> On Jun 17, Alan Cox wrote:
>
> > > If I copy a large (650Mb) file to a different filename, then perform
> > > a compare on them, I encounter random, different miscompares. I've
> > > seen this on several 2.2.9 and 2.2.10 systems.
> >
> > Ok people is this
> >
> > SCSI or IDE (or a mix)
I'm _not_ seeing errors; see below.
System 1 is P200 w/IDE, two AC22100H drives, striped swap. Bootup says:
PIIX3: IDE controller on PCI bus 00 dev 39
PIIX3: not 100% native mode: will probe irqs later
ide0: BM-DMA at 0xe800-0xe807, BIOS settings: hda:pio, hdb:pio
ide1: BM-DMA at 0xe808-0xe80f, BIOS settings: hdc:pio, hdd:pio
hda: WDC AC22100H, ATA DISK drive
hdc: WDC AC22100H, ATA DISK drive
BTW, this is with IDE DMA _OFF_. It's normally on, but when I upgraded to
2.2, it is now off by default and I forgot to turn it on.
System 2 is Alpha EV45 w/ncr53c810 and Micropolis SCSI-II. Bootup says:
ncr53c8xx: at PCI bus 0, device 6, function 0
ncr53c8xx: 53c810 detected
ncr53c810-0: rev=0x02, base=0x4200000, io_port=0x8000, irq=11
ncr53c810-0: ID 7, Fast-10, Parity Checking
ncr53c810-0: restart (scsi reset).
scsi0 : ncr53c8xx - version 3.2
scsi : 1 host.
Vendor: MICROP Model: 3243-19 1128RV Rev: 28RV
Type: Direct-Access ANSI SCSI revision: 02
Detected scsi disk sda at scsi0, channel 0, id 0, lun 0
Vendor: SONY Model: CD-R CDU920S Rev: 2.0b
Type: WORM ANSI SCSI revision: 02
ncr53c810-0-<0,0>: tagged command queue depth set to 8
scsi : detected 1 SCSI disk total.
ncr53c810-0-<0,*>: FAST-10 SCSI 10.0 MB/s (100 ns, offset 8)
SCSI device sda: hdwr sector= 512 bytes. Sectors= 8388315 [4095 MB] [4.1 GB]
> > What compiler was used
Linux version 2.2.9 (bryan@terror) (gcc version 2.7.2.3) #3 Thu May 27 11:45:04 PDT 1999
and
Linux version 2.2.9 (bryan@neptune.terran) (gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)) #5 Mon May 24 22:33:44 PDT 1999
> On Jun 17, Philip Gladstone wrote:
>
> > Ahah -- when I try this with a 120MB file, I get non-random miscompares!
...
> I've done some similar attempts with 110MB and 220MB files
> but I had no `success' (read: no compare errors!).
I ran several tests on system 1 with 100000k files from both /dev/zero and
/dev/urandom, copied to and from both the same drive and the other drive.
I never got any compare errors with cmp -l.
I also tried it on system 2 with 120000k files from one of my raw
partitions and also never saw a mismatch with cmp -l.
Again: I did not see any problem. System 1 has been running for 23 days
under busy load and with lots of disk activity (G's of data touched).
-bp
-- # bryan at terran dot org # http://www.terran.org/~bryan
- 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/