Re: [PATCH rdma-next v4 0/4] Define and use mana queues for CQs and WQs

[Leon Romanovsky <leon@kernel.org>]

On Mon, Apr 08, 2024 at 01:47:45PM +0000, Konstantin Taranov wrote:
> > From: Leon Romanovsky <leon@kernel.org>
> > On Mon, Apr 08, 2024 at 12:50:12PM +0000, Konstantin Taranov wrote:
> > > > From: Leon Romanovsky <leon@kernel.org> On Mon, Apr 08, 2024 at
> > > > 02:14:02AM -0700, Konstantin Taranov wrote:
> > > > > From: Konstantin Taranov <kotaranov@microsoft.com>
> > > > >
> > > > > This patch series aims to reduce code duplication by introducing a
> > > > > notion of mana ib queues and corresponding helpers to create and
> > > > > destroy them.
> > > > >
> > > > > v3->v4:
> > > > > * Removed debug prints in patches, as asked by Leon
> > > > >
> > > > > v2->v3:
> > > > > * [in 4/4] Do not define an additional struct for a raw qp
> > > > >
> > > > > v1->v2:
> > > > > * [in 1/4] Added a comment about the ignored return value
> > > > > * [in 2/4] Replaced RDMA:mana_ib to RDMA/mana_ib in the subject
> > > > > * [in 4/4] Renamed mana_ib_raw_qp to mana_ib_raw_sq
> > > > >
> > > > > Konstantin Taranov (4):
> > > > >   RDMA/mana_ib: Introduce helpers to create and destroy mana queues
> > > > >   RDMA/mana_ib: Use struct mana_ib_queue for CQs
> > > > >   RDMA/mana_ib: Use struct mana_ib_queue for WQs
> > > > >   RDMA/mana_ib: Use struct mana_ib_queue for RAW QPs
> > > > >
> > > > >  drivers/infiniband/hw/mana/cq.c      | 52 +++-------------
> > > > >  drivers/infiniband/hw/mana/main.c    | 39 ++++++++++++
> > > > >  drivers/infiniband/hw/mana/mana_ib.h | 26 ++++----
> > > > >  drivers/infiniband/hw/mana/qp.c      | 93 +++++++++-------------------
> > > > >  drivers/infiniband/hw/mana/wq.c      | 33 ++--------
> > > > >  5 files changed, 96 insertions(+), 147 deletions(-)
> > > >
> > > > It doesn't apply.
> > > >
> > >
> > > I guess there was some mis-synchronisation between us.
> > > I see that you have already applied the patch 6 days ago:
> > > https://nam06.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgit.
> > >
> > kernel.org%2Fpub%2Fscm%2Flinux%2Fkernel%2Fgit%2Frdma%2Frdma.git%2
> > Flog%
> > >
> > 2F&data=05%7C02%7Ckotaranov%40microsoft.com%7C09ea6de381194295c
> > 4ae08dc
> > >
> > 57cbe121%7C72f988bf86f141af91ab2d7cd011db47%7C1%7C0%7C63848178
> > 04102717
> > >
> > 33%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMz
> > IiLCJBTiI
> > >
> > 6Ik1haWwiLCJXVCI6Mn0%3D%7C0%7C%7C%7C&sdata=jwGGhmatHqdN4bW
> > Xc%2FtyXtubD
> > > ZxxCXpnyL26S5lEKd0%3D&reserved=0
> > >
> > > I am sorry for sending a newer version after the patch has been applied.
> > > I have not checked this before sending.
> > > I can take care of useless debug prints in a future cleanup patch.
> > 
> > Please rebase your series, and resend.
> 
> Sorry for a confusion. I mean you have already applied this patch series (v3) 6 days ago.
> See commits:
> 46f5be7cd4bceb3a503c544b3dab7b75fe4bb96b
> 60a7ac0b8bec5df9764b7460ffee91fc981e8a31
> 688bac28e3dc9eb795ae8ea5aa40cb637e289faa
> f10242b3da908dc9d4bfa040e6511a5b86522499
> 
> As a result, I cannot rebase. I could send a completely new patch that removes some debug prints.

Yes, sorry for not being clear. Please send a cleanup patch.

Thanks