From: Dmitry Osipenko <digetx@gmail.com>
To: Joerg Roedel <joro@8bytes.org>,
Robin Murphy <robin.murphy@arm.com>,
Thierry Reding <thierry.reding@gmail.com>,
Jonathan Hunter <jonathanh@nvidia.com>,
Rob Herring <robh+dt@kernel.org>
Cc: iommu@lists.linux-foundation.org, devicetree@vger.kernel.org,
linux-tegra@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH v3 05/19] iommu/tegra: gart: Optimize mapping / unmapping performance
Date: Sat, 18 Aug 2018 18:54:16 +0300 [thread overview]
Message-ID: <20180818155430.5586-6-digetx@gmail.com> (raw)
In-Reply-To: <20180818155430.5586-1-digetx@gmail.com>
Currently GART writes one page entry at a time. More optimal would be to
aggregate the writes and flush BUS buffer in the end, this gives map/unmap
10-40% performance boost (depending on size of mapping) in comparison to
flushing after each page entry update.
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
---
drivers/iommu/tegra-gart.c | 12 ++++++++++--
1 file changed, 10 insertions(+), 2 deletions(-)
diff --git a/drivers/iommu/tegra-gart.c b/drivers/iommu/tegra-gart.c
index f6cf5cd5aaca..86a855c0d031 100644
--- a/drivers/iommu/tegra-gart.c
+++ b/drivers/iommu/tegra-gart.c
@@ -287,7 +287,6 @@ static int gart_iommu_map(struct iommu_domain *domain, unsigned long iova,
}
}
gart_set_pte(gart, iova, GART_PTE(pfn));
- FLUSH_GART_REGS(gart);
spin_unlock_irqrestore(&gart->pte_lock, flags);
return 0;
}
@@ -304,7 +303,6 @@ static size_t gart_iommu_unmap(struct iommu_domain *domain, unsigned long iova,
spin_lock_irqsave(&gart->pte_lock, flags);
gart_set_pte(gart, iova, 0);
- FLUSH_GART_REGS(gart);
spin_unlock_irqrestore(&gart->pte_lock, flags);
return bytes;
}
@@ -370,6 +368,14 @@ static int gart_iommu_of_xlate(struct device *dev,
return 0;
}
+static void gart_iommu_sync(struct iommu_domain *domain)
+{
+ struct gart_domain *gart_domain = to_gart_domain(domain);
+ struct gart_device *gart = gart_domain->gart;
+
+ FLUSH_GART_REGS(gart);
+}
+
static const struct iommu_ops gart_iommu_ops = {
.capable = gart_iommu_capable,
.domain_alloc = gart_iommu_domain_alloc,
@@ -384,6 +390,8 @@ static const struct iommu_ops gart_iommu_ops = {
.iova_to_phys = gart_iommu_iova_to_phys,
.pgsize_bitmap = GART_IOMMU_PGSIZES,
.of_xlate = gart_iommu_of_xlate,
+ .iotlb_sync_map = gart_iommu_sync,
+ .iotlb_sync = gart_iommu_sync,
};
static int tegra_gart_suspend(struct device *dev)
--
2.18.0
next prev parent reply other threads:[~2018-08-18 15:55 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-08-18 15:54 [PATCH v3 00/19] IOMMU: Tegra GART driver clean up and optimization Dmitry Osipenko
2018-08-18 15:54 ` [PATCH v3 01/19] iommu/tegra: gart: Remove pr_fmt and clean up includes Dmitry Osipenko
2018-08-18 15:54 ` [PATCH v3 02/19] iommu/tegra: gart: Clean up driver probe errors handling Dmitry Osipenko
2018-08-18 15:54 ` [PATCH v3 03/19] iommu/tegra: gart: Ignore devices without IOMMU phandle in DT Dmitry Osipenko
2018-08-18 15:54 ` [PATCH v3 04/19] iommu: Introduce iotlb_sync_map callback Dmitry Osipenko
2018-08-18 15:54 ` Dmitry Osipenko [this message]
2018-08-18 15:54 ` [PATCH v3 06/19] dt-bindings: memory: tegra: Squash tegra20-gart into tegra20-mc Dmitry Osipenko
2018-08-20 19:12 ` Rob Herring
2018-08-20 19:27 ` Dmitry Osipenko
2018-08-20 19:35 ` Dmitry Osipenko
2018-08-28 10:47 ` Thierry Reding
2018-08-28 13:09 ` Dmitry Osipenko
2018-08-18 15:54 ` [PATCH v3 07/19] ARM: dts: tegra20: Update Memory Controller node to the new binding Dmitry Osipenko
2018-08-18 15:54 ` [PATCH v3 08/19] memory: tegra: Don't invoke Tegra30+ specific memory timing setup on Tegra20 Dmitry Osipenko
2018-08-18 15:54 ` [PATCH v3 09/19] memory: tegra: Adapt to Tegra20 device-tree binding changes Dmitry Osipenko
2018-09-03 21:06 ` Marcel Ziswiler
2018-09-04 8:56 ` Dmitry Osipenko
2018-08-18 15:54 ` [PATCH v3 10/19] memory: tegra: Read client ID on GART page fault Dmitry Osipenko
2018-08-18 15:54 ` [PATCH v3 11/19] iommu/tegra: gart: Integrate with Memory Controller driver Dmitry Osipenko
2018-08-18 15:54 ` [PATCH v3 12/19] iommu/tegra: gart: Fix spinlock recursion Dmitry Osipenko
2018-08-18 15:54 ` [PATCH v3 13/19] iommu/tegra: gart: Fix NULL pointer dereference Dmitry Osipenko
2018-08-18 15:54 ` [PATCH v3 14/19] iommu/tegra: gart: Allow only one active domain at a time Dmitry Osipenko
2018-08-18 15:54 ` [PATCH v3 15/19] iommu/tegra: gart: Don't use managed resources Dmitry Osipenko
2018-08-18 15:54 ` [PATCH v3 16/19] iommu/tegra: gart: Prepend error/debug messages with "GART:" Dmitry Osipenko
2018-08-18 15:54 ` [PATCH v3 17/19] iommu/tegra: gart: Don't detach devices from inactive domains Dmitry Osipenko
2018-08-18 15:54 ` [PATCH v3 18/19] iommu/tegra: gart: Simplify clients-tracking code Dmitry Osipenko
2018-08-18 15:54 ` [PATCH v3 19/19] iommu/tegra: gart: Perform code refactoring Dmitry Osipenko
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=20180818155430.5586-6-digetx@gmail.com \
--to=digetx@gmail.com \
--cc=devicetree@vger.kernel.org \
--cc=iommu@lists.linux-foundation.org \
--cc=jonathanh@nvidia.com \
--cc=joro@8bytes.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-tegra@vger.kernel.org \
--cc=robh+dt@kernel.org \
--cc=robin.murphy@arm.com \
--cc=thierry.reding@gmail.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).