Re: [PATCH v2] drm/amdgpu: check a user-provided number of BOs in list

[Alex Deucher]

On Wed, Apr 23, 2025 at 10:29 AM Christian König
<christian.koenig@xxxxxxx> wrote:
>
> On 4/22/25 18:26, Deucher, Alexander wrote:
> > [Public]
> >
> >> -----Original Message-----
> >> From: Alex Deucher <alexdeucher@xxxxxxxxx>
> >> Sent: Tuesday, April 22, 2025 9:46 AM
> >> To: Koenig, Christian <Christian.Koenig@xxxxxxx>
> >> Cc: Denis Arefev <arefev@xxxxxxxxx>; Deucher, Alexander
> >> <Alexander.Deucher@xxxxxxx>; David Airlie <airlied@xxxxxxxxx>; Simona Vetter
> >> <simona@xxxxxxxx>; Andrey Grodzovsky <andrey.grodzovsky@xxxxxxx>;
> >> Chunming Zhou <david1.zhou@xxxxxxx>; amd-gfx@xxxxxxxxxxxxxxxxxxxxx; dri-
> >> devel@xxxxxxxxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; lvc-
> >> project@xxxxxxxxxxxxxxxx; stable@xxxxxxxxxxxxxxx
> >> Subject: Re: [PATCH v2] drm/amdgpu: check a user-provided number of BOs in list
> >>
> >> Applied.  Thanks!
> >
> > This change beaks the following IGT tests:
> >
> > igt@amdgpu/amd_vcn@vcn-decoder-create-decode-destroy@vcn-decoder-create
> > igt@amdgpu/amd_vcn@vcn-decoder-create-decode-destroy@vcn-decoder-decode
> > igt@amdgpu/amd_vcn@vcn-decoder-create-decode-destroy@vcn-decoder-destroy
> > igt@amdgpu/amd_jpeg_dec@amdgpu_cs_jpeg_decode
> > igt@amdgpu/amd_cs_nop@cs-nops-with-nop-compute0@cs-nop-with-nop-compute0
> > igt@amdgpu/amd_cs_nop@cs-nops-with-sync-compute0@cs-nop-with-sync-compute0
> > igt@amdgpu/amd_cs_nop@cs-nops-with-fork-compute0@cs-nop-with-fork-compute0
> > igt@amdgpu/amd_cs_nop@cs-nops-with-sync-fork-compute0@cs-nop-with-sync-fork-compute0
> > igt@amdgpu/amd_basic@userptr-with-ip-dma@userptr
> > igt@amdgpu/amd_basic@cs-compute-with-ip-compute@cs-compute
> > igt@amdgpu/amd_basic@cs-sdma-with-ip-dma@cs-sdma
> > igt@amdgpu/amd_basic@eviction-test-with-ip-dma@eviction_test
> > igt@amdgpu/amd_cp_dma_misc@gtt_to_vram-amdgpu_hw_ip_compute0
> > igt@amdgpu/amd_cp_dma_misc@vram_to_gtt-amdgpu_hw_ip_compute0
> > igt@amdgpu/amd_cp_dma_misc@vram_to_vram-amdgpu_hw_ip_compute0
>
>
> Could it be that we used BO list with zero entries for those?

Yes.  Dropping the 0 check fixed them.  E.g.,

+       if (in->bo_number > USHRT_MAX)
+               return -EINVAL;

Alex

>
> Christian.
>
> >
> > Alex
> >
> >>
> >> On Tue, Apr 22, 2025 at 5:13 AM Koenig, Christian <Christian.Koenig@xxxxxxx>
> >> wrote:
> >>>
> >>> [AMD Official Use Only - AMD Internal Distribution Only]
> >>>
> >>> Reviewed-by: Christian König <christian.koenig@xxxxxxx>
> >>>
> >>> ________________________________________
> >>> Von: Denis Arefev <arefev@xxxxxxxxx>
> >>> Gesendet: Freitag, 18. April 2025 10:31
> >>> An: Deucher, Alexander
> >>> Cc: Koenig, Christian; David Airlie; Simona Vetter; Andrey Grodzovsky;
> >>> Chunming Zhou; amd-gfx@xxxxxxxxxxxxxxxxxxxxx;
> >>> dri-devel@xxxxxxxxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx;
> >>> lvc-project@xxxxxxxxxxxxxxxx; stable@xxxxxxxxxxxxxxx
> >>> Betreff: [PATCH v2] drm/amdgpu: check a user-provided number of BOs in
> >>> list
> >>>
> >>> The user can set any value to the variable ‘bo_number’, via the ioctl
> >>> command DRM_IOCTL_AMDGPU_BO_LIST. This will affect the arithmetic
> >>> expression ‘in->bo_number * in->bo_info_size’, which is prone to
> >>> overflow. Add a valid value check.
> >>>
> >>> Found by Linux Verification Center (linuxtesting.org) with SVACE.
> >>>
> >>> Fixes: 964d0fbf6301 ("drm/amdgpu: Allow to create BO lists in CS ioctl
> >>> v3")
> >>> Cc: stable@xxxxxxxxxxxxxxx
> >>> Signed-off-by: Denis Arefev <arefev@xxxxxxxxx>
> >>> ---
> >>> V1 -> V2:
> >>> Set a reasonable limit 'USHRT_MAX' for 'bo_number' it as Christian
> >>> König <christian.koenig@xxxxxxx> suggested
> >>>
> >>>  drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c | 3 +++
> >>>  1 file changed, 3 insertions(+)
> >>>
> >>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
> >>> b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
> >>> index 702f6610d024..85f7ee1e085d 100644
> >>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
> >>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
> >>> @@ -189,6 +189,9 @@ int amdgpu_bo_create_list_entry_array(struct
> >> drm_amdgpu_bo_list_in *in,
> >>>         struct drm_amdgpu_bo_list_entry *info;
> >>>         int r;
> >>>
> >>> +       if (!in->bo_number || in->bo_number > USHRT_MAX)
> >>> +               return -EINVAL;
> >>> +
> >>>         info = kvmalloc_array(in->bo_number, info_size, GFP_KERNEL);
> >>>         if (!info)
> >>>                 return -ENOMEM;
> >>> --
> >>> 2.43.0
> >>>
>