Re: [PATCH v2] rdma: Enable ib_alloc_cq to spread work over a device's comp_vectors

From: Jason Gunthorpe <jgg@ziepe.ca>
To: Leon Romanovsky <leon@kernel.org>
Cc: Chuck Lever <chuck.lever@oracle.com>,
	linux-rdma@vger.kernel.org, linux-nfs@vger.kernel.org,
	linux-cifs@vger.kernel.org, v9fs-developer@lists.sourceforge.net
Subject: Re: [PATCH v2] rdma: Enable ib_alloc_cq to spread work over a device's comp_vectors
Date: Mon, 29 Jul 2019 11:19:45 -0300	[thread overview]
Message-ID: <20190729141945.GF17990@ziepe.ca> (raw)
In-Reply-To: <20190729054933.GK4674@mtr-leonro.mtl.com>

On Mon, Jul 29, 2019 at 08:49:33AM +0300, Leon Romanovsky wrote:
> > +/**
> > + * ib_alloc_cq_any: Allocate kernel CQ
> > + * @dev: The IB device
> > + * @private: Private data attached to the CQE
> > + * @nr_cqe: Number of CQEs in the CQ
> > + * @poll_ctx: Context used for polling the CQ
> > + */
> > +static inline struct ib_cq *ib_alloc_cq_any(struct ib_device *dev,
> > +					    void *private, int nr_cqe,
> > +					    enum ib_poll_context poll_ctx)
> > +{
> > +	return __ib_alloc_cq_any(dev, private, nr_cqe, poll_ctx,
> > +				 KBUILD_MODNAME);
> > +}
> 
> This should be macro and not inline function, because compiler can be
> instructed do not inline functions (there is kconfig option for that)
> and it will cause that wrong KBUILD_MODNAME will be inserted (ib_core
> instead of ulp).

No, it can't, a static inline can only be de-inlined within the same
compilation unit, so KBUILD_MODNAME can never be mixed up.

You only need to use macros of the value changes within the
compilation unit.

Jason