Hi Marcelo, Richard, Ivan, Jay,
On Thu, Jan 10, 2002 at 06:30:07PM -0200, Marcelo Tosatti wrote:
> pre3:
>
[...]
> - Miata dma corruption workaround (Richard Henderson)
[...]
> pre2:
[...]
> - Alpha fixes (Jay Estabrook)
[...]
To be honest, I expected to see the patch from Ivan go in.
He came up with the solution to disable scatter-gather for pci dma, after I
sent a patch which justs worked around the page boundary cross bug for
ide-dma on the miata/pyxis. Disabling SG for PCI DMA on miata is no problem,
as we can directly map a 2GB window of memory into PCI space.
A second patch also made sure it works for ISA (which also support 32bis on
pyxis), so the floppy controller is happy again.
This second patch is the only thing I see when having a quick look at the
2.4.18pre3 patch, and Ivan should be credited for it as far as I know.
Is this an oversight? Or is there seom magic bit switched which I overlooked
which switches off SG PCI DMA on pyxis?
Regards,
-- Kurt Garloff <garloff@suse.de> Eindhoven, NL GPG key: See mail header, key servers Linux kernel development SuSE GmbH, Nuernberg, DE SCSI, Security
This archive was generated by hypermail 2b29 : Tue Jan 15 2002 - 21:00:37 EST