From: Matthew Brost <matthew.brost@intel.com>
To: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Cc: <intel-xe@lists.freedesktop.org>
Subject: Re: [PATCH] drm/xe: Remove the pat_index field from xe_vma_op_map
Date: Tue, 14 May 2024 20:20:26 +0000 [thread overview]
Message-ID: <ZkPHiqwNPbqwzdUG@DUT025-TGLU.fm.intel.com> (raw)
In-Reply-To: <20240514110341.3581829-1-himal.prasad.ghimiray@intel.com>
On Tue, May 14, 2024 at 04:33:41PM +0530, Himal Prasad Ghimiray wrote:
> The pat_index field within xe_vma_op_map facilitates the creation of new
> VMAs for DRM_GPUVA_OP_MAP as a GPU baseva operation. Rather than
> initializing this field in vm_bind_ioctl_ops_create and subsequently
> utilizing it in vm_bind_ioctl_ops_parse, eliminate pat_index from
> xe_vma_op_map altogether. Instead, directly pass the user-provided
> pat_index to vm_bind_ioctl_ops_parse.
>
The patch doesn't seem to change any behaviors so can you elborate on
reasoning? I'm hesitant to accept changes to this code path aside from
bug fixes as I have a fairly large rework going on in this area [1] [2].
Matt
[1] https://patchwork.freedesktop.org/series/133034/
[2] https://patchwork.freedesktop.org/series/125608/
> Cc: Matthew Brost <matthew.brost@intel.com>
> Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
> ---
> drivers/gpu/drm/xe/xe_vm.c | 16 +++++++---------
> drivers/gpu/drm/xe/xe_vm_types.h | 2 --
> 2 files changed, 7 insertions(+), 11 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
> index c5b1694b292f..a11685e9546b 100644
> --- a/drivers/gpu/drm/xe/xe_vm.c
> +++ b/drivers/gpu/drm/xe/xe_vm.c
> @@ -2055,8 +2055,7 @@ static void print_op(struct xe_device *xe, struct drm_gpuva_op *op)
> static struct drm_gpuva_ops *
> vm_bind_ioctl_ops_create(struct xe_vm *vm, struct xe_bo *bo,
> u64 bo_offset_or_userptr, u64 addr, u64 range,
> - u32 operation, u32 flags,
> - u32 prefetch_region, u16 pat_index)
> + u32 operation, u32 flags, u32 prefetch_region)
> {
> struct drm_gem_object *obj = bo ? &bo->ttm.base : NULL;
> struct drm_gpuva_ops *ops;
> @@ -2117,7 +2116,6 @@ vm_bind_ioctl_ops_create(struct xe_vm *vm, struct xe_bo *bo,
> flags & DRM_XE_VM_BIND_FLAG_READONLY;
> op->map.is_null = flags & DRM_XE_VM_BIND_FLAG_NULL;
> op->map.dumpable = flags & DRM_XE_VM_BIND_FLAG_DUMPABLE;
> - op->map.pat_index = pat_index;
> } else if (__op->op == DRM_GPUVA_OP_PREFETCH) {
> op->prefetch.region = prefetch_region;
> }
> @@ -2278,7 +2276,7 @@ static int xe_vma_op_commit(struct xe_vm *vm, struct xe_vma_op *op)
>
>
> static int vm_bind_ioctl_ops_parse(struct xe_vm *vm, struct xe_exec_queue *q,
> - struct drm_gpuva_ops *ops,
> + struct drm_gpuva_ops *ops, u16 pat_index,
> struct xe_sync_entry *syncs, u32 num_syncs,
> struct xe_vma_ops *vops, bool last)
> {
> @@ -2322,8 +2320,7 @@ static int vm_bind_ioctl_ops_parse(struct xe_vm *vm, struct xe_exec_queue *q,
> flags |= op->map.dumpable ?
> VMA_CREATE_FLAG_DUMPABLE : 0;
>
> - vma = new_vma(vm, &op->base.map, op->map.pat_index,
> - flags);
> + vma = new_vma(vm, &op->base.map, pat_index, flags);
> if (IS_ERR(vma))
> return PTR_ERR(vma);
>
> @@ -3242,15 +3239,16 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
>
> ops[i] = vm_bind_ioctl_ops_create(vm, bos[i], obj_offset,
> addr, range, op, flags,
> - prefetch_region, pat_index);
> + prefetch_region);
> if (IS_ERR(ops[i])) {
> err = PTR_ERR(ops[i]);
> ops[i] = NULL;
> goto unwind_ops;
> }
>
> - err = vm_bind_ioctl_ops_parse(vm, q, ops[i], syncs, num_syncs,
> - &vops, i == args->num_binds - 1);
> + err = vm_bind_ioctl_ops_parse(vm, q, ops[i], pat_index,
> + syncs, num_syncs, &vops,
> + i == args->num_binds - 1);
> if (err)
> goto unwind_ops;
> }
> diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h
> index ce1a63a5e3e7..e13400dd5ee8 100644
> --- a/drivers/gpu/drm/xe/xe_vm_types.h
> +++ b/drivers/gpu/drm/xe/xe_vm_types.h
> @@ -284,8 +284,6 @@ struct xe_vma_op_map {
> bool is_null;
> /** @dumpable: whether BO is dumped on GPU hang */
> bool dumpable;
> - /** @pat_index: The pat index to use for this operation. */
> - u16 pat_index;
> };
>
> /** struct xe_vma_op_remap - VMA remap operation */
> --
> 2.25.1
>
prev parent reply other threads:[~2024-05-14 20:21 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-05-14 11:03 [PATCH] drm/xe: Remove the pat_index field from xe_vma_op_map Himal Prasad Ghimiray
2024-05-14 10:54 ` ✓ CI.Patch_applied: success for " Patchwork
2024-05-14 10:54 ` ✓ CI.checkpatch: " Patchwork
2024-05-14 10:55 ` ✓ CI.KUnit: " Patchwork
2024-05-14 11:07 ` ✓ CI.Build: " Patchwork
2024-05-14 11:09 ` ✓ CI.Hooks: " Patchwork
2024-05-14 11:11 ` ✓ CI.checksparse: " Patchwork
2024-05-14 11:43 ` ✓ CI.BAT: " Patchwork
2024-05-14 13:42 ` ✓ CI.FULL: " Patchwork
2024-05-14 20:20 ` Matthew Brost [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ZkPHiqwNPbqwzdUG@DUT025-TGLU.fm.intel.com \
--to=matthew.brost@intel.com \
--cc=himal.prasad.ghimiray@intel.com \
--cc=intel-xe@lists.freedesktop.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox