lkml.org 
[lkml]   [2000]   [Mar]   [27]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    Date
    From
    SubjectRe: VIA MVP3 gives bad IDE UDMA33 performace ?!
    Harald Koenig wrote:
    >
    > is there any chance to get better IDE performace for VIA MVP3 chipset ?
    >

    I am observing some IDE-related probs with 2.3.99pre3 as well. Variable
    throughput and a complete crash.

    It's a BP6 with a couple of Seagates on the UDMA66 channel. When I use
    'hdparm -d1 -A1' I see 22 MBytes/sec from 'hdparm -t'. But after a
    little I/O (a couple more 'hdparm -t' commands is enough) the throughput
    falls away to 6 MB/s.

    I assume that the drive is doing a soft reset and losing the '-A1'
    setting (hdparm -K1 gets an I/O error). This theory may be incorrect
    because 'readahead = 8 (on)' remains true even when it's running
    slowly.

    PIIX4: IDE controller on PCI bus 00 dev 39
    PIIX4: not 100% native mode: will probe irqs later
    ide0: BM-DMA at 0xf000-0xf007, BIOS settings: hda:pio, hdb:pio
    ide1: BM-DMA at 0xf008-0xf00f, BIOS settings: hdc:pio, hdd:pio
    HPT366: onboard version of chipset, pin1=1 pin2=2
    HPT366: IDE controller on PCI bus 00 dev 98
    HPT366: not 100% native mode: will probe irqs later
    ide2: BM-DMA at 0xe000-0xe007, BIOS settings: hde:DMA, hdf:DMA
    HPT366: IDE controller on PCI bus 00 dev 99
    HPT366: not 100% native mode: will probe irqs later
    ide3: BM-DMA at 0xec00-0xec07, BIOS settings: hdg:pio, hdh:pio

    hdf: ST313021A, ATA DISK drive
    hdf: ST313021A, 12419MB w/512kB Cache, CHS=25232/16/63, UDMA(66)

    [ hde and hdf are both on ide2 ]

    [root@mnm ide]# hdparm -q -A1 -q -d1 /dev/hdf
    [root@mnm ide]# hdparm -t /dev/hdf

    /dev/hdf:
    Timing buffered disk reads: 64 MB in 2.91 seconds =21.99 MB/sec

    *** OK, 22MB/s

    [root@mnm ide]# hdparm -T /dev/hdf

    /dev/hdf:
    Timing buffer-cache reads: 128 MB in 1.93 seconds =66.32 MB/sec
    [root@mnm ide]# hdparm -t /dev/hdf

    /dev/hdf:
    Timing buffered disk reads: 64 MB in 10.23 seconds = 6.26 MB/sec

    *** Slowed down to 6 MB/s

    [root@mnm ide]# hdparm -I /dev/hdf

    /dev/hdf:

    Model=TS130312 A , FwRev=.330 ,
    SerialNo=C60T3BPX
    Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% }
    RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
    BuffType=0(?), BuffSize=512kB, MaxMultSect=32, MultSect=16
    DblWordIO=no, maxPIO=2(fast), DMA=yes, maxDMA=2(fast)
    CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=25434228
    WARNING 8920164 ORPHANED SECTORS :: KERNEL REPORTING ERROR
    tDMA={min:120,rec:120}, DMA modes: mword0 mword1 mword2
    IORDY=on/off, tPIO={min:240,w/IORDY:120}, PIO modes: mode3 mode4
    UDMA modes: mode0 mode1 mode2 mode3 *mode4
    Drive Supports : ATA-1 ATA-2 ATA-3 ATA-4 ATA-5

    [root@mnm ide]# hdparm /dev/hdf

    /dev/hdf:
    multcount = 16 (on)
    I/O support = 1 (32-bit)
    unmaskirq = 0 (off)
    using_dma = 1 (on)
    keepsettings = 0 (off)
    nowerr = 0 (off)
    readonly = 0 (off)
    readahead = 8 (on)
    geometry = 25232/16/63, sectors = 25434228, start = 0
    [root@mnm ide]# hdparm -q -A1 -q -d1 /dev/hdf
    [root@mnm ide]# hdparm /dev/hdf

    /dev/hdf:
    multcount = 16 (on)
    I/O support = 1 (32-bit)
    unmaskirq = 0 (off)
    using_dma = 1 (on)
    keepsettings = 0 (off)
    nowerr = 0 (off)
    readonly = 0 (off)
    readahead = 8 (on)
    geometry = 25232/16/63, sectors = 25434228, start = 0

    ** Output unchanged across readahead reset.

    I tried '-X66' and the disk system locked up altogether but the OS kept
    running. After finding another /sbin/init (handy app, that) I tried
    -X34.

    This crashes the OS altogether. I used 'hdparm -q -A1 -q -d1 -X34
    /dev/hdf' followed by 'hdparm -t /dev/hdf' and the console reported:

    hdf: timeout waiting for DMA
    ide_dmaproc: chipset supported ide_dma_timeout func only: 14

    And the system froze.


    Andre, this freeze is 100% reproducible, so I can run any tests/patches
    as required.


    --
    -akpm-

    -
    To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
    the body of a message to majordomo@vger.rutgers.edu
    Please read the FAQ at http://www.tux.org/lkml/

    \
     
     \ /
      Last update: 2005-03-22 13:57    [W:0.026 / U:29.788 seconds]
    ©2003-2016 Jasper Spaans. hosted at Digital OceanAdvertise on this site