1 |
On Saturday 09 November 2002 20:36, Michael R. Gayeski wrote: |
2 |
> Are you using 80-wire cables for your drive? The motherboard will |
3 |
> politely refuse to operate at UDMA 100 without the new cables. Also, |
4 |
|
5 |
I have no idea about the AMD driver, but my VIA driver creates a |
6 |
/proc/ide/via, which contains a lot of useful info on the controller, |
7 |
including what type of cables are detected. |
8 |
|
9 |
> I believe there is a kernel option to force UDMA100 at boot, but I |
10 |
> seem to remember it not being recommended. |
11 |
|
12 |
>From Documentation/Configure.help (on 2.4.20-pre10-ac2): |
13 |
|
14 |
Generic PCI bus-master DMA support |
15 |
CONFIG_BLK_DEV_IDEDMA_PCI |
16 |
If your PCI system uses IDE drive(s) (as opposed to SCSI, say) and |
17 |
is capable of bus-master DMA operation (most Pentium PCI systems), |
18 |
you will want to say Y here to reduce CPU overhead. You can then use |
19 |
the "hdparm" utility to enable DMA for drives for which it was not |
20 |
enabled automatically. By default, DMA is not enabled automatically |
21 |
for these drives, but you can change that by saying Y to the |
22 |
following question "Use DMA by default when available". You can get |
23 |
[..snip..] |
24 |
Use PCI DMA by default when available |
25 |
CONFIG_IDEDMA_PCI_AUTO |
26 |
Prior to kernel version 2.1.112, Linux used to automatically use |
27 |
DMA for IDE drives and chipsets which support it. Due to concerns |
28 |
about a couple of cases where buggy hardware may have caused damage, |
29 |
the default is now to NOT use DMA automatically. To revert to the |
30 |
previous behaviour, say Y to this question. |
31 |
|
32 |
If you suspect your hardware is at all flakey, say N here. |
33 |
Do NOT email the IDE kernel people regarding this issue! |
34 |
|
35 |
It is normally safe to answer Y to this question unless your |
36 |
motherboard uses a VIA VP2 chipset, in which case you should say N. |
37 |
|
38 |
|
39 |
It shouldn't be a problem at all. |
40 |
|
41 |
Regards, |
42 |
Dewet |
43 |
|
44 |
|
45 |
-- |
46 |
gentoo-dev@g.o mailing list |