linux-kernel - Re: [RFC 0/8] Copy Offload with Peer-to-Peer PCI Memory

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed, 19 Apr 2017 11:25:49 +1000
From:   Benjamin Herrenschmidt <benh@...nel.crashing.org>
To:     Jason Gunthorpe <jgunthorpe@...idianresearch.com>,
        Dan Williams <dan.j.williams@...el.com>
Cc:     Logan Gunthorpe <logang@...tatee.com>,
        Bjorn Helgaas <helgaas@...nel.org>,
        Christoph Hellwig <hch@....de>,
        Sagi Grimberg <sagi@...mberg.me>,
        "James E.J. Bottomley" <jejb@...ux.vnet.ibm.com>,
        "Martin K. Petersen" <martin.petersen@...cle.com>,
        Jens Axboe <axboe@...nel.dk>,
        Steve Wise <swise@...ngridcomputing.com>,
        Stephen Bates <sbates@...thlin.com>,
        Max Gurtovoy <maxg@...lanox.com>,
        Keith Busch <keith.busch@...el.com>, linux-pci@...r.kernel.org,
        linux-scsi <linux-scsi@...r.kernel.org>,
        linux-nvme@...ts.infradead.org, linux-rdma@...r.kernel.org,
        linux-nvdimm <linux-nvdimm@...1.01.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        Jerome Glisse <jglisse@...hat.com>
Subject: Re: [RFC 0/8] Copy Offload with Peer-to-Peer PCI Memory

On Tue, 2017-04-18 at 17:21 -0600, Jason Gunthorpe wrote:
> Splitting the sgl is different from iommu batching.
> 
> As an example, an O_DIRECT write of 1 MB with a single 4K P2P page in
> the middle.
> 
> The optimum behavior is to allocate a 1MB-4K iommu range and fill it
> with the CPU memory. Then return a SGL with three entires, two
> pointing into the range and one to the p2p.
> 
> It is creating each range which tends to be expensive, so creating
> two
> ranges (or worse, if every SGL created a range it would be 255) is
> very undesired.

I think it's easier to get us started to just use a helper and
stick it in the existing sglist processing loop of the architecture.

As we noticed, stacking dma_ops is actually non-trivial and opens quite
the can of worms.

As Jerome mentioned, you can end up with IOs ops containing an sglist
that is a collection of memory and GPU pages for example.

Cheers,
Ben.