search for: nvif_vmm_pfnmap_v0_v

...or (i = 0; i < npages; ++i) { - struct page *page; - - if (!(range->hmm_pfns[i] & HMM_PFN_VALID)) { - ioctl_addr[i] = 0; - continue; - } - - page = hmm_pfn_to_page(range->hmm_pfns[i]); - if (is_device_private_page(page)) - ioctl_addr[i] = nouveau_dmem_page_addr(page) | - NVIF_VMM_PFNMAP_V0_V | - NVIF_VMM_PFNMAP_V0_VRAM; - else - ioctl_addr[i] = page_to_phys(page) | - NVIF_VMM_PFNMAP_V0_V | - NVIF_VMM_PFNMAP_V0_HOST; - if (range->hmm_pfns[i] & HMM_PFN_WRITE) - ioctl_addr[i] |= NVIF_VMM_PFNMAP_V0_W; + if (!(range->hmm_pfns[0] & HMM_PFN_VALID)) { + ioctl_a...

...or (i = 0; i < npages; ++i) { - struct page *page; - - if (!(range->hmm_pfns[i] & HMM_PFN_VALID)) { - ioctl_addr[i] = 0; - continue; - } - - page = hmm_pfn_to_page(range->hmm_pfns[i]); - if (is_device_private_page(page)) - ioctl_addr[i] = nouveau_dmem_page_addr(page) | - NVIF_VMM_PFNMAP_V0_V | - NVIF_VMM_PFNMAP_V0_VRAM; - else - ioctl_addr[i] = page_to_phys(page) | - NVIF_VMM_PFNMAP_V0_V | - NVIF_VMM_PFNMAP_V0_HOST; - if (range->hmm_pfns[i] & HMM_PFN_WRITE) - ioctl_addr[i] |= NVIF_VMM_PFNMAP_V0_W; + if (!(range->hmm_pfns[0] & HMM_PFN_VALID)) { + ioctl_a...

...mp; HMM_PFN_VALID)) { > + ioctl_addr[i] = 0; > continue; > + } Can't we rely on the caller pre-zeroing the array? > + page = hmm_pfn_to_page(range->hmm_pfns[i]); > + if (is_device_private_page(page)) > + ioctl_addr[i] = nouveau_dmem_page_addr(page) | > + NVIF_VMM_PFNMAP_V0_V | > + NVIF_VMM_PFNMAP_V0_VRAM; > + else > + ioctl_addr[i] = page_to_phys(page) | > + NVIF_VMM_PFNMAP_V0_V | > + NVIF_VMM_PFNMAP_V0_HOST; > + if (range->hmm_pfns[i] & HMM_PFN_WRITE) > + ioctl_addr[i] |= NVIF_VMM_PFNMAP_V0_W; Now that this routine isn...

On 8/10/19 4:13 AM, Christoph Hellwig wrote: > On something vaguely related to this patch: > > You use the NVIF_VMM_PFNMAP_V0_V* defines from nvif/if000c.h, which are > a little odd as we only ever set these bits, but they also don't seem > to appear to be in values that are directly fed to the hardware. > > On the other hand mmu/vmm.h defines a set of NVIF_VMM_PFNMAP_V0_* Yes, I see NVKM_VMM_PFN_* > c...

The goal for this series is to introduce the hmm_range_fault() output array flags HMM_PFN_PMD and HMM_PFN_PUD. This allows a device driver to know that a given 4K PFN is actually mapped by the CPU using either a PMD sized or PUD sized CPU page table entry and therefore the device driver can safely map system memory using larger device MMU PTEs. The series is based on 5.8.0-rc3 and is intended for

...zeroing? I think it makes sense that this routine fully sets the output array and does not assume pre-initialize > > + page = hmm_pfn_to_page(range->hmm_pfns[i]); > > + if (is_device_private_page(page)) > > + ioctl_addr[i] = nouveau_dmem_page_addr(page) | > > + NVIF_VMM_PFNMAP_V0_V | > > + NVIF_VMM_PFNMAP_V0_VRAM; > > + else > > + ioctl_addr[i] = page_to_phys(page) | > > + NVIF_VMM_PFNMAP_V0_V | > > + NVIF_VMM_PFNMAP_V0_HOST; > > + if (range->hmm_pfns[i] & HMM_PFN_WRITE) > > + ioctl_addr[i] |= NVIF_VMM_PFNMAP_...

...ouveau/nouveau_svm.c b/drivers/gpu/drm/nouveau/nouveau_svm.c index c68e9317cf0740..cf0d9bd61bebf9 100644 --- a/drivers/gpu/drm/nouveau/nouveau_svm.c +++ b/drivers/gpu/drm/nouveau/nouveau_svm.c @@ -379,7 +379,6 @@ static const u64 nouveau_svm_pfn_values[HMM_PFN_VALUE_MAX] = { [HMM_PFN_ERROR ] = ~NVIF_VMM_PFNMAP_V0_V, [HMM_PFN_NONE ] = NVIF_VMM_PFNMAP_V0_NONE, - [HMM_PFN_SPECIAL] = ~NVIF_VMM_PFNMAP_V0_V, }; /* Issue fault replay for GPU to retry accesses that faulted previously. */ diff --git a/include/linux/hmm.h b/include/linux/hmm.h index 0df27dd03d53d7..81c302c884c0e3 100644 --- a/include/linux/hmm...

...ouveau/nouveau_svm.c b/drivers/gpu/drm/nouveau/nouveau_svm.c index c68e9317cf0740..cf0d9bd61bebf9 100644 --- a/drivers/gpu/drm/nouveau/nouveau_svm.c +++ b/drivers/gpu/drm/nouveau/nouveau_svm.c @@ -379,7 +379,6 @@ static const u64 nouveau_svm_pfn_values[HMM_PFN_VALUE_MAX] = { [HMM_PFN_ERROR ] = ~NVIF_VMM_PFNMAP_V0_V, [HMM_PFN_NONE ] = NVIF_VMM_PFNMAP_V0_NONE, - [HMM_PFN_SPECIAL] = ~NVIF_VMM_PFNMAP_V0_V, }; /* Issue fault replay for GPU to retry accesses that faulted previously. */ diff --git a/include/linux/hmm.h b/include/linux/hmm.h index 0df27dd03d53d7..81c302c884c0e3 100644 --- a/include/linux/hmm...

The goal for this series is to introduce the hmm_pfn_to_map_order() function. This allows a device driver to know that a given 4K PFN is actually mapped by the CPU using a larger sized CPU page table entry and therefore the device driver can safely map system memory using larger device MMU PTEs. The series is based on 5.8.0-rc3 and is intended for Jason Gunthorpe's hmm tree. These were

...ad(page); + args->p.page = page_shift(page); + args->p.size = 1UL << args->p.page; + args->p.addr &= ~(args->p.size - 1); + } if (is_device_private_page(page)) - ioctl_addr[0] = nouveau_dmem_page_addr(page) | + args->p.phys[0] = nouveau_dmem_page_addr(page) | NVIF_VMM_PFNMAP_V0_V | NVIF_VMM_PFNMAP_V0_VRAM; else - ioctl_addr[0] = page_to_phys(page) | + args->p.phys[0] = page_to_phys(page) | NVIF_VMM_PFNMAP_V0_V | NVIF_VMM_PFNMAP_V0_HOST; if (range->hmm_pfns[0] & HMM_PFN_WRITE) - ioctl_addr[0] |= NVIF_VMM_PFNMAP_V0_W; + args->p.phys[0] |= N...

...; + goto done; > + } > + > dpage = nouveau_dmem_page_alloc_locked(drm); > if (!dpage) > goto out; > @@ -560,6 +566,7 @@ static unsigned long nouveau_dmem_migrate_copy_one(struct nouveau_drm *drm, > goto out_free_page; > } > > +done: > *pfn = NVIF_VMM_PFNMAP_V0_V | NVIF_VMM_PFNMAP_V0_VRAM | > ((paddr >> PAGE_SHIFT) << NVIF_VMM_PFNMAP_V0_ADDR_SHIFT); > if (src & MIGRATE_PFN_WRITE) > @@ -615,6 +622,7 @@ nouveau_dmem_migrate_vma(struct nouveau_drm *drm, > struct migrate_vma args = { > .vma = vma, > .start =...

From: Jason Gunthorpe <jgg at mellanox.com> The API is a bit complicated for the uses we actually have, and disucssions for simplifying have come up a number of times. This small series removes the customizable pfn format and simplifies the return code of hmm_range_fault() All the drivers are adjusted to process in the simplified format. I would appreciated tested-by's for the two

hmm_range_fault() returns an array of page frame numbers and flags for how the pages are mapped in the requested process' page tables. The PFN can be used to get the struct page with hmm_pfn_to_page() and the page size order can be determined with compound_order(page) but if the page is larger than order 0 (PAGE_SIZE), there is no indication that the page is mapped using a larger page size. To

...->hmm_pfns[i] & HMM_PFN_VALID)) { + ioctl_addr[i] = 0; continue; + } - addr = nouveau_dmem_page_addr(page); - range->pfns[i] &= ((1UL << range->pfn_shift) - 1); - range->pfns[i] |= (addr >> PAGE_SHIFT) << range->pfn_shift; - range->pfns[i] |= NVIF_VMM_PFNMAP_V0_VRAM; + page = hmm_pfn_to_page(range->hmm_pfns[i]); + if (is_device_private_page(page)) + ioctl_addr[i] = nouveau_dmem_page_addr(page) | + NVIF_VMM_PFNMAP_V0_V | + NVIF_VMM_PFNMAP_V0_VRAM; + else + ioctl_addr[i] = page_to_phys(page) | + NVIF_VMM_PFNMAP_V0_V | + NVIF_VMM_PFNM...

On 8/8/19 12:07 AM, Christoph Hellwig wrote: > On Wed, Aug 07, 2019 at 08:02:14AM -0700, Ralph Campbell wrote: >> When memory is migrated to the GPU it is likely to be accessed by GPU >> code soon afterwards. Instead of waiting for a GPU fault, map the >> migrated memory into the GPU page tables with the same access permissions >> as the source CPU page table entries.

...(page == NULL) - continue; - - if (!is_device_private_page(page)) - continue; - - addr = nouveau_dmem_page_addr(page); - range->pfns[i] &= ((1UL << range->pfn_shift) - 1); - range->pfns[i] |= (addr >> PAGE_SHIFT) << range->pfn_shift; - range->pfns[i] |= NVIF_VMM_PFNMAP_V0_VRAM; - } -} diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.h b/drivers/gpu/drm/nouveau/nouveau_dmem.h index 92394be5d64923..db3b59b210af13 100644 --- a/drivers/gpu/drm/nouveau/nouveau_dmem.h +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.h @@ -37,9 +37,8 @@ int nouveau_dmem_migrate_vma(struct nouveau...

...> + ioctl_addr[i] = 0; > continue; > + } > > - addr = nouveau_dmem_page_addr(page); > - range->pfns[i] &= ((1UL << range->pfn_shift) - 1); > - range->pfns[i] |= (addr >> PAGE_SHIFT) << range->pfn_shift; > - range->pfns[i] |= NVIF_VMM_PFNMAP_V0_VRAM; > + page = hmm_pfn_to_page(range->hmm_pfns[i]); > + if (is_device_private_page(page)) > + ioctl_addr[i] = nouveau_dmem_page_addr(page) | > + NVIF_VMM_PFNMAP_V0_V | > + NVIF_VMM_PFNMAP_V0_VRAM; > + else > + ioctl_addr[i] = page_to_phys(page) | > + NV...

From: Jason Gunthorpe <jgg at mellanox.com> The API is a bit complicated for the uses we actually have, and disucssions for simplifying have come up a number of times. This small series removes the customizable pfn format and simplifies the return code of hmm_range_fault() All the drivers are adjusted to process in the simplified format. I would appreciated tested-by's for the two

...m_fault(struct nvif_notify *notify) */ while (fi < fn) { struct nouveau_svm_fault *fault = buffer->fault[fi++]; - pi = (fault->addr - range.start) >> PAGE_SHIFT; + pi = (fault->addr - args.i.p.addr) >> PAGE_SHIFT; if (ret || - !(range.pfns[pi] & NVIF_VMM_PFNMAP_V0_V) || - (!(range.pfns[pi] & NVIF_VMM_PFNMAP_V0_W) && + !(args.phys[pi] & NVIF_VMM_PFNMAP_V0_V) || + (!(args.phys[pi] & NVIF_VMM_PFNMAP_V0_W) && fault->access != 0 && fault->access != 3)) { nouveau_svm_fault_cancel_fault(svm, f...

...read. >> Why not "pfns + i"? > > OK, will do in v2. > Should I convert to "dma_addrs + nr_dma" too? I'll fix it up for v3 of the migrate_vma series. This is a leftover from passing an args structure. On something vaguely related to this patch: You use the NVIF_VMM_PFNMAP_V0_V* defines from nvif/if000c.h, which are a little odd as we only ever set these bits, but they also don't seem to appear to be in values that are directly fed to the hardware. On the other hand mmu/vmm.h defines a set of NVIF_VMM_PFNMAP_V0_* constants with similar names and identical values, and...