| Message ID | 9bd208e9187db7a934428fe44ae0906028a8c9c3.1544469820.git.robin.murphy@arm.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | arm64: dma-mapping: Fix FORCE_CONTIGUOUS buffer clearing | expand |
On Mon, Dec 10, 2018 at 07:33:31PM +0000, Robin Murphy wrote: > We need to invalidate the caches *before* clearing the buffer via the > non-cacheable alias, else in the worst case __dma_flush_area() may > write back dirty lines over the top of our nice new zeros. > > Fixes: dd65a941f6ba ("arm64: dma-mapping: clear buffers allocated with FORCE_CONTIGUOUS flag") > Signed-off-by: Robin Murphy <robin.murphy@arm.com> > --- > arch/arm64/mm/dma-mapping.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c > index a3ac26284845..a53704406099 100644 > --- a/arch/arm64/mm/dma-mapping.c > +++ b/arch/arm64/mm/dma-mapping.c > @@ -429,9 +429,9 @@ static void *__iommu_alloc_attrs(struct device *dev, size_t size, > prot, > __builtin_return_address(0)); > if (addr) { > - memset(addr, 0, size); > if (!coherent) > __dma_flush_area(page_to_virt(page), iosize); > + memset(addr, 0, size); > } else { > iommu_dma_unmap_page(dev, *handle, iosize, 0, attrs); > dma_release_from_contiguous(dev, page, Oh crikey, well spotted! Acked-by: Will Deacon <will.deacon@arm.com> Catalin, please take this with a Cc stable. Will
On Mon, Dec 10, 2018 at 07:36:18PM +0000, Will Deacon wrote: > On Mon, Dec 10, 2018 at 07:33:31PM +0000, Robin Murphy wrote: > > We need to invalidate the caches *before* clearing the buffer via the > > non-cacheable alias, else in the worst case __dma_flush_area() may > > write back dirty lines over the top of our nice new zeros. > > > > Fixes: dd65a941f6ba ("arm64: dma-mapping: clear buffers allocated with FORCE_CONTIGUOUS flag") > > Signed-off-by: Robin Murphy <robin.murphy@arm.com> > > --- > > arch/arm64/mm/dma-mapping.c | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c > > index a3ac26284845..a53704406099 100644 > > --- a/arch/arm64/mm/dma-mapping.c > > +++ b/arch/arm64/mm/dma-mapping.c > > @@ -429,9 +429,9 @@ static void *__iommu_alloc_attrs(struct device *dev, size_t size, > > prot, > > __builtin_return_address(0)); > > if (addr) { > > - memset(addr, 0, size); > > if (!coherent) > > __dma_flush_area(page_to_virt(page), iosize); > > + memset(addr, 0, size); > > } else { > > iommu_dma_unmap_page(dev, *handle, iosize, 0, attrs); > > dma_release_from_contiguous(dev, page, > > Oh crikey, well spotted! > > Acked-by: Will Deacon <will.deacon@arm.com> > > Catalin, please take this with a Cc stable. Queued. Thanks.
On 12/11/2018 1:03 AM, Robin Murphy wrote: > We need to invalidate the caches *before* clearing the buffer via the > non-cacheable alias, else in the worst case __dma_flush_area() may > write back dirty lines over the top of our nice new zeros. > > Fixes: dd65a941f6ba ("arm64: dma-mapping: clear buffers allocated with FORCE_CONTIGUOUS flag") > Signed-off-by: Robin Murphy <robin.murphy@arm.com> > --- > arch/arm64/mm/dma-mapping.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c > index a3ac26284845..a53704406099 100644 > --- a/arch/arm64/mm/dma-mapping.c > +++ b/arch/arm64/mm/dma-mapping.c > @@ -429,9 +429,9 @@ static void *__iommu_alloc_attrs(struct device *dev, size_t size, > prot, > __builtin_return_address(0)); > if (addr) { > - memset(addr, 0, size); > if (!coherent) > __dma_flush_area(page_to_virt(page), iosize); > + memset(addr, 0, size); For coherent case, we still skip flushing/invalidating dirty kernel mappings. Isn't it a problem ? Shouldn't it be, --- a/arch/arm64/mm/dma-mapping.c +++ b/arch/arm64/mm/dma-mapping.c @@ -774,9 +774,8 @@ static void *__iommu_alloc_attrs(struct device *dev, size_t size, prot, __builtin_return_address(0)); if (addr) { + __dma_flush_area(page_to_virt(page), iosize); memset(addr, 0, size); - if (!coherent) - __dma_flush_area(page_to_virt(page), iosize); } else { iommu_dma_unmap_page(dev, *handle, iosize, 0, attrs); dma_release_from_contiguous(dev, page, > } else { > iommu_dma_unmap_page(dev, *handle, iosize, 0, attrs); > dma_release_from_contiguous(dev, page, > Chintan
On Wed, Dec 12, 2018 at 01:01:41PM +0530, Chintan Pandya wrote: >> if (addr) { >> - memset(addr, 0, size); >> if (!coherent) >> __dma_flush_area(page_to_virt(page), iosize); >> + memset(addr, 0, size); > > For coherent case, we still skip flushing/invalidating dirty kernel > mappings. Isn't it a problem ? For the dma coherent case there is no need to writeback and/or invalidate the kernel mapping.
diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c index a3ac26284845..a53704406099 100644 --- a/arch/arm64/mm/dma-mapping.c +++ b/arch/arm64/mm/dma-mapping.c @@ -429,9 +429,9 @@ static void *__iommu_alloc_attrs(struct device *dev, size_t size, prot, __builtin_return_address(0)); if (addr) { - memset(addr, 0, size); if (!coherent) __dma_flush_area(page_to_virt(page), iosize); + memset(addr, 0, size); } else { iommu_dma_unmap_page(dev, *handle, iosize, 0, attrs); dma_release_from_contiguous(dev, page,
We need to invalidate the caches *before* clearing the buffer via the non-cacheable alias, else in the worst case __dma_flush_area() may write back dirty lines over the top of our nice new zeros. Fixes: dd65a941f6ba ("arm64: dma-mapping: clear buffers allocated with FORCE_CONTIGUOUS flag") Signed-off-by: Robin Murphy <robin.murphy@arm.com> --- arch/arm64/mm/dma-mapping.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)