Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Wed, 9 Sep 2020 10:05:18 -0300
From:   Jason Gunthorpe <jgg@ziepe.ca>
To:     Christoph Hellwig <hch@infradead.org>
Cc:     Ming Mao <maoming.maoming@huawei.com>,
        linux-kernel@vger.kernel.org, kvm@vger.kernel.org,
        linux-mm@kvack.org, alex.williamson@redhat.com,
        akpm@linux-foundation.org, cohuck@redhat.com,
        jianjay.zhou@huawei.com, weidong.huang@huawei.com,
        peterx@redhat.com, aarcange@redhat.com, wangyunjian@huawei.com,
        willy@infradead.org, jhubbard@nvidia.com
Subject: Re: [PATCH V4 1/2] vfio dma_map/unmap: optimized for hugetlbfs pages
Message-ID: <20200909130518.GE87483@ziepe.ca>
References: <20200908133204.1338-1-maoming.maoming@huawei.com>
 <20200908133204.1338-2-maoming.maoming@huawei.com>
 <20200909080114.GA8321@infradead.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20200909080114.GA8321@infradead.org>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed, Sep 09, 2020 at 09:01:14AM +0100, Christoph Hellwig wrote:
> I really don't think this approach is any good.  You workaround
> a deficiency in the pin_user_pages API in one particular caller for
> one particular use case.

RDMA has the same basic issues, this should should not be solved with
workarounds in VFIO - a common API would be good

> I think you'd rather want either:
> 
>  (1) a FOLL_HUGEPAGE flag for the pin_user_pages API family that returns
>      a single struct page for any kind of huge page, which would also
>      benefit all kinds of other users rather than adding these kinds of
>      hacks to vfio.

How to use? The VMAs can have mixed page sizes so the caller would
have to somehow switch and call twice? Not sure this is faster.

>  (2) add a bvec version of the API that returns a variable size
>      "extent"

This is the best one, I think.. The IOMMU setup can have multiple page
sizes, so having largest contiguous blocks pre-computed should speed
that up.

vfio should be a win to use a sgl rather than a page list?

Especially if we can also reduce the number of pages pinned by only
pinning head pages..

> I had started on (2) a while ago, and here is branch with my code (which
> is broken and fails test, but might be a start):
> 
> http://git.infradead.org/users/hch/misc.git/shortlog/refs/heads/gup-bvec
> 
> But for now I wonder if (1) is the better start, which could still be
> reused to (2) later.

What about some 'pin_user_page_sgl' as a stepping stone?

Switching from that point to bvec seems like a smaller step?

Jason