[PATCH v2 0/6] MediaTek IOMMU improve tlb flush performance in map/unmap
Will Deacon
will at kernel.org
Wed Nov 25 07:27:19 EST 2020
On Thu, Nov 19, 2020 at 02:18:30PM +0800, Yong Wu wrote:
> This patchset is to improve tlb flushing performance in iommu_map/unmap
> for MediaTek IOMMU.
>
> For iommu_map, currently MediaTek IOMMU use IO_PGTABLE_QUIRK_TLBI_ON_MAP
> to do tlb_flush for each a memory chunk. this is so unnecessary. we could
> improve it by tlb flushing one time at the end of iommu_map.
>
> For iommu_unmap, currently we have already improve this performance by
> gather. But the current gather should take care its granule size. if the
> granule size is different, it will do tlb flush and gather again. Our HW
> don't care about granule size. thus I add a flag(granule_ignore) for this
> case.
>
> After this patchset, we could achieve only tlb flushing once in iommu_map
> and iommu_unmap.
>
> Regardless of sg, for each a segment, I did a simple test:
>
> size = 20 * SZ_1M;
> /* the worst case, all are 4k mapping. */
> ret = iommu_map(domain, 0x5bb02000, 0x123f1000, size, IOMMU_READ);
> iommu_unmap(domain, 0x5bb02000, size);
>
> This is the comparing time(unit is us):
> original-time after-improve
> map-20M 59943 2347
> unmap-20M 264 36
>
> This patchset also flush tlb once in the iommu_map_sg case.
>
> patch [1/6][2/6][3/6] are for map while the others are for unmap.
>
> change note:
> v2: Refactor all the code.
> base on v5.10-rc1.
Robin -- please can you take a look at this series? I'm not sure I'll
have time to get to it at the moment and you understand how this stuff
is supposed to work.
Cheers,
Will
More information about the Linux-mediatek
mailing list