Hi Matan, On 3/31/20 1:12 PM, Matan Azrad wrote: > Since a virtq configuration may be changed in disable state it is better to > recreate a virtq becoming enabled. > This series adding this behaviour to the mlx5 driver for vDPA. > > Matan Azrad (3): > vdpa/mlx5: manage virtqs by array > vdpa/mlx5: separate virtq stop > vdpa/mlx5: recteate a virtq becoming enabled > > drivers/vdpa/mlx5/mlx5_vdpa.c | 47 +++++++-------- > drivers/vdpa/mlx5/mlx5_vdpa.h | 54 +++++++++++++++-- > drivers/vdpa/mlx5/mlx5_vdpa_lm.c | 57 ++++++++---------- > drivers/vdpa/mlx5/mlx5_vdpa_steer.c | 103 ++++++++++++++++---------------- > drivers/vdpa/mlx5/mlx5_vdpa_virtq.c | 113 > +++++++++++++++++++++++++++--------- > 5 files changed, 231 insertions(+), 143 deletions(-) >
I wanted to apply your series, but it fails to apply: $ git am -3 vdpa-mlx5-recteate-a-virtq-becoming-enabled.patch Applying: vdpa/mlx5: manage virtqs by array error: sha1 information is lacking or useless (drivers/vdpa/mlx5/mlx5_vdpa.c). error: could not build fake ancestor Patch failed at 0001 vdpa/mlx5: manage virtqs by array hint: Use 'git am --show-current-patch' to see the failed patch When you have resolved this problem, run "git am --continue". If you prefer to skip this patch, run "git am --skip" instead. To restore the original branch and stop patching, run "git am --abort". Could you please rebase it on top of dpdk-next-virtio/master? While doing that, if you could take fix the minor issues/typos I * reported, that would be great! Thanks, Maxime
