Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Simona Vetter <simona.vetter@ffwll.ch>
To: Liviu Dudau <liviu.dudau@arm.com>
Cc: "Simona Vetter" <simona.vetter@ffwll.ch>,
	"DRI Development" <dri-devel@lists.freedesktop.org>,
	"Intel Xe Development" <intel-xe@lists.freedesktop.org>,
	"Adrián Larumbe" <adrian.larumbe@collabora.com>,
	"Boris Brezillon" <boris.brezillon@collabora.com>,
	"Steven Price" <steven.price@arm.com>,
	"Simona Vetter" <simona.vetter@intel.com>
Subject: Re: [PATCH 2/2] drm/panthor: Fix UAF in panthor_gem_create_with_handle() debugfs code
Date: Wed, 9 Jul 2025 15:17:29 +0200	[thread overview]
Message-ID: <aG5r6UIAzZ6PI1kU@phenom.ffwll.local> (raw)
In-Reply-To: <aGzjA0bunrqwf6Pp@e110455-lin.cambridge.arm.com>

On Tue, Jul 08, 2025 at 10:21:07AM +0100, Liviu Dudau wrote:
> Hi Simona,
> 
> On Mon, Jul 07, 2025 at 05:18:14PM +0200, Simona Vetter wrote:
> > The object is potentially already gone after the drm_gem_object_put().
> > In general the object should be fully constructed before calling
> > drm_gem_handle_create(), except the debugfs tracking uses a separate
> > lock and list and separate flag to denotate whether the object is
> > actually initilized.
> > 
> > Since I'm touching this all anyway simplify this by only adding the
> > object to the debugfs when it's ready for that, which allows us to
> > delete that separate flag. panthor_gem_debugfs_bo_rm() already checks
> > whether we've actually been added to the list or this is some error
> > path cleanup.
> 
> Thanks for the cleanup, just minor nits.
> 
> > 
> > v2: Fix build issues for !CONFIG_DEBUGFS (Adrián)
> > 
> > Fixes: a3707f53eb3f ("drm/panthor: show device-wide list of DRM GEM objects over DebugFS")
> > Cc: Adrián Larumbe <adrian.larumbe@collabora.com>
> > Cc: Boris Brezillon <boris.brezillon@collabora.com>
> > Cc: Steven Price <steven.price@arm.com>
> > Cc: Liviu Dudau <liviu.dudau@arm.com>
> > Signed-off-by: Simona Vetter <simona.vetter@intel.com>
> > Signed-off-by: Simona Vetter <simona.vetter@ffwll.ch>
> > ---
> >  drivers/gpu/drm/panthor/panthor_gem.c | 34 ++++++++++++++-------------
> >  drivers/gpu/drm/panthor/panthor_gem.h |  3 ---
> >  2 files changed, 18 insertions(+), 19 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/panthor/panthor_gem.c b/drivers/gpu/drm/panthor/panthor_gem.c
> > index 7c00fd77758b..8232f91f3bf7 100644
> > --- a/drivers/gpu/drm/panthor/panthor_gem.c
> > +++ b/drivers/gpu/drm/panthor/panthor_gem.c
> > @@ -16,10 +16,14 @@
> >  #include "panthor_mmu.h"
> >  
> >  #ifdef CONFIG_DEBUG_FS
> > -static void panthor_gem_debugfs_bo_add(struct panthor_device *ptdev,
> > -				       struct panthor_gem_object *bo)
> > +static void panthor_gem_debugfs_bo_init(struct panthor_gem_object *bo)
> >  {
> >  	INIT_LIST_HEAD(&bo->debugfs.node);
> > +}
> 
> Would be nice to have an empty line here separating the functions.
> 
> > +static void panthor_gem_debugfs_bo_add(struct panthor_gem_object *bo)
> > +{
> > +	struct panthor_device *ptdev = container_of(bo->base.base.dev,
> > +						    struct panthor_device, base);
> >  
> >  	bo->debugfs.creator.tgid = current->group_leader->pid;
> >  	get_task_comm(bo->debugfs.creator.process_name, current->group_leader);
> > @@ -44,14 +48,13 @@ static void panthor_gem_debugfs_bo_rm(struct panthor_gem_object *bo)
> >  
> >  static void panthor_gem_debugfs_set_usage_flags(struct panthor_gem_object *bo, u32 usage_flags)
> >  {
> > -	bo->debugfs.flags = usage_flags | PANTHOR_DEBUGFS_GEM_USAGE_FLAG_INITIALIZED;
> > +	bo->debugfs.flags = usage_flags;
> > +	panthor_gem_debugfs_bo_add(bo);
> >  }
> >  #else
> > -static void panthor_gem_debugfs_bo_add(struct panthor_device *ptdev,
> > -				       struct panthor_gem_object *bo)
> > -{}
> >  static void panthor_gem_debugfs_bo_rm(struct panthor_gem_object *bo) {}
> >  static void panthor_gem_debugfs_set_usage_flags(struct panthor_gem_object *bo, u32 usage_flags) {}
> > +static void panthor_gem_debugfs_bo_init(struct panthor_gem_object *bo) {}
> >  #endif
> >  
> >  static void panthor_gem_free_object(struct drm_gem_object *obj)
> > @@ -246,7 +249,7 @@ struct drm_gem_object *panthor_gem_create_object(struct drm_device *ddev, size_t
> >  	drm_gem_gpuva_set_lock(&obj->base.base, &obj->gpuva_list_lock);
> >  	mutex_init(&obj->label.lock);
> >  
> > -	panthor_gem_debugfs_bo_add(ptdev, obj);
> > +	panthor_gem_debugfs_bo_init(obj);
> >  
> >  	return &obj->base.base;
> >  }
> > @@ -285,6 +288,12 @@ panthor_gem_create_with_handle(struct drm_file *file,
> >  		bo->base.base.resv = bo->exclusive_vm_root_gem->resv;
> >  	}
> >  
> > +	/*
> > +	 * No explicit flags are needed in the call below, since the
> > +	 * function internally sets the INITIALIZED bit for us.
> > +	 */
> 
> This comment is no longer needed (or not in this form).
> 
> > +	panthor_gem_debugfs_set_usage_flags(bo, 0);
> > +
> >  	/*
> >  	 * Allocate an id of idr table where the obj is registered
> >  	 * and handle has the id what user can see.
> > @@ -296,12 +305,6 @@ panthor_gem_create_with_handle(struct drm_file *file,
> >  	/* drop reference from allocate - handle holds it now. */
> >  	drm_gem_object_put(&shmem->base);
> >  
> > -	/*
> > -	 * No explicit flags are needed in the call below, since the
> > -	 * function internally sets the INITIALIZED bit for us.
> > -	 */
> > -	panthor_gem_debugfs_set_usage_flags(bo, 0);
> > -
> >  	return ret;
> >  }
> >  
> > @@ -387,7 +390,7 @@ static void panthor_gem_debugfs_bo_print(struct panthor_gem_object *bo,
> >  	unsigned int refcount = kref_read(&bo->base.base.refcount);
> >  	char creator_info[32] = {};
> >  	size_t resident_size;
> > -	u32 gem_usage_flags = bo->debugfs.flags & (u32)~PANTHOR_DEBUGFS_GEM_USAGE_FLAG_INITIALIZED;
> > +	u32 gem_usage_flags = bo->debugfs.flags;
> >  	u32 gem_state_flags = 0;
> >  
> >  	/* Skip BOs being destroyed. */
> > @@ -436,8 +439,7 @@ void panthor_gem_debugfs_print_bos(struct panthor_device *ptdev,
> >  
> >  	scoped_guard(mutex, &ptdev->gems.lock) {
> >  		list_for_each_entry(bo, &ptdev->gems.node, debugfs.node) {
> > -			if (bo->debugfs.flags & PANTHOR_DEBUGFS_GEM_USAGE_FLAG_INITIALIZED)
> > -				panthor_gem_debugfs_bo_print(bo, m, &totals);
> > +			panthor_gem_debugfs_bo_print(bo, m, &totals);
> >  		}
> >  	}
> >  
> > diff --git a/drivers/gpu/drm/panthor/panthor_gem.h b/drivers/gpu/drm/panthor/panthor_gem.h
> > index 4dd732dcd59f..8fc7215e9b90 100644
> > --- a/drivers/gpu/drm/panthor/panthor_gem.h
> > +++ b/drivers/gpu/drm/panthor/panthor_gem.h
> > @@ -35,9 +35,6 @@ enum panthor_debugfs_gem_usage_flags {
> >  
> >  	/** @PANTHOR_DEBUGFS_GEM_USAGE_FLAG_FW_MAPPED: BO is mapped on the FW VM. */
> >  	PANTHOR_DEBUGFS_GEM_USAGE_FLAG_FW_MAPPED = BIT(PANTHOR_DEBUGFS_GEM_USAGE_FW_MAPPED_BIT),
> > -
> > -	/** @PANTHOR_DEBUGFS_GEM_USAGE_FLAG_INITIALIZED: BO is ready for DebugFS display. */
> > -	PANTHOR_DEBUGFS_GEM_USAGE_FLAG_INITIALIZED = BIT(31),
> >  };
> >  
> >  /**
> > -- 
> > 2.49.0
> >
> 
> Otherwise, patch looks good to me!
> 
> Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>

Thanks for the review, will resend.
-Sima

> 
> Best regards,
> Liviu
> 
> 
> -- 
> ====================
> | I would like to |
> | fix the world,  |
> | but they're not |
> | giving me the   |
>  \ source code!  /
>   ---------------
>     ¯\_(ツ)_/¯

-- 
Simona Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch

  reply	other threads:[~2025-07-09 13:17 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-07-07 15:18 [PATCH 1/2] drm/gem: Fix race in drm_gem_handle_create_tail() Simona Vetter
2025-07-07 15:18 ` [PATCH 2/2] drm/panthor: Fix UAF in panthor_gem_create_with_handle() debugfs code Simona Vetter
2025-07-08  9:21   ` Liviu Dudau
2025-07-09 13:17     ` Simona Vetter [this message]
2025-07-09 13:52   ` [PATCH] " Simona Vetter
2025-07-09 15:48     ` Steven Price
2025-07-10  8:53       ` Simona Vetter
2025-07-10  9:23         ` Steven Price
2025-07-07 16:03 ` ✗ CI.checkpatch: warning for series starting with [1/2] drm/gem: Fix race in drm_gem_handle_create_tail() Patchwork
2025-07-07 16:04 ` ✓ CI.KUnit: success " Patchwork
2025-07-07 16:19 ` ✗ CI.checksparse: warning " Patchwork
2025-07-07 16:58 ` ✓ Xe.CI.BAT: success " Patchwork
2025-07-07 19:19 ` ✓ Xe.CI.Full: " Patchwork
2025-07-09 13:54 ` [PATCH 1/2] " Simona Vetter
2025-07-09 13:57 ` ✗ CI.checkpatch: warning for series starting with [1/2] drm/gem: Fix race in drm_gem_handle_create_tail() (rev2) Patchwork
2025-07-09 13:58 ` ✓ CI.KUnit: success " Patchwork
2025-07-09 14:13 ` ✗ CI.checksparse: warning " Patchwork
2025-07-09 14:38 ` ✓ Xe.CI.BAT: success " Patchwork
2025-07-09 17:11 ` ✗ Xe.CI.Full: failure " 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=aG5r6UIAzZ6PI1kU@phenom.ffwll.local \
    --to=simona.vetter@ffwll.ch \
    --cc=adrian.larumbe@collabora.com \
    --cc=boris.brezillon@collabora.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=liviu.dudau@arm.com \
    --cc=simona.vetter@intel.com \
    --cc=steven.price@arm.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