From: Robin Murphy <robin.murphy@arm.com>
To: John Garry <john.garry@huawei.com>,
damien.lemoal@opensource.wdc.com, joro@8bytes.org,
will@kernel.org, jejb@linux.ibm.com, martin.petersen@oracle.com,
hch@lst.de, m.szyprowski@samsung.com
Cc: linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-ide@vger.kernel.org, iommu@lists.linux-foundation.org,
iommu@lists.linux.dev, linux-scsi@vger.kernel.org,
linuxarm@huawei.com
Subject: Re: [PATCH v4 1/5] dma-mapping: Add dma_opt_mapping_size()
Date: Tue, 28 Jun 2022 12:23:01 +0100 [thread overview]
Message-ID: <bbca5df5-8681-d6d9-201d-3d48b34e3001@arm.com> (raw)
In-Reply-To: <1656343521-62897-2-git-send-email-john.garry@huawei.com>
On 2022-06-27 16:25, John Garry wrote:
> Streaming DMA mapping involving an IOMMU may be much slower for larger
> total mapping size. This is because every IOMMU DMA mapping requires an
> IOVA to be allocated and freed. IOVA sizes above a certain limit are not
> cached, which can have a big impact on DMA mapping performance.
>
> Provide an API for device drivers to know this "optimal" limit, such that
> they may try to produce mapping which don't exceed it.
>
> Signed-off-by: John Garry <john.garry@huawei.com>
> Reviewed-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
> ---
> Documentation/core-api/dma-api.rst | 9 +++++++++
> include/linux/dma-map-ops.h | 1 +
> include/linux/dma-mapping.h | 5 +++++
> kernel/dma/mapping.c | 12 ++++++++++++
> 4 files changed, 27 insertions(+)
>
> diff --git a/Documentation/core-api/dma-api.rst b/Documentation/core-api/dma-api.rst
> index 6d6d0edd2d27..b3cd9763d28b 100644
> --- a/Documentation/core-api/dma-api.rst
> +++ b/Documentation/core-api/dma-api.rst
> @@ -204,6 +204,15 @@ Returns the maximum size of a mapping for the device. The size parameter
> of the mapping functions like dma_map_single(), dma_map_page() and
> others should not be larger than the returned value.
>
> +::
> +
> + size_t
> + dma_opt_mapping_size(struct device *dev);
> +
> +Returns the maximum optimal size of a mapping for the device. Mapping large
> +buffers may take longer so device drivers are advised to limit total DMA
> +streaming mappings length to the returned value.
Nit: I'm not sure "advised" is necessarily the right thing to say in
general - that's only really true for a caller who cares about
throughput of churning through short-lived mappings more than anything
else, and doesn't take a significant hit overall from splitting up
larger requests. I do think it's good to clarify the exact context of
"optimal" here, but I'd prefer to be objectively clear that it's for
workloads where the up-front mapping overhead dominates.
Thanks,
Robin.
> +
> ::
>
> bool
> diff --git a/include/linux/dma-map-ops.h b/include/linux/dma-map-ops.h
> index 0d5b06b3a4a6..98ceba6fa848 100644
> --- a/include/linux/dma-map-ops.h
> +++ b/include/linux/dma-map-ops.h
> @@ -69,6 +69,7 @@ struct dma_map_ops {
> int (*dma_supported)(struct device *dev, u64 mask);
> u64 (*get_required_mask)(struct device *dev);
> size_t (*max_mapping_size)(struct device *dev);
> + size_t (*opt_mapping_size)(void);
> unsigned long (*get_merge_boundary)(struct device *dev);
> };
>
> diff --git a/include/linux/dma-mapping.h b/include/linux/dma-mapping.h
> index dca2b1355bb1..fe3849434b2a 100644
> --- a/include/linux/dma-mapping.h
> +++ b/include/linux/dma-mapping.h
> @@ -144,6 +144,7 @@ int dma_set_mask(struct device *dev, u64 mask);
> int dma_set_coherent_mask(struct device *dev, u64 mask);
> u64 dma_get_required_mask(struct device *dev);
> size_t dma_max_mapping_size(struct device *dev);
> +size_t dma_opt_mapping_size(struct device *dev);
> bool dma_need_sync(struct device *dev, dma_addr_t dma_addr);
> unsigned long dma_get_merge_boundary(struct device *dev);
> struct sg_table *dma_alloc_noncontiguous(struct device *dev, size_t size,
> @@ -266,6 +267,10 @@ static inline size_t dma_max_mapping_size(struct device *dev)
> {
> return 0;
> }
> +static inline size_t dma_opt_mapping_size(struct device *dev)
> +{
> + return 0;
> +}
> static inline bool dma_need_sync(struct device *dev, dma_addr_t dma_addr)
> {
> return false;
> diff --git a/kernel/dma/mapping.c b/kernel/dma/mapping.c
> index db7244291b74..1bfe11b1edb6 100644
> --- a/kernel/dma/mapping.c
> +++ b/kernel/dma/mapping.c
> @@ -773,6 +773,18 @@ size_t dma_max_mapping_size(struct device *dev)
> }
> EXPORT_SYMBOL_GPL(dma_max_mapping_size);
>
> +size_t dma_opt_mapping_size(struct device *dev)
> +{
> + const struct dma_map_ops *ops = get_dma_ops(dev);
> + size_t size = SIZE_MAX;
> +
> + if (ops && ops->opt_mapping_size)
> + size = ops->opt_mapping_size();
> +
> + return min(dma_max_mapping_size(dev), size);
> +}
> +EXPORT_SYMBOL_GPL(dma_opt_mapping_size);
> +
> bool dma_need_sync(struct device *dev, dma_addr_t dma_addr)
> {
> const struct dma_map_ops *ops = get_dma_ops(dev);
next prev parent reply other threads:[~2022-06-28 11:23 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-06-27 15:25 [PATCH v4 0/5] DMA mapping changes for SCSI core John Garry
2022-06-27 15:25 ` [PATCH v4 1/5] dma-mapping: Add dma_opt_mapping_size() John Garry
2022-06-28 11:23 ` Robin Murphy [this message]
2022-06-28 11:27 ` John Garry
2022-06-29 11:57 ` John Garry
2022-06-27 15:25 ` [PATCH v4 2/5] dma-iommu: Add iommu_dma_opt_mapping_size() John Garry
2022-06-28 10:56 ` Robin Murphy
2022-06-27 15:25 ` [PATCH v4 3/5] scsi: core: Cap shost max_sectors according to DMA mapping limits only once John Garry
2022-06-27 15:25 ` [PATCH v4 4/5] scsi: scsi_transport_sas: Cap shost max_sectors according to DMA optimal mapping limit John Garry
2022-06-27 15:25 ` [PATCH v4 5/5] libata-scsi: Cap ata_device->max_sectors according to shost->max_sectors John Garry
2022-06-27 23:24 ` Damien Le Moal
2022-06-28 7:54 ` John Garry
2022-06-28 9:14 ` Damien Le Moal
2022-06-28 11:33 ` John Garry
2022-06-29 5:40 ` Christoph Hellwig
2022-06-29 5:58 ` Damien Le Moal
2022-06-29 7:43 ` John Garry
2022-06-29 8:24 ` Damien Le Moal
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=bbca5df5-8681-d6d9-201d-3d48b34e3001@arm.com \
--to=robin.murphy@arm.com \
--cc=damien.lemoal@opensource.wdc.com \
--cc=hch@lst.de \
--cc=iommu@lists.linux-foundation.org \
--cc=iommu@lists.linux.dev \
--cc=jejb@linux.ibm.com \
--cc=john.garry@huawei.com \
--cc=joro@8bytes.org \
--cc=linux-doc@vger.kernel.org \
--cc=linux-ide@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=linuxarm@huawei.com \
--cc=m.szyprowski@samsung.com \
--cc=martin.petersen@oracle.com \
--cc=will@kernel.org \
/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).