Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Matthew Brost <matthew.brost@intel.com>
To: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
Cc: <intel-xe@lists.freedesktop.org>, <shekhar.chauhan@intel.com>,
	<harish.chegondi@intel.com>
Subject: Re: [PATCH v8 4/8] drm/xe: Reject vm_ops->access on BO mappings of an unplugged device
Date: Thu, 19 Nov 2026 14:36:04 -0800	[thread overview]
Message-ID: <av951Pol7BED74zP@gsse-cloud1.jf.intel.com> (raw)
In-Reply-To: <20261119002737.3899877-14-umesh.nerlige.ramappa@intel.com>

On Wed, Nov 18, 2026 at 04:27:42PM -0800, Umesh Nerlige Ramappa wrote:
> A userspace mapping of a BO can outlive the device, e.g. across unbind
> or a PCI slot reset. After drm_dev_unplug() the fault handler only
> installs a dummy page, but ptrace, /proc/<pid>/mem and
> process_vm_readv() can't use get_user_pages() on these VM_PFNMAP vmas
> and fall back to vm_ops->access(), i.e. xe_bo_vm_access().
> 
> xe_bo_vm_access() doesn't check whether the device is still there. For
> a BO in VRAM, xe_ttm_access_memory() then either goes through the VRAM
> iomap, which is unmapped on remove, or submits a migrate job after the
> GT has been torn down. This applies to GEM mmaps and to the OA buffer
> mmap of an OA stream that outlives the device.
> 
> Enter a drm_dev_enter() section in xe_bo_vm_access() and fail with
> -ENODEV once the device is unplugged, like xe_bo_cpu_fault() does.
> 
> Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
> Cc: Matthew Brost <matthew.brost@intel.com>

I'd add a fixes tag as Sashiko suggests.

Patch looks correct though:
Reviewed-by: Matthew Brost <matthew.brost@intel.com>

> ---
>  drivers/gpu/drm/xe/xe_bo.c | 15 +++++++++++++--
>  1 file changed, 13 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/gpu/drm/xe/xe_bo.c b/drivers/gpu/drm/xe/xe_bo.c
> index 42a6aa1a2e4a..a74c38df2bdd 100644
> --- a/drivers/gpu/drm/xe/xe_bo.c
> +++ b/drivers/gpu/drm/xe/xe_bo.c
> @@ -2345,9 +2345,20 @@ static int xe_bo_vm_access(struct vm_area_struct *vma, unsigned long addr,
>  	struct ttm_buffer_object *ttm_bo = vma->vm_private_data;
>  	struct xe_bo *bo = ttm_to_xe_bo(ttm_bo);
>  	struct xe_device *xe = xe_bo_device(bo);
> +	int idx, ret;
>  
> -	guard(xe_pm_runtime)(xe);
> -	return ttm_bo_vm_access(vma, addr, buf, len, write);
> +	/*
> +	 * A mapping can outlive the device. Once it's unplugged, VRAM is unmapped and the
> +	 * migrate engines are gone, so refuse access, like the fault handler does.
> +	 */
> +	if (!drm_dev_enter(&xe->drm, &idx))
> +		return -ENODEV;
> +
> +	scoped_guard(xe_pm_runtime, xe)
> +		ret = ttm_bo_vm_access(vma, addr, buf, len, write);
> +
> +	drm_dev_exit(idx);
> +	return ret;
>  }
>  
>  /**
> -- 
> 2.55.0
> 

  parent reply	other threads:[~2026-11-19 22:36 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-19  0:27 [PATCH v8 0/8] Cleanup OA stream on PCI unbind/remove Umesh Nerlige Ramappa
2026-11-19  0:27 ` [PATCH v8 1/8] drm/xe/oa: Ignore a HW tail pointer outside the OA buffer Umesh Nerlige Ramappa
2026-11-19  0:39   ` sashiko-bot
2026-11-19  0:27 ` [PATCH v8 2/8] drm/xe/oa: Use the drm device address_space for the OA stream fd Umesh Nerlige Ramappa
2026-11-19  0:39   ` sashiko-bot
2026-11-19  0:27 ` [PATCH v8 3/8] drm/xe/oa: Wait for in-flight OA batches before stream HW teardown Umesh Nerlige Ramappa
2026-11-19  0:27 ` [PATCH v8 4/8] drm/xe: Reject vm_ops->access on BO mappings of an unplugged device Umesh Nerlige Ramappa
2026-11-19  0:39   ` sashiko-bot
2026-11-19 22:36   ` Matthew Brost [this message]
2026-11-19  0:27 ` [PATCH v8 5/8] drm/xe/oa: Fence OA ioctls and mmap against device unplug and wedging Umesh Nerlige Ramappa
2026-11-19  0:27 ` [PATCH v8 6/8] drm/xe/oa: Take the runtime PM reference in xe_oa_destroy_locked() Umesh Nerlige Ramappa
2026-11-19  0:27 ` [PATCH v8 7/8] drm/xe/oa: Free the OA buffer and the stream after the stream destroy Umesh Nerlige Ramappa
2026-11-19  0:48   ` sashiko-bot
2026-11-19  0:27 ` [PATCH v8 8/8] drm/xe/oa: Quiesce OA streams on driver unbind and PCI reset Umesh Nerlige Ramappa
2026-11-19  1:11   ` sashiko-bot
2026-11-19  6:05 ` ✓ CI.KUnit: success for Cleanup OA stream on PCI unbind/remove (rev8) Patchwork
2026-11-19  6:46 ` ✓ Xe.CI.BAT: " 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=av951Pol7BED74zP@gsse-cloud1.jf.intel.com \
    --to=matthew.brost@intel.com \
    --cc=harish.chegondi@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=shekhar.chauhan@intel.com \
    --cc=umesh.nerlige.ramappa@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