[Intel-xe] [igt-dev] [PATCH i-g-t 03/16] drm-uapi/xe_drm: Separate VM_BIND's operation and flag, align with latest uapi
Matthew Brost
matthew.brost at intel.com
Tue Sep 19 15:31:34 UTC 2023
On Tue, Sep 19, 2023 at 10:19:46AM -0400, Rodrigo Vivi wrote:
> From: Francois Dugast <francois.dugast at intel.com>
>
> Align with commit ("drm/xe/uapi: Separate VM_BIND's operation and flag")
>
> Signed-off-by: Francois Dugast <francois.dugast at intel.com>
> Signed-off-by: Rodrigo Vivi <rodrigo.vivi at intel.com>
Reviewed-by: Matthew Brost <matthew.brost at intel.com>
> ---
> include/drm-uapi/xe_drm.h | 14 ++++++++------
> lib/intel_batchbuffer.c | 11 +++++++----
> lib/xe/xe_ioctl.c | 31 ++++++++++++++++---------------
> lib/xe/xe_ioctl.h | 6 +++---
> lib/xe/xe_util.c | 9 ++++++---
> tests/intel/xe_exec_basic.c | 2 +-
> tests/intel/xe_exec_threads.c | 2 +-
> tests/intel/xe_vm.c | 16 +++++++++-------
> 8 files changed, 51 insertions(+), 40 deletions(-)
>
> diff --git a/include/drm-uapi/xe_drm.h b/include/drm-uapi/xe_drm.h
> index f96c84a98..078edd9f8 100644
> --- a/include/drm-uapi/xe_drm.h
> +++ b/include/drm-uapi/xe_drm.h
> @@ -663,8 +663,10 @@ struct drm_xe_vm_bind_op {
> #define XE_VM_BIND_OP_RESTART 0x3
> #define XE_VM_BIND_OP_UNMAP_ALL 0x4
> #define XE_VM_BIND_OP_PREFETCH 0x5
> + /** @op: Bind operation to perform */
> + __u32 op;
>
> -#define XE_VM_BIND_FLAG_READONLY (0x1 << 16)
> +#define XE_VM_BIND_FLAG_READONLY (0x1 << 0)
> /*
> * A bind ops completions are always async, hence the support for out
> * sync. This flag indicates the allocation of the memory for new page
> @@ -689,12 +691,12 @@ struct drm_xe_vm_bind_op {
> * configured in the VM and must be set if the VM is configured with
> * DRM_XE_VM_CREATE_ASYNC_BIND_OPS and not in an error state.
> */
> -#define XE_VM_BIND_FLAG_ASYNC (0x1 << 17)
> +#define XE_VM_BIND_FLAG_ASYNC (0x1 << 1)
> /*
> * Valid on a faulting VM only, do the MAP operation immediately rather
> * than deferring the MAP to the page fault handler.
> */
> -#define XE_VM_BIND_FLAG_IMMEDIATE (0x1 << 18)
> +#define XE_VM_BIND_FLAG_IMMEDIATE (0x1 << 2)
> /*
> * When the NULL flag is set, the page tables are setup with a special
> * bit which indicates writes are dropped and all reads return zero. In
> @@ -702,9 +704,9 @@ struct drm_xe_vm_bind_op {
> * operations, the BO handle MBZ, and the BO offset MBZ. This flag is
> * intended to implement VK sparse bindings.
> */
> -#define XE_VM_BIND_FLAG_NULL (0x1 << 19)
> - /** @op: Operation to perform (lower 16 bits) and flags (upper 16 bits) */
> - __u32 op;
> +#define XE_VM_BIND_FLAG_NULL (0x1 << 3)
> + /** @flags: Bind flags */
> + __u32 flags;
>
> /** @mem_region: Memory region to prefetch VMA to, instance not a mask */
> __u32 region;
> diff --git a/lib/intel_batchbuffer.c b/lib/intel_batchbuffer.c
> index e7b1b755f..6e668d28c 100644
> --- a/lib/intel_batchbuffer.c
> +++ b/lib/intel_batchbuffer.c
> @@ -1281,7 +1281,8 @@ void intel_bb_destroy(struct intel_bb *ibb)
> }
>
> static struct drm_xe_vm_bind_op *xe_alloc_bind_ops(struct intel_bb *ibb,
> - uint32_t op, uint32_t region)
> + uint32_t op, uint32_t flags,
> + uint32_t region)
> {
> struct drm_i915_gem_exec_object2 **objects = ibb->objects;
> struct drm_xe_vm_bind_op *bind_ops, *ops;
> @@ -1298,6 +1299,7 @@ static struct drm_xe_vm_bind_op *xe_alloc_bind_ops(struct intel_bb *ibb,
> ops->obj = objects[i]->handle;
>
> ops->op = op;
> + ops->flags = flags;
> ops->obj_offset = 0;
> ops->addr = objects[i]->offset;
> ops->range = objects[i]->rsvd1;
> @@ -1323,9 +1325,10 @@ static void __unbind_xe_objects(struct intel_bb *ibb)
>
> if (ibb->num_objects > 1) {
> struct drm_xe_vm_bind_op *bind_ops;
> - uint32_t op = XE_VM_BIND_OP_UNMAP | XE_VM_BIND_FLAG_ASYNC;
> + uint32_t op = XE_VM_BIND_OP_UNMAP;
> + uint32_t flags = XE_VM_BIND_FLAG_ASYNC;
>
> - bind_ops = xe_alloc_bind_ops(ibb, op, 0);
> + bind_ops = xe_alloc_bind_ops(ibb, op, flags, 0);
> xe_vm_bind_array(ibb->fd, ibb->vm_id, 0, bind_ops,
> ibb->num_objects, syncs, 2);
> free(bind_ops);
> @@ -2354,7 +2357,7 @@ __xe_bb_exec(struct intel_bb *ibb, uint64_t flags, bool sync)
>
> syncs[0].handle = syncobj_create(ibb->fd, 0);
> if (ibb->num_objects > 1) {
> - bind_ops = xe_alloc_bind_ops(ibb, XE_VM_BIND_OP_MAP | XE_VM_BIND_FLAG_ASYNC, 0);
> + bind_ops = xe_alloc_bind_ops(ibb, XE_VM_BIND_OP_MAP, XE_VM_BIND_FLAG_ASYNC, 0);
> xe_vm_bind_array(ibb->fd, ibb->vm_id, 0, bind_ops,
> ibb->num_objects, syncs, 1);
> free(bind_ops);
> diff --git a/lib/xe/xe_ioctl.c b/lib/xe/xe_ioctl.c
> index 730dcfd16..48cd185de 100644
> --- a/lib/xe/xe_ioctl.c
> +++ b/lib/xe/xe_ioctl.c
> @@ -67,7 +67,7 @@ void xe_vm_unbind_all_async(int fd, uint32_t vm, uint32_t exec_queue,
> uint32_t num_syncs)
> {
> __xe_vm_bind_assert(fd, vm, exec_queue, bo, 0, 0, 0,
> - XE_VM_BIND_OP_UNMAP_ALL | XE_VM_BIND_FLAG_ASYNC,
> + XE_VM_BIND_OP_UNMAP_ALL, XE_VM_BIND_FLAG_ASYNC,
> sync, num_syncs, 0, 0);
> }
>
> @@ -91,8 +91,8 @@ void xe_vm_bind_array(int fd, uint32_t vm, uint32_t exec_queue,
>
> int __xe_vm_bind(int fd, uint32_t vm, uint32_t exec_queue, uint32_t bo,
> uint64_t offset, uint64_t addr, uint64_t size, uint32_t op,
> - struct drm_xe_sync *sync, uint32_t num_syncs, uint32_t region,
> - uint64_t ext)
> + uint32_t flags, struct drm_xe_sync *sync, uint32_t num_syncs,
> + uint32_t region, uint64_t ext)
> {
> struct drm_xe_vm_bind bind = {
> .extensions = ext,
> @@ -103,6 +103,7 @@ int __xe_vm_bind(int fd, uint32_t vm, uint32_t exec_queue, uint32_t bo,
> .bind.range = size,
> .bind.addr = addr,
> .bind.op = op,
> + .bind.flags = flags,
> .bind.region = region,
> .num_syncs = num_syncs,
> .syncs = (uintptr_t)sync,
> @@ -117,11 +118,11 @@ int __xe_vm_bind(int fd, uint32_t vm, uint32_t exec_queue, uint32_t bo,
>
> void __xe_vm_bind_assert(int fd, uint32_t vm, uint32_t exec_queue, uint32_t bo,
> uint64_t offset, uint64_t addr, uint64_t size,
> - uint32_t op, struct drm_xe_sync *sync,
> + uint32_t op, uint32_t flags, struct drm_xe_sync *sync,
> uint32_t num_syncs, uint32_t region, uint64_t ext)
> {
> igt_assert_eq(__xe_vm_bind(fd, vm, exec_queue, bo, offset, addr, size,
> - op, sync, num_syncs, region, ext), 0);
> + op, flags, sync, num_syncs, region, ext), 0);
> }
>
> void xe_vm_bind(int fd, uint32_t vm, uint32_t bo, uint64_t offset,
> @@ -129,7 +130,7 @@ void xe_vm_bind(int fd, uint32_t vm, uint32_t bo, uint64_t offset,
> struct drm_xe_sync *sync, uint32_t num_syncs)
> {
> __xe_vm_bind_assert(fd, vm, 0, bo, offset, addr, size,
> - XE_VM_BIND_OP_MAP, sync, num_syncs, 0, 0);
> + XE_VM_BIND_OP_MAP, 0, sync, num_syncs, 0, 0);
> }
>
> void xe_vm_unbind(int fd, uint32_t vm, uint64_t offset,
> @@ -137,7 +138,7 @@ void xe_vm_unbind(int fd, uint32_t vm, uint64_t offset,
> struct drm_xe_sync *sync, uint32_t num_syncs)
> {
> __xe_vm_bind_assert(fd, vm, 0, 0, offset, addr, size,
> - XE_VM_BIND_OP_UNMAP, sync, num_syncs, 0, 0);
> + XE_VM_BIND_OP_UNMAP, 0, sync, num_syncs, 0, 0);
> }
>
> void xe_vm_prefetch_async(int fd, uint32_t vm, uint32_t exec_queue, uint64_t offset,
> @@ -146,7 +147,7 @@ void xe_vm_prefetch_async(int fd, uint32_t vm, uint32_t exec_queue, uint64_t off
> uint32_t region)
> {
> __xe_vm_bind_assert(fd, vm, exec_queue, 0, offset, addr, size,
> - XE_VM_BIND_OP_PREFETCH | XE_VM_BIND_FLAG_ASYNC,
> + XE_VM_BIND_OP_PREFETCH, XE_VM_BIND_FLAG_ASYNC,
> sync, num_syncs, region, 0);
> }
>
> @@ -155,7 +156,7 @@ void xe_vm_bind_async(int fd, uint32_t vm, uint32_t exec_queue, uint32_t bo,
> struct drm_xe_sync *sync, uint32_t num_syncs)
> {
> __xe_vm_bind_assert(fd, vm, exec_queue, bo, offset, addr, size,
> - XE_VM_BIND_OP_MAP | XE_VM_BIND_FLAG_ASYNC, sync,
> + XE_VM_BIND_OP_MAP, XE_VM_BIND_FLAG_ASYNC, sync,
> num_syncs, 0, 0);
> }
>
> @@ -165,7 +166,7 @@ void xe_vm_bind_async_flags(int fd, uint32_t vm, uint32_t exec_queue, uint32_t b
> uint32_t flags)
> {
> __xe_vm_bind_assert(fd, vm, exec_queue, bo, offset, addr, size,
> - XE_VM_BIND_OP_MAP | XE_VM_BIND_FLAG_ASYNC | flags,
> + XE_VM_BIND_OP_MAP, XE_VM_BIND_FLAG_ASYNC | flags,
> sync, num_syncs, 0, 0);
> }
>
> @@ -174,7 +175,7 @@ void xe_vm_bind_userptr_async(int fd, uint32_t vm, uint32_t exec_queue,
> struct drm_xe_sync *sync, uint32_t num_syncs)
> {
> __xe_vm_bind_assert(fd, vm, exec_queue, 0, userptr, addr, size,
> - XE_VM_BIND_OP_MAP_USERPTR | XE_VM_BIND_FLAG_ASYNC,
> + XE_VM_BIND_OP_MAP_USERPTR, XE_VM_BIND_FLAG_ASYNC,
> sync, num_syncs, 0, 0);
> }
>
> @@ -184,7 +185,7 @@ void xe_vm_bind_userptr_async_flags(int fd, uint32_t vm, uint32_t exec_queue,
> uint32_t num_syncs, uint32_t flags)
> {
> __xe_vm_bind_assert(fd, vm, exec_queue, 0, userptr, addr, size,
> - XE_VM_BIND_OP_MAP_USERPTR | XE_VM_BIND_FLAG_ASYNC |
> + XE_VM_BIND_OP_MAP_USERPTR, XE_VM_BIND_FLAG_ASYNC |
> flags, sync, num_syncs, 0, 0);
> }
>
> @@ -193,7 +194,7 @@ void xe_vm_unbind_async(int fd, uint32_t vm, uint32_t exec_queue,
> struct drm_xe_sync *sync, uint32_t num_syncs)
> {
> __xe_vm_bind_assert(fd, vm, exec_queue, 0, offset, addr, size,
> - XE_VM_BIND_OP_UNMAP | XE_VM_BIND_FLAG_ASYNC, sync,
> + XE_VM_BIND_OP_UNMAP, XE_VM_BIND_FLAG_ASYNC, sync,
> num_syncs, 0, 0);
> }
>
> @@ -205,8 +206,8 @@ static void __xe_vm_bind_sync(int fd, uint32_t vm, uint32_t bo, uint64_t offset,
> .handle = syncobj_create(fd, 0),
> };
>
> - __xe_vm_bind_assert(fd, vm, 0, bo, offset, addr, size, op, &sync, 1, 0,
> - 0);
> + __xe_vm_bind_assert(fd, vm, 0, bo, offset, addr, size, op, 0, &sync, 1,
> + 0, 0);
>
> igt_assert(syncobj_wait(fd, &sync.handle, 1, INT64_MAX, 0, NULL));
> syncobj_destroy(fd, sync.handle);
> diff --git a/lib/xe/xe_ioctl.h b/lib/xe/xe_ioctl.h
> index 6c281b3bf..f0e4109dc 100644
> --- a/lib/xe/xe_ioctl.h
> +++ b/lib/xe/xe_ioctl.h
> @@ -19,11 +19,11 @@ uint32_t xe_cs_prefetch_size(int fd);
> uint32_t xe_vm_create(int fd, uint32_t flags, uint64_t ext);
> int __xe_vm_bind(int fd, uint32_t vm, uint32_t exec_queue, uint32_t bo,
> uint64_t offset, uint64_t addr, uint64_t size, uint32_t op,
> - struct drm_xe_sync *sync, uint32_t num_syncs, uint32_t region,
> - uint64_t ext);
> + uint32_t flags, struct drm_xe_sync *sync, uint32_t num_syncs,
> + uint32_t region, uint64_t ext);
> void __xe_vm_bind_assert(int fd, uint32_t vm, uint32_t exec_queue, uint32_t bo,
> uint64_t offset, uint64_t addr, uint64_t size,
> - uint32_t op, struct drm_xe_sync *sync,
> + uint32_t op, uint32_t flags, struct drm_xe_sync *sync,
> uint32_t num_syncs, uint32_t region, uint64_t ext);
> void xe_vm_bind(int fd, uint32_t vm, uint32_t bo, uint64_t offset,
> uint64_t addr, uint64_t size,
> diff --git a/lib/xe/xe_util.c b/lib/xe/xe_util.c
> index 2f9ffe2f1..5fa4d4610 100644
> --- a/lib/xe/xe_util.c
> +++ b/lib/xe/xe_util.c
> @@ -116,7 +116,7 @@ static struct drm_xe_vm_bind_op *xe_alloc_bind_ops(struct igt_list_head *obj_lis
> {
> struct drm_xe_vm_bind_op *bind_ops, *ops;
> struct xe_object *obj;
> - uint32_t num_objects = 0, i = 0, op;
> + uint32_t num_objects = 0, i = 0, op, flags;
>
> igt_list_for_each_entry(obj, obj_list, link)
> num_objects++;
> @@ -134,13 +134,16 @@ static struct drm_xe_vm_bind_op *xe_alloc_bind_ops(struct igt_list_head *obj_lis
> ops = &bind_ops[i];
>
> if (obj->bind_op == XE_OBJECT_BIND) {
> - op = XE_VM_BIND_OP_MAP | XE_VM_BIND_FLAG_ASYNC;
> + op = XE_VM_BIND_OP_MAP;
> + flags = XE_VM_BIND_FLAG_ASYNC;
> ops->obj = obj->handle;
> } else {
> - op = XE_VM_BIND_OP_UNMAP | XE_VM_BIND_FLAG_ASYNC;
> + op = XE_VM_BIND_OP_UNMAP;
> + flags = XE_VM_BIND_FLAG_ASYNC;
> }
>
> ops->op = op;
> + ops->flags = flags;
> ops->obj_offset = 0;
> ops->addr = obj->offset;
> ops->range = obj->size;
> diff --git a/tests/intel/xe_exec_basic.c b/tests/intel/xe_exec_basic.c
> index a4414e052..e29398aaa 100644
> --- a/tests/intel/xe_exec_basic.c
> +++ b/tests/intel/xe_exec_basic.c
> @@ -170,7 +170,7 @@ test_exec(int fd, struct drm_xe_engine_class_instance *eci,
> if (flags & SPARSE)
> __xe_vm_bind_assert(fd, vm[i], bind_exec_queues[i],
> 0, 0, sparse_addr[i], bo_size,
> - XE_VM_BIND_OP_MAP |
> + XE_VM_BIND_OP_MAP,
> XE_VM_BIND_FLAG_ASYNC |
> XE_VM_BIND_FLAG_NULL, sync,
> 1, 0, 0);
> diff --git a/tests/intel/xe_exec_threads.c b/tests/intel/xe_exec_threads.c
> index 12e76874e..1f9af894f 100644
> --- a/tests/intel/xe_exec_threads.c
> +++ b/tests/intel/xe_exec_threads.c
> @@ -609,7 +609,7 @@ test_legacy_mode(int fd, uint32_t vm, uint64_t addr, uint64_t userptr,
> if (rebind_error_inject == i)
> __xe_vm_bind_assert(fd, vm, bind_exec_queues[e],
> 0, 0, addr, bo_size,
> - XE_VM_BIND_OP_UNMAP |
> + XE_VM_BIND_OP_UNMAP,
> XE_VM_BIND_FLAG_ASYNC |
> INJECT_ERROR, sync_all,
> n_exec_queues, 0, 0);
> diff --git a/tests/intel/xe_vm.c b/tests/intel/xe_vm.c
> index 4952ea786..f96305851 100644
> --- a/tests/intel/xe_vm.c
> +++ b/tests/intel/xe_vm.c
> @@ -316,7 +316,7 @@ static void userptr_invalid(int fd)
> vm = xe_vm_create(fd, 0, 0);
> munmap(data, size);
> ret = __xe_vm_bind(fd, vm, 0, 0, to_user_pointer(data), 0x40000,
> - size, XE_VM_BIND_OP_MAP_USERPTR, NULL, 0, 0, 0);
> + size, XE_VM_BIND_OP_MAP_USERPTR, 0, NULL, 0, 0, 0);
> igt_assert(ret == -EFAULT);
>
> xe_vm_destroy(fd, vm);
> @@ -437,7 +437,7 @@ static void vm_async_ops_err(int fd, bool destroy)
> if (i == N_BINDS / 8) /* Inject error on this bind */
> __xe_vm_bind_assert(fd, vm, 0, bo, 0,
> addr + i * bo_size * 2,
> - bo_size, XE_VM_BIND_OP_MAP |
> + bo_size, XE_VM_BIND_OP_MAP,
> XE_VM_BIND_FLAG_ASYNC |
> INJECT_ERROR, &sync, 1, 0, 0);
> else
> @@ -451,7 +451,7 @@ static void vm_async_ops_err(int fd, bool destroy)
> if (i == N_BINDS / 8)
> __xe_vm_bind_assert(fd, vm, 0, 0, 0,
> addr + i * bo_size * 2,
> - bo_size, XE_VM_BIND_OP_UNMAP |
> + bo_size, XE_VM_BIND_OP_UNMAP,
> XE_VM_BIND_FLAG_ASYNC |
> INJECT_ERROR, &sync, 1, 0, 0);
> else
> @@ -465,7 +465,7 @@ static void vm_async_ops_err(int fd, bool destroy)
> if (i == N_BINDS / 8)
> __xe_vm_bind_assert(fd, vm, 0, bo, 0,
> addr + i * bo_size * 2,
> - bo_size, XE_VM_BIND_OP_MAP |
> + bo_size, XE_VM_BIND_OP_MAP,
> XE_VM_BIND_FLAG_ASYNC |
> INJECT_ERROR, &sync, 1, 0, 0);
> else
> @@ -479,7 +479,7 @@ static void vm_async_ops_err(int fd, bool destroy)
> if (i == N_BINDS / 8)
> __xe_vm_bind_assert(fd, vm, 0, 0, 0,
> addr + i * bo_size * 2,
> - bo_size, XE_VM_BIND_OP_UNMAP |
> + bo_size, XE_VM_BIND_OP_UNMAP,
> XE_VM_BIND_FLAG_ASYNC |
> INJECT_ERROR, &sync, 1, 0, 0);
> else
> @@ -928,7 +928,8 @@ test_bind_array(int fd, struct drm_xe_engine_class_instance *eci, int n_execs,
> bind_ops[i].range = bo_size;
> bind_ops[i].addr = addr;
> bind_ops[i].tile_mask = 0x1 << eci->gt_id;
> - bind_ops[i].op = XE_VM_BIND_OP_MAP | XE_VM_BIND_FLAG_ASYNC;
> + bind_ops[i].op = XE_VM_BIND_OP_MAP;
> + bind_ops[i].flags = XE_VM_BIND_FLAG_ASYNC;
> bind_ops[i].region = 0;
> bind_ops[i].reserved[0] = 0;
> bind_ops[i].reserved[1] = 0;
> @@ -972,7 +973,8 @@ test_bind_array(int fd, struct drm_xe_engine_class_instance *eci, int n_execs,
>
> for (i = 0; i < n_execs; ++i) {
> bind_ops[i].obj = 0;
> - bind_ops[i].op = XE_VM_BIND_OP_UNMAP | XE_VM_BIND_FLAG_ASYNC;
> + bind_ops[i].op = XE_VM_BIND_OP_UNMAP;
> + bind_ops[i].flags = XE_VM_BIND_FLAG_ASYNC;
> }
>
> syncobj_reset(fd, &sync[0].handle, 1);
> --
> 2.41.0
>
More information about the Intel-xe
mailing list