1 |
ok folks... |
2 |
|
3 |
|
4 |
hdparm -v /dev/hda1 |
5 |
|
6 |
/dev/hda1: |
7 |
multcount = 16 (on) |
8 |
IO_support = 0 (default 16-bit) |
9 |
unmaskirq = 0 (off) |
10 |
using_dma = 1 (on) |
11 |
keepsettings = 0 (off) |
12 |
readonly = 0 (off) |
13 |
readahead = 8 (on) |
14 |
geometry = 16708/16/63, sectors = 767950848, start = 0 |
15 |
|
16 |
if i need to change the parameters, please let me know |
17 |
|
18 |
|
19 |
-=g=- |
20 |
On 11/23/05, Gustavo Zacarias <gustavoz@g.o> wrote: |
21 |
> |
22 |
> -----BEGIN PGP SIGNED MESSAGE----- |
23 |
> Hash: SHA1 |
24 |
> |
25 |
> David S. Miller wrote: |
26 |
> |
27 |
> > I'm suspecting rather IDE transfer corruption under load. |
28 |
> > A good datapoint would be if the IDE disk is using DMA or |
29 |
> > PIO mode, and if DMA to retry the emerge with DMA disabled. |
30 |
> > Also, there are other funky IDE feature optimization settings |
31 |
> > that might now be safe on those Ultra5/10 CMD64x IDE chips. |
32 |
> > |
33 |
> > I seem to remember Gentoo turned on all the fancy IDE features |
34 |
> > via hdparm on bootup on sparc64 boxes, correct me if I'm wrong. |
35 |
> |
36 |
> hdparm is never called on startup by default, be it either the LiveCD or |
37 |
> a new installation unless the users adds it to the startup. |
38 |
> CONFIG_BLK_DEV_IDEDMA_PCI is Y on the LiveCD kernel, but |
39 |
> CONFIG_IDEDMA_PCI_AUTO isn't so it shouldn't fire up unless the user |
40 |
> wants so via hdparm. |
41 |
> |
42 |
> - -- |
43 |
> Gustavo Zacarias |
44 |
> Gentoo/SPARC monkey |
45 |
> -----BEGIN PGP SIGNATURE----- |
46 |
> Version: GnuPG v1.4.2-ecc0.1.6 (GNU/Linux) |
47 |
> Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org |
48 |
> |
49 |
> iD8DBQFDhOOiV3G/IBCn/JARAgU7AJ0QtxUs39xaw8lUtyyLYUPxnnf7WwCbBn0E |
50 |
> WJM0+W5qfNvw5nPy9bX+Ejg= |
51 |
> =RIrw |
52 |
> -----END PGP SIGNATURE----- |
53 |
> -- |
54 |
> gentoo-sparc@g.o mailing list |
55 |
> |
56 |
> |