RE: Setting UDMA on ide drives

Thomas Molina (tmolina@home.com)
Thu, 28 Oct 1999 07:03:13 -0500 (CDT)


On Thu, 28 Oct 1999, Michael Cummins wrote:

> It's more important we know what your UDMA chipset is.
> You could be having the same problem I used to have.
> I have an ALI M15xx chipset, and I need to patch the kernel to allow DMA
> switching.
> So what's your chipset, and does the kernel correctly identify it on boot?
> Michael Cummins mailto:mjcumm@isa.mim.com.au

The motherboard is a TMC TI5VG+. For the chipset my user's manual says
the MVP3 has built-in PCI EIDE with support for UDMA/33. lspci reports
the following:

00:07.1 IDE interface: VIA Technologies, Inc. VT82C586 IDE [Apollo] (rev
06) (prog-if 8a)
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B-
Status: Cap- 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 64 set
Region 4: I/O ports at e000

I've seen the chips it mentions, but haven't seen any of VIA's docs
mentioning capabilities. I should visit VIA's website. I think some
people on lkml consider the chipset broken.

Looking further, I read Documentation/Configure.help and found some
interesting info there. I also see that both CONFIG_BLK_DEV_IDEPCI, and
CONFIG_BLK_DEV_IDEDMA_PCI are not set. I have this memory of reading
about those needing to be unset in order to make ide-scsi emulation
work. If that is true I may be stuck choosing between UDMA and ide-scsi
emulation (needed for my CD-RW to work).

-
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/