I have the following situation on my VIA KX133 based Athlon system:
IDE performance is abysmal. I have exactly one IDE target in the system,
a Maxtor 53073U6. According to a posting I saw here I set idebus=33 at
boot time. Now look at this:
VP_IDE: IDE controller on PCI bus 00 dev 39
VP_IDE: not 100% native mode: will probe irqs later
ide0: BM-DMA at 0xd000-0xd007, BIOS settings: hda:DMA, hdb:pio
ide0: VIA Bus-Master (U)DMA Timing Config Success
ide1: BM-DMA at 0xd008-0xd00f, BIOS settings: hdc:pio, hdd:pio
ide1: VIA Bus-Master (U)DMA Timing Config Success
hda: Maxtor 53073U6, ATA DISK drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
hda: Maxtor 53073U6, 29311MB w/2048kB Cache, CHS=3736/255/63
Looks to me as if DMA should be on for hda, right? hdparm tells me
otherwise:
/dev/hda:
multcount = 0 (off)
I/O support = 0 (default 16-bit)
unmaskirq = 0 (off)
using_dma = 0 (off)
keepsettings = 0 (off)
nowerr = 0 (off)
readonly = 0 (off)
readahead = 8 (on)
geometry = 3736/255/63, sectors = 60030432, start = 0
When untarring a big tar, the system becomes unresponsive and I have
40% system CPU and more.
So I tried to set multcount to 8, use I/O support 1 or 3 or enable DMA,
but in each case the system starts to read bogus data from disk, i.e.
binaries segfault randomly, MD5 checksums change, gzip -t gets wrong
checksums.
Any ideas? Could this be a heat issue? Windows seems to work fine using
DMA on the same disk. Mhh.
Felix
-
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/
This archive was generated by hypermail 2b29 : Sun Apr 23 2000 - 21:00:09 EST