On Wed, Jun 19, 2019 at 9:13 AM Andy Shevchenko
<andriy.shevchenko@xxxxxxxxxxxxxxx> wrote:
>
> On Wed, Mar 20, 2019 at 04:31:17PM -0500, Bjorn Helgaas wrote:
> > On Wed, Mar 20, 2019 at 11:13 AM Andy Shevchenko
> > <andriy.shevchenko@xxxxxxxxxxxxxxx> wrote:
> > >
> > > After the commit cf65a0f6f6ff
> > >
> > > ("dma-mapping: move all DMA mapping code to kernel/dma")
> > >
> > > some of the files are referring to outdated information, i.e. old file names
> > > of DMA mapping sources.
> > >
> > > Fix it here.
>
> Bjorn, thanks for review, my answers below.
>
> > > * This function checks if the reserved crashkernel is allowed on the specific
> > > * IA64 machine flavour. Machines without an IO TLB use swiotlb and require
> > > * some memory below 4 GB (i.e. in 32 bit area), see the implementation of
> > > - * lib/swiotlb.c. The hpzx1 architecture has an IO TLB but cannot use that
> > > + * kernel/dma/swiotlb.c. The hpzx1 architecture has an IO TLB but cannot use that
> > > * in kdump case. See the comment in sba_init() in sba_iommu.c.
> >
> > Is the point here that just that if you lack an IOTLB and want devices
> > to be able to reach system memory above 4GB, you need a bounce buffer
> > below 4GB? If so, maybe we could just say *that* instead of a
> > nebulous reference to "the implementation of */swiotlb.c", which
> > doesn't tell you what part of the implementation is relevant.
>
> This patch is about broken links.
Oh, of course! Sorry, I don't know what I was thinking. I was
wasting my time (and yours) with comments about things you're not
changing, sorry about that.
Bjorn
