| Message ID | cover.1536764440.git.robin.murphy@arm.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | iommu: Avoid DMA ops domain refcount contention | expand |
Hi Robin, On Wed, Sep 12, 2018 at 04:24:11PM +0100, Robin Murphy wrote: > John raised the issue[1] that we have some unnecessary refcount contention > in the DMA ops path which shows scalability problems now that we have more > real high-performance hardware using iommu-dma. The x86 IOMMU drivers are > sidestepping this by stashing domain references in archdata, but since > that's not very nice for architecture-agnostic code, I think it's time to > look at a generic API-level solution. > > These are a couple of quick patches based on the idea I had back when > first implementing iommu-dma, but didn't have any way to justify at the > time. However, the reports of 10-25% better networking performance on v1 > suggest that it's very worthwhile (and far more significant than I ever > would have guessed). > > As far as merging goes, I don't mind at all whether this goes via IOMMU, > or via dma-mapping provided Joerg's happy to ack it. I think it makes most sense for Joerg to take this series via his tree. Anyway, I've been running them on my TX2 box and things are happy enough, so: Tested-by: Will Deacon <will.deacon@arm.com> Will
On 14/09/2018 13:48, Will Deacon wrote: > Hi Robin, > Hi Robin, I just spoke with Dongdong and we will test this version also so that we may provide a "Tested-by" tag. Thanks, John > On Wed, Sep 12, 2018 at 04:24:11PM +0100, Robin Murphy wrote: >> John raised the issue[1] that we have some unnecessary refcount contention >> in the DMA ops path which shows scalability problems now that we have more >> real high-performance hardware using iommu-dma. The x86 IOMMU drivers are >> sidestepping this by stashing domain references in archdata, but since >> that's not very nice for architecture-agnostic code, I think it's time to >> look at a generic API-level solution. >> >> These are a couple of quick patches based on the idea I had back when >> first implementing iommu-dma, but didn't have any way to justify at the >> time. However, the reports of 10-25% better networking performance on v1 >> suggest that it's very worthwhile (and far more significant than I ever >> would have guessed). >> >> As far as merging goes, I don't mind at all whether this goes via IOMMU, >> or via dma-mapping provided Joerg's happy to ack it. > > I think it makes most sense for Joerg to take this series via his tree. > > Anyway, I've been running them on my TX2 box and things are happy enough, > so: > > Tested-by: Will Deacon <will.deacon@arm.com> > > Will > > . >
On Fri, Sep 14, 2018 at 01:48:59PM +0100, Will Deacon wrote: > > As far as merging goes, I don't mind at all whether this goes via IOMMU, > > or via dma-mapping provided Joerg's happy to ack it. > > I think it makes most sense for Joerg to take this series via his tree. FYI, I have WIP patches to move the arm dma-iommu wrappers to common code: http://git.infradead.org/users/hch/misc.git/shortlog/refs/heads/dma-maybe-coherent which will require some synchronization of the involved trees. In the end it is iommu code, so the actual iommu patches should probably go into the iommu tree after all, but it might have to pull in the dma-mapping branch for that. Or we just punt it until the next merge window. Not sure how fast Tom needs the common dma-iommu code for the x86 AMD iommu conversion.
>Not sure how fast Tom needs the common dma-iommu code for the x86 AMD iommu conversion. I am currently busy working on something else and won't be able to do/test the x86 AMD iommu conversion anytime soon. So I don't need the common dma-iommu code anytime soon. On 17 September 2018 at 14:33, Christoph Hellwig <hch@lst.de> wrote: > On Fri, Sep 14, 2018 at 01:48:59PM +0100, Will Deacon wrote: >> > As far as merging goes, I don't mind at all whether this goes via IOMMU, >> > or via dma-mapping provided Joerg's happy to ack it. >> >> I think it makes most sense for Joerg to take this series via his tree. > > FYI, I have WIP patches to move the arm dma-iommu wrappers to common > code: > > http://git.infradead.org/users/hch/misc.git/shortlog/refs/heads/dma-maybe-coherent > > which will require some synchronization of the involved trees. In the > end it is iommu code, so the actual iommu patches should probably > go into the iommu tree after all, but it might have to pull in the > dma-mapping branch for that. Or we just punt it until the next merge > window. Not sure how fast Tom needs the common dma-iommu code for > the x86 AMD iommu conversion.
On 17/09/2018 12:20, John Garry wrote: > On 14/09/2018 13:48, Will Deacon wrote: >> Hi Robin, >> > > Hi Robin, > > I just spoke with Dongdong and we will test this version also so that we > may provide a "Tested-by" tag. > I tested this, so for series: Tested-by: John Garry <john.garry@huawei.com> Thanks, John > Thanks, > John > >> On Wed, Sep 12, 2018 at 04:24:11PM +0100, Robin Murphy wrote: >>> John raised the issue[1] that we have some unnecessary refcount >>> contention >>> in the DMA ops path which shows scalability problems now that we have >>> more >>> real high-performance hardware using iommu-dma. The x86 IOMMU drivers >>> are >>> sidestepping this by stashing domain references in archdata, but since >>> that's not very nice for architecture-agnostic code, I think it's >>> time to >>> look at a generic API-level solution. >>> >>> These are a couple of quick patches based on the idea I had back when >>> first implementing iommu-dma, but didn't have any way to justify at the >>> time. However, the reports of 10-25% better networking performance on v1 >>> suggest that it's very worthwhile (and far more significant than I ever >>> would have guessed). >>> >>> As far as merging goes, I don't mind at all whether this goes via IOMMU, >>> or via dma-mapping provided Joerg's happy to ack it. >> >> I think it makes most sense for Joerg to take this series via his tree. >> >> Anyway, I've been running them on my TX2 box and things are happy enough, >> so: >> >> Tested-by: Will Deacon <will.deacon@arm.com> >> >> Will >> >> . >> > > > _______________________________________________ > iommu mailing list > iommu@lists.linux-foundation.org > https://lists.linuxfoundation.org/mailman/listinfo/iommu > > . >
On Wed, Sep 12, 2018 at 04:24:11PM +0100, Robin Murphy wrote: > Robin Murphy (3): > iommu: Add fast hook for getting DMA domains > iommu/dma: Use fast DMA domain lookup > arm64/dma-mapping: Mildly optimise non-coherent IOMMU ops Applied, thanks Robin.