I can also kind of confirm that. Likewise installed this weekend under
my good old 2.0.25 kernel. Noticed at once that hdparm -X34 hung the
machine and didn't try that again! It replaced a 3yr old seagate which
was running happily with those settings, and using them:
Model=FUJITSU MPB3032ATU E, FwRev=4010, SerialNo=03117294
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% }
RawCHS=6704/15/63, TrkSize=0, SectSize=0, ECCbytes=4
BuffType=0(?), BuffSize=0kB, MaxMultSect=16, MultSect=off
DblWordIO=no, maxPIO=2(fast), DMA=yes, maxDMA=0(slow)
CurCHS=6704/15/63, CurSects=6335280, LBA=yes, LBAsects=6335280
tDMA={min:120,rec:120}, DMA modes: mword0 mword1 *mword2
IORDY=yes, tPIO={min:240,w/IORDY:120}, PIO modes: mode3 mode4
Currently running in dma mode and getting superb results from bonnie
(4 - 7 MB/s in various tests) on my old T1. Roughly the same drive
went into a BX board, where it turns over in 6-11 MB/s under bonnie
with the same drive settings and the same (binary) kernel.
/dev/hda:
Timing buffer-cache reads: 64 MB in 2.56 seconds =25.00 MB/sec
Timing buffered disk reads: 32 MB in 4.36 seconds = 7.34 MB/sec
The readings are 114.29MB/s and 10.45MB/s on the bx. Hic.
> 2.0.34-pre6 (Red Hat 5.1 installation)
> 2.0.35 (Red Hat 5.1 installation)
>
> The following errors show up on "random" sectors of the drive during the
> boot of the kernel if UDMA is turned on:
Uh .. you are braver than me. I didn't even try once I saw the lockup with
-X34. At first I thought it was a compatibility problem with the seagate that
I had on the same controller then. Nope.
> Turning off UDMA via the motherboard cleared up the problem, and no
> sectors show up "bad" when looked at with "norton" or similar disk
> utilities.
Via the motherboard? Does the motherboard have that kind of option? Oh well,
maybe yours does.
> On Mon, 3 Aug 1998, Jaime E. Gomez wrote:
>
> > I am having problems with a hard disk Fujitsu MPB3064AT UDMA and the
> > kernel 2.0.34 (Red Hat 5.1 and Slackwere 3.5).
Peter
-
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.altern.org/andrebalsa/doc/lkml-faq.html