[Openchrome-devel] PCI bus corruption?

[Yuval Hager]

On Wednesday 26 November 2008, Yuval Hager wrote:
> On Wednesday 26 November 2008, Peter Stuge wrote:
> > My question is if there are a few strategic places in openchrome
> > where MMIO accesses could be logged and/or validated to actually be
> > within the device's MMIO range, to find a starting point for further
> > debugging of this problem?
>
> I've narrowed this down to:
>
> Index: src/via_lvds.c
> ===================================================================
> --- src/via_lvds.c	(revision 686)
> +++ src/via_lvds.c	(working copy)
> @@ -86,7 +86,10 @@ ViaLVDSDFPPower(ScrnInfoPtr pScrn, Bool on)
>
>      if (on) {
>          /* Turn DFP High/Low pad on. */
> +        DEBUG(xf86DrvMsg(pScrn->scrnIndex, X_INFO, "YH: ViaLVDSDFPPower
> NOT WRITING!\n"));
> +	/*
>          hwp->writeSeq(hwp, 0x2A, hwp->readSeq(hwp, 0x2A) | 0x0F);
> +	*/
>      } else {
>          /* Turn DFP High/Low pad off. */
>          hwp->writeSeq(hwp, 0x2A, hwp->readSeq(hwp, 0x2A) & 0x0F);
>
>
> With this patch applied, I cannot recreate the problem.
>

I can confirm this is working perfectly for the last 3 days. The wireless is 
working flawlessly through endless DPMS activations and deactivations, 
suspend-to-ram, and a few suspend-to-disk.
However, I am quite certain this is not the "right" fix - I would love to know 
if you have figured out the problem and the correct solution for it.

> When I logout (X shutdown?) the whole system hangs, but I am not sure this
> is a result of this change - it might have been the X.org upgrade.
>

Scratch that. It might have been a problem in my testing. This doesn't happen 
anymore.

Cheers,

--yuval
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part.
Url : http://wiki.openchrome.org/pipermail/openchrome-devel/attachments/20081130/e9e4ffc4/attachment.bin