Christoph Hellwig writes:
> On Wed, Oct 31, 2018 at 12:09:29AM +1100, Benjamin Herrenschmidt wrote:
>> On Tue, 2018-10-30 at 13:58 +0100, Christoph Hellwig wrote:
>> > Please take my patch instead. We have a kernel polcity to not keep
>> > dead code around, and everyone including Linus and the a
On Wed, Oct 31, 2018 at 12:09:29AM +1100, Benjamin Herrenschmidt wrote:
> On Tue, 2018-10-30 at 13:58 +0100, Christoph Hellwig wrote:
> > Please take my patch instead. We have a kernel polcity to not keep
> > dead code around, and everyone including Linus and the attending IBMers
> > confirmed thi
On Tue, 2018-10-30 at 13:58 +0100, Christoph Hellwig wrote:
> Please take my patch instead. We have a kernel polcity to not keep
> dead code around, and everyone including Linus and the attending IBMers
> confirmed this.
Let's call a cat a cat ... ;-)
It's not *dead* code. It's code that is used
Please take my patch instead. We have a kernel polcity to not keep
dead code around, and everyone including Linus and the attending IBMers
confirmed this.
On Tue, Oct 30, 2018 at 10:02:03PM +1100, Alistair Popple wrote:
> The NPU IOMMU is setup to mirror the parent PCIe device IOMMU
> setup. Ther
The NPU IOMMU is setup to mirror the parent PCIe device IOMMU
setup. Therefore it does not make sense to call dma operations such as
dma_map_page, etc. directly on these devices. The existing dma-ops
simply print a warning if they are ever called, however this is
unnecessary and the warnings are li
