* clean up the bypass code in intel-iommu a little @ 2019-04-10 16:14 ` Christoph Hellwig 0 siblings, 0 replies; 10+ messages in thread From: Christoph Hellwig @ 2019-04-10 16:14 UTC (permalink / raw) To: David Woodhouse, Joerg Roedel Cc: iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA Hi Dave and Joerg, below are three relatively simple patches to clean up the bypass case and make it share more code with the dma-direct implementation. ^ permalink raw reply [flat|nested] 10+ messages in thread
* clean up the bypass code in intel-iommu a little @ 2019-04-10 16:14 ` Christoph Hellwig 0 siblings, 0 replies; 10+ messages in thread From: Christoph Hellwig @ 2019-04-10 16:14 UTC (permalink / raw) To: David Woodhouse, Joerg Roedel; +Cc: iommu Hi Dave and Joerg, below are three relatively simple patches to clean up the bypass case and make it share more code with the dma-direct implementation. _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu ^ permalink raw reply [flat|nested] 10+ messages in thread
[parent not found: <20190410161408.7119-1-hch-jcswGhMUV9g@public.gmane.org>]
* [PATCH 1/3] intel-iommu: clean up iommu_no_mapping @ 2019-04-10 16:14 ` Christoph Hellwig 0 siblings, 0 replies; 10+ messages in thread From: Christoph Hellwig @ 2019-04-10 16:14 UTC (permalink / raw) To: David Woodhouse, Joerg Roedel Cc: iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA Invert the return value to avoid double negatives, use a bool instead of int as the return value, and reduce some indentation after early returns. Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org> --- drivers/iommu/intel-iommu.c | 47 +++++++++++++++++-------------------- 1 file changed, 21 insertions(+), 26 deletions(-) diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c index 28cb713d728c..121a0cb450df 100644 --- a/drivers/iommu/intel-iommu.c +++ b/drivers/iommu/intel-iommu.c @@ -3606,45 +3606,40 @@ struct dmar_domain *get_valid_domain_for_dev(struct device *dev) } /* Check if the dev needs to go through non-identity map and unmap process.*/ -static int iommu_no_mapping(struct device *dev) +static bool iommu_need_mapping(struct device *dev) { int found; if (iommu_dummy(dev)) - return 1; + return false; if (!iommu_identity_mapping) - return 0; + return true; found = identity_mapping(dev); if (found) { if (iommu_should_identity_map(dev, 0)) - return 1; - else { - /* - * 32 bit DMA is removed from si_domain and fall back - * to non-identity mapping. - */ - dmar_remove_one_dev_info(dev); - dev_info(dev, "32bit DMA uses non-identity mapping\n"); - return 0; - } + return false; + + /* + * 32 bit DMA is removed from si_domain and fall back to + * non-identity mapping. + */ + dmar_remove_one_dev_info(dev); + dev_info(dev, "32bit DMA uses non-identity mapping\n"); } else { /* * In case of a detached 64 bit DMA device from vm, the device * is put into si_domain for identity mapping. */ - if (iommu_should_identity_map(dev, 0)) { - int ret; - ret = domain_add_dev_info(si_domain, dev); - if (!ret) { - dev_info(dev, "64bit DMA uses identity mapping\n"); - return 1; - } + if (iommu_should_identity_map(dev, 0) && + !domain_add_dev_info(si_domain, dev)) { + dev_info(dev, "64bit DMA uses identity mapping\n"); + return false; } } - return 0; + return true; } static dma_addr_t __intel_map_single(struct device *dev, phys_addr_t paddr, @@ -3660,7 +3655,7 @@ static dma_addr_t __intel_map_single(struct device *dev, phys_addr_t paddr, BUG_ON(dir == DMA_NONE); - if (iommu_no_mapping(dev)) + if (!iommu_need_mapping(dev)) return paddr; domain = get_valid_domain_for_dev(dev); @@ -3731,7 +3726,7 @@ static void intel_unmap(struct device *dev, dma_addr_t dev_addr, size_t size) struct intel_iommu *iommu; struct page *freelist; - if (iommu_no_mapping(dev)) + if (!iommu_need_mapping(dev)) return; domain = find_domain(dev); @@ -3782,7 +3777,7 @@ static void *intel_alloc_coherent(struct device *dev, size_t size, size = PAGE_ALIGN(size); order = get_order(size); - if (!iommu_no_mapping(dev)) + if (iommu_need_mapping(dev)) flags &= ~(GFP_DMA | GFP_DMA32); else if (dev->coherent_dma_mask < dma_get_required_mask(dev)) { if (dev->coherent_dma_mask < DMA_BIT_MASK(32)) @@ -3796,7 +3791,7 @@ static void *intel_alloc_coherent(struct device *dev, size_t size, page = dma_alloc_from_contiguous(dev, count, order, flags & __GFP_NOWARN); - if (page && iommu_no_mapping(dev) && + if (page && !iommu_need_mapping(dev) && page_to_phys(page) + size > dev->coherent_dma_mask) { dma_release_from_contiguous(dev, page, count); page = NULL; @@ -3878,7 +3873,7 @@ static int intel_map_sg(struct device *dev, struct scatterlist *sglist, int nele struct intel_iommu *iommu; BUG_ON(dir == DMA_NONE); - if (iommu_no_mapping(dev)) + if (!iommu_need_mapping(dev)) return intel_nontranslate_map_sg(dev, sglist, nelems, dir); domain = get_valid_domain_for_dev(dev); -- 2.20.1 ^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH 1/3] intel-iommu: clean up iommu_no_mapping @ 2019-04-10 16:14 ` Christoph Hellwig 0 siblings, 0 replies; 10+ messages in thread From: Christoph Hellwig @ 2019-04-10 16:14 UTC (permalink / raw) To: David Woodhouse, Joerg Roedel; +Cc: iommu Invert the return value to avoid double negatives, use a bool instead of int as the return value, and reduce some indentation after early returns. Signed-off-by: Christoph Hellwig <hch@lst.de> --- drivers/iommu/intel-iommu.c | 47 +++++++++++++++++-------------------- 1 file changed, 21 insertions(+), 26 deletions(-) diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c index 28cb713d728c..121a0cb450df 100644 --- a/drivers/iommu/intel-iommu.c +++ b/drivers/iommu/intel-iommu.c @@ -3606,45 +3606,40 @@ struct dmar_domain *get_valid_domain_for_dev(struct device *dev) } /* Check if the dev needs to go through non-identity map and unmap process.*/ -static int iommu_no_mapping(struct device *dev) +static bool iommu_need_mapping(struct device *dev) { int found; if (iommu_dummy(dev)) - return 1; + return false; if (!iommu_identity_mapping) - return 0; + return true; found = identity_mapping(dev); if (found) { if (iommu_should_identity_map(dev, 0)) - return 1; - else { - /* - * 32 bit DMA is removed from si_domain and fall back - * to non-identity mapping. - */ - dmar_remove_one_dev_info(dev); - dev_info(dev, "32bit DMA uses non-identity mapping\n"); - return 0; - } + return false; + + /* + * 32 bit DMA is removed from si_domain and fall back to + * non-identity mapping. + */ + dmar_remove_one_dev_info(dev); + dev_info(dev, "32bit DMA uses non-identity mapping\n"); } else { /* * In case of a detached 64 bit DMA device from vm, the device * is put into si_domain for identity mapping. */ - if (iommu_should_identity_map(dev, 0)) { - int ret; - ret = domain_add_dev_info(si_domain, dev); - if (!ret) { - dev_info(dev, "64bit DMA uses identity mapping\n"); - return 1; - } + if (iommu_should_identity_map(dev, 0) && + !domain_add_dev_info(si_domain, dev)) { + dev_info(dev, "64bit DMA uses identity mapping\n"); + return false; } } - return 0; + return true; } static dma_addr_t __intel_map_single(struct device *dev, phys_addr_t paddr, @@ -3660,7 +3655,7 @@ static dma_addr_t __intel_map_single(struct device *dev, phys_addr_t paddr, BUG_ON(dir == DMA_NONE); - if (iommu_no_mapping(dev)) + if (!iommu_need_mapping(dev)) return paddr; domain = get_valid_domain_for_dev(dev); @@ -3731,7 +3726,7 @@ static void intel_unmap(struct device *dev, dma_addr_t dev_addr, size_t size) struct intel_iommu *iommu; struct page *freelist; - if (iommu_no_mapping(dev)) + if (!iommu_need_mapping(dev)) return; domain = find_domain(dev); @@ -3782,7 +3777,7 @@ static void *intel_alloc_coherent(struct device *dev, size_t size, size = PAGE_ALIGN(size); order = get_order(size); - if (!iommu_no_mapping(dev)) + if (iommu_need_mapping(dev)) flags &= ~(GFP_DMA | GFP_DMA32); else if (dev->coherent_dma_mask < dma_get_required_mask(dev)) { if (dev->coherent_dma_mask < DMA_BIT_MASK(32)) @@ -3796,7 +3791,7 @@ static void *intel_alloc_coherent(struct device *dev, size_t size, page = dma_alloc_from_contiguous(dev, count, order, flags & __GFP_NOWARN); - if (page && iommu_no_mapping(dev) && + if (page && !iommu_need_mapping(dev) && page_to_phys(page) + size > dev->coherent_dma_mask) { dma_release_from_contiguous(dev, page, count); page = NULL; @@ -3878,7 +3873,7 @@ static int intel_map_sg(struct device *dev, struct scatterlist *sglist, int nele struct intel_iommu *iommu; BUG_ON(dir == DMA_NONE); - if (iommu_no_mapping(dev)) + if (!iommu_need_mapping(dev)) return intel_nontranslate_map_sg(dev, sglist, nelems, dir); domain = get_valid_domain_for_dev(dev); -- 2.20.1 _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu ^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH 2/3] intel-iommu: use dma_direct for bypass devices @ 2019-04-10 16:14 ` Christoph Hellwig 0 siblings, 0 replies; 10+ messages in thread From: Christoph Hellwig @ 2019-04-10 16:14 UTC (permalink / raw) To: David Woodhouse, Joerg Roedel Cc: iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA The intel-iommu driver currently has a partial reimplementation of the direct mapping code for devices that use pass through mode. Replace that code with calls to the relevant dma_direct routines at the highest level. This means we have exactly the same behvior as the dma direct code itself, and can prepare for eventually only attaching the intel_iommu ops to devices that actually need dynamic iommu mappings. Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org> --- drivers/iommu/intel-iommu.c | 71 ++++++++++++++++--------------------- 1 file changed, 31 insertions(+), 40 deletions(-) diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c index 121a0cb450df..497217cb0880 100644 --- a/drivers/iommu/intel-iommu.c +++ b/drivers/iommu/intel-iommu.c @@ -3655,9 +3655,6 @@ static dma_addr_t __intel_map_single(struct device *dev, phys_addr_t paddr, BUG_ON(dir == DMA_NONE); - if (!iommu_need_mapping(dev)) - return paddr; - domain = get_valid_domain_for_dev(dev); if (!domain) return DMA_MAPPING_ERROR; @@ -3706,15 +3703,20 @@ static dma_addr_t intel_map_page(struct device *dev, struct page *page, enum dma_data_direction dir, unsigned long attrs) { - return __intel_map_single(dev, page_to_phys(page) + offset, size, - dir, *dev->dma_mask); + if (iommu_need_mapping(dev)) + return __intel_map_single(dev, page_to_phys(page) + offset, + size, dir, *dev->dma_mask); + return dma_direct_map_page(dev, page, offset, size, dir, attrs); } static dma_addr_t intel_map_resource(struct device *dev, phys_addr_t phys_addr, size_t size, enum dma_data_direction dir, unsigned long attrs) { - return __intel_map_single(dev, phys_addr, size, dir, *dev->dma_mask); + if (iommu_need_mapping(dev)) + return __intel_map_single(dev, phys_addr, size, dir, + *dev->dma_mask); + return dma_direct_map_resource(dev, phys_addr, size, dir, attrs); } static void intel_unmap(struct device *dev, dma_addr_t dev_addr, size_t size) @@ -3726,9 +3728,6 @@ static void intel_unmap(struct device *dev, dma_addr_t dev_addr, size_t size) struct intel_iommu *iommu; struct page *freelist; - if (!iommu_need_mapping(dev)) - return; - domain = find_domain(dev); BUG_ON(!domain); @@ -3764,7 +3763,17 @@ static void intel_unmap_page(struct device *dev, dma_addr_t dev_addr, size_t size, enum dma_data_direction dir, unsigned long attrs) { - intel_unmap(dev, dev_addr, size); + if (iommu_need_mapping(dev)) + intel_unmap(dev, dev_addr, size); + else + dma_direct_unmap_page(dev, dev_addr, size, dir, attrs); +} + +static void intel_unmap_resource(struct device *dev, dma_addr_t dev_addr, + size_t size, enum dma_data_direction dir, unsigned long attrs) +{ + if (iommu_need_mapping(dev)) + intel_unmap(dev, dev_addr, size); } static void *intel_alloc_coherent(struct device *dev, size_t size, @@ -3774,28 +3783,18 @@ static void *intel_alloc_coherent(struct device *dev, size_t size, struct page *page = NULL; int order; + if (!iommu_need_mapping(dev)) + return dma_direct_alloc(dev, size, dma_handle, flags, attrs); + size = PAGE_ALIGN(size); order = get_order(size); - - if (iommu_need_mapping(dev)) - flags &= ~(GFP_DMA | GFP_DMA32); - else if (dev->coherent_dma_mask < dma_get_required_mask(dev)) { - if (dev->coherent_dma_mask < DMA_BIT_MASK(32)) - flags |= GFP_DMA; - else - flags |= GFP_DMA32; - } + flags &= ~(GFP_DMA | GFP_DMA32); if (gfpflags_allow_blocking(flags)) { unsigned int count = size >> PAGE_SHIFT; page = dma_alloc_from_contiguous(dev, count, order, flags & __GFP_NOWARN); - if (page && !iommu_need_mapping(dev) && - page_to_phys(page) + size > dev->coherent_dma_mask) { - dma_release_from_contiguous(dev, page, count); - page = NULL; - } } if (!page) @@ -3821,6 +3820,9 @@ static void intel_free_coherent(struct device *dev, size_t size, void *vaddr, int order; struct page *page = virt_to_page(vaddr); + if (!iommu_need_mapping(dev)) + return dma_direct_free(dev, size, vaddr, dma_handle, attrs); + size = PAGE_ALIGN(size); order = get_order(size); @@ -3838,6 +3840,9 @@ static void intel_unmap_sg(struct device *dev, struct scatterlist *sglist, struct scatterlist *sg; int i; + if (!iommu_need_mapping(dev)) + return dma_direct_unmap_sg(dev, sglist, nelems, dir, attrs); + for_each_sg(sglist, sg, nelems, i) { nrpages += aligned_nrpages(sg_dma_address(sg), sg_dma_len(sg)); } @@ -3845,20 +3850,6 @@ static void intel_unmap_sg(struct device *dev, struct scatterlist *sglist, intel_unmap(dev, startaddr, nrpages << VTD_PAGE_SHIFT); } -static int intel_nontranslate_map_sg(struct device *hddev, - struct scatterlist *sglist, int nelems, int dir) -{ - int i; - struct scatterlist *sg; - - for_each_sg(sglist, sg, nelems, i) { - BUG_ON(!sg_page(sg)); - sg->dma_address = sg_phys(sg); - sg->dma_length = sg->length; - } - return nelems; -} - static int intel_map_sg(struct device *dev, struct scatterlist *sglist, int nelems, enum dma_data_direction dir, unsigned long attrs) { @@ -3874,7 +3865,7 @@ static int intel_map_sg(struct device *dev, struct scatterlist *sglist, int nele BUG_ON(dir == DMA_NONE); if (!iommu_need_mapping(dev)) - return intel_nontranslate_map_sg(dev, sglist, nelems, dir); + return dma_direct_map_sg(dev, sglist, nelems, dir, attrs); domain = get_valid_domain_for_dev(dev); if (!domain) @@ -3924,7 +3915,7 @@ static const struct dma_map_ops intel_dma_ops = { .map_page = intel_map_page, .unmap_page = intel_unmap_page, .map_resource = intel_map_resource, - .unmap_resource = intel_unmap_page, + .unmap_resource = intel_unmap_resource, .dma_supported = dma_direct_supported, }; -- 2.20.1 ^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH 2/3] intel-iommu: use dma_direct for bypass devices @ 2019-04-10 16:14 ` Christoph Hellwig 0 siblings, 0 replies; 10+ messages in thread From: Christoph Hellwig @ 2019-04-10 16:14 UTC (permalink / raw) To: David Woodhouse, Joerg Roedel; +Cc: iommu The intel-iommu driver currently has a partial reimplementation of the direct mapping code for devices that use pass through mode. Replace that code with calls to the relevant dma_direct routines at the highest level. This means we have exactly the same behvior as the dma direct code itself, and can prepare for eventually only attaching the intel_iommu ops to devices that actually need dynamic iommu mappings. Signed-off-by: Christoph Hellwig <hch@lst.de> --- drivers/iommu/intel-iommu.c | 71 ++++++++++++++++--------------------- 1 file changed, 31 insertions(+), 40 deletions(-) diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c index 121a0cb450df..497217cb0880 100644 --- a/drivers/iommu/intel-iommu.c +++ b/drivers/iommu/intel-iommu.c @@ -3655,9 +3655,6 @@ static dma_addr_t __intel_map_single(struct device *dev, phys_addr_t paddr, BUG_ON(dir == DMA_NONE); - if (!iommu_need_mapping(dev)) - return paddr; - domain = get_valid_domain_for_dev(dev); if (!domain) return DMA_MAPPING_ERROR; @@ -3706,15 +3703,20 @@ static dma_addr_t intel_map_page(struct device *dev, struct page *page, enum dma_data_direction dir, unsigned long attrs) { - return __intel_map_single(dev, page_to_phys(page) + offset, size, - dir, *dev->dma_mask); + if (iommu_need_mapping(dev)) + return __intel_map_single(dev, page_to_phys(page) + offset, + size, dir, *dev->dma_mask); + return dma_direct_map_page(dev, page, offset, size, dir, attrs); } static dma_addr_t intel_map_resource(struct device *dev, phys_addr_t phys_addr, size_t size, enum dma_data_direction dir, unsigned long attrs) { - return __intel_map_single(dev, phys_addr, size, dir, *dev->dma_mask); + if (iommu_need_mapping(dev)) + return __intel_map_single(dev, phys_addr, size, dir, + *dev->dma_mask); + return dma_direct_map_resource(dev, phys_addr, size, dir, attrs); } static void intel_unmap(struct device *dev, dma_addr_t dev_addr, size_t size) @@ -3726,9 +3728,6 @@ static void intel_unmap(struct device *dev, dma_addr_t dev_addr, size_t size) struct intel_iommu *iommu; struct page *freelist; - if (!iommu_need_mapping(dev)) - return; - domain = find_domain(dev); BUG_ON(!domain); @@ -3764,7 +3763,17 @@ static void intel_unmap_page(struct device *dev, dma_addr_t dev_addr, size_t size, enum dma_data_direction dir, unsigned long attrs) { - intel_unmap(dev, dev_addr, size); + if (iommu_need_mapping(dev)) + intel_unmap(dev, dev_addr, size); + else + dma_direct_unmap_page(dev, dev_addr, size, dir, attrs); +} + +static void intel_unmap_resource(struct device *dev, dma_addr_t dev_addr, + size_t size, enum dma_data_direction dir, unsigned long attrs) +{ + if (iommu_need_mapping(dev)) + intel_unmap(dev, dev_addr, size); } static void *intel_alloc_coherent(struct device *dev, size_t size, @@ -3774,28 +3783,18 @@ static void *intel_alloc_coherent(struct device *dev, size_t size, struct page *page = NULL; int order; + if (!iommu_need_mapping(dev)) + return dma_direct_alloc(dev, size, dma_handle, flags, attrs); + size = PAGE_ALIGN(size); order = get_order(size); - - if (iommu_need_mapping(dev)) - flags &= ~(GFP_DMA | GFP_DMA32); - else if (dev->coherent_dma_mask < dma_get_required_mask(dev)) { - if (dev->coherent_dma_mask < DMA_BIT_MASK(32)) - flags |= GFP_DMA; - else - flags |= GFP_DMA32; - } + flags &= ~(GFP_DMA | GFP_DMA32); if (gfpflags_allow_blocking(flags)) { unsigned int count = size >> PAGE_SHIFT; page = dma_alloc_from_contiguous(dev, count, order, flags & __GFP_NOWARN); - if (page && !iommu_need_mapping(dev) && - page_to_phys(page) + size > dev->coherent_dma_mask) { - dma_release_from_contiguous(dev, page, count); - page = NULL; - } } if (!page) @@ -3821,6 +3820,9 @@ static void intel_free_coherent(struct device *dev, size_t size, void *vaddr, int order; struct page *page = virt_to_page(vaddr); + if (!iommu_need_mapping(dev)) + return dma_direct_free(dev, size, vaddr, dma_handle, attrs); + size = PAGE_ALIGN(size); order = get_order(size); @@ -3838,6 +3840,9 @@ static void intel_unmap_sg(struct device *dev, struct scatterlist *sglist, struct scatterlist *sg; int i; + if (!iommu_need_mapping(dev)) + return dma_direct_unmap_sg(dev, sglist, nelems, dir, attrs); + for_each_sg(sglist, sg, nelems, i) { nrpages += aligned_nrpages(sg_dma_address(sg), sg_dma_len(sg)); } @@ -3845,20 +3850,6 @@ static void intel_unmap_sg(struct device *dev, struct scatterlist *sglist, intel_unmap(dev, startaddr, nrpages << VTD_PAGE_SHIFT); } -static int intel_nontranslate_map_sg(struct device *hddev, - struct scatterlist *sglist, int nelems, int dir) -{ - int i; - struct scatterlist *sg; - - for_each_sg(sglist, sg, nelems, i) { - BUG_ON(!sg_page(sg)); - sg->dma_address = sg_phys(sg); - sg->dma_length = sg->length; - } - return nelems; -} - static int intel_map_sg(struct device *dev, struct scatterlist *sglist, int nelems, enum dma_data_direction dir, unsigned long attrs) { @@ -3874,7 +3865,7 @@ static int intel_map_sg(struct device *dev, struct scatterlist *sglist, int nele BUG_ON(dir == DMA_NONE); if (!iommu_need_mapping(dev)) - return intel_nontranslate_map_sg(dev, sglist, nelems, dir); + return dma_direct_map_sg(dev, sglist, nelems, dir, attrs); domain = get_valid_domain_for_dev(dev); if (!domain) @@ -3924,7 +3915,7 @@ static const struct dma_map_ops intel_dma_ops = { .map_page = intel_map_page, .unmap_page = intel_unmap_page, .map_resource = intel_map_resource, - .unmap_resource = intel_unmap_page, + .unmap_resource = intel_unmap_resource, .dma_supported = dma_direct_supported, }; -- 2.20.1 _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu ^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH 3/3] intel-iommu: don't clear GFP_DMA and GFP_DMA32 flags @ 2019-04-10 16:14 ` Christoph Hellwig 0 siblings, 0 replies; 10+ messages in thread From: Christoph Hellwig @ 2019-04-10 16:14 UTC (permalink / raw) To: David Woodhouse, Joerg Roedel Cc: iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA We already do this in the caller. Signed-off-by: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org> --- drivers/iommu/intel-iommu.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c index 497217cb0880..b7fd297a3c23 100644 --- a/drivers/iommu/intel-iommu.c +++ b/drivers/iommu/intel-iommu.c @@ -3788,7 +3788,6 @@ static void *intel_alloc_coherent(struct device *dev, size_t size, size = PAGE_ALIGN(size); order = get_order(size); - flags &= ~(GFP_DMA | GFP_DMA32); if (gfpflags_allow_blocking(flags)) { unsigned int count = size >> PAGE_SHIFT; -- 2.20.1 ^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH 3/3] intel-iommu: don't clear GFP_DMA and GFP_DMA32 flags @ 2019-04-10 16:14 ` Christoph Hellwig 0 siblings, 0 replies; 10+ messages in thread From: Christoph Hellwig @ 2019-04-10 16:14 UTC (permalink / raw) To: David Woodhouse, Joerg Roedel; +Cc: iommu We already do this in the caller. Signed-off-by: Christoph Hellwig <hch@lst.de> --- drivers/iommu/intel-iommu.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c index 497217cb0880..b7fd297a3c23 100644 --- a/drivers/iommu/intel-iommu.c +++ b/drivers/iommu/intel-iommu.c @@ -3788,7 +3788,6 @@ static void *intel_alloc_coherent(struct device *dev, size_t size, size = PAGE_ALIGN(size); order = get_order(size); - flags &= ~(GFP_DMA | GFP_DMA32); if (gfpflags_allow_blocking(flags)) { unsigned int count = size >> PAGE_SHIFT; -- 2.20.1 _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu ^ permalink raw reply related [flat|nested] 10+ messages in thread
* Re: clean up the bypass code in intel-iommu a little @ 2019-04-11 15:53 ` Joerg Roedel 0 siblings, 0 replies; 10+ messages in thread From: Joerg Roedel @ 2019-04-11 15:53 UTC (permalink / raw) To: Christoph Hellwig Cc: iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA, David Woodhouse, ashok.raj-ral2JQCrhuEAvxtiuMwx3w On Wed, Apr 10, 2019 at 06:14:05PM +0200, Christoph Hellwig wrote: > below are three relatively simple patches to clean up the > bypass case and make it share more code with the dma-direct > implementation. These look simple and straightforward to me, applied. Can you please also Cc LKML on iommu-patches in the future? It makes the patches easier to dig out of the email archive. Also please Cc the Intel people I added here for future intel-iommu patches (we should probably more formalize their role on the driver). Thanks, Joerg ^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: clean up the bypass code in intel-iommu a little @ 2019-04-11 15:53 ` Joerg Roedel 0 siblings, 0 replies; 10+ messages in thread From: Joerg Roedel @ 2019-04-11 15:53 UTC (permalink / raw) To: Christoph Hellwig; +Cc: iommu, David Woodhouse, ashok.raj On Wed, Apr 10, 2019 at 06:14:05PM +0200, Christoph Hellwig wrote: > below are three relatively simple patches to clean up the > bypass case and make it share more code with the dma-direct > implementation. These look simple and straightforward to me, applied. Can you please also Cc LKML on iommu-patches in the future? It makes the patches easier to dig out of the email archive. Also please Cc the Intel people I added here for future intel-iommu patches (we should probably more formalize their role on the driver). Thanks, Joerg _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu ^ permalink raw reply [flat|nested] 10+ messages in thread
end of thread, other threads:[~2019-04-11 15:54 UTC | newest]
Thread overview: 10+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2019-04-10 16:14 clean up the bypass code in intel-iommu a little Christoph Hellwig
2019-04-10 16:14 ` Christoph Hellwig
[not found] ` <20190410161408.7119-1-hch-jcswGhMUV9g@public.gmane.org>
2019-04-10 16:14 ` [PATCH 1/3] intel-iommu: clean up iommu_no_mapping Christoph Hellwig
2019-04-10 16:14 ` Christoph Hellwig
2019-04-10 16:14 ` [PATCH 2/3] intel-iommu: use dma_direct for bypass devices Christoph Hellwig
2019-04-10 16:14 ` Christoph Hellwig
2019-04-10 16:14 ` [PATCH 3/3] intel-iommu: don't clear GFP_DMA and GFP_DMA32 flags Christoph Hellwig
2019-04-10 16:14 ` Christoph Hellwig
2019-04-11 15:53 ` clean up the bypass code in intel-iommu a little Joerg Roedel
2019-04-11 15:53 ` Joerg Roedel
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.