RE: [PATCH rdma-next 4/4] RDMA/verbs: Simplify modify QP check

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 




> -----Original Message-----
> From: linux-rdma-owner@xxxxxxxxxxxxxxx [mailto:linux-rdma-
> owner@xxxxxxxxxxxxxxx] On Behalf Of Dennis Dalessandro
> Sent: Monday, March 12, 2018 8:08 AM
> To: Leon Romanovsky <leon@xxxxxxxxxx>; Doug Ledford
> <dledford@xxxxxxxxxx>; Jason Gunthorpe <jgg@xxxxxxxxxxxx>
> Cc: Leon Romanovsky <leonro@xxxxxxxxxxxx>; RDMA mailing list <linux-
> rdma@xxxxxxxxxxxxxxx>; Maor Gottlieb <maorg@xxxxxxxxxxxx>
> Subject: Re: [PATCH rdma-next 4/4] RDMA/verbs: Simplify modify QP check
> 
> On 3/11/2018 7:51 AM, Leon Romanovsky wrote:
> > From: Leon Romanovsky <leonro@xxxxxxxxxxxx>
> >
> > All callers to ib_modify_qp_is_ok() provides enum ib_qp_state makes
> > the checks of out-of-scope redundant. Let's remove them together with
> > updating function signature to return boolean result.
> 
> I'm not so sure this is true. Rdmavt just pulls the current and new state and
> passes but doesn't check. I'm fine adding the checks to rdmavt if all the other
> drivers do their own but I would think it's best to leave the checks in the
> common/core function and remove the duplicated code checks from all the
> drivers.
Mlx5 driver performs QP state transition check and actual transition under a lock as modify_qp can be called from multiple user level threads.
Therefore once similar functionality is available in core, it is desired to remove from the HCA drivers.

��.n��������+%������w��{.n�����{���fk��ܨ}���Ơz�j:+v�����w����ޙ��&�)ߡ�a����z�ޗ���ݢj��w�f




[Index of Archives]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Photo]     [Yosemite News]     [Yosemite Photos]     [Linux Kernel]     [Linux SCSI]     [XFree86]

  Powered by Linux