Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Nirmoy Das <nirmoy.das@linux.intel.com>
To: Nirmoy Das <nirmoy.das@intel.com>, intel-xe@lists.freedesktop.org
Cc: lionel Landwerlin <lionel.g.landwerlin@intel.com>,
	Matt Roper <matthew.d.roper@intel.com>
Subject: Re: [PATCH v3 2/7] drm/xe: Consolidate setting PTE_AE into one place
Date: Tue, 16 Apr 2024 16:33:27 +0200	[thread overview]
Message-ID: <cea6d4cc-fb44-46c8-a790-4d1e590b7745@linux.intel.com> (raw)
In-Reply-To: <20240415145214.25641-3-nirmoy.das@intel.com>


On 4/15/2024 4:52 PM, Nirmoy Das wrote:
> Currently decision to set PTE_AE is spread between xe_pt
> and xe_vm files and there is no reason to be keep it that
> way. Consolidate the logic for better maintainability.
>
> Atomics is not expected on userptr memory so this patch
> also making sure PTE_AE is only applied when a buffer object
> exist.

Synced with Lionel on this and the expectation is not correct for mesa. KMD

have to allow device atomics if UMD ask for it using the VM bind flag on 
userptr buffers too. I will correct this patch.


Regards,

Nirmoy

>
> Signed-off-by: Nirmoy Das <nirmoy.das@intel.com>
> ---
>   drivers/gpu/drm/xe/xe_pt.c | 4 +---
>   drivers/gpu/drm/xe/xe_vm.c | 7 ++++---
>   2 files changed, 5 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c
> index 5b7930f46cf3..7dc13a8bb44f 100644
> --- a/drivers/gpu/drm/xe/xe_pt.c
> +++ b/drivers/gpu/drm/xe/xe_pt.c
> @@ -597,7 +597,6 @@ static int
>   xe_pt_stage_bind(struct xe_tile *tile, struct xe_vma *vma,
>   		 struct xe_vm_pgtable_update *entries, u32 *num_entries)
>   {
> -	struct xe_device *xe = tile_to_xe(tile);
>   	struct xe_bo *bo = xe_vma_bo(vma);
>   	bool is_devmem = !xe_vma_is_userptr(vma) && bo &&
>   		(xe_bo_is_vram(bo) || xe_bo_is_stolen_devmem(bo));
> @@ -619,8 +618,7 @@ xe_pt_stage_bind(struct xe_tile *tile, struct xe_vma *vma,
>   	struct xe_pt *pt = xe_vma_vm(vma)->pt_root[tile->id];
>   	int ret;
>   
> -	if ((vma->gpuva.flags & XE_VMA_ATOMIC_PTE_BIT) &&
> -	    (is_devmem || !IS_DGFX(xe)))
> +	if (vma->gpuva.flags & XE_VMA_ATOMIC_PTE_BIT)
>   		xe_walk.default_pte |= XE_USM_PPGTT_PTE_AE;
>   
>   	if (is_devmem) {
> diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
> index 2dbba55e7785..b1dcaa35b6cc 100644
> --- a/drivers/gpu/drm/xe/xe_vm.c
> +++ b/drivers/gpu/drm/xe/xe_vm.c
> @@ -806,9 +806,6 @@ static struct xe_vma *xe_vma_create(struct xe_vm *vm,
>   	for_each_tile(tile, vm->xe, id)
>   		vma->tile_mask |= 0x1 << id;
>   
> -	if (GRAPHICS_VER(vm->xe) >= 20 || vm->xe->info.platform == XE_PVC)
> -		vma->gpuva.flags |= XE_VMA_ATOMIC_PTE_BIT;
> -
>   	vma->pat_index = pat_index;
>   
>   	if (bo) {
> @@ -816,6 +813,10 @@ static struct xe_vma *xe_vma_create(struct xe_vm *vm,
>   
>   		xe_bo_assert_held(bo);
>   
> +		if (vm->xe->info.has_atomic_enable_pte_bit &&
> +		    (xe_bo_is_vram(bo) || !IS_DGFX(vm->xe)))
> +			vma->gpuva.flags |= XE_VMA_ATOMIC_PTE_BIT;
> +
>   		vm_bo = drm_gpuvm_bo_obtain(vma->gpuva.vm, &bo->ttm.base);
>   		if (IS_ERR(vm_bo)) {
>   			xe_vma_free(vma);

  reply	other threads:[~2024-04-16 14:33 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-15 14:52 [PATCH v3 0/7] Enable device atomics with a VM bind flag Nirmoy Das
2024-04-15 14:52 ` [PATCH v3 1/7] drm/xe: Introduce has_atomic_enable_pte_bit device info Nirmoy Das
2024-04-19 16:06   ` Zeng, Oak
2024-04-15 14:52 ` [PATCH v3 2/7] drm/xe: Consolidate setting PTE_AE into one place Nirmoy Das
2024-04-16 14:33   ` Nirmoy Das [this message]
2024-04-19 18:35   ` Zeng, Oak
2024-04-22  8:18     ` Nirmoy Das
2024-04-15 14:52 ` [PATCH v3 3/7] drm/xe: Add function to check if BO has single placement Nirmoy Das
2024-04-15 14:52 ` [PATCH v3 4/7] drm/xe: Move vm bind bo validation to a helper function Nirmoy Das
2024-04-16  0:55   ` Matthew Brost
2024-04-16 13:32     ` Nirmoy Das
2024-04-19 20:14   ` Zeng, Oak
2024-04-15 14:52 ` [PATCH v3 5/7] drm/xe: Introduce has_device_atomics_on_smem device info Nirmoy Das
2024-04-19 20:24   ` Zeng, Oak
2024-04-15 14:52 ` [PATCH v3 6/7] drm/xe/uapi: Introduce VMA bind flag for device atomics Nirmoy Das
2024-04-19  7:16   ` Lionel Landwerlin
2024-04-22  8:39     ` Nirmoy Das
2024-04-19 21:04   ` Zeng, Oak
2024-04-22 10:12     ` Nirmoy Das
2024-04-22 21:39       ` Zeng, Oak
2024-04-23 12:33         ` Nirmoy Das
2024-04-15 14:52 ` [PATCH v3 7/7] drm/xe/uapi: Add a query flag for has_device_atomics_on_smem Nirmoy Das
2024-04-19  7:08   ` Lionel Landwerlin
2024-04-22  8:53     ` Nirmoy Das
2024-04-19 21:06   ` Zeng, Oak
2024-04-15 21:19 ` ✓ CI.Patch_applied: success for Enable device atomics with a VM bind flag (rev3) Patchwork
2024-04-15 21:19 ` ✓ CI.checkpatch: " Patchwork
2024-04-15 21:21 ` ✓ CI.KUnit: " Patchwork
2024-04-15 21:37 ` ✓ CI.Build: " Patchwork
2024-04-15 21:40 ` ✓ CI.Hooks: " Patchwork
2024-04-15 21:41 ` ✓ CI.checksparse: " Patchwork
2024-04-15 22:22 ` ✗ CI.BAT: failure " Patchwork
2024-04-16 13:46 ` ✓ CI.FULL: success " Patchwork
2024-04-19  7:17 ` [PATCH v3 0/7] Enable device atomics with a VM bind flag Lionel Landwerlin
2024-04-22 10:13   ` Nirmoy Das
2024-04-22 14:50 ` Souza, Jose

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=cea6d4cc-fb44-46c8-a790-4d1e590b7745@linux.intel.com \
    --to=nirmoy.das@linux.intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=lionel.g.landwerlin@intel.com \
    --cc=matthew.d.roper@intel.com \
    --cc=nirmoy.das@intel.com \
    /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