[PATCH v6 1/3] drm/i915/vma: Fix UAF on destroy against retire race

Mon Mar 4 15:52:33 UTC 2024

On 3/1/2024 8:29 AM, Janusz Krzysztofik wrote:
> Object debugging tools were sporadically reporting illegal attempts to
> free a still active i915 VMA object when parking a GT believed to be idle.
>
> [161.359441] ODEBUG: free active (active state 0) object: ffff88811643b958 object type: i915_active hint: __i915_vma_active+0x0/0x50 [i915]
> [161.360082] WARNING: CPU: 5 PID: 276 at lib/debugobjects.c:514 debug_print_object+0x80/0xb0
> ...
> [161.360304] CPU: 5 PID: 276 Comm: kworker/5:2 Not tainted 6.5.0-rc1-CI_DRM_13375-g003f860e5577+ #1
> [161.360314] Hardware name: Intel Corporation Rocket Lake Client Platform/RocketLake S UDIMM 6L RVP, BIOS RKLSFWI1.R00.3173.A03.2204210138 04/21/2022
> [161.360322] Workqueue: i915-unordered __intel_wakeref_put_work [i915]
> [161.360592] RIP: 0010:debug_print_object+0x80/0xb0
> ...
> [161.361347] debug_object_free+0xeb/0x110
> [161.361362] i915_active_fini+0x14/0x130 [i915]
> [161.361866] release_references+0xfe/0x1f0 [i915]
> [161.362543] i915_vma_parked+0x1db/0x380 [i915]
> [161.363129] __gt_park+0x121/0x230 [i915]
> [161.363515] ____intel_wakeref_put_last+0x1f/0x70 [i915]
>
> That has been tracked down to be happening when another thread is
> deactivating the VMA inside __active_retire() helper, after the VMA's
> active counter has been already decremented to 0, but before deactivation
> of the VMA's object is reported to the object debugging tool.
>
> We could prevent from that race by serializing i915_active_fini() with
> __active_retire() via ref->tree_lock, but that wouldn't stop the VMA from
> being used, e.g. from __i915_vma_retire() called at the end of
> __active_retire(), after that VMA has been already freed by a concurrent
> i915_vma_destroy() on return from the i915_active_fini().  Then, we should
> rather fix the issue at the VMA level, not in i915_active.
>
> Since __i915_vma_parked() is called from __gt_park() on last put of the
> GT's wakeref, the issue could be addressed by holding the GT wakeref long
> enough for __active_retire() to complete before that wakeref is released
> and the GT parked.
>
> I believe the issue was introduced by commit d93939730347 ("drm/i915:
> Remove the vma refcount") which moved a call to i915_active_fini() from
> a dropped i915_vma_release(), called on last put of the removed VMA kref,
> to i915_vma_parked() processing path called on last put of a GT wakeref.
> However, its visibility to the object debugging tool was suppressed by a
> bug in i915_active that was fixed two weeks later with commit e92eb246feb9
> ("drm/i915/active: Fix missing debug object activation").
>
> A VMA associated with a request doesn't acquire a GT wakeref by itself.
> Instead, it depends on a wakeref held directly by the request's active
> intel_context for a GT associated with its VM, and indirectly on that
> intel_context's engine wakeref if the engine belongs to the same GT as the
> VMA's VM.  Those wakerefs are released asynchronously to VMA deactivation.
>
> Fix the issue by getting a wakeref for the VMA's GT when activating it,
> and putting that wakeref only after the VMA is deactivated.  However,
> exclude global GTT from that processing path, otherwise the GPU never goes
> idle.  Since __i915_vma_retire() may be called from atomic contexts, use
> async variant of wakeref put.  Also, to avoid circular locking dependency,
> take care of acquiring the wakeref before VM mutex when both are needed.
>
> v6: Since __i915_vma_active/retire() callbacks are not serialized, storing
>      a wakeref tracking handle inside struct i915_vma is not safe, and
>      there is no other good place for that.  Use untracked variants of
>      intel_gt_pm_get/put_async().
> v5: Replace "tile" with "GT" across commit description (Rodrigo),
>    - avoid mentioning multi-GT case in commit description (Rodrigo),
>    - explain why we need to take a temporary wakeref unconditionally inside
>      i915_vma_pin_ww() (Rodrigo).
> v4: Refresh on top of commit 5e4e06e4087e ("drm/i915: Track gt pm
>      wakerefs") (Andi),
>    - for more easy backporting, split out removal of former insufficient
>      workarounds and move them to separate patches (Nirmoy).
>    - clean up commit message and description a bit.
> v3: Identify root cause more precisely, and a commit to blame,
>    - identify and drop former workarounds,
>    - update commit message and description.
> v2: Get the wakeref before VM mutex to avoid circular locking dependency,
>    - drop questionable Fixes: tag.
>
> Fixes: d93939730347 ("drm/i915: Remove the vma refcount")
> Closes: https://gitlab.freedesktop.org/drm/intel/issues/8875
> Signed-off-by: Janusz Krzysztofik <janusz.krzysztofik at linux.intel.com>
> Cc: Thomas Hellström <thomas.hellstrom at linux.intel.com>
> Cc: Nirmoy Das <nirmoy.das at intel.com>
> Cc: Andi Shyti <andi.shyti at linux.intel.com>
> Cc: Rodrigo Vivi <rodrigo.vivi at intel.com>
> Cc: stable at vger.kernel.org # v5.19+
> ---
>   drivers/gpu/drm/i915/i915_vma.c | 26 +++++++++++++++++++-------
>   1 file changed, 19 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
> index d09aad34ba37f..ffe81fe338f7e 100644
> --- a/drivers/gpu/drm/i915/i915_vma.c
> +++ b/drivers/gpu/drm/i915/i915_vma.c
> @@ -34,6 +34,7 @@
>   #include "gt/intel_engine.h"
>   #include "gt/intel_engine_heartbeat.h"
>   #include "gt/intel_gt.h"
> +#include "gt/intel_gt_pm.h"
>   #include "gt/intel_gt_requests.h"
>   #include "gt/intel_tlb.h"
>   
> @@ -103,12 +104,25 @@ static inline struct i915_vma *active_to_vma(struct i915_active *ref)
>   
>   static int __i915_vma_active(struct i915_active *ref)
>   {
> -	return i915_vma_tryget(active_to_vma(ref)) ? 0 : -ENOENT;
> +	struct i915_vma *vma = active_to_vma(ref);
> +
> +	if (!i915_vma_tryget(vma))
> +		return -ENOENT;
> +
> +	if (!i915_vma_is_ggtt(vma))
> +		intel_gt_pm_get_untracked(vma->vm->gt);
Please add the v6 comment here as well, otherwise this series is:

Reviewed-by: Nirmoy Das <nirmoy.das at intel.com>


> +
> +	return 0;
>   }
>   
>   static void __i915_vma_retire(struct i915_active *ref)
>   {
> -	i915_vma_put(active_to_vma(ref));
> +	struct i915_vma *vma = active_to_vma(ref);
> +
> +	if (!i915_vma_is_ggtt(vma))
> +		intel_gt_pm_put_async_untracked(vma->vm->gt);
> +
> +	i915_vma_put(vma);
>   }
>   
>   static struct i915_vma *
> @@ -1404,7 +1418,7 @@ int i915_vma_pin_ww(struct i915_vma *vma, struct i915_gem_ww_ctx *ww,
>   	struct i915_vma_work *work = NULL;
>   	struct dma_fence *moving = NULL;
>   	struct i915_vma_resource *vma_res = NULL;
> -	intel_wakeref_t wakeref = 0;
> +	intel_wakeref_t wakeref;
>   	unsigned int bound;
>   	int err;
>   
> @@ -1424,8 +1438,7 @@ int i915_vma_pin_ww(struct i915_vma *vma, struct i915_gem_ww_ctx *ww,
>   	if (err)
>   		return err;
>   
> -	if (flags & PIN_GLOBAL)
> -		wakeref = intel_runtime_pm_get(&vma->vm->i915->runtime_pm);
> +	wakeref = intel_runtime_pm_get(&vma->vm->i915->runtime_pm);
>   
>   	if (flags & vma->vm->bind_async_flags) {
>   		/* lock VM */
> @@ -1561,8 +1574,7 @@ int i915_vma_pin_ww(struct i915_vma *vma, struct i915_gem_ww_ctx *ww,
>   	if (work)
>   		dma_fence_work_commit_imm(&work->base);
>   err_rpm:
> -	if (wakeref)
> -		intel_runtime_pm_put(&vma->vm->i915->runtime_pm, wakeref);
> +	intel_runtime_pm_put(&vma->vm->i915->runtime_pm, wakeref);
>   
>   	if (moving)
>   		dma_fence_put(moving);