Re: [PATCH] IB/ipath: use arch_phys_wc_add() and require PAT disabled

[Doug Ledford <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>]

[-- Attachment #1: Type: text/plain, Size: 2917 bytes --]

On Wed, 2015-04-29 at 09:01 -0700, Luis R. Rodriguez wrote:
> On Mon, Apr 27, 2015 at 11:38 AM, Doug Ledford <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> wrote:
> > On Mon, 2015-04-27 at 09:46 -0700, Luis R. Rodriguez wrote:
> >> On Wed, Apr 22, 2015 at 12:26 PM, Luis R. Rodriguez
> >> <mcgrof-3uybbJdB1yH774rrrx3eTA@public.gmane.org> wrote:
> >> > From: "Luis R. Rodriguez" <mcgrof-IBi9RG/b67k@public.gmane.org>
> >> >
> >> > We are burrying direct access to MTRR code support on
> >> > x86 in order to take advantage of PAT. In the future we
> >> > also want to make the default behaviour of ioremap_nocache()
> >> > to use strong UC, use of mtrr_add() on those systems
> >> > would make write-combining void.
> >> >
> >> > In order to help both enable us to later make strong
> >> > UC default and in order to phase out direct MTRR access
> >> > code port the driver over to arch_phys_wc_add() and
> >> > annotate that the device driver requires systems to
> >> > boot with PAT disabled, with the nopat kernel parameter.
> >> >
> >> > This is a worthy compromise given that the ipath device
> >> > driver powers the old HTX bus cards that only work in
> >> > AMD systems, while the newer IB/qib device driver
> >> > powers all PCI-e cards. The ipath device driver is
> >> > obsolete, hardware hard to find and because of this
> >> > this its a reasonable compromise to make to require
> >> > users of ipath to boot with nopat.
> >>
> >> Hey folks, I realize its being discussed whether or not to remove the
> >> driver entirely from the kernel but in the meantime, is this a
> >> reasonable compromise ?
> >
> > [ trimmed Cc: list to probably the only people that care ]
> >
> > I would think so.
> 
> OK great, can this be merged then? I am waiting for this patch to help
> move forward with deprecating MTRR upstream.

I'm generally OK with picking this up and merging it in.  However, it
relies on pat_enabled being an exported symbol, and that isn't the case.
Are you submitting an additional patch via a different tree to make that
symbol exported to modules?

> > I think we might as well mark this driver as
> > deprecated and put a tentative date on removal while we are at it.
> 
> OK great, that can be done as a separate patch.
> 
> > Mike, any specific input here?  I would suggest mark it deprecated with
> > a planned removal sometime in late 2015/early 2016.
> 
> I will note that feature-removal file was removed from Linux so if we
> want to deprecate something now we just do it.
> 
>  Luis
> --
> To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html


-- 
Doug Ledford <dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
              GPG KeyID: 0E572FDD



[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 819 bytes --]