[Intel-xe] [PATCH 2/2] drm/xe: Make bind engines safe
Matthew Brost
matthew.brost at intel.com
Thu Jul 20 05:12:04 UTC 2023
We currently have a race between bind engines which can result in
corrupted page tables leading to faults.
A simple example:
bind A 0x0000-0x1000, engine A, has unsatisfied in-fence
bind B 0x1000-0x2000, engine B, no in-fences
exec A uses 0x1000-0x2000
Bind B will pass bind A and exec A will fault. This occurs as bind A
programs the root of the page table in a bind job which is held up by an
in-fence. Bind B in this case just programs a leaf entry of the
structure.
To fix use range-fence utility to track cross bind engine conflicts. In
the above example bind A would insert an dependency into the range-fence
tree with a key of 0x0-0x7fffffffff, bind B would find that dependency
and its bind job would scheduled behind the unsatisfied in-fence and
bind A's job.
Reviewed-by: Maarten Lankhorst<maarten.lankhorst at linux.intel.com>
Co-developed-by: Thomas Hellström <thomas.hellstrom at linux.intel.com>
Signed-off-by: Matthew Brost <matthew.brost at intel.com>
---
drivers/gpu/drm/xe/xe_migrate.c | 2 +
drivers/gpu/drm/xe/xe_migrate.h | 8 +++
drivers/gpu/drm/xe/xe_pt.c | 115 +++++++++++++++++++++++++++++++
drivers/gpu/drm/xe/xe_vm.c | 8 +++
drivers/gpu/drm/xe/xe_vm_types.h | 7 ++
5 files changed, 140 insertions(+)
diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 4c6f299fd024..bc7dac4e2086 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -1074,6 +1074,7 @@ xe_migrate_update_pgtables_cpu(struct xe_migrate *m,
return ERR_PTR(-ETIME);
if (ops->pre_commit) {
+ pt_update->job = NULL;
err = ops->pre_commit(pt_update);
if (err)
return ERR_PTR(err);
@@ -1295,6 +1296,7 @@ xe_migrate_update_pgtables(struct xe_migrate *m,
goto err_job;
if (ops->pre_commit) {
+ pt_update->job = job;
err = ops->pre_commit(pt_update);
if (err)
goto err_job;
diff --git a/drivers/gpu/drm/xe/xe_migrate.h b/drivers/gpu/drm/xe/xe_migrate.h
index 204337ea3b4e..0d62aff6421c 100644
--- a/drivers/gpu/drm/xe/xe_migrate.h
+++ b/drivers/gpu/drm/xe/xe_migrate.h
@@ -69,6 +69,14 @@ struct xe_migrate_pt_update {
const struct xe_migrate_pt_update_ops *ops;
/** @vma: The vma we're updating the pagetable for. */
struct xe_vma *vma;
+ /** @job: The job if a GPU page-table update. NULL otherwise */
+ struct xe_sched_job *job;
+ /** @start: Start of update for the range fence */
+ u64 start;
+ /** @last: Last of update for the range fence */
+ u64 last;
+ /** @tile_id: Tile ID of the update */
+ u8 tile_id;
};
struct xe_migrate *xe_migrate_init(struct xe_tile *tile);
diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c
index 851ea7c01b91..a458224a88ca 100644
--- a/drivers/gpu/drm/xe/xe_pt.c
+++ b/drivers/gpu/drm/xe/xe_pt.c
@@ -1116,6 +1116,53 @@ struct xe_pt_migrate_pt_update {
bool locked;
};
+/*
+ * This function adds the needed dependencies to a page-table update job
+ * to make sure racing jobs for separate bind engines don't race writing
+ * to the same page-table range, wreaking havoc. Initially use a single
+ * fence for the entire VM. An optimization would use smaller granularity.
+ */
+static int xe_pt_vm_dependencies(struct xe_sched_job *job,
+ struct xe_range_fence_tree *rftree,
+ u64 start, u64 last)
+{
+ struct xe_range_fence *rtfence;
+ struct dma_fence *fence;
+ int err;
+
+ rtfence = xe_range_fence_tree_first(rftree, start, last);
+ while (rtfence) {
+ fence = rtfence->fence;
+
+ if (!dma_fence_is_signaled(fence)) {
+ /*
+ * Is this a CPU update? GPU is busy updating, so return
+ * an error
+ */
+ if (!job)
+ return -ETIME;
+
+ dma_fence_get(fence);
+ err = drm_sched_job_add_dependency(&job->drm, fence);
+ if (err)
+ return err;
+ }
+
+ rtfence = xe_range_fence_tree_next(rtfence, start, last);
+ }
+
+ return 0;
+}
+
+static int xe_pt_pre_commit(struct xe_migrate_pt_update *pt_update)
+{
+ struct xe_range_fence_tree *rftree =
+ &xe_vma_vm(pt_update->vma)->rftree[pt_update->tile_id];
+
+ return xe_pt_vm_dependencies(pt_update->job, rftree,
+ pt_update->start, pt_update->last);
+}
+
static int xe_pt_userptr_pre_commit(struct xe_migrate_pt_update *pt_update)
{
struct xe_pt_migrate_pt_update *userptr_update =
@@ -1123,6 +1170,13 @@ static int xe_pt_userptr_pre_commit(struct xe_migrate_pt_update *pt_update)
struct xe_vma *vma = pt_update->vma;
unsigned long notifier_seq = vma->userptr.notifier_seq;
struct xe_vm *vm = xe_vma_vm(vma);
+ int err = xe_pt_vm_dependencies(pt_update->job,
+ &vm->rftree[pt_update->tile_id],
+ pt_update->start,
+ pt_update->last);
+
+ if (err)
+ return err;
userptr_update->locked = false;
@@ -1161,6 +1215,7 @@ static int xe_pt_userptr_pre_commit(struct xe_migrate_pt_update *pt_update)
static const struct xe_migrate_pt_update_ops bind_ops = {
.populate = xe_vm_populate_pgtable,
+ .pre_commit = xe_pt_pre_commit,
};
static const struct xe_migrate_pt_update_ops userptr_bind_ops = {
@@ -1258,6 +1313,27 @@ static int invalidation_fence_init(struct xe_gt *gt,
return ret && ret != -ENOENT ? ret : 0;
}
+static void xe_pt_calc_rfence_interval(struct xe_vma *vma,
+ struct xe_pt_migrate_pt_update *update,
+ struct xe_vm_pgtable_update *entries,
+ u32 num_entries)
+{
+ int i, level = 0;
+
+ for (i = 0; i < num_entries; i++) {
+ const struct xe_vm_pgtable_update *entry = &entries[i];
+
+ if (entry->pt->level > level)
+ level = entry->pt->level;
+ }
+
+ /* Greedy (non-optimal) calculation but simple */
+ update->base.start = ALIGN_DOWN(xe_vma_start(vma),
+ 0x1ull << xe_pt_shift(level));
+ update->base.last = ALIGN(xe_vma_end(vma),
+ 0x1ull << xe_pt_shift(level)) - 1;
+}
+
/**
* __xe_pt_bind_vma() - Build and connect a page-table tree for the vma
* address range.
@@ -1290,6 +1366,7 @@ __xe_pt_bind_vma(struct xe_tile *tile, struct xe_vma *vma, struct xe_engine *e,
.base = {
.ops = xe_vma_is_userptr(vma) ? &userptr_bind_ops : &bind_ops,
.vma = vma,
+ .tile_id = tile->id,
},
.bind = true,
};
@@ -1297,6 +1374,7 @@ __xe_pt_bind_vma(struct xe_tile *tile, struct xe_vma *vma, struct xe_engine *e,
u32 num_entries;
struct dma_fence *fence;
struct invalidation_fence *ifence = NULL;
+ struct xe_range_fence *rfence;
int err;
bind_pt_update.locked = false;
@@ -1313,6 +1391,8 @@ __xe_pt_bind_vma(struct xe_tile *tile, struct xe_vma *vma, struct xe_engine *e,
XE_BUG_ON(num_entries > ARRAY_SIZE(entries));
xe_vm_dbg_print_entries(tile_to_xe(tile), entries, num_entries);
+ xe_pt_calc_rfence_interval(vma, &bind_pt_update, entries,
+ num_entries);
/*
* If rebind, we have to invalidate TLB on !LR vms to invalidate
@@ -1333,6 +1413,12 @@ __xe_pt_bind_vma(struct xe_tile *tile, struct xe_vma *vma, struct xe_engine *e,
return ERR_PTR(-ENOMEM);
}
+ rfence = kzalloc(sizeof(*rfence), GFP_KERNEL);
+ if (!rfence) {
+ kfree(ifence);
+ return ERR_PTR(-ENOMEM);
+ }
+
fence = xe_migrate_update_pgtables(tile->migrate,
vm, xe_vma_bo(vma),
e ? e : vm->eng[tile->id],
@@ -1342,6 +1428,14 @@ __xe_pt_bind_vma(struct xe_tile *tile, struct xe_vma *vma, struct xe_engine *e,
if (!IS_ERR(fence)) {
bool last_munmap_rebind = vma->gpuva.flags & XE_VMA_LAST_REBIND;
LLIST_HEAD(deferred);
+ int err;
+
+ err = xe_range_fence_insert(&vm->rftree[tile->id], rfence,
+ &xe_range_fence_kfree_ops,
+ bind_pt_update.base.start,
+ bind_pt_update.base.last, fence);
+ if (err)
+ dma_fence_wait(fence, false);
/* TLB invalidation must be done before signaling rebind */
if (ifence) {
@@ -1380,6 +1474,7 @@ __xe_pt_bind_vma(struct xe_tile *tile, struct xe_vma *vma, struct xe_engine *e,
queue_work(vm->xe->ordered_wq,
&vm->preempt.rebind_work);
} else {
+ kfree(rfence);
kfree(ifence);
if (bind_pt_update.locked)
up_read(&vm->userptr.notifier_lock);
@@ -1589,6 +1684,7 @@ xe_pt_commit_unbind(struct xe_vma *vma,
static const struct xe_migrate_pt_update_ops unbind_ops = {
.populate = xe_migrate_clear_pgtable_callback,
+ .pre_commit = xe_pt_pre_commit,
};
static const struct xe_migrate_pt_update_ops userptr_unbind_ops = {
@@ -1626,12 +1722,15 @@ __xe_pt_unbind_vma(struct xe_tile *tile, struct xe_vma *vma, struct xe_engine *e
.ops = xe_vma_is_userptr(vma) ? &userptr_unbind_ops :
&unbind_ops,
.vma = vma,
+ .tile_id = tile->id,
},
};
struct xe_vm *vm = xe_vma_vm(vma);
u32 num_entries;
struct dma_fence *fence = NULL;
struct invalidation_fence *ifence;
+ struct xe_range_fence *rfence;
+
LLIST_HEAD(deferred);
xe_bo_assert_held(xe_vma_bo(vma));
@@ -1645,11 +1744,19 @@ __xe_pt_unbind_vma(struct xe_tile *tile, struct xe_vma *vma, struct xe_engine *e
XE_BUG_ON(num_entries > ARRAY_SIZE(entries));
xe_vm_dbg_print_entries(tile_to_xe(tile), entries, num_entries);
+ xe_pt_calc_rfence_interval(vma, &unbind_pt_update, entries,
+ num_entries);
ifence = kzalloc(sizeof(*ifence), GFP_KERNEL);
if (!ifence)
return ERR_PTR(-ENOMEM);
+ rfence = kzalloc(sizeof(*rfence), GFP_KERNEL);
+ if (!rfence) {
+ kfree(ifence);
+ return ERR_PTR(-ENOMEM);
+ }
+
/*
* Even if we were already evicted and unbind to destroy, we need to
* clear again here. The eviction may have updated pagetables at a
@@ -1664,6 +1771,13 @@ __xe_pt_unbind_vma(struct xe_tile *tile, struct xe_vma *vma, struct xe_engine *e
if (!IS_ERR(fence)) {
int err;
+ err = xe_range_fence_insert(&vm->rftree[tile->id], rfence,
+ &xe_range_fence_kfree_ops,
+ unbind_pt_update.base.start,
+ unbind_pt_update.base.last, fence);
+ if (err)
+ dma_fence_wait(fence, false);
+
/* TLB invalidation must be done before signaling unbind */
err = invalidation_fence_init(tile->primary_gt, ifence, fence, vma);
if (err) {
@@ -1685,6 +1799,7 @@ __xe_pt_unbind_vma(struct xe_tile *tile, struct xe_vma *vma, struct xe_engine *e
unbind_pt_update.locked ? &deferred : NULL);
vma->tile_present &= ~BIT(tile->id);
} else {
+ kfree(rfence);
kfree(ifence);
}
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index fc2a09087d98..f8e6f1fda1fe 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -1229,6 +1229,9 @@ struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags)
INIT_LIST_HEAD(&vm->preempt.engines);
vm->preempt.min_run_period_ms = 10; /* FIXME: Wire up to uAPI */
+ for_each_tile(tile, xe, id)
+ xe_range_fence_tree_init(&vm->rftree[id]);
+
INIT_LIST_HEAD(&vm->extobj.list);
if (!(flags & XE_VM_FLAG_MIGRATION))
@@ -1348,6 +1351,8 @@ struct xe_vm *xe_vm_create(struct xe_device *xe, u32 flags)
drm_gpuva_manager_destroy(&vm->mgr);
err_put:
dma_resv_fini(&vm->resv);
+ for_each_tile(tile, xe, id)
+ xe_range_fence_tree_fini(&vm->rftree[id]);
kfree(vm);
if (!(flags & XE_VM_FLAG_MIGRATION))
xe_device_mem_access_put(xe);
@@ -1490,6 +1495,9 @@ void xe_vm_close_and_put(struct xe_vm *vm)
xe->usm.num_vm_in_non_fault_mode--;
mutex_unlock(&xe->usm.lock);
+ for_each_tile(tile, xe, id)
+ xe_range_fence_tree_fini(&vm->rftree[id]);
+
xe_vm_put(vm);
}
diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h
index 379c5d617d72..bcbcad1929d6 100644
--- a/drivers/gpu/drm/xe/xe_vm_types.h
+++ b/drivers/gpu/drm/xe/xe_vm_types.h
@@ -15,6 +15,7 @@
#include "xe_device_types.h"
#include "xe_pt_types.h"
+#include "xe_range_fence.h"
struct async_op_fence;
struct xe_bo;
@@ -182,6 +183,12 @@ struct xe_vm {
*/
struct work_struct destroy_work;
+ /**
+ * @rftree: range fence tree to track updates to page table structure.
+ * Used to implement conflict tracking between independent bind engines.
+ */
+ struct xe_range_fence_tree rftree[XE_MAX_TILES_PER_DEVICE];
+
/** @extobj: bookkeeping for external objects. Protected by the vm lock */
struct {
/** @enties: number of external BOs attached this VM */
--
2.34.1
More information about the Intel-xe
mailing list