lkml.org 
[lkml]   [2011]   [Nov]   [8]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH] pci: More PRI/PASID cleanup
From
Date
On Tue, 2011-11-08 at 18:17 +0100, Roedel, Joerg wrote:
> On Tue, Nov 08, 2011 at 09:44:30AM -0700, Alex Williamson wrote:
>
> > bit 0 (PCI_PASID_ENABLE) is reserved in the CAP register...
>
> Is it? Which spec are you using? In my version it is not reserved but
> states if it is supported to set the enable-bit.

Latest I can find is the March 31, 2011 PASID ECN, which just lists that
bit as reserved.

> > Which means we need to check CTRL, not CAP to see if it was previously
> > enabled... or maybe this check is entirely wrong and we're was trying to
> > see if enable is supported.
>
> I will check how this looks in my test environment.
>
> > And nobody exposes PCI_PASID_ENABLE because it doesn't exist as a
> > capability.
> >
> > It's easy to see this if the bit definitions are named appropriately and
> > specified per register instead of being lumped together as "close
> > enough". Thanks,
>
> I don't object against your renames as long as it doesn't cause
> merge-conflicts with what I plan to send upstream.

I can drop it if need be, was just trying to do some cleanup on the
consistency of pci_reg.h before adding a bunch more defines to help
bounds checking and parsing for vfio-pci. Unless my spec is outdated,
it seems like there's more than an aesthetic change here though, so
resolving the conflicts with your latest work might be warranted.
Thanks,

Alex



\
 
 \ /
  Last update: 2011-11-08 18:35    [W:0.045 / U:0.360 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site