From: Christoph Hellwig <hch@lst.de>
To: John Garry <john.garry@huawei.com>
Cc: will@kernel.org, linux-kernel@vger.kernel.org,
iommu@lists.linux-foundation.org, robin.murphy@arm.com,
hch@lst.de, liyihang6@hisilicon.com
Subject: Re: [RFC PATCH] dma-iommu: Add iommu_dma_max_mapping_size()
Date: Tue, 17 May 2022 10:38:34 +0200 [thread overview]
Message-ID: <20220517083834.GA16965@lst.de> (raw)
In-Reply-To: <1652706361-92557-1-git-send-email-john.garry@huawei.com>
On Mon, May 16, 2022 at 09:06:01PM +0800, John Garry wrote:
> For streaming DMA mappings involving an IOMMU and whose IOVA len regularly
> exceeds the IOVA rcache upper limit (meaning that they are not cached),
> performance can be reduced.
>
> Add the IOMMU callback for DMA mapping API dma_max_mapping_size(), which
> allows the drivers to know the mapping limit and thus limit the requested
> IOVA lengths.
>
> This resolves the performance issue originally reported in [0] for a SCSI
> HBA driver which was regularly mapping SGLs which required IOVAs in
> excess of the IOVA caching limit. In this case the block layer limits the
> max sectors per request - as configured in __scsi_init_queue() - which
> will limit the total SGL length the driver tries to map and in turn limits
> IOVA lengths requested.
>
> [0] https://lore.kernel.org/linux-iommu/20210129092120.1482-1-thunder.leizhen@huawei.com/
>
> Signed-off-by: John Garry <john.garry@huawei.com>
> ---
> Sending as an RFC as iommu_dma_max_mapping_size() is a soft limit, and not
> a hard limit which I expect is the semantics of dma_map_ops.max_mapping_size
>
> diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c
> index 09f6e1c0f9c0..e2d5205cde37 100644
> --- a/drivers/iommu/dma-iommu.c
> +++ b/drivers/iommu/dma-iommu.c
> @@ -1442,6 +1442,21 @@ static unsigned long iommu_dma_get_merge_boundary(struct device *dev)
> return (1UL << __ffs(domain->pgsize_bitmap)) - 1;
> }
>
> + if (!domain)
> + return 0;
> +
> + cookie = domain->iova_cookie;
> + if (!cookie || cookie->type != IOMMU_DMA_IOVA_COOKIE)
> + return 0;
Can these conditions even be true here?
> +static inline unsigned long iova_rcache_range(void)
> +{
> + return 0;
> +}
Given that IOMMU_DMA select IOMMU_IOVA there is no need for this stub.
Otherwise this looks sensible to me.
_______________________________________________
iommu mailing list
iommu@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/iommu
WARNING: multiple messages have this Message-ID (diff)
From: Christoph Hellwig <hch@lst.de>
To: John Garry <john.garry@huawei.com>
Cc: joro@8bytes.org, will@kernel.org, hch@lst.de,
robin.murphy@arm.com, m.szyprowski@samsung.com,
chenxiang66@hisilicon.com, thunder.leizhen@huawei.com,
iommu@lists.linux-foundation.org, linux-kernel@vger.kernel.org,
liyihang6@hisilicon.com
Subject: Re: [RFC PATCH] dma-iommu: Add iommu_dma_max_mapping_size()
Date: Tue, 17 May 2022 10:38:34 +0200 [thread overview]
Message-ID: <20220517083834.GA16965@lst.de> (raw)
In-Reply-To: <1652706361-92557-1-git-send-email-john.garry@huawei.com>
On Mon, May 16, 2022 at 09:06:01PM +0800, John Garry wrote:
> For streaming DMA mappings involving an IOMMU and whose IOVA len regularly
> exceeds the IOVA rcache upper limit (meaning that they are not cached),
> performance can be reduced.
>
> Add the IOMMU callback for DMA mapping API dma_max_mapping_size(), which
> allows the drivers to know the mapping limit and thus limit the requested
> IOVA lengths.
>
> This resolves the performance issue originally reported in [0] for a SCSI
> HBA driver which was regularly mapping SGLs which required IOVAs in
> excess of the IOVA caching limit. In this case the block layer limits the
> max sectors per request - as configured in __scsi_init_queue() - which
> will limit the total SGL length the driver tries to map and in turn limits
> IOVA lengths requested.
>
> [0] https://lore.kernel.org/linux-iommu/20210129092120.1482-1-thunder.leizhen@huawei.com/
>
> Signed-off-by: John Garry <john.garry@huawei.com>
> ---
> Sending as an RFC as iommu_dma_max_mapping_size() is a soft limit, and not
> a hard limit which I expect is the semantics of dma_map_ops.max_mapping_size
>
> diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c
> index 09f6e1c0f9c0..e2d5205cde37 100644
> --- a/drivers/iommu/dma-iommu.c
> +++ b/drivers/iommu/dma-iommu.c
> @@ -1442,6 +1442,21 @@ static unsigned long iommu_dma_get_merge_boundary(struct device *dev)
> return (1UL << __ffs(domain->pgsize_bitmap)) - 1;
> }
>
> + if (!domain)
> + return 0;
> +
> + cookie = domain->iova_cookie;
> + if (!cookie || cookie->type != IOMMU_DMA_IOVA_COOKIE)
> + return 0;
Can these conditions even be true here?
> +static inline unsigned long iova_rcache_range(void)
> +{
> + return 0;
> +}
Given that IOMMU_DMA select IOMMU_IOVA there is no need for this stub.
Otherwise this looks sensible to me.
next prev parent reply other threads:[~2022-05-17 8:38 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-05-16 13:06 [RFC PATCH] dma-iommu: Add iommu_dma_max_mapping_size() John Garry via iommu
2022-05-16 13:06 ` John Garry
2022-05-17 8:38 ` Christoph Hellwig [this message]
2022-05-17 8:38 ` Christoph Hellwig
2022-05-17 9:02 ` John Garry via iommu
2022-05-17 9:02 ` John Garry
2022-05-17 9:11 ` Christoph Hellwig
2022-05-17 9:11 ` Christoph Hellwig
2022-05-17 10:40 ` Robin Murphy
2022-05-17 10:40 ` Robin Murphy
2022-05-17 11:26 ` John Garry via iommu
2022-05-17 11:26 ` John Garry
2022-05-17 12:02 ` Robin Murphy
2022-05-17 12:02 ` Robin Murphy
2022-05-17 13:50 ` John Garry via iommu
2022-05-17 13:50 ` John Garry
2022-05-18 13:13 ` Christoph Hellwig
2022-05-18 13:13 ` Christoph Hellwig
2022-05-20 8:51 ` Joerg Roedel
2022-05-20 8:51 ` Joerg Roedel
2022-05-18 13:12 ` Christoph Hellwig
2022-05-18 13:12 ` Christoph Hellwig
2022-05-18 13:45 ` Robin Murphy
2022-05-18 13:45 ` Robin Murphy
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=20220517083834.GA16965@lst.de \
--to=hch@lst.de \
--cc=iommu@lists.linux-foundation.org \
--cc=john.garry@huawei.com \
--cc=linux-kernel@vger.kernel.org \
--cc=liyihang6@hisilicon.com \
--cc=robin.murphy@arm.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.