Re: [PATCH v2 8/8] habanalabs: enable 64-bit DMA mask in POWER9

From: Oded Gabbay
Date: Sat Jun 15 2019 - 08:18:02 EST


On Tue, Jun 11, 2019 at 6:26 PM Greg KH <gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
>
> On Tue, Jun 11, 2019 at 08:17:53AM -0700, Christoph Hellwig wrote:
> > On Tue, Jun 11, 2019 at 11:58:57AM +0200, Greg KH wrote:
> > > That feels like a big hack. ppc doesn't have any "what arch am I
> > > running on?" runtime call? Did you ask on the ppc64 mailing list? I'm
> > > ok to take this for now, but odds are you need a better fix for this
> > > sometime...
> >
> > That isn't the worst part of it. The whole idea of checking what I'm
> > running to set a dma mask just doesn't make any sense at all.
>
> Oded, I thought I asked if there was a dma call you should be making to
> keep this type of check from being needed. What happened to that? As
> Christoph points out, none of this should be needed, which is what I
> thought I originally said :)
>
> thanks,
>
> greg k-h

Hi Greg,
So after the dust has settled a bit, do you think it is reasonable to
add this patch upstream ?

As Benjamin and Oliver mentioned, there is no better-looking/standard
way to solve this, considering my device's limitations.
AFAICS, the only way is either this hack, or the kernel module parameter method.

I'll of course monitor the PPC code upstream and if they will manage
to push a fix to their current DMA mask limitation (that will allow
setting dma mask of 48 bits and without setting bit 59 in outbound
transactions), I will modify my code accordingly and then this hack
won't be necessary. But for now, it is what it is.

What do you think ?

Oded