From: Keith Busch <[email protected]>
To: Christoph Hellwig <[email protected]>
Cc: Keith Busch <[email protected]>,
[email protected], [email protected],
[email protected], [email protected],
[email protected], Alexander Viro <[email protected]>,
Kernel Team <[email protected]>
Subject: Re: [PATCHv3 0/7] dma mapping optimisations
Date: Tue, 9 Aug 2022 10:46:04 -0600 [thread overview]
Message-ID: <YvKPTGf56v/[email protected]> (raw)
In-Reply-To: <[email protected]>
On Tue, Aug 09, 2022 at 08:46:13AM +0200, Christoph Hellwig wrote:
> - the DMA streaming ops (dma_map_*) are only intended for short term
> mappings, not for long living ones that can lead to starvation.
> Especially with swiotlb, but also with some IOMMUs. I think this
> needs to be changed to an API that can allocate DMA mapped memory
> using dma_alloc_noncoherent/dma_alloc_noncontigious for the device
> and then give access to that to the user instead
I am a bit reluctant to require a new memory allocator to use the feature. I'm
also generally not concerned about the odd resource constrained IOMMU. User
space drivers pre-map all their memory resources up front, and this is
essentially the same concept.
For swiotlb, though, we can error out the mapping if the requested memory uses
swiotlb with the device: the driver's .dma_map() can return ENOTSUPP if
is_swiotlb_buffer() is true. Would that be more acceptable?
next prev parent reply other threads:[~2022-08-09 16:46 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-08-05 16:24 [PATCHv3 0/7] dma mapping optimisations Keith Busch
2022-08-05 16:24 ` [PATCHv3 1/7] blk-mq: add ops to dma map bvec Keith Busch
2022-08-05 16:24 ` [PATCHv3 2/7] file: " Keith Busch
2022-08-08 0:21 ` Dave Chinner
2022-08-08 1:13 ` Matthew Wilcox
2022-08-08 2:15 ` Dave Chinner
2022-08-08 2:49 ` Matthew Wilcox
2022-08-08 7:31 ` Dave Chinner
2022-08-08 15:28 ` Keith Busch
2022-08-08 10:14 ` Pavel Begunkov
2022-08-05 16:24 ` [PATCHv3 3/7] iov_iter: introduce type for preregistered dma tags Keith Busch
2022-08-05 16:24 ` [PATCHv3 4/7] block: add dma tag bio type Keith Busch
2022-08-05 16:24 ` [PATCHv3 5/7] io_uring: introduce file slot release helper Keith Busch
2022-08-05 16:24 ` [PATCHv3 6/7] io_uring: add support for dma pre-mapping Keith Busch
2022-08-05 16:24 ` [PATCHv3 7/7] nvme-pci: implement dma_map support Keith Busch
2022-08-09 6:46 ` [PATCHv3 0/7] dma mapping optimisations Christoph Hellwig
2022-08-09 14:18 ` Keith Busch
2022-08-09 18:39 ` Christoph Hellwig
2022-08-09 16:46 ` Keith Busch [this message]
2022-08-09 18:41 ` Christoph Hellwig
2022-08-10 18:05 ` Keith Busch
2022-08-11 7:22 ` Christoph Hellwig
2022-08-31 21:19 ` Keith Busch
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=YvKPTGf56v/[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
/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