From: Will Deacon <will@kernel.org>
To: Yong Wu <yong.wu@mediatek.com>
Cc: Joerg Roedel <joro@8bytes.org>,
Robin Murphy <robin.murphy@arm.com>,
Matthias Brugger <matthias.bgg@gmail.com>,
Krzysztof Kozlowski <krzk@kernel.org>,
Tomasz Figa <tfiga@google.com>,
linux-mediatek@lists.infradead.org, srv_heupstream@mediatek.com,
linux-kernel@vger.kernel.org,
linux-arm-kernel@lists.infradead.org,
iommu@lists.linux-foundation.org, youlin.pei@mediatek.com,
Nicolas Boichat <drinkcat@chromium.org>,
anan.sun@mediatek.com, chao.hao@mediatek.com,
Greg Kroah-Hartman <gregkh@google.com>,
kernel-team@android.com, Christoph Hellwig <hch@infradead.org>,
David Laight <David.Laight@ACULAB.COM>
Subject: Re: [PATCH v4 0/7] MediaTek IOMMU improve tlb flush performance in map/unmap
Date: Fri, 22 Jan 2021 19:32:27 +0000 [thread overview]
Message-ID: <20210122193226.GD25471@willie-the-truck> (raw)
In-Reply-To: <20210107122909.16317-1-yong.wu@mediatek.com>
On Thu, Jan 07, 2021 at 08:29:02PM +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 gather the range in our file.
>
> 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 [1/7][2/7][3/7] are for map while the others are for unmap.
>
> change note:
> v4: a. base on v5.11-rc1.
> b. Add a little helper _iommu_map.
> c. Fix a build fail for tegra-gart.c. I didn't notice there is another place
> call gart_iommu_sync_map.
> d. Switch gather->end to the read end address("start + end - 1").
>
> v3: https://lore.kernel.org/linux-iommu/20201216103607.23050-1-yong.wu@mediatek.com/#r
> Refactor the unmap flow suggested by Robin.
>
> v2: https://lore.kernel.org/linux-iommu/20201119061836.15238-1-yong.wu@mediatek.com/
> Refactor all the code.
> base on v5.10-rc1.
>
> Yong Wu (7):
> iommu: Move iotlb_sync_map out from __iommu_map
> iommu: Add iova and size as parameters in iotlb_sync_map
> iommu/mediatek: Add iotlb_sync_map to sync whole the iova range
> iommu: Switch gather->end to the inclusive end
> iommu/io-pgtable: Allow io_pgtable_tlb ops optional
> iommu/mediatek: Gather iova in iommu_unmap to achieve tlb sync once
> iommu/mediatek: Remove the tlb-ops for v7s
For the series:
Acked-by: Will Deacon <will@kernel.org>
Joerg -- how would you like to handle merging this? I suppose either you
could host a separate branch that I could merge if needed, or I could
include this in my pull to you, or something else.
Please let me know what you prefer,
Cheers,
Will
next prev parent reply other threads:[~2021-01-22 20:19 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-01-07 12:29 [PATCH v4 0/7] MediaTek IOMMU improve tlb flush performance in map/unmap Yong Wu
2021-01-07 12:29 ` [PATCH v4 1/7] iommu: Move iotlb_sync_map out from __iommu_map Yong Wu
2021-02-02 1:07 ` Doug Anderson
2021-01-07 12:29 ` [PATCH v4 2/7] iommu: Add iova and size as parameters in iotlb_sync_map Yong Wu
2021-01-07 12:29 ` [PATCH v4 3/7] iommu/mediatek: Add iotlb_sync_map to sync whole the iova range Yong Wu
2021-01-07 12:29 ` [PATCH v4 4/7] iommu: Switch gather->end to the inclusive end Yong Wu
2021-01-18 18:41 ` Robin Murphy
2021-01-07 12:29 ` [PATCH v4 5/7] iommu/io-pgtable: Allow io_pgtable_tlb ops optional Yong Wu
2021-01-18 18:42 ` Robin Murphy
2021-01-07 12:29 ` [PATCH v4 6/7] iommu/mediatek: Gather iova in iommu_unmap to achieve tlb sync once Yong Wu
2021-01-18 18:44 ` Robin Murphy
2021-01-07 12:29 ` [PATCH v4 7/7] iommu/mediatek: Remove the tlb-ops for v7s Yong Wu
2021-01-18 18:46 ` Robin Murphy
2021-01-22 19:32 ` Will Deacon [this message]
2021-01-27 13:19 ` [PATCH v4 0/7] MediaTek IOMMU improve tlb flush performance in map/unmap Will Deacon
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210122193226.GD25471@willie-the-truck \
--to=will@kernel.org \
--cc=David.Laight@ACULAB.COM \
--cc=anan.sun@mediatek.com \
--cc=chao.hao@mediatek.com \
--cc=drinkcat@chromium.org \
--cc=gregkh@google.com \
--cc=hch@infradead.org \
--cc=iommu@lists.linux-foundation.org \
--cc=joro@8bytes.org \
--cc=kernel-team@android.com \
--cc=krzk@kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mediatek@lists.infradead.org \
--cc=matthias.bgg@gmail.com \
--cc=robin.murphy@arm.com \
--cc=srv_heupstream@mediatek.com \
--cc=tfiga@google.com \
--cc=yong.wu@mediatek.com \
--cc=youlin.pei@mediatek.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).