[PATCH v3 8/8] nouveau/svm: Implement atomic SVM access

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Some NVIDIA GPUs do not support direct atomic access to system memory
via PCIe. Instead this must be emulated by granting the GPU exclusive
access to the memory. This is achieved by replacing CPU page table
entries with special swap entries that fault on userspace access.

The driver then grants the GPU permission to update the page undergoing
atomic access via the GPU page tables. When CPU access to the page is
required a CPU fault is raised which calls into the device driver via
MMU notifiers to revoke the atomic access. The original page table
entries are then restored allowing CPU access to proceed.

Signed-off-by: Alistair Popple <apopple@xxxxxxxxxx>
---
 drivers/gpu/drm/nouveau/include/nvif/if000c.h |  1 +
 drivers/gpu/drm/nouveau/nouveau_svm.c         | 88 ++++++++++++++++---
 drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h |  1 +
 .../drm/nouveau/nvkm/subdev/mmu/vmmgp100.c    |  6 ++
 4 files changed, 83 insertions(+), 13 deletions(-)

diff --git a/drivers/gpu/drm/nouveau/include/nvif/if000c.h b/drivers/gpu/drm/nouveau/include/nvif/if000c.h
index d6dd40f21eed..9c7ff56831c5 100644
--- a/drivers/gpu/drm/nouveau/include/nvif/if000c.h
+++ b/drivers/gpu/drm/nouveau/include/nvif/if000c.h
@@ -77,6 +77,7 @@ struct nvif_vmm_pfnmap_v0 {
 #define NVIF_VMM_PFNMAP_V0_APER                           0x00000000000000f0ULL
 #define NVIF_VMM_PFNMAP_V0_HOST                           0x0000000000000000ULL
 #define NVIF_VMM_PFNMAP_V0_VRAM                           0x0000000000000010ULL
+#define NVIF_VMM_PFNMAP_V0_A				  0x0000000000000004ULL
 #define NVIF_VMM_PFNMAP_V0_W                              0x0000000000000002ULL
 #define NVIF_VMM_PFNMAP_V0_V                              0x0000000000000001ULL
 #define NVIF_VMM_PFNMAP_V0_NONE                           0x0000000000000000ULL
diff --git a/drivers/gpu/drm/nouveau/nouveau_svm.c b/drivers/gpu/drm/nouveau/nouveau_svm.c
index cd7b47c946cf..630c4a7bcb55 100644
--- a/drivers/gpu/drm/nouveau/nouveau_svm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_svm.c
@@ -35,6 +35,7 @@
 #include <linux/sched/mm.h>
 #include <linux/sort.h>
 #include <linux/hmm.h>
+#include <linux/rmap.h>
 
 struct nouveau_svm {
 	struct nouveau_drm *drm;
@@ -421,9 +422,9 @@ nouveau_svm_fault_cmp(const void *a, const void *b)
 		return ret;
 	if ((ret = (s64)fa->addr - fb->addr))
 		return ret;
-	/*XXX: atomic? */
-	return (fa->access == 0 || fa->access == 3) -
-	       (fb->access == 0 || fb->access == 3);
+	/* Atomic access (2) has highest priority */
+	return (-1*(fa->access == 2) + (fa->access == 0 || fa->access == 3)) -
+	       (-1*(fb->access == 2) + (fb->access == 0 || fb->access == 3));
 }
 
 static void
@@ -555,10 +556,58 @@ static void nouveau_hmm_convert_pfn(struct nouveau_drm *drm,
 		args->p.phys[0] |= NVIF_VMM_PFNMAP_V0_W;
 }
 
+static int nouveau_atomic_range_fault(struct nouveau_svmm *svmm,
+			       struct nouveau_drm *drm,
+			       struct nouveau_pfnmap_args *args, u32 size,
+			       unsigned long hmm_flags, struct mm_struct *mm)
+{
+	struct page *page;
+	unsigned long start = args->p.addr;
+	struct vm_area_struct *vma;
+	int ret = 0;
+
+	mmap_read_lock(mm);
+	vma = find_vma_intersection(mm, start, start + size);
+	if (!vma || !(vma->vm_flags & VM_WRITE)) {
+		ret = -EPERM;
+		goto out;
+	}
+
+	make_device_exclusive_range(mm, start, start + PAGE_SIZE, &page);
+	if (!page) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	/* Map the page on the GPU. */
+	args->p.page = 12;
+	args->p.size = PAGE_SIZE;
+	args->p.addr = start;
+	args->p.phys[0] = page_to_phys(page) |
+		NVIF_VMM_PFNMAP_V0_V |
+		NVIF_VMM_PFNMAP_V0_W |
+		NVIF_VMM_PFNMAP_V0_A |
+		NVIF_VMM_PFNMAP_V0_HOST;
+
+	mutex_lock(&svmm->mutex);
+	svmm->vmm->vmm.object.client->super = true;
+	ret = nvif_object_ioctl(&svmm->vmm->vmm.object, args, size, NULL);
+	svmm->vmm->vmm.object.client->super = false;
+	mutex_unlock(&svmm->mutex);
+
+	set_page_dirty(page);
+	unlock_page(page);
+	put_page(page);
+
+out:
+	mmap_read_unlock(mm);
+	return ret;
+}
+
 static int nouveau_range_fault(struct nouveau_svmm *svmm,
 			       struct nouveau_drm *drm,
 			       struct nouveau_pfnmap_args *args, u32 size,
-			       unsigned long hmm_flags,
+			       unsigned long hmm_flags, int atomic,
 			       struct svm_notifier *notifier)
 {
 	unsigned long timeout =
@@ -608,12 +657,18 @@ static int nouveau_range_fault(struct nouveau_svmm *svmm,
 		break;
 	}
 
-	nouveau_hmm_convert_pfn(drm, &range, args);
+	if (atomic) {
+		mutex_unlock(&svmm->mutex);
+		ret = nouveau_atomic_range_fault(svmm, drm, args,
+						size, hmm_flags, mm);
+	} else {
+		nouveau_hmm_convert_pfn(drm, &range, args);
 
-	svmm->vmm->vmm.object.client->super = true;
-	ret = nvif_object_ioctl(&svmm->vmm->vmm.object, args, size, NULL);
-	svmm->vmm->vmm.object.client->super = false;
-	mutex_unlock(&svmm->mutex);
+		svmm->vmm->vmm.object.client->super = true;
+		ret = nvif_object_ioctl(&svmm->vmm->vmm.object, args, size, NULL);
+		svmm->vmm->vmm.object.client->super = false;
+		mutex_unlock(&svmm->mutex);
+	}
 
 out:
 	mmu_interval_notifier_remove(&notifier->notifier);
@@ -637,7 +692,7 @@ nouveau_svm_fault(struct nvif_notify *notify)
 	unsigned long hmm_flags;
 	u64 inst, start, limit;
 	int fi, fn;
-	int replay = 0, ret;
+	int replay = 0, atomic = 0, ret;
 
 	/* Parse available fault buffer entries into a cache, and update
 	 * the GET pointer so HW can reuse the entries.
@@ -718,12 +773,15 @@ nouveau_svm_fault(struct nvif_notify *notify)
 		/*
 		 * Determine required permissions based on GPU fault
 		 * access flags.
-		 * XXX: atomic?
 		 */
 		switch (buffer->fault[fi]->access) {
 		case 0: /* READ. */
 			hmm_flags = HMM_PFN_REQ_FAULT;
 			break;
+		case 2: /* ATOMIC. */
+			hmm_flags = HMM_PFN_REQ_FAULT | HMM_PFN_REQ_WRITE;
+			atomic = true;
+			break;
 		case 3: /* PREFETCH. */
 			hmm_flags = 0;
 			break;
@@ -740,7 +798,7 @@ nouveau_svm_fault(struct nvif_notify *notify)
 
 		notifier.svmm = svmm;
 		ret = nouveau_range_fault(svmm, svm->drm, &args.i,
-					sizeof(args), hmm_flags, &notifier);
+			sizeof(args), hmm_flags, atomic, &notifier);
 		mmput(mm);
 
 		limit = args.i.p.addr + args.i.p.size;
@@ -760,7 +818,11 @@ nouveau_svm_fault(struct nvif_notify *notify)
 			     !(args.phys[0] & NVIF_VMM_PFNMAP_V0_V)) ||
 			    (buffer->fault[fi]->access != 0 /* READ. */ &&
 			     buffer->fault[fi]->access != 3 /* PREFETCH. */ &&
-			     !(args.phys[0] & NVIF_VMM_PFNMAP_V0_W)))
+			     !(args.phys[0] & NVIF_VMM_PFNMAP_V0_W)) ||
+			    (buffer->fault[fi]->access != 0 /* READ. */ &&
+			     buffer->fault[fi]->access != 1 /* WRITE. */ &&
+			     buffer->fault[fi]->access != 3 /* PREFETCH. */ &&
+			     !(args.phys[0] & NVIF_VMM_PFNMAP_V0_A)))
 				break;
 		}
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h
index a2b179568970..f6188aa9171c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h
@@ -178,6 +178,7 @@ void nvkm_vmm_unmap_region(struct nvkm_vmm *, struct nvkm_vma *);
 #define NVKM_VMM_PFN_APER                                 0x00000000000000f0ULL
 #define NVKM_VMM_PFN_HOST                                 0x0000000000000000ULL
 #define NVKM_VMM_PFN_VRAM                                 0x0000000000000010ULL
+#define NVKM_VMM_PFN_A					  0x0000000000000004ULL
 #define NVKM_VMM_PFN_W                                    0x0000000000000002ULL
 #define NVKM_VMM_PFN_V                                    0x0000000000000001ULL
 #define NVKM_VMM_PFN_NONE                                 0x0000000000000000ULL
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgp100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgp100.c
index 236db5570771..f02abd9cb4dd 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgp100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmgp100.c
@@ -88,6 +88,9 @@ gp100_vmm_pgt_pfn(struct nvkm_vmm *vmm, struct nvkm_mmu_pt *pt,
 		if (!(*map->pfn & NVKM_VMM_PFN_W))
 			data |= BIT_ULL(6); /* RO. */
 
+		if (!(*map->pfn & NVKM_VMM_PFN_A))
+			data |= BIT_ULL(7); /* Atomic disable. */
+
 		if (!(*map->pfn & NVKM_VMM_PFN_VRAM)) {
 			addr = *map->pfn >> NVKM_VMM_PFN_ADDR_SHIFT;
 			addr = dma_map_page(dev, pfn_to_page(addr), 0,
@@ -322,6 +325,9 @@ gp100_vmm_pd0_pfn(struct nvkm_vmm *vmm, struct nvkm_mmu_pt *pt,
 		if (!(*map->pfn & NVKM_VMM_PFN_W))
 			data |= BIT_ULL(6); /* RO. */
 
+		if (!(*map->pfn & NVKM_VMM_PFN_A))
+			data |= BIT_ULL(7); /* Atomic disable. */
+
 		if (!(*map->pfn & NVKM_VMM_PFN_VRAM)) {
 			addr = *map->pfn >> NVKM_VMM_PFN_ADDR_SHIFT;
 			addr = dma_map_page(dev, pfn_to_page(addr), 0,
-- 
2.20.1





[Index of Archives]     [Kernel Newbies]     [Security]     [Netfilter]     [Bugtraq]     [Linux FS]     [Yosemite Forum]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Samba]     [Video 4 Linux]     [Device Mapper]     [Linux Resources]

  Powered by Linux