* [PATCH v1 0/2] iommu/vt-d: boost the mapping process @ 2021-09-15 15:21 Longpeng(Mike) 2021-09-15 15:21 ` [PATCH v1 1/2] iommu/vt-d: convert the return type of first_pte_in_page to bool Longpeng(Mike) 2021-09-15 15:21 ` [PATCH v1 2/2] iommu/vt-d: avoid duplicated removing in __domain_mapping Longpeng(Mike) 0 siblings, 2 replies; 4+ messages in thread From: Longpeng(Mike) @ 2021-09-15 15:21 UTC (permalink / raw) To: dwmw2, baolu.lu, joro, will Cc: Longpeng(Mike), iommu, arei.gonglei, linux-kernel Hi guys, We found that the __domain_mapping() would take too long when the memory region is too large, we try to make it faster in this patchset. The performance number can be found in PATCH 2, please review when you free, thanks. Longpeng(Mike) (2): iommu/vt-d: convert the return type of first_pte_in_page to bool iommu/vt-d: avoid duplicated removing in __domain_mapping drivers/iommu/intel/iommu.c | 12 +++++++----- include/linux/intel-iommu.h | 8 +++++++- 2 files changed, 14 insertions(+), 6 deletions(-) -- 1.8.3.1 _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu ^ permalink raw reply [flat|nested] 4+ messages in thread
* [PATCH v1 1/2] iommu/vt-d: convert the return type of first_pte_in_page to bool 2021-09-15 15:21 [PATCH v1 0/2] iommu/vt-d: boost the mapping process Longpeng(Mike) @ 2021-09-15 15:21 ` Longpeng(Mike) 2021-09-15 15:21 ` [PATCH v1 2/2] iommu/vt-d: avoid duplicated removing in __domain_mapping Longpeng(Mike) 1 sibling, 0 replies; 4+ messages in thread From: Longpeng(Mike) @ 2021-09-15 15:21 UTC (permalink / raw) To: dwmw2, baolu.lu, joro, will Cc: Longpeng(Mike), iommu, arei.gonglei, linux-kernel first_pte_in_page() returns boolean value, so let's convert its return type to bool. Signed-off-by: Longpeng(Mike) <longpeng2@huawei.com> --- include/linux/intel-iommu.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h index 05a65eb..a590b00 100644 --- a/include/linux/intel-iommu.h +++ b/include/linux/intel-iommu.h @@ -708,7 +708,7 @@ static inline bool dma_pte_superpage(struct dma_pte *pte) return (pte->val & DMA_PTE_LARGE_PAGE); } -static inline int first_pte_in_page(struct dma_pte *pte) +static inline bool first_pte_in_page(struct dma_pte *pte) { return !((unsigned long)pte & ~VTD_PAGE_MASK); } -- 1.8.3.1 _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu ^ permalink raw reply related [flat|nested] 4+ messages in thread
* [PATCH v1 2/2] iommu/vt-d: avoid duplicated removing in __domain_mapping 2021-09-15 15:21 [PATCH v1 0/2] iommu/vt-d: boost the mapping process Longpeng(Mike) 2021-09-15 15:21 ` [PATCH v1 1/2] iommu/vt-d: convert the return type of first_pte_in_page to bool Longpeng(Mike) @ 2021-09-15 15:21 ` Longpeng(Mike) 2021-09-30 14:14 ` Lu Baolu 1 sibling, 1 reply; 4+ messages in thread From: Longpeng(Mike) @ 2021-09-15 15:21 UTC (permalink / raw) To: dwmw2, baolu.lu, joro, will Cc: Longpeng(Mike), iommu, arei.gonglei, linux-kernel __domain_mapping() always removes the pages in the range from 'iov_pfn' to 'end_pfn', but the 'end_pfn' is always the last pfn of the range that the caller wants to map. This would introduce too many duplicated removing and leads the map operation take too long, for example: Map iova=0x100000,nr_pages=0x7d61800 iov_pfn: 0x100000, end_pfn: 0x7e617ff iov_pfn: 0x140000, end_pfn: 0x7e617ff iov_pfn: 0x180000, end_pfn: 0x7e617ff iov_pfn: 0x1c0000, end_pfn: 0x7e617ff iov_pfn: 0x200000, end_pfn: 0x7e617ff ... it takes about 50ms in total. We can reduce the cost by recalculate the 'end_pfn' and limit it to the boundary of the end of the pte page. Map iova=0x100000,nr_pages=0x7d61800 iov_pfn: 0x100000, end_pfn: 0x13ffff iov_pfn: 0x140000, end_pfn: 0x17ffff iov_pfn: 0x180000, end_pfn: 0x1bffff iov_pfn: 0x1c0000, end_pfn: 0x1fffff iov_pfn: 0x200000, end_pfn: 0x23ffff ... it only need 9ms now. Signed-off-by: Longpeng(Mike) <longpeng2@huawei.com> --- drivers/iommu/intel/iommu.c | 12 +++++++----- include/linux/intel-iommu.h | 6 ++++++ 2 files changed, 13 insertions(+), 5 deletions(-) diff --git a/drivers/iommu/intel/iommu.c b/drivers/iommu/intel/iommu.c index d75f59a..87cbf34 100644 --- a/drivers/iommu/intel/iommu.c +++ b/drivers/iommu/intel/iommu.c @@ -2354,12 +2354,18 @@ static void switch_to_super_page(struct dmar_domain *domain, return -ENOMEM; first_pte = pte; + lvl_pages = lvl_to_nr_pages(largepage_lvl); + BUG_ON(nr_pages < lvl_pages); + /* It is large page*/ if (largepage_lvl > 1) { unsigned long end_pfn; + unsigned long pages_to_remove; pteval |= DMA_PTE_LARGE_PAGE; - end_pfn = ((iov_pfn + nr_pages) & level_mask(largepage_lvl)) - 1; + pages_to_remove = min_t(unsigned long, nr_pages, + nr_pte_to_next_page(pte) * lvl_pages); + end_pfn = iov_pfn + pages_to_remove - 1; switch_to_super_page(domain, iov_pfn, end_pfn, largepage_lvl); } else { pteval &= ~(uint64_t)DMA_PTE_LARGE_PAGE; @@ -2381,10 +2387,6 @@ static void switch_to_super_page(struct dmar_domain *domain, WARN_ON(1); } - lvl_pages = lvl_to_nr_pages(largepage_lvl); - - BUG_ON(nr_pages < lvl_pages); - nr_pages -= lvl_pages; iov_pfn += lvl_pages; phys_pfn += lvl_pages; diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h index a590b00..4bff70c 100644 --- a/include/linux/intel-iommu.h +++ b/include/linux/intel-iommu.h @@ -713,6 +713,12 @@ static inline bool first_pte_in_page(struct dma_pte *pte) return !((unsigned long)pte & ~VTD_PAGE_MASK); } +static inline int nr_pte_to_next_page(struct dma_pte *pte) +{ + return first_pte_in_page(pte) ? BIT_ULL(VTD_STRIDE_SHIFT) : + (struct dma_pte *)VTD_PAGE_ALIGN((unsigned long)pte) - pte; +} + extern struct dmar_drhd_unit * dmar_find_matched_drhd_unit(struct pci_dev *dev); extern int dmar_find_matched_atsr_unit(struct pci_dev *dev); -- 1.8.3.1 _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu ^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH v1 2/2] iommu/vt-d: avoid duplicated removing in __domain_mapping 2021-09-15 15:21 ` [PATCH v1 2/2] iommu/vt-d: avoid duplicated removing in __domain_mapping Longpeng(Mike) @ 2021-09-30 14:14 ` Lu Baolu 0 siblings, 0 replies; 4+ messages in thread From: Lu Baolu @ 2021-09-30 14:14 UTC (permalink / raw) To: Longpeng(Mike), dwmw2, joro, will; +Cc: iommu, arei.gonglei, linux-kernel Hi Longpeng, On 2021/9/15 23:21, Longpeng(Mike) wrote: > __domain_mapping() always removes the pages in the range from > 'iov_pfn' to 'end_pfn', but the 'end_pfn' is always the last pfn > of the range that the caller wants to map. > > This would introduce too many duplicated removing and leads the > map operation take too long, for example: > > Map iova=0x100000,nr_pages=0x7d61800 > iov_pfn: 0x100000, end_pfn: 0x7e617ff > iov_pfn: 0x140000, end_pfn: 0x7e617ff > iov_pfn: 0x180000, end_pfn: 0x7e617ff > iov_pfn: 0x1c0000, end_pfn: 0x7e617ff > iov_pfn: 0x200000, end_pfn: 0x7e617ff > ... > it takes about 50ms in total. > > We can reduce the cost by recalculate the 'end_pfn' and limit it > to the boundary of the end of the pte page. > > Map iova=0x100000,nr_pages=0x7d61800 > iov_pfn: 0x100000, end_pfn: 0x13ffff > iov_pfn: 0x140000, end_pfn: 0x17ffff > iov_pfn: 0x180000, end_pfn: 0x1bffff > iov_pfn: 0x1c0000, end_pfn: 0x1fffff > iov_pfn: 0x200000, end_pfn: 0x23ffff > ... > it only need 9ms now. > > Signed-off-by: Longpeng(Mike) <longpeng2@huawei.com> The 0day robot reports below compiling warning when building a 32-bit kernel (make W=1 ARCH=i386): All errors (new ones prefixed by >>): In file included from drivers/gpu/drm/i915/i915_drv.h:43, from drivers/gpu/drm/i915/display/intel_display_types.h:47, from drivers/gpu/drm/i915/display/intel_dsi.h:30, from <command-line>: include/linux/intel-iommu.h: In function 'nr_pte_to_next_page': >> include/linux/intel-iommu.h:719:3: error: cast to pointer from integer of different size [-Werror=int-to-pointer-cast] 719 | (struct dma_pte *)VTD_PAGE_ALIGN((unsigned long)pte) - pte; | ^ cc1: all warnings being treated as errors vim +719 include/linux/intel-iommu.h 715 716 static inline int nr_pte_to_next_page(struct dma_pte *pte) 717 { 718 return first_pte_in_page(pte) ? BIT_ULL(VTD_STRIDE_SHIFT) : > 719 (struct dma_pte *)VTD_PAGE_ALIGN((unsigned long)pte) - pte; 720 } 721 Can you please take a look at this? Best regards, baolu > --- > drivers/iommu/intel/iommu.c | 12 +++++++----- > include/linux/intel-iommu.h | 6 ++++++ > 2 files changed, 13 insertions(+), 5 deletions(-) > > diff --git a/drivers/iommu/intel/iommu.c b/drivers/iommu/intel/iommu.c > index d75f59a..87cbf34 100644 > --- a/drivers/iommu/intel/iommu.c > +++ b/drivers/iommu/intel/iommu.c > @@ -2354,12 +2354,18 @@ static void switch_to_super_page(struct dmar_domain *domain, > return -ENOMEM; > first_pte = pte; > > + lvl_pages = lvl_to_nr_pages(largepage_lvl); > + BUG_ON(nr_pages < lvl_pages); > + > /* It is large page*/ > if (largepage_lvl > 1) { > unsigned long end_pfn; > + unsigned long pages_to_remove; > > pteval |= DMA_PTE_LARGE_PAGE; > - end_pfn = ((iov_pfn + nr_pages) & level_mask(largepage_lvl)) - 1; > + pages_to_remove = min_t(unsigned long, nr_pages, > + nr_pte_to_next_page(pte) * lvl_pages); > + end_pfn = iov_pfn + pages_to_remove - 1; > switch_to_super_page(domain, iov_pfn, end_pfn, largepage_lvl); > } else { > pteval &= ~(uint64_t)DMA_PTE_LARGE_PAGE; > @@ -2381,10 +2387,6 @@ static void switch_to_super_page(struct dmar_domain *domain, > WARN_ON(1); > } > > - lvl_pages = lvl_to_nr_pages(largepage_lvl); > - > - BUG_ON(nr_pages < lvl_pages); > - > nr_pages -= lvl_pages; > iov_pfn += lvl_pages; > phys_pfn += lvl_pages; > diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h > index a590b00..4bff70c 100644 > --- a/include/linux/intel-iommu.h > +++ b/include/linux/intel-iommu.h > @@ -713,6 +713,12 @@ static inline bool first_pte_in_page(struct dma_pte *pte) > return !((unsigned long)pte & ~VTD_PAGE_MASK); > } > > +static inline int nr_pte_to_next_page(struct dma_pte *pte) > +{ > + return first_pte_in_page(pte) ? BIT_ULL(VTD_STRIDE_SHIFT) : > + (struct dma_pte *)VTD_PAGE_ALIGN((unsigned long)pte) - pte; > +} > + > extern struct dmar_drhd_unit * dmar_find_matched_drhd_unit(struct pci_dev *dev); > extern int dmar_find_matched_atsr_unit(struct pci_dev *dev); > > _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu ^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2021-09-30 14:14 UTC | newest] Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2021-09-15 15:21 [PATCH v1 0/2] iommu/vt-d: boost the mapping process Longpeng(Mike) 2021-09-15 15:21 ` [PATCH v1 1/2] iommu/vt-d: convert the return type of first_pte_in_page to bool Longpeng(Mike) 2021-09-15 15:21 ` [PATCH v1 2/2] iommu/vt-d: avoid duplicated removing in __domain_mapping Longpeng(Mike) 2021-09-30 14:14 ` Lu Baolu
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).