From: Leon Romanovsky <leon@kernel•org>
To: Jason Gunthorpe <jgg@ziepe•ca>
Cc: "Matthew Wilcox" <willy@infradead•org>,
"David Hildenbrand" <david@redhat•com>,
"Robin Murphy" <robin.murphy@arm•com>,
"Marek Szyprowski" <m.szyprowski@samsung•com>,
"Christoph Hellwig" <hch@lst•de>,
"Jonathan Corbet" <corbet@lwn•net>,
"Madhavan Srinivasan" <maddy@linux•ibm.com>,
"Michael Ellerman" <mpe@ellerman•id.au>,
"Nicholas Piggin" <npiggin@gmail•com>,
"Christophe Leroy" <christophe.leroy@csgroup•eu>,
"Joerg Roedel" <joro@8bytes•org>, "Will Deacon" <will@kernel•org>,
"Michael S. Tsirkin" <mst@redhat•com>,
"Jason Wang" <jasowang@redhat•com>,
"Xuan Zhuo" <xuanzhuo@linux•alibaba.com>,
"Eugenio Pérez" <eperezma@redhat•com>,
"Alexander Potapenko" <glider@google•com>,
"Marco Elver" <elver@google•com>,
"Dmitry Vyukov" <dvyukov@google•com>,
"Masami Hiramatsu" <mhiramat@kernel•org>,
"Mathieu Desnoyers" <mathieu.desnoyers@efficios•com>,
"Jérôme Glisse" <jglisse@redhat•com>,
"Andrew Morton" <akpm@linux-foundation•org>,
linux-doc@vger•kernel.org, linux-kernel@vger•kernel.org,
linuxppc-dev@lists•ozlabs.org, iommu@lists•linux.dev,
virtualization@lists•linux.dev, kasan-dev@googlegroups•com,
linux-trace-kernel@vger•kernel.org, linux-mm@kvack•org
Subject: Re: [PATCH 0/8] dma-mapping: migrate to physical address-based API
Date: Thu, 31 Jul 2025 09:01:17 +0300 [thread overview]
Message-ID: <20250731060117.GR402218@unreal> (raw)
In-Reply-To: <20250730142818.GL26511@ziepe.ca>
On Wed, Jul 30, 2025 at 11:28:18AM -0300, Jason Gunthorpe wrote:
> On Wed, Jul 30, 2025 at 04:40:26PM +0300, Leon Romanovsky wrote:
<...>
> > The most reasonable way to prevent DMA_ATTR_SKIP_CPU_SYNC leakage is to
> > introduce new DMA attribute (let's call it DMA_ATTR_MMIO for now) and
> > pass it to both dma_map_phys() and dma_iova_link(). This flag will
> > indicate that p2p type is PCI_P2PDMA_MAP_THRU_HOST_BRIDGE and call to
> > right callbacks which will set IOMMU_MMIO flag and skip CPU sync,
>
> So the idea is if the memory is non-cachable, no-KVA you'd call
> dma_iova_link(phys_addr, DMA_ATTR_MMIO) and dma_map_phys(phys_addr,
> DMA_ATTR_MMIO) ?
Yes
>
> And then internally the dma_ops and dma_iommu would use the existing
> map_page/map_resource variations based on the flag, thus ensuring that
> MMIO is never kmap'd or cache flushed?
>
> dma_map_resource is really then just
> dma_map_phys(phys_addr, DMA_ATTR_MMIO)?
>
> I like this, I think it well addresses the concerns.
Yes, I had this idea and implementation before. :(
>
> Jason
>
next prev parent reply other threads:[~2025-07-31 6:01 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <CGME20250625131920eucas1p271b196cde042bd39ac08fb12beff5baf@eucas1p2.samsung.com>
2025-06-25 13:18 ` [PATCH 0/8] dma-mapping: migrate to physical address-based API Leon Romanovsky
2025-06-25 13:18 ` [PATCH 1/8] dma-debug: refactor to use physical addresses for page mapping Leon Romanovsky
2025-06-25 13:18 ` [PATCH 2/8] dma-mapping: rename trace_dma_*map_page to trace_dma_*map_phys Leon Romanovsky
2025-06-25 13:19 ` [PATCH 3/8] iommu/dma: rename iommu_dma_*map_page to iommu_dma_*map_phys Leon Romanovsky
2025-06-25 13:19 ` [PATCH 4/8] dma-mapping: convert dma_direct_*map_page to be phys_addr_t based Leon Romanovsky
2025-06-25 13:19 ` [PATCH 5/8] kmsan: convert kmsan_handle_dma to use physical addresses Leon Romanovsky
2025-06-26 17:43 ` Alexander Potapenko
2025-06-26 18:45 ` Leon Romanovsky
2025-06-27 16:28 ` Alexander Potapenko
2025-06-25 13:19 ` [PATCH 6/8] dma-mapping: fail early if physical address is mapped through platform callback Leon Romanovsky
2025-07-25 20:04 ` Robin Murphy
2025-07-27 6:30 ` Leon Romanovsky
2025-06-25 13:19 ` [PATCH 7/8] dma-mapping: export new dma_*map_phys() interface Leon Romanovsky
2025-06-25 13:19 ` [PATCH 8/8] mm/hmm: migrate to physical address-based DMA mapping API Leon Romanovsky
2025-07-15 13:24 ` Will Deacon
2025-07-15 13:58 ` Leon Romanovsky
2025-06-27 13:44 ` [PATCH 0/8] dma-mapping: migrate to physical address-based API Marek Szyprowski
2025-06-27 17:02 ` Leon Romanovsky
2025-06-30 13:38 ` Christoph Hellwig
2025-07-08 10:27 ` Marek Szyprowski
2025-07-08 11:00 ` Leon Romanovsky
2025-07-08 11:45 ` Marek Szyprowski
2025-07-08 12:06 ` Leon Romanovsky
2025-07-08 12:56 ` Marek Szyprowski
2025-07-08 15:57 ` Marek Szyprowski
2025-07-30 11:11 ` Robin Murphy
2025-07-30 13:40 ` Leon Romanovsky
2025-07-30 14:28 ` Jason Gunthorpe
2025-07-31 6:01 ` Leon Romanovsky [this message]
2025-07-30 16:32 ` Marek Szyprowski
2025-07-31 17:37 ` Matthew Wilcox
2025-08-03 15:59 ` Jason Gunthorpe
2025-08-04 3:37 ` Matthew Wilcox
2025-08-05 15:36 ` Jason Gunthorpe
2025-07-06 6:00 ` Leon Romanovsky
2025-07-25 20:05 ` Robin Murphy
2025-07-29 14:03 ` Jason Gunthorpe
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=20250731060117.GR402218@unreal \
--to=leon@kernel$(echo .)org \
--cc=akpm@linux-foundation$(echo .)org \
--cc=christophe.leroy@csgroup$(echo .)eu \
--cc=corbet@lwn$(echo .)net \
--cc=david@redhat$(echo .)com \
--cc=dvyukov@google$(echo .)com \
--cc=elver@google$(echo .)com \
--cc=eperezma@redhat$(echo .)com \
--cc=glider@google$(echo .)com \
--cc=hch@lst$(echo .)de \
--cc=iommu@lists$(echo .)linux.dev \
--cc=jasowang@redhat$(echo .)com \
--cc=jgg@ziepe$(echo .)ca \
--cc=jglisse@redhat$(echo .)com \
--cc=joro@8bytes$(echo .)org \
--cc=kasan-dev@googlegroups$(echo .)com \
--cc=linux-doc@vger$(echo .)kernel.org \
--cc=linux-kernel@vger$(echo .)kernel.org \
--cc=linux-mm@kvack$(echo .)org \
--cc=linux-trace-kernel@vger$(echo .)kernel.org \
--cc=linuxppc-dev@lists$(echo .)ozlabs.org \
--cc=m.szyprowski@samsung$(echo .)com \
--cc=maddy@linux$(echo .)ibm.com \
--cc=mathieu.desnoyers@efficios$(echo .)com \
--cc=mhiramat@kernel$(echo .)org \
--cc=mpe@ellerman$(echo .)id.au \
--cc=mst@redhat$(echo .)com \
--cc=npiggin@gmail$(echo .)com \
--cc=robin.murphy@arm$(echo .)com \
--cc=virtualization@lists$(echo .)linux.dev \
--cc=will@kernel$(echo .)org \
--cc=willy@infradead$(echo .)org \
--cc=xuanzhuo@linux$(echo .)alibaba.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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