Re: [PATCH rdma-next] IB/mlx5: Fix long EEH recover time with NVMe offloads

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

 



On Wed, Dec 19, 2018 at 04:20:30PM +0000, Jason Gunthorpe wrote:
> On Wed, Dec 19, 2018 at 05:50:08PM +0200, Leon Romanovsky wrote:
>
> > > Generally you don't also put #ifdefs around the structure members when
> > > doing this... ie that is an unnecessary micro optimization.
> >
> > In mlx5_ib, I can easily drop "#ifdef
> > CONFIG_INFINIBAND_ON_DEMAND_PAGING", but what about "struct
> > ib_ucontext"? Are you going to accept it?
>
> I'm tempted to drop the entire config??
>
> Can you justify why we have it?

I can't

>
> > Anyway, can we please progress with current patch and continue
> > discussion?
>
> I'll look again, but I don't see why it should be so ugly..

Does the rest of subsystem already in so great shape that a couple of
ifs which improves performance looks ugly?

I'm working onto prepare followup patch.

>
> Jason

Attachment: signature.asc
Description: PGP signature


[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