* [PATCH] dma-direct: Restore arch_dma_alloc() for the DMA_ATTR_NO_KERNEL_MAPPING case
[not found] <CGME20260930135751eucas1p1c3b8b79ed87c3bf6f473d26e4663e7e4@eucas1p1.samsung.com>
@ 2026-09-30 13:57 ` Marek Szyprowski
2026-10-09 14:24 ` Marek Szyprowski
0 siblings, 1 reply; 2+ messages in thread
From: Marek Szyprowski @ 2026-09-30 13:57 UTC (permalink / raw)
To: iommu, linux-kernel, linux-arm-kernel
Cc: Marek Szyprowski, Robin Murphy, Christoph Hellwig,
Aneesh Kumar K . V, Russell King, Claude Opus 5 . 5
Commit 849facea92fa ("dma-direct: simplify the DMA_ATTR_NO_KERNEL_MAPPING
handling") moved the DMA_ATTR_NO_KERNEL_MAPPING shortcut in
dma_direct_alloc() in front of the arch_dma_alloc() call. Architectures
providing their own DMA coherent allocator (CONFIG_ARCH_HAS_DMA_ALLOC)
therefore no longer see such requests. ARM 32bit is one of such. Lack of
the call to arch_dma_alloc() results in the lack of proper cache
management on HighMem pages.
Take the dma-direct shortcut only for coherent devices or when the
architecture doesn't provide arch_dma_alloc(). While touching this,
move that check into separate function to avoid duplication in alloc
and free paths.
Fixes: 849facea92fa ("dma-direct: simplify the DMA_ATTR_NO_KERNEL_MAPPING handling")
Reported-by: Aneesh Kumar K.V <aneesh.kumar@kernel.org>
Closes: https://lore.kernel.org/all/yq5ams2akznl.fsf@kernel.org/
Assisted-by: Claude Opus 5.5 <noreply@anthropic.com>
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
---
kernel/dma/direct.c | 22 ++++++++++++++++++----
1 file changed, 18 insertions(+), 4 deletions(-)
diff --git a/kernel/dma/direct.c b/kernel/dma/direct.c
index da665ca22d5c..a31bd3129f3f 100644
--- a/kernel/dma/direct.c
+++ b/kernel/dma/direct.c
@@ -179,6 +179,22 @@ static struct page *dma_direct_alloc_from_pool(struct device *dev, size_t size,
return page;
}
+/*
+ * Check if the allocation can skip the kernel mapping and return the page
+ * pointer as the cookie.
+ */
+static bool dma_direct_use_no_mapping(struct device *dev, unsigned long attrs)
+{
+ if ((attrs & (DMA_ATTR_NO_KERNEL_MAPPING | __DMA_ATTR_ALLOC_CC_SHARED)) !=
+ DMA_ATTR_NO_KERNEL_MAPPING)
+ return false;
+ if (is_swiotlb_for_alloc(dev))
+ return false;
+ if (IS_ENABLED(CONFIG_ARCH_HAS_DMA_ALLOC) && !dev_is_dma_coherent(dev))
+ return false;
+ return true;
+}
+
static void *dma_direct_alloc_no_mapping(struct device *dev, size_t size,
dma_addr_t *dma_handle, gfp_t gfp)
{
@@ -226,8 +242,7 @@ void *dma_direct_alloc(struct device *dev, size_t size,
if (attrs & DMA_ATTR_NO_WARN)
gfp |= __GFP_NOWARN;
- if (((attrs & (DMA_ATTR_NO_KERNEL_MAPPING | __DMA_ATTR_ALLOC_CC_SHARED)) ==
- DMA_ATTR_NO_KERNEL_MAPPING) && !is_swiotlb_for_alloc(dev))
+ if (dma_direct_use_no_mapping(dev, attrs))
return dma_direct_alloc_no_mapping(dev, size, dma_handle, gfp);
if (!dev_is_dma_coherent(dev)) {
@@ -374,8 +389,7 @@ void dma_direct_free(struct device *dev, size_t size,
if (attrs & __DMA_ATTR_ALLOC_CC_SHARED)
mark_mem_encrypted = true;
- if (((attrs & (DMA_ATTR_NO_KERNEL_MAPPING | __DMA_ATTR_ALLOC_CC_SHARED)) ==
- DMA_ATTR_NO_KERNEL_MAPPING) && !is_swiotlb_for_alloc(dev)) {
+ if (dma_direct_use_no_mapping(dev, attrs)) {
/* cpu_addr is a struct page cookie, not a kernel address */
dma_free_contiguous(dev, cpu_addr, size);
return;
--
2.43.0
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] dma-direct: Restore arch_dma_alloc() for the DMA_ATTR_NO_KERNEL_MAPPING case
2026-09-30 13:57 ` [PATCH] dma-direct: Restore arch_dma_alloc() for the DMA_ATTR_NO_KERNEL_MAPPING case Marek Szyprowski
@ 2026-10-09 14:24 ` Marek Szyprowski
0 siblings, 0 replies; 2+ messages in thread
From: Marek Szyprowski @ 2026-10-09 14:24 UTC (permalink / raw)
To: iommu, linux-kernel, linux-arm-kernel
Cc: Robin Murphy, Christoph Hellwig, Aneesh Kumar K . V, Russell King,
Claude Opus 5 . 5
On 30.09.2026 15:57, Marek Szyprowski wrote:
> Commit 849facea92fa ("dma-direct: simplify the DMA_ATTR_NO_KERNEL_MAPPING
> handling") moved the DMA_ATTR_NO_KERNEL_MAPPING shortcut in
> dma_direct_alloc() in front of the arch_dma_alloc() call. Architectures
> providing their own DMA coherent allocator (CONFIG_ARCH_HAS_DMA_ALLOC)
> therefore no longer see such requests. ARM 32bit is one of such. Lack of
> the call to arch_dma_alloc() results in the lack of proper cache
> management on HighMem pages.
>
> Take the dma-direct shortcut only for coherent devices or when the
> architecture doesn't provide arch_dma_alloc(). While touching this,
> move that check into separate function to avoid duplication in alloc
> and free paths.
>
> Fixes: 849facea92fa ("dma-direct: simplify the DMA_ATTR_NO_KERNEL_MAPPING handling")
> Reported-by: Aneesh Kumar K.V <aneesh.kumar@kernel.org>
> Closes: https://lore.kernel.org/all/yq5ams2akznl.fsf@kernel.org/
> Assisted-by: Claude Opus 5.5 <noreply@anthropic.com>
> Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Applied to dma-mapping-for-next, because it looks that this case has no
active users, so it is not the urgent fix.
> ---
> kernel/dma/direct.c | 22 ++++++++++++++++++----
> 1 file changed, 18 insertions(+), 4 deletions(-)
>
> diff --git a/kernel/dma/direct.c b/kernel/dma/direct.c
> index da665ca22d5c..a31bd3129f3f 100644
> --- a/kernel/dma/direct.c
> +++ b/kernel/dma/direct.c
> @@ -179,6 +179,22 @@ static struct page *dma_direct_alloc_from_pool(struct device *dev, size_t size,
> return page;
> }
>
> +/*
> + * Check if the allocation can skip the kernel mapping and return the page
> + * pointer as the cookie.
> + */
> +static bool dma_direct_use_no_mapping(struct device *dev, unsigned long attrs)
> +{
> + if ((attrs & (DMA_ATTR_NO_KERNEL_MAPPING | __DMA_ATTR_ALLOC_CC_SHARED)) !=
> + DMA_ATTR_NO_KERNEL_MAPPING)
> + return false;
> + if (is_swiotlb_for_alloc(dev))
> + return false;
> + if (IS_ENABLED(CONFIG_ARCH_HAS_DMA_ALLOC) && !dev_is_dma_coherent(dev))
> + return false;
> + return true;
> +}
> +
> static void *dma_direct_alloc_no_mapping(struct device *dev, size_t size,
> dma_addr_t *dma_handle, gfp_t gfp)
> {
> @@ -226,8 +242,7 @@ void *dma_direct_alloc(struct device *dev, size_t size,
> if (attrs & DMA_ATTR_NO_WARN)
> gfp |= __GFP_NOWARN;
>
> - if (((attrs & (DMA_ATTR_NO_KERNEL_MAPPING | __DMA_ATTR_ALLOC_CC_SHARED)) ==
> - DMA_ATTR_NO_KERNEL_MAPPING) && !is_swiotlb_for_alloc(dev))
> + if (dma_direct_use_no_mapping(dev, attrs))
> return dma_direct_alloc_no_mapping(dev, size, dma_handle, gfp);
>
> if (!dev_is_dma_coherent(dev)) {
> @@ -374,8 +389,7 @@ void dma_direct_free(struct device *dev, size_t size,
> if (attrs & __DMA_ATTR_ALLOC_CC_SHARED)
> mark_mem_encrypted = true;
>
> - if (((attrs & (DMA_ATTR_NO_KERNEL_MAPPING | __DMA_ATTR_ALLOC_CC_SHARED)) ==
> - DMA_ATTR_NO_KERNEL_MAPPING) && !is_swiotlb_for_alloc(dev)) {
> + if (dma_direct_use_no_mapping(dev, attrs)) {
> /* cpu_addr is a struct page cookie, not a kernel address */
> dma_free_contiguous(dev, cpu_addr, size);
> return;
Best regards
--
Marek Szyprowski, PhD
Samsung R&D Institute Poland
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-10-09 14:25 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <CGME20260930135751eucas1p1c3b8b79ed87c3bf6f473d26e4663e7e4@eucas1p1.samsung.com>
2026-09-30 13:57 ` [PATCH] dma-direct: Restore arch_dma_alloc() for the DMA_ATTR_NO_KERNEL_MAPPING case Marek Szyprowski
2026-10-09 14:24 ` Marek Szyprowski
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox