[PATCH linux dev-4.10 13/16] Revert "drivers/fsi: GPIO stability changes for Cronus/Hostboot"

[Andrew Jeffery]

On Thu, 2018-02-15 at 11:46 -0600, Christopher Bostic wrote:
> 
> On 2/15/18 6:36 AM, Andrew Jeffery wrote:
> > This reverts commit cecf390c294cd6cf1080ed45cf0966691879750d.
> > 
> > The cost of reading OCC hwmon attributes is reduced by roughly 50%[0] when we
> > do not incur large delays between cycles in the FSI GPIO master. No impact was
> > observed on the stability of Hostboot by reverting the delays, though
> > admittedly it was not tested with Cronus.
> 
> Hi Andrew,
> 
> The delay values you are reverting were chosen because they fixed highly 
> intermittent FSI ECC bus failures flagged during heavy bus traffic 
> generated via Cronus.   I'd not feel comfortable with this update unless 
> it can be verified this is no longer an issue.
> 

Yes. I anticipated this would be the response. I put the patch in the
series to demonstrate the potential performance gain of verifying that
we no longer hit the issue. Maybe we can chat offline about how we
might do that.

Cheers,

Andrew
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 801 bytes
Desc: This is a digitally signed message part
URL: <http://lists.ozlabs.org/pipermail/openbmc/attachments/20180216/8dbe487a/attachment.sig>