From: Ralph Campbell Subject: [PATCH 1/6] nouveau/hmm: map pages after migration Date: Fri, 8 May 2020 12:20:04 -0700 Message-ID: <20200508192009.15302-2-rcampbell@nvidia.com> References: <20200508192009.15302-1-rcampbell@nvidia.com> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Return-path: In-Reply-To: <20200508192009.15302-1-rcampbell@nvidia.com> Sender: linux-kernel-owner@vger.kernel.org To: nouveau@lists.freedesktop.org, linux-rdma@vger.kernel.org, linux-mm@kvack.org, linux-kselftest@vger.kernel.org, linux-kernel@vger.kernel.org Cc: Jerome Glisse , John Hubbard , Christoph Hellwig , Jason Gunthorpe , Ben Skeggs , Andrew Morton , Shuah Khan , Ralph Campbell List-Id: nouveau.vger.kernel.org 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. This preserves copy on write semantics. Signed-off-by: Ralph Campbell Cc: Christoph Hellwig Cc: Jason Gunthorpe Cc: "J=C3=A9r=C3=B4me Glisse" Cc: Ben Skeggs --- drivers/gpu/drm/nouveau/nouveau_dmem.c | 46 +++++++++++++------- drivers/gpu/drm/nouveau/nouveau_dmem.h | 2 + drivers/gpu/drm/nouveau/nouveau_svm.c | 59 +++++++++++++++++++++++++- drivers/gpu/drm/nouveau/nouveau_svm.h | 5 +++ 4 files changed, 95 insertions(+), 17 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouve= au/nouveau_dmem.c index 3364904eccff..b29b01a56d07 100644 --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c @@ -25,12 +25,14 @@ #include "nouveau_dma.h" #include "nouveau_mem.h" #include "nouveau_bo.h" +#include "nouveau_svm.h" =20 #include #include #include #include #include +#include =20 #include #include @@ -561,10 +563,11 @@ nouveau_dmem_init(struct nouveau_drm *drm) } =20 static unsigned long nouveau_dmem_migrate_copy_one(struct nouveau_drm *drm= , - unsigned long src, dma_addr_t *dma_addr) + unsigned long src, dma_addr_t *dma_addr, u64 *pfn) { struct device *dev =3D drm->dev->dev; struct page *dpage, *spage; + unsigned long paddr; =20 spage =3D migrate_pfn_to_page(src); if (!spage || !(src & MIGRATE_PFN_MIGRATE)) @@ -572,17 +575,21 @@ static unsigned long nouveau_dmem_migrate_copy_one(st= ruct nouveau_drm *drm, =20 dpage =3D nouveau_dmem_page_alloc_locked(drm); if (!dpage) - return 0; + goto out; =20 *dma_addr =3D dma_map_page(dev, spage, 0, PAGE_SIZE, DMA_BIDIRECTIONAL); if (dma_mapping_error(dev, *dma_addr)) goto out_free_page; =20 + paddr =3D nouveau_dmem_page_addr(dpage); if (drm->dmem->migrate.copy_func(drm, 1, NOUVEAU_APER_VRAM, - nouveau_dmem_page_addr(dpage), NOUVEAU_APER_HOST, - *dma_addr)) + paddr, NOUVEAU_APER_HOST, *dma_addr)) goto out_dma_unmap; =20 + *pfn =3D NVIF_VMM_PFNMAP_V0_V | NVIF_VMM_PFNMAP_V0_VRAM | + ((paddr >> PAGE_SHIFT) << NVIF_VMM_PFNMAP_V0_ADDR_SHIFT); + if (src & MIGRATE_PFN_WRITE) + *pfn |=3D NVIF_VMM_PFNMAP_V0_W; return migrate_pfn(page_to_pfn(dpage)) | MIGRATE_PFN_LOCKED; =20 out_dma_unmap: @@ -590,18 +597,20 @@ static unsigned long nouveau_dmem_migrate_copy_one(st= ruct nouveau_drm *drm, out_free_page: nouveau_dmem_page_free_locked(drm, dpage); out: + *pfn =3D NVIF_VMM_PFNMAP_V0_NONE; return 0; } =20 static void nouveau_dmem_migrate_chunk(struct nouveau_drm *drm, - struct migrate_vma *args, dma_addr_t *dma_addrs) + struct nouveau_svmm *svmm, struct migrate_vma *args, + dma_addr_t *dma_addrs, u64 *pfns) { struct nouveau_fence *fence; unsigned long addr =3D args->start, nr_dma =3D 0, i; =20 for (i =3D 0; addr < args->end; i++) { args->dst[i] =3D nouveau_dmem_migrate_copy_one(drm, args->src[i], - dma_addrs + nr_dma); + dma_addrs + nr_dma, pfns + i); if (args->dst[i]) nr_dma++; addr +=3D PAGE_SIZE; @@ -610,20 +619,18 @@ static void nouveau_dmem_migrate_chunk(struct nouveau= _drm *drm, nouveau_fence_new(drm->dmem->migrate.chan, false, &fence); migrate_vma_pages(args); nouveau_dmem_fence_done(&fence); + nouveau_pfns_map(svmm, args->vma->vm_mm, args->start, pfns, i); =20 while (nr_dma--) { dma_unmap_page(drm->dev->dev, dma_addrs[nr_dma], PAGE_SIZE, DMA_BIDIRECTIONAL); } - /* - * FIXME optimization: update GPU page table to point to newly migrated - * memory. - */ migrate_vma_finalize(args); } =20 int nouveau_dmem_migrate_vma(struct nouveau_drm *drm, + struct nouveau_svmm *svmm, struct vm_area_struct *vma, unsigned long start, unsigned long end) @@ -635,7 +642,8 @@ nouveau_dmem_migrate_vma(struct nouveau_drm *drm, .vma =3D vma, .start =3D start, }; - unsigned long c, i; + unsigned long i; + u64 *pfns; int ret =3D -ENOMEM; =20 args.src =3D kcalloc(max, sizeof(*args.src), GFP_KERNEL); @@ -649,19 +657,25 @@ nouveau_dmem_migrate_vma(struct nouveau_drm *drm, if (!dma_addrs) goto out_free_dst; =20 - for (i =3D 0; i < npages; i +=3D c) { - c =3D min(SG_MAX_SINGLE_ALLOC, npages); - args.end =3D start + (c << PAGE_SHIFT); + pfns =3D nouveau_pfns_alloc(max); + if (!pfns) + goto out_free_dma; + + for (i =3D 0; i < npages; i +=3D max) { + args.end =3D start + (max << PAGE_SHIFT); ret =3D migrate_vma_setup(&args); if (ret) - goto out_free_dma; + goto out_free_pfns; =20 if (args.cpages) - nouveau_dmem_migrate_chunk(drm, &args, dma_addrs); + nouveau_dmem_migrate_chunk(drm, svmm, &args, dma_addrs, + pfns); args.start =3D args.end; } =20 ret =3D 0; +out_free_pfns: + nouveau_pfns_free(pfns); out_free_dma: kfree(dma_addrs); out_free_dst: diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.h b/drivers/gpu/drm/nouve= au/nouveau_dmem.h index db3b59b210af..64da5d3635c8 100644 --- a/drivers/gpu/drm/nouveau/nouveau_dmem.h +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.h @@ -25,6 +25,7 @@ struct drm_device; struct drm_file; struct nouveau_drm; +struct nouveau_svmm; struct hmm_range; =20 #if IS_ENABLED(CONFIG_DRM_NOUVEAU_SVM) @@ -34,6 +35,7 @@ void nouveau_dmem_suspend(struct nouveau_drm *); void nouveau_dmem_resume(struct nouveau_drm *); =20 int nouveau_dmem_migrate_vma(struct nouveau_drm *drm, + struct nouveau_svmm *svmm, struct vm_area_struct *vma, unsigned long start, unsigned long end); diff --git a/drivers/gpu/drm/nouveau/nouveau_svm.c b/drivers/gpu/drm/nouvea= u/nouveau_svm.c index 407e34a5c0ab..22f054f7ee3e 100644 --- a/drivers/gpu/drm/nouveau/nouveau_svm.c +++ b/drivers/gpu/drm/nouveau/nouveau_svm.c @@ -70,6 +70,12 @@ struct nouveau_svm { #define SVM_DBG(s,f,a...) NV_DEBUG((s)->drm, "svm: "f"\n", ##a) #define SVM_ERR(s,f,a...) NV_WARN((s)->drm, "svm: "f"\n", ##a) =20 +struct nouveau_pfnmap_args { + struct nvif_ioctl_v0 i; + struct nvif_ioctl_mthd_v0 m; + struct nvif_vmm_pfnmap_v0 p; +}; + struct nouveau_ivmm { struct nouveau_svmm *svmm; u64 inst; @@ -187,7 +193,8 @@ nouveau_svmm_bind(struct drm_device *dev, void *data, addr =3D max(addr, vma->vm_start); next =3D min(vma->vm_end, end); /* This is a best effort so we ignore errors */ - nouveau_dmem_migrate_vma(cli->drm, vma, addr, next); + nouveau_dmem_migrate_vma(cli->drm, cli->svm.svmm, vma, addr, + next); addr =3D next; } =20 @@ -814,6 +821,56 @@ nouveau_svm_fault(struct nvif_notify *notify) return NVIF_NOTIFY_KEEP; } =20 +static struct nouveau_pfnmap_args * +nouveau_pfns_to_args(void *pfns) +{ + return container_of(pfns, struct nouveau_pfnmap_args, p.phys); +} + +u64 * +nouveau_pfns_alloc(unsigned long npages) +{ + struct nouveau_pfnmap_args *args; + + args =3D kzalloc(struct_size(args, p.phys, npages), GFP_KERNEL); + if (!args) + return NULL; + + args->i.type =3D NVIF_IOCTL_V0_MTHD; + args->m.method =3D NVIF_VMM_V0_PFNMAP; + args->p.page =3D PAGE_SHIFT; + + return args->p.phys; +} + +void +nouveau_pfns_free(u64 *pfns) +{ + struct nouveau_pfnmap_args *args =3D nouveau_pfns_to_args(pfns); + + kfree(args); +} + +void +nouveau_pfns_map(struct nouveau_svmm *svmm, struct mm_struct *mm, + unsigned long addr, u64 *pfns, unsigned long npages) +{ + struct nouveau_pfnmap_args *args =3D nouveau_pfns_to_args(pfns); + int ret; + + args->p.addr =3D addr; + args->p.size =3D npages << PAGE_SHIFT; + + mutex_lock(&svmm->mutex); + + svmm->vmm->vmm.object.client->super =3D true; + ret =3D nvif_object_ioctl(&svmm->vmm->vmm.object, args, sizeof(*args) + + npages * sizeof(args->p.phys[0]), NULL); + svmm->vmm->vmm.object.client->super =3D false; + + mutex_unlock(&svmm->mutex); +} + static void nouveau_svm_fault_buffer_fini(struct nouveau_svm *svm, int id) { diff --git a/drivers/gpu/drm/nouveau/nouveau_svm.h b/drivers/gpu/drm/nouvea= u/nouveau_svm.h index e839d8189461..f0fcd1b72e8b 100644 --- a/drivers/gpu/drm/nouveau/nouveau_svm.h +++ b/drivers/gpu/drm/nouveau/nouveau_svm.h @@ -18,6 +18,11 @@ void nouveau_svmm_fini(struct nouveau_svmm **); int nouveau_svmm_join(struct nouveau_svmm *, u64 inst); void nouveau_svmm_part(struct nouveau_svmm *, u64 inst); int nouveau_svmm_bind(struct drm_device *, void *, struct drm_file *); + +u64 *nouveau_pfns_alloc(unsigned long npages); +void nouveau_pfns_free(u64 *pfns); +void nouveau_pfns_map(struct nouveau_svmm *svmm, struct mm_struct *mm, + unsigned long addr, u64 *pfns, unsigned long npages); #else /* IS_ENABLED(CONFIG_DRM_NOUVEAU_SVM) */ static inline void nouveau_svm_init(struct nouveau_drm *drm) {} static inline void nouveau_svm_fini(struct nouveau_drm *drm) {} --=20 2.20.1