Another NVMe failure, this time with AER info

[Bjorn Helgaas]

On Fri, May 11, 2018 at 11:42:42AM -0600, Keith Busch wrote:
> On Fri, May 11, 2018 at 11:26:11AM -0600, Keith Busch wrote:
> > I trust you know the offsets here, but it's hard to tell what this
> > is doing with hard-coded addresses. Just to be safe and for clarity,
> > I recommend the 'CAP_*+<offset>' with a mask.
> > 
> > For example, disabling ASPM L1.2 can look like:
> > 
> >  # setpci -s <B:D.f> CAP_PM+8.l=0:4
> 
> My mistake above: CAP_PM is a different capability, not the intended
> one. It looks like setpci doesn't even have a convenient symbol for the
> L1 PM extended capability, so the hard-coded offsets are the only way
> for this setting. Sorry about the mistake.

No problem, would be a nice janitor job to add more of those symbolic
offsets to setpci.  I recently added more cap ID decoding to lspci, but
didn't think about updating these things for setpci.

Using the symbols would definitely make this easier and better!