* Re: [PATCH 3/5] Move dma_ops from archdata into struct device [not found] ` <58a325aa-8df7-6a2e-fadb-5d9031d8a1a0@sandisk.com> @ 2016-12-09 18:22 ` Christoph Hellwig 2016-12-09 19:13 ` David Woodhouse [not found] ` <1481310814.98151.11.camel@infradead.org> 0 siblings, 2 replies; 3+ messages in thread From: Christoph Hellwig @ 2016-12-09 18:22 UTC (permalink / raw) To: Bart Van Assche Cc: linux-arch, Sagi Grimberg, linux-rdma@vger.kernel.org, linux-kernel, virtualization, Doug Ledford, David Woodhouse, Christoph Hellwig We'll need a bit of a wieder audience for this I think.. On Wed, Dec 07, 2016 at 05:11:28PM -0800, Bart Van Assche wrote: > Additionally, introduce set_dma_ops(). A later patch will introduce a > call to that function in the RDMA drivers that will be modified to use > dma_noop_ops. This looks good to me, and we had a lot of talk about this for other purposes for a while. ^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH 3/5] Move dma_ops from archdata into struct device 2016-12-09 18:22 ` [PATCH 3/5] Move dma_ops from archdata into struct device Christoph Hellwig @ 2016-12-09 19:13 ` David Woodhouse [not found] ` <1481310814.98151.11.camel@infradead.org> 1 sibling, 0 replies; 3+ messages in thread From: David Woodhouse @ 2016-12-09 19:13 UTC (permalink / raw) To: Christoph Hellwig, Bart Van Assche Cc: linux-arch, Sagi Grimberg, linux-rdma@vger.kernel.org, linux-kernel, virtualization, Doug Ledford [-- Attachment #1.1: Type: text/plain, Size: 721 bytes --] On Fri, 2016-12-09 at 19:22 +0100, Christoph Hellwig wrote: > We'll need a bit of a wieder audience for this I think.. > > On Wed, Dec 07, 2016 at 05:11:28PM -0800, Bart Van Assche wrote: > > Additionally, introduce set_dma_ops(). A later patch will introduce a > > call to that function in the RDMA drivers that will be modified to use > > dma_noop_ops. > > This looks good to me, and we had a lot of talk about this for other > purposes for a while. Hm, I'm not convinced we want per-device dma_ops. What we want is per- device IOMMU ops, and any dma_ops are just a generic or platform- specific (in some cases) wrapper around those. We shouldn't normally need per-device DMA ops at all. -- dwmw2 [-- Attachment #1.2: smime.p7s --] [-- Type: application/x-pkcs7-signature, Size: 5760 bytes --] [-- Attachment #2: Type: text/plain, Size: 183 bytes --] _______________________________________________ Virtualization mailing list Virtualization@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/virtualization ^ permalink raw reply [flat|nested] 3+ messages in thread
[parent not found: <1481310814.98151.11.camel@infradead.org>]
* Re: [PATCH 3/5] Move dma_ops from archdata into struct device [not found] ` <1481310814.98151.11.camel@infradead.org> @ 2016-12-09 19:46 ` Bart Van Assche 0 siblings, 0 replies; 3+ messages in thread From: Bart Van Assche @ 2016-12-09 19:46 UTC (permalink / raw) To: David Woodhouse, Christoph Hellwig Cc: linux-arch, Sagi Grimberg, linux-rdma@vger.kernel.org, linux-kernel, virtualization, Doug Ledford On 12/09/2016 11:13 AM, David Woodhouse wrote: > On Fri, 2016-12-09 at 19:22 +0100, Christoph Hellwig wrote: >> We'll need a bit of a wieder audience for this I think.. >> >> On Wed, Dec 07, 2016 at 05:11:28PM -0800, Bart Van Assche wrote: >>> Additionally, introduce set_dma_ops(). A later patch will introduce a >>> call to that function in the RDMA drivers that will be modified to use >>> dma_noop_ops. >> >> This looks good to me, and we had a lot of talk about this for other >> purposes for a while. > > Hm, I'm not convinced we want per-device dma_ops. What we want is per- > device IOMMU ops, and any dma_ops are just a generic or platform- > specific (in some cases) wrapper around those. We shouldn't normally > need per-device DMA ops at all. Hello David, Can you recommend an approach for e.g. the qib driver (drivers/infiniband/hw/qib)? That driver uses the CPU (PIO) instead of DMA to transfer data to a PCIe device. Sorry but I don't see how per-device IOMMU ops would allow to avoid that e.g. a cache flush is triggered before PIO starts. Bart. ^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2016-12-09 19:46 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <07c07529-4636-fafb-2598-7358d8a1460d@sandisk.com>
[not found] ` <58a325aa-8df7-6a2e-fadb-5d9031d8a1a0@sandisk.com>
2016-12-09 18:22 ` [PATCH 3/5] Move dma_ops from archdata into struct device Christoph Hellwig
2016-12-09 19:13 ` David Woodhouse
[not found] ` <1481310814.98151.11.camel@infradead.org>
2016-12-09 19:46 ` Bart Van Assche
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).