Re: [PATCH rdma-next v2 0/4] Mark create/destroy address handle calls as atomic/sleepable

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

 



On Wed, Dec 12, 2018 at 11:09:04AM +0200, Gal Pressman wrote:
> Hi Jason and Doug,
> This patchset adds a new 'flags' field to create/destroy address handle flow
> and a flag that marks whether the callback is executed in a sleepable context
> in order to allow drivers to wait for device completion instead of busy polling
> when possible.
> 
> Patches #1, #2 add the flag to create/destroy AH callbacks and change all the
> drivers, patches #3, #4 change bnxt_re driver to make use of the flag and not
> poll for firmware completion.
> 
> Note that I left the bool function parameter in the bnxt_re internal qplib
> calls because that's consistent with all other functions, changed it anywhere
> else.
> 
> Changelog:
> v1->v2:
> * Rebased on top of rdmavt callbacks fixes by Kamal and adjust the callbacks accordingly
> * Use flags and an enum instead of bool flag
> * Make bnxt_re_create_shadow_qp_ah call create AH as sleepable (changed from non-sleepable)
> * Make bnxt_re_destroy_qp call destroy AH as sleepable (changed from non-sleepable)

I rebased this and applied it to for-next

Thanks,
Jason



[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