Re: [PATCH 0/4] RDMA/srp: Handle dev_set_name() failure

[Jason Gunthorpe <jgg@ziepe.ca>]

On Sun, Aug 28, 2022 at 12:50:28PM -0700, Bart Van Assche wrote:
> On 8/28/22 03:04, Leon Romanovsky wrote:
> > On Thu, Aug 25, 2022 at 02:38:56PM -0700, Bart Van Assche wrote:
> > > This patch series includes one patch that handles dev_set_name() failure and
> > > three refactoring patches. Please consider these patches for the next merge
> > > window.
> > 
> > You confuse me. "next merge window" means that patches are targeted to
> > -next, but you added stable@... tag and didn't add any Fixes lines.
> > 
> > I applied everything to rdma-next and removed stable@ tag.
> 
> Hi Leon,
> 
> Although it's not a big deal for this patch series, please do not modify patches
> without agreement from the patch author.
> 
> As far as I know adding a Fixes: tag if a Cc: stable tag is present is not required
> by any document in the Documentation/ directory?
> 
> I had not added a Fixes: tag because the issue fixed by patch 3/3 was introduced
> by the commit that added the ib_srp driver to the kernel tree. So it would be fine
> to backport the first three patches of this series to all older kernel versions to
> which the patches can be backported.

Generally I always want to see the Fixes tag because it aides everyone
in the ecosystem to know when they should consider the patch for
backporting with a simple uniform algorithm.

So marking the first commit as the thing being fixed is expected.

Jason