Re: IO/MMIO 2.4 ATA/IDE driver recore near complete

Re: IO/MMIO 2.4 ATA/IDE driver recore near complete

[Benjamin Herrenschmidt]

>Please note BMDMA for MMIO is not native since there appears to be
>pci-posting errors under x86.

Can you elaborate ? I had to deal with driver bugs related to
PCI wr. posting in the past. I assume you know what you are doing,
but just in case, you missed a SW bug, I may be able to help.

Ben.

IO/MMIO 2.4 ATA/IDE driver recore near complete

[Andre Hedrick]

Uniform Multi-Platform E-IDE driver Revision: 6.31
ide: Assuming 33MHz system bus speed for PIO modes
PIIX3: IDE controller on PCI bus 00 dev 39
PIIX3: chipset revision 0
PIIX3: not 100% native mode: will probe irqs later
    ide0: BM-DMA at 0xffa0-0xffa7, BIOS settings: hda:pio, hdb:pio
    ide1: BM-DMA at 0xffa8-0xffaf, BIOS settings: hdc:pio, hdd:pio
SiI680: IDE controller on PCI bus 00 dev 90
SiI680: chipset revision 1
SiI680: not 100% native mode: will probe irqs later
SiI680: mmio capable at 0xe080df00
SiI680: BASE CLOCK == 2X PCI
    ide2: MMIO-DMA at 0xe080df00-0xe080df07, BIOS settings: hde:pio, hdf:pio
    ide3: MMIO-DMA at 0xe080df08-0xe080df0f, BIOS settings: hdg:pio, hdh:pio
hda: ATAPI 44X CDROM, ATAPI CD/DVD-ROM drive
hdb: CREATIVEDVD5240E-1, ATAPI CD/DVD-ROM drive
hde: Maxtor 4G160J8, ATA DISK drive
hdg: Maxtor 4G160H8, ATA DISK drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide2 at 0xe080dfc0-0xe080dfc7,0xe080dfca on irq 9
ide3 at 0xe080df80-0xe080df87,0xe080df8a on irq 9
hde: host protected area => 1
hde: 320173056 sectors (163929 MB) w/2048KiB Cache, CHS=19929/255/63
hdg: host protected area => 1
hdg: 320173056 sectors (163929 MB) w/2048KiB Cache, CHS=19929/255/63
hda: ATAPI 40X CD-ROM drive, 128kB Cache, (U)DMA
Uniform CD-ROM driver Revision: 3.12
hdb: ATAPI 32X DVD-ROM drive, 512kB Cache, DMA

Please note BMDMA for MMIO is not native since there appears to be
pci-posting errors under x86.

Cheers,

Andre Hedrick
LAD Storage Consulting Group

Re: IO/MMIO 2.4 ATA/IDE driver recore near complete

[Ingo Oeser]

On Fri, May 17, 2002 at 04:15:15AM +0100, Benjamin Herrenschmidt wrote:
> >Please note BMDMA for MMIO is not native since there appears to be
> >pci-posting errors under x86.
> 
> Can you elaborate ? I had to deal with driver bugs related to
> PCI wr. posting in the past. I assume you know what you are doing,
> but just in case, you missed a SW bug, I may be able to help.

I'm also interested in these issues, since I plan to use BMDMA
into MMIO areas in my own driver. So maybe we should discuss this
on the list? ;-)

Also the machines/chipsets, where this happens would be nice to
know. How to detect and workaround such errors is another
challenging topic I would like to discuss.

Thanks and Regards

Ingo Oeser
-- 
Science is what we can tell a computer. Art is everything else. --- D.E.Knuth

Re: IO/MMIO 2.4 ATA/IDE driver recore near complete

[Anssi Saari]

I wonder if this would help me with my audio CD writing problem which
seems to use PIO mode and slow my system to a snails pace?

Re: IO/MMIO 2.4 ATA/IDE driver recore near complete

[Jeff Garzik]

On Sun, May 19, 2002 at 01:33:24PM +0300, Anssi Saari wrote:
> I wonder if this would help me with my audio CD writing problem which
> seems to use PIO mode and slow my system to a snails pace?

Nope.  It depends on your ATA controller, and the vast majority of
controllers currently on the market use PIO.

You are thinking about PIO versus DMA, which are the two data transfer
modes of ATA.  We are talking about a situation which replaces PIO with
MMIO, but otherwise looks and acts similarly to PIO+DMA.

	Jeff