From: "Ghimiray, Himal Prasad" <himal.prasad.ghimiray@intel.com>
To: "Thomas Hellström" <thomas.hellstrom@linux.intel.com>,
intel-xe@lists.freedesktop.org
Cc: Matthew Brost <matthew.brost@intel.com>,
"Falkowski, John" <john.falkowski@intel.com>,
"Mrozek, Michal" <michal.mrozek@intel.com>
Subject: Re: [PATCH 2/2] drm/xe/uapi: Hide the madvise autoreset behind a VM_BIND flag
Date: Tue, 14 Oct 2025 14:06:36 +0530 [thread overview]
Message-ID: <38422801-586a-49ff-b6f9-1f22ba3f5430@intel.com> (raw)
In-Reply-To: <20251014082756.6858-2-thomas.hellstrom@linux.intel.com>
On 14-10-2025 13:57, Thomas Hellström wrote:
> The madvise implementation currently resets the SVM madvise if the
> underlying CPU map is unmapped. This is in an attempt to mimic the
> CPU madvice behaviour. However, it's not clear that this is a desired
> behaviour since if the end app user relies on it for malloc()ed
> objects or stack objects, it may not work as intended.
>
> Instead of having the autoreset functionality being a direct
> application-facing implicit UAPI, make the UMD explicitly choose
> this behaviour if it wants to expose it by introducing
> DRM_XE_VM_BIND_FLAG_MADVISE_AUTORESET, and add a semantics
> description.
>
> Fixes: a2eb8aec3ebe ("drm/xe: Reset VMA attributes to default in SVM garbage collector")
> Cc: Matthew Brost <matthew.brost@intel.com>
> Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
> Cc: "Falkowski, John" <john.falkowski@intel.com>
> Cc: "Mrozek, Michal" <michal.mrozek@intel.com>
> Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
LGTM
Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com
> ---
> drivers/gpu/drm/xe/xe_svm.c | 5 +++++
> drivers/gpu/drm/xe/xe_vm.c | 12 +++++++++---
> drivers/gpu/drm/xe/xe_vm_types.h | 1 +
> include/uapi/drm/xe_drm.h | 15 +++++++++++++++
> 4 files changed, 30 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c
> index b268ee0d2271..f8773cbc2025 100644
> --- a/drivers/gpu/drm/xe/xe_svm.c
> +++ b/drivers/gpu/drm/xe/xe_svm.c
> @@ -302,6 +302,11 @@ static int xe_svm_range_set_default_attr(struct xe_vm *vm, u64 range_start, u64
> if (!vma)
> return -EINVAL;
>
> + if (!(vma->gpuva.flags & XE_VMA_MADV_AUTORESET)) {
> + drm_dbg(&vm->xe->drm, "Skipping madvise reset for vma.\n");
> + return 0;
> + }
> +
> if (xe_vma_has_default_mem_attrs(vma))
> return 0;
>
> diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
> index 9c29608aa3bc..d8a9d8090b37 100644
> --- a/drivers/gpu/drm/xe/xe_vm.c
> +++ b/drivers/gpu/drm/xe/xe_vm.c
> @@ -646,7 +646,8 @@ static void xe_vma_ops_incr_pt_update_ops(struct xe_vma_ops *vops, u8 tile_mask,
> XE_VMA_READ_ONLY | \
> XE_VMA_DUMPABLE | \
> XE_VMA_SYSTEM_ALLOCATOR | \
> - DRM_GPUVA_SPARSE)
> + DRM_GPUVA_SPARSE | \
> + XE_VMA_MADV_AUTORESET)
>
> static void xe_vm_populate_rebind(struct xe_vma_op *op, struct xe_vma *vma,
> u8 tile_mask)
> @@ -2296,6 +2297,8 @@ vm_bind_ioctl_ops_create(struct xe_vm *vm, struct xe_vma_ops *vops,
> op->map.vma_flags |= XE_VMA_SYSTEM_ALLOCATOR;
> if (flags & DRM_XE_VM_BIND_FLAG_DUMPABLE)
> op->map.vma_flags |= XE_VMA_DUMPABLE;
> + if (flags & DRM_XE_VM_BIND_FLAG_MADVISE_AUTORESET)
> + op->map.vma_flags |= XE_VMA_MADV_AUTORESET;
> op->map.pat_index = pat_index;
> op->map.invalidate_on_bind =
> __xe_vm_needs_clear_scratch_pages(vm, flags);
> @@ -3279,7 +3282,8 @@ ALLOW_ERROR_INJECTION(vm_bind_ioctl_ops_execute, ERRNO);
> DRM_XE_VM_BIND_FLAG_NULL | \
> DRM_XE_VM_BIND_FLAG_DUMPABLE | \
> DRM_XE_VM_BIND_FLAG_CHECK_PXP | \
> - DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR)
> + DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR | \
> + DRM_XE_VM_BIND_FLAG_MADVISE_AUTORESET)
>
> #ifdef TEST_VM_OPS_ERROR
> #define SUPPORTED_FLAGS (SUPPORTED_FLAGS_STUB | FORCE_OP_ERROR)
> @@ -3394,7 +3398,9 @@ static int vm_bind_ioctl_check_args(struct xe_device *xe, struct xe_vm *vm,
> XE_IOCTL_DBG(xe, (prefetch_region != DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC &&
> !(BIT(prefetch_region) & xe->info.mem_region_mask))) ||
> XE_IOCTL_DBG(xe, obj &&
> - op == DRM_XE_VM_BIND_OP_UNMAP)) {
> + op == DRM_XE_VM_BIND_OP_UNMAP) ||
> + XE_IOCTL_DBG(xe, (flags & DRM_XE_VM_BIND_FLAG_MADVISE_AUTORESET) &&
> + (!is_cpu_addr_mirror || op != DRM_XE_VM_BIND_OP_MAP))) {
> err = -EINVAL;
> goto free_bind_ops;
> }
> diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h
> index a3b422b27ae8..d6e2a0fdd4b3 100644
> --- a/drivers/gpu/drm/xe/xe_vm_types.h
> +++ b/drivers/gpu/drm/xe/xe_vm_types.h
> @@ -46,6 +46,7 @@ struct xe_vm_pgtable_update_op;
> #define XE_VMA_PTE_COMPACT (DRM_GPUVA_USERBITS << 7)
> #define XE_VMA_DUMPABLE (DRM_GPUVA_USERBITS << 8)
> #define XE_VMA_SYSTEM_ALLOCATOR (DRM_GPUVA_USERBITS << 9)
> +#define XE_VMA_MADV_AUTORESET (DRM_GPUVA_USERBITS << 10)
>
> /**
> * struct xe_vma_mem_attr - memory attributes associated with vma
> diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h
> index 40ff19f52a8d..98bc8858e4bd 100644
> --- a/include/uapi/drm/xe_drm.h
> +++ b/include/uapi/drm/xe_drm.h
> @@ -1013,6 +1013,20 @@ struct drm_xe_vm_destroy {
> * valid on VMs with DRM_XE_VM_CREATE_FLAG_FAULT_MODE set. The CPU address
> * mirror flag are only valid for DRM_XE_VM_BIND_OP_MAP operations, the BO
> * handle MBZ, and the BO offset MBZ.
> + * - %DRM_XE_VM_BIND_FLAG_MADVISE_AUTORESET - Can be used in combination with
> + * %DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR to reset madvises when the underlying
> + * CPU address space range is unmapped (typically with munmap(2) or brk(2)).
> + * The madvise values set with %DRM_IOCTL_XE_MADVISE are reset to the values
> + * that were present immediately after the %DRM_IOCTL_XE_VM_BIND.
> + * The reset GPU virtual address range is the intersection of the range bound
> + * using %DRM_IOCTL_XE_VM_BIND and the virtual CPU address space range
> + * unmapped.
> + * This functionality is present to mimic the behaviour of CPU address space
> + * madvises set using madvise(2), which are typically reset on unmap.
> + * Note: free(3) may or may not call munmap(2) and/or brk(2), and may thus
> + * not invoke autoreset. Neither will stack variables going out of scope.
> + * Therefore it's recommended to always explicitly reset the madvises when
> + * freeing the memory backing a region used in a %DRM_IOCTL_XE_MADVISE call.
> *
> * The @prefetch_mem_region_instance for %DRM_XE_VM_BIND_OP_PREFETCH can also be:
> * - %DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC, which ensures prefetching occurs in
> @@ -1119,6 +1133,7 @@ struct drm_xe_vm_bind_op {
> #define DRM_XE_VM_BIND_FLAG_DUMPABLE (1 << 3)
> #define DRM_XE_VM_BIND_FLAG_CHECK_PXP (1 << 4)
> #define DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR (1 << 5)
> +#define DRM_XE_VM_BIND_FLAG_MADVISE_AUTORESET (1 << 6)
> /** @flags: Bind flags */
> __u32 flags;
>
next prev parent reply other threads:[~2025-10-14 8:36 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-14 8:27 [PATCH 1/2] drm/xe: Retain vma flags when recreating and splitting vmas for madvise Thomas Hellström
2025-10-14 8:27 ` [PATCH 2/2] drm/xe/uapi: Hide the madvise autoreset behind a VM_BIND flag Thomas Hellström
2025-10-14 8:36 ` Ghimiray, Himal Prasad [this message]
2025-10-14 8:35 ` ✓ CI.KUnit: success for series starting with [1/2] drm/xe: Retain vma flags when recreating and splitting vmas for madvise Patchwork
2025-10-14 9:12 ` ✓ Xe.CI.BAT: " Patchwork
2025-10-14 17:39 ` ✓ Xe.CI.Full: " Patchwork
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=38422801-586a-49ff-b6f9-1f22ba3f5430@intel.com \
--to=himal.prasad.ghimiray@intel.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=john.falkowski@intel.com \
--cc=matthew.brost@intel.com \
--cc=michal.mrozek@intel.com \
--cc=thomas.hellstrom@linux.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