linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Catalin Marinas <catalin.marinas@arm.com>
To: Linus Torvalds <torvalds@linux-foundation.org>,
	Arnd Bergmann <arnd@arndb.de>, Christoph Hellwig <hch@lst.de>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Will Deacon <will@kernel.org>, Marc Zyngier <maz@kernel.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Herbert Xu <herbert@gondor.apana.org.au>,
	Ard Biesheuvel <ardb@kernel.org>,
	Isaac Manjarres <isaacmanjarres@google.com>,
	Saravana Kannan <saravanak@google.com>,
	Alasdair Kergon <agk@redhat.com>, Daniel Vetter <daniel@ffwll.ch>,
	Joerg Roedel <joro@8bytes.org>, Mark Brown <broonie@kernel.org>,
	Mike Snitzer <snitzer@kernel.org>,
	"Rafael J. Wysocki" <rafael@kernel.org>,
	Robin Murphy <robin.murphy@arm.com>,
	linux-mm@kvack.org, iommu@lists.linux.dev,
	linux-arm-kernel@lists.infradead.org
Subject: [PATCH v3 02/13] dma-mapping: Force bouncing if the kmalloc() size is not cacheline-aligned
Date: Sun,  6 Nov 2022 22:01:32 +0000	[thread overview]
Message-ID: <20221106220143.2129263-3-catalin.marinas@arm.com> (raw)
In-Reply-To: <20221106220143.2129263-1-catalin.marinas@arm.com>

For direct DMA, if the size is small enough to have originated from a
kmalloc() cache below ARCH_DMA_MINALIGN, check its alignment against
cache_line_size() and bounce if necessary. For larger sizes, it is the
responsibility of the DMA API caller to ensure proper alignment.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Robin Murphy <robin.murphy@arm.com>
---
 include/linux/dma-map-ops.h | 27 +++++++++++++++++++++++++++
 kernel/dma/direct.h         |  3 ++-
 2 files changed, 29 insertions(+), 1 deletion(-)

diff --git a/include/linux/dma-map-ops.h b/include/linux/dma-map-ops.h
index d678afeb8a13..785f7aa90f57 100644
--- a/include/linux/dma-map-ops.h
+++ b/include/linux/dma-map-ops.h
@@ -8,6 +8,7 @@
 
 #include <linux/dma-mapping.h>
 #include <linux/pgtable.h>
+#include <linux/slab.h>
 
 struct cma;
 
@@ -275,6 +276,32 @@ static inline bool dev_is_dma_coherent(struct device *dev)
 }
 #endif /* CONFIG_ARCH_HAS_DMA_COHERENCE_H */
 
+/*
+ * Check whether the given size, assuming it is for a kmalloc()'ed object, is
+ * safe for non-coherent DMA or needs bouncing.
+ */
+static inline bool dma_kmalloc_needs_bounce(struct device *dev, size_t size,
+					    enum dma_data_direction dir)
+{
+	/*
+	 * No need for bouncing if coherent DMA or the direction is
+	 * DMA_TO_DEVICE.
+	 */
+	if (!IS_ENABLED(CONFIG_DMA_BOUNCE_UNALIGNED_KMALLOC) ||
+	    dir == DMA_TO_DEVICE || dev_is_dma_coherent(dev))
+		return false;
+
+	/*
+	 * Larger kmalloc() sizes are guaranteed to be aligned to
+	 * ARCH_DMA_MINALIGN.
+	 */
+	if (size >= 2 * ARCH_DMA_MINALIGN ||
+	    IS_ALIGNED(kmalloc_size_roundup(size), dma_get_cache_alignment()))
+		return false;
+
+	return true;
+}
+
 void *arch_dma_alloc(struct device *dev, size_t size, dma_addr_t *dma_handle,
 		gfp_t gfp, unsigned long attrs);
 void arch_dma_free(struct device *dev, size_t size, void *cpu_addr,
diff --git a/kernel/dma/direct.h b/kernel/dma/direct.h
index e38ffc5e6bdd..97ec892ea0b5 100644
--- a/kernel/dma/direct.h
+++ b/kernel/dma/direct.h
@@ -94,7 +94,8 @@ static inline dma_addr_t dma_direct_map_page(struct device *dev,
 		return swiotlb_map(dev, phys, size, dir, attrs);
 	}
 
-	if (unlikely(!dma_capable(dev, dma_addr, size, true))) {
+	if (unlikely(!dma_capable(dev, dma_addr, size, true)) ||
+	    dma_kmalloc_needs_bounce(dev, size, dir)) {
 		if (is_pci_p2pdma_page(page))
 			return DMA_MAPPING_ERROR;
 		if (is_swiotlb_active(dev))


  parent reply	other threads:[~2022-11-06 22:02 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-11-06 22:01 [PATCH v3 00/13] mm, dma, arm64: Reduce ARCH_KMALLOC_MINALIGN to 8 Catalin Marinas
2022-11-06 22:01 ` [PATCH v3 01/13] mm/slab: Decouple ARCH_KMALLOC_MINALIGN from ARCH_DMA_MINALIGN Catalin Marinas
2022-11-06 22:01 ` Catalin Marinas [this message]
2022-11-07  9:43   ` [PATCH v3 02/13] dma-mapping: Force bouncing if the kmalloc() size is not cacheline-aligned Christoph Hellwig
2022-11-06 22:01 ` [PATCH v3 03/13] iommu/dma: Force bouncing of the " Catalin Marinas
2022-11-07  9:46   ` Christoph Hellwig
2022-11-07 10:54     ` Catalin Marinas
2022-11-07 13:26       ` Robin Murphy
2022-11-08 10:51         ` Catalin Marinas
2022-11-08 11:40           ` Robin Murphy
2022-11-08  7:50       ` Christoph Hellwig
2022-11-14 23:23   ` Isaac Manjarres
2022-11-15 11:48     ` Catalin Marinas
2022-11-06 22:01 ` [PATCH v3 04/13] mm/slab: Allow kmalloc() minimum alignment fallback to dma_get_cache_alignment() Catalin Marinas
2022-11-07  0:50   ` kernel test robot
2022-11-07  9:22     ` Catalin Marinas
2022-11-07  1:51   ` kernel test robot
2022-11-06 22:01 ` [PATCH v3 05/13] mm/slab: Simplify create_kmalloc_cache() args and make it static Catalin Marinas
2022-11-06 22:01 ` [PATCH v3 06/13] dma: Allow the smaller cache_line_size() returned by dma_get_cache_alignment() Catalin Marinas
2022-11-06 22:01 ` [PATCH v3 07/13] drivers/base: Use ARCH_DMA_MINALIGN instead of ARCH_KMALLOC_MINALIGN Catalin Marinas
2022-11-06 22:01 ` [PATCH v3 08/13] drivers/gpu: " Catalin Marinas
2022-11-06 22:01 ` [PATCH v3 09/13] drivers/usb: " Catalin Marinas
2022-11-06 22:01 ` [PATCH v3 10/13] drivers/spi: " Catalin Marinas
2022-11-07 12:58   ` Mark Brown
2022-11-06 22:01 ` [PATCH v3 11/13] crypto: " Catalin Marinas
2022-11-07  2:22   ` Herbert Xu
2022-11-07  9:05     ` Catalin Marinas
2022-11-07  9:12       ` Herbert Xu
2022-11-07  9:38         ` Catalin Marinas
2022-11-06 22:01 ` [PATCH v3 12/13] drivers/md: " Catalin Marinas
2022-11-06 22:01 ` [PATCH v3 13/13] dma: arm64: Add CONFIG_DMA_BOUNCE_UNALIGNED_KMALLOC and enable it for arm64 Catalin Marinas
2022-11-07 13:03   ` Robin Murphy
2022-11-07 14:38     ` Christoph Hellwig
2022-11-07 15:24       ` Robin Murphy
2022-11-08  9:52     ` Catalin Marinas
2022-11-08 10:03       ` Christoph Hellwig
2022-11-30 18:48         ` Isaac Manjarres
2022-11-30 23:32           ` Alexander Graf
2023-04-20 11:51             ` Petr Tesařík
2023-03-16 18:38 ` [PATCH v3 00/13] mm, dma, arm64: Reduce ARCH_KMALLOC_MINALIGN to 8 Isaac Manjarres
2023-04-19 16:06   ` Catalin Marinas
2023-04-20  9:52     ` Petr Tesarik
2023-04-20 17:43       ` Catalin Marinas
2023-05-15 19:09     ` Isaac Manjarres
2023-05-16 17:19       ` Catalin Marinas
2023-05-16 18:19         ` Isaac Manjarres

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=20221106220143.2129263-3-catalin.marinas@arm.com \
    --to=catalin.marinas@arm.com \
    --cc=agk@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=ardb@kernel.org \
    --cc=arnd@arndb.de \
    --cc=broonie@kernel.org \
    --cc=daniel@ffwll.ch \
    --cc=gregkh@linuxfoundation.org \
    --cc=hch@lst.de \
    --cc=herbert@gondor.apana.org.au \
    --cc=iommu@lists.linux.dev \
    --cc=isaacmanjarres@google.com \
    --cc=joro@8bytes.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-mm@kvack.org \
    --cc=maz@kernel.org \
    --cc=rafael@kernel.org \
    --cc=robin.murphy@arm.com \
    --cc=saravanak@google.com \
    --cc=snitzer@kernel.org \
    --cc=torvalds@linux-foundation.org \
    --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).