        max    current
    cylinders    16383    16383
    heads        16    16
    sectors/track    63    63
    --
    CHS current addressable sectors:   16514064
    LBA    user addressable sectors:  117210240
    LBA48  user addressable sectors:  117210240
    device size with M = 1024*1024:       57231 MBytes
    device size with M = 1000*1000:       60011 MBytes (60 GB)
Capabilities:
    LBA, IORDY(can be disabled)
    Standby timer values: spec'd by Vendor, no device specific minimum
    R/W multiple sector transfer: Max = 16    Current = 16
    Advanced power management level: 128
    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:
       *    SMART feature set
            Security Mode feature set
       *    Power Management feature set
       *    Write cache
       *    Look-ahead
       *    Host Protected Area feature set
       *    WRITE_BUFFER command
       *    READ_BUFFER command
       *    NOP cmd
       *    DOWNLOAD_MICROCODE
       *    Advanced Power Management feature set
            Power-Up In Standby feature set
       *    SET_FEATURES required to spinup after power up
            Address Offset Reserved Area Boot
            SET_MAX security extension
       *    Automatic Acoustic Management feature set
       *    48-bit Address feature set
       *    Device Configuration Overlay feature set
       *    Mandatory FLUSH_CACHE
       *    FLUSH_CACHE_EXT
       *    SMART error logging
       *    SMART self-test
       *    General Purpose Logging feature set
Security: 
    Master password revision code = 65534
        supported
    not    enabled
    not    locked
    not    frozen
    not    expired: security count
    not    supported: enhanced erase
    50min for SECURITY ERASE UNIT. 
HW reset results:
    CBLID- above Vih
    Device num = 0 determined by the jumper
Checksum: correct

I've tried to use sdparm, but it hasn't worked

[root@marcin-laptop /]# sdparm --set=ARRE=1 /dev/sda
    /dev/sda: ATA       HTS541060G9AT00   MB3V
change_mode_page: failed setting page: Read write error recovery
[root@marcin-laptop /]# sdparm --set=ARRE=1 --save /dev/sda
    /dev/sda: ATA       HTS541060G9AT00   MB3V
change_mode_page: mode page indicates it is not savable but
    '--save' option given (try without it)

I've tried to solve this with http://smartmontools.sourceforge.net/#testinghelp, but changing type (-d scsi or -d sat), editing rc.conf and rebooting didn't work (what a surprise...).

Any ideas (TIA, if any)?

dabski · 2008-04-15 00:20:07

Looks like your already using DMA, if I recall correctly DMA is on by default now.

DMA: mdma0 mdma1 mdma2 udma0 udma1 udma2 udma3 udma4 *udma5

lilsirecho · 2008-04-15 01:54:41

The test performed by hdparm is probably only applicable to HDD devices and may give erroneous results when applied to sata devices with sdx ID.

Some reports indicate hdparm isn't a good measure with raid devices as well , even if HDD variety.

But I could be wrong!

marcin7427 · 2008-04-15 09:24:19

the thing is, I can't change some values (e.g. ARRE or AWRE; I have one of these settled as '0', and I've read, that the best option is keeping them both as '1'); I don't mind if ARRE and AWRE are enabled or not (for my, they may indicate 'I love Satan' value...), I only care about my HD performance - and that is dramatically poor (I forgot to mention - I'd extended RAM by 1 GiB, what makes something like 1,37 GiB + 128 MiB on graphics), after 3 weeks (and installing some applications) I wasn't able to comfortably copying 350 MiB files, browsing internet (Firefox) and listen to music (mpd) in the same time; thus, my impression of low HD efficiency is not caused by wrong way of hdparm measuring, I pressume :) (just for the record, I use fluxbox...)

anyway, thanks for your help

P.S. is it possible, that all that mess is caused by default kernel? AFAIR, current kernel does not support sata by default

Last edited by marcin7427 (2008-04-15 09:36:54)

cnshzj007 · 2008-04-15 11:04:06

[zhangjing@Archlinux007 ~]$ sudo hdparm -Tt /dev/sda

/dev/sda:
Timing cached reads: 2000 MB in 2.00 seconds = 1000.30 MB/sec
Timing buffered disk reads: 76 MB in 3.06 seconds = 24.81 MB/sec

This is the quiet mode in bios. Dell D630.

.:B:. · 2008-04-15 14:41:54

Marcin: Check dmesg for any pointers about the UDMA mode. Libata sometimes misdetects how the HD is connected (for laptops it's mostly though a ZIF) and will erroneously set it at
UDMA 33 (!). Check this thread for pointers.

I had (and still have) to patch my kernel for that. Another solution is to use the old ide layer - but both require you recompile your kernel.

Last edited by B (2008-04-15 14:42:36)

marcin7427 · 2008-04-16 14:11:34

in the very beginning I'd like to thank you, guys, for response.

the problem with my sata was likely caused by

kernel /vmlinuz26 root=/dev/sda10 NOAPIC IRQPOLL vga=773 ro

it was supposed to solve my problem with internet connection (which was unstable by then) and did its job well, but apparently also messed up my sata support

to sum up, nOOb question

.:B:. · 2008-04-16 15:07:43

Well we cannot help you if you don't verify any of the questions we raised above... The forum's psychic has his day off.

dmz · 2008-10-30 00:06:07

B wrote:

Marcin: Check dmesg for any pointers about the UDMA mode. Libata sometimes misdetects how the HD is connected (for laptops it's mostly though a ZIF) and will erroneously set it at
UDMA 33 (!). Check this thread for pointers.
I had (and still have) to patch my kernel for that. Another solution is to use the old ide layer - but both require you recompile your kernel.

I am having the same problem, with one of my disks, and dmesg tells me this:

  DMA      0x00000000 -> 0x00001000
  DMA zone: 3967 pages, LIFO batch:0
    ide0: BM-DMA at 0xf900-0xf907
hda: UDMA/66 mode selected
ahci 0000:00:12.0: controller can't do 64bit DMA, forcing 32bit
ata1: SATA max UDMA/133 abar m1024@0xfe02f000 port 0xfe02f100 irq 22
ata2: SATA max UDMA/133 abar m1024@0xfe02f000 port 0xfe02f180 irq 22
ata3: SATA max UDMA/133 abar m1024@0xfe02f000 port 0xfe02f200 irq 22
ata4: SATA max UDMA/133 abar m1024@0xfe02f000 port 0xfe02f280 irq 22
ata1.00: ATA-8: SAMSUNG HD501LJ, CR100-11, max UDMA7
ata1.00: configured for UDMA/133
ata2.00: ATA-7: ST3320820AS, 3.AAD, max UDMA/133
ata2.00: configured for UDMA/133
ata3.00: ATA-8: SAMSUNG HD501LJ, CR100-12, max UDMA7
ata3.00: configured for UDMA/133

I have no idea what to do now though, could you please help me out?

marcin7427 · 2008-11-01 09:52:12

If I may suggest... I'm not sure, what is wrong with your system, but I can tell you which config works on my PC (don't treat this as a perfect solution).
/boot/grub/menu.lst

# (0) Arch Linux
title Arch Linux
root (hd0,6)
kernel /vmlinuz26 root=/dev/sda9 sata noapic acpi=off vga=773 ro
initrd /kernel26.img

I'm sure you've got different hardware, but you may check this out. Also, make sure it won't affect other functions (I've got some network issues, zum Beispiel).
Good luck!

deej · 2008-11-01 12:28:01

If this helps, here's a copy of the relevant section of my 'menu.lst':

# (0) Arch Linux
title  Arch Linux
root   (hd0,0)
kernel /vmlinuz26 root=/dev/disk/by-uuid/8f5da52f-9be0-4420-b621-0ee5457786d5 ro pci=nomsi iommu=noaperture
initrd /kernel26.img

I've always used the 'pci=nomsi' to enable my sata II drive. The 'iommu=noaperture' is just a hack for my BIOS.

Here are my timings [ for what they're worth]:

bash-3.2# hdparm -Tt /dev/sda

/dev/sda:
 Timing cached reads:   2512 MB in  2.00 seconds = 1256.98 MB/sec
 Timing buffered disk reads:  190 MB in  3.01 seconds =  63.10 MB/sec

Deej

Last edited by deej (2008-11-01 12:29:22)

Arch Linux

#1 2008-04-11 17:12:36

sata, low HDD performance

#2 2008-04-15 00:20:07

Re: sata, low HDD performance

#3 2008-04-15 01:54:41

Re: sata, low HDD performance

#4 2008-04-15 09:24:19

Re: sata, low HDD performance

#5 2008-04-15 11:04:06

Re: sata, low HDD performance

#6 2008-04-15 14:41:54

Re: sata, low HDD performance

#7 2008-04-16 14:11:34

Re: sata, low HDD performance

#8 2008-04-16 15:07:43

Re: sata, low HDD performance

#9 2008-10-30 00:06:07

Re: sata, low HDD performance

#10 2008-11-01 09:52:12

Re: sata, low HDD performance

#11 2008-11-01 12:28:01

Re: sata, low HDD performance

Board footer