On Wed, Dec 19, 2018 at 07:24:56AM +1100, Stephen Rothwell wrote:
> Hi all,
>
> On Thu, 13 Dec 2018 11:55:29 +1100 Stephen Rothwell <sfr@xxxxxxxxxxxxxxxx> wrote:
> >
> > On Wed, 12 Dec 2018 19:45:53 +0200 Leon Romanovsky <leon@xxxxxxxxxx> wrote:
> > >
> > > From: Leon Romanovsky <leonro@xxxxxxxxxxxx>
> > >
> > > Handle atomic was left as unimplemented from 2013,
> > > remove the code till this part will be developed.
> > >
> > > Remove the dead code by simplifying SW completion logic which is
> > > supposed to be the same for send and receive paths.
> > >
> > > Fixes: e126ba97dba9 ("mlx5: Add driver for Mellanox Connect-IB adapters")
> > > Reported-by: Stephen Rothwell <sfr@xxxxxxxxxxxxxxxx>
> > > Signed-off-by: Leon Romanovsky <leonro@xxxxxxxxxxxx>
> > > Changelog v0->v1:
> > > * Simplified sw_send_comp() and sw_recv_comp() routines
> > > * Deleted last_poll variable, which is not really needed and was
> > > for unnedde optimization, because the logic should be the same
> > > for mlx4 and mlx5 devices.
> >
> > I have applied this to linux-next for today and it fixes the build
> > failure.
> >
> > Tested-by: Stephen Rothwell <sfr@xxxxxxxxxxxxxxxx> # compile tested
>
> I am still applying this patch each day ... is this, or some other fix,
> going to be applied to the rdma tree?
Queued to for-next thanks
Jason
