Re: [PATCH v1 09/10] iommufd/iova_bitmap: Dynamic pinning on iova_bitmap_set()

[Jason Gunthorpe <jgg@nvidia.com>]

On Tue, Jun 25, 2024 at 04:27:24PM +0100, Joao Martins wrote:
> On 25/06/2024 13:27, Jason Gunthorpe wrote:
> > On Tue, Jun 25, 2024 at 12:13:15PM +0100, Joao Martins wrote:
> >> On 25/06/2024 04:38, Tian, Kevin wrote:
> >>>> From: Joao Martins <joao.m.martins@oracle.com>
> >>>> Sent: Saturday, June 22, 2024 2:43 AM
> >>>>
> >>>> Link: https://lore.kernel.org/linux-iommu/6b90f949-48da-4cb3-ad9a-
> >>>> ed54f1351a9a@oracle.com/
> >>>> Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
> >>>
> >>> missed a fixed tag?
> >>
> >> Technically it is:
> >>
> >> Fixes: 2780025e01e2 ("iommufd/iova_bitmap: Handle recording beyond the mapped
> >> pages")
> >>
> >> Though I deliberately didn't add it, because we need the patch afterwards to be
> >> fully fixed in iova_bitmap_for_each() callers. Not quite sure what to do in
> >> these cases
> > 
> > Add the Fixes tag to both is better than not having the tag at all..
> > 
> Right -- but do I propagate that to all the dependencies? That was sort of what
> I was wondering. Technically it's patches 6-10 would then get the
> Fixes tag.

Well, if the patch actually fixes something, even it is only part of
something then it should have the tag.

If it is just refactoring to prepare then no it doesn't.

> cases'. For just classifying the commit it fixes, then I just need on this
> patch. Putting on the second one was more for stable tree, and in that case it
> would actually be all from 6-10 not just 10.

Fixes tag is not for dependencies, that has to be sorted out some
other way..

Jason