| Message ID | 20230523225336.2642008-1-olvaffe@gmail.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [v2] amdgpu: validate drm_amdgpu_gem_va addrs | expand |
Am 24.05.23 um 00:53 schrieb Chia-I Wu: > Validate drm_amdgpu_gem_va addrs in amdgpu_gem_va_ioctl. > amdgpu_vm_bo_replace_map no longer needs to validate (and its > validations were insufficient either). amdgpu_vm_bo_map has internal > users and its validations are kept. No, please keep all validation inside amdgpu_vm.c code. See the offset is unused or might have a different meaning for some use cases, so validating it here is actually not correct. Christian. > > This is motivated by OOB access in amdgpu_vm_update_range when > offset_in_bo+map_size overflows. > > Userspace (radeonsi and radv) seems fine as well. > > v2: keep the validations in amdgpu_vm_bo_map > > Fixes: 9f7eb5367d00 ("drm/amdgpu: actually use the VM map parameters") > Signed-off-by: Chia-I Wu <olvaffe@gmail.com> > --- > drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 15 +++++++++++++++ > drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 8 +------- > 2 files changed, 16 insertions(+), 7 deletions(-) > > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c > index d8e683688daab..36d5adfdf0f69 100644 > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c > @@ -681,6 +681,21 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, > uint64_t vm_size; > int r = 0; > > + if (args->va_address & ~PAGE_MASK || args->offset_in_bo & ~PAGE_MASK || > + args->map_size & ~PAGE_MASK) { > + dev_dbg(dev->dev, "unaligned va_address 0x%LX, offset_in_bo 0x%LX, or map_size 0x%LX\n", > + args->va_address, args->offset_in_bo, args->map_size); > + return -EINVAL; > + } > + > + if (args->map_size == 0 || > + args->va_address + args->map_size < args->va_address || > + args->offset_in_bo + args->map_size < args->offset_in_bo) { > + dev_dbg(dev->dev, "invalid map_size 0x%LX (va_address 0x%LX, offset_in_bo 0x%LX)\n", > + args->map_size, args->va_address, args->offset_in_bo); > + return -EINVAL; > + } > + > if (args->va_address < AMDGPU_VA_RESERVED_SIZE) { > dev_dbg(dev->dev, > "va_address 0x%LX is in reserved area 0x%LX\n", > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c > index b9441ab457ea7..6307baaa136cf 100644 > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c > @@ -1501,15 +1501,9 @@ int amdgpu_vm_bo_replace_map(struct amdgpu_device *adev, > uint64_t eaddr; > int r; > > - /* validate the parameters */ > - if (saddr & ~PAGE_MASK || offset & ~PAGE_MASK || > - size == 0 || size & ~PAGE_MASK) > - return -EINVAL; > - > /* make sure object fit at this offset */ > eaddr = saddr + size - 1; > - if (saddr >= eaddr || > - (bo && offset + size > amdgpu_bo_size(bo)) || > + if ((bo && offset + size > amdgpu_bo_size(bo)) || > (eaddr >= adev->vm_manager.max_pfn << AMDGPU_GPU_PAGE_SHIFT)) > return -EINVAL; >
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index d8e683688daab..36d5adfdf0f69 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -681,6 +681,21 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, uint64_t vm_size; int r = 0; + if (args->va_address & ~PAGE_MASK || args->offset_in_bo & ~PAGE_MASK || + args->map_size & ~PAGE_MASK) { + dev_dbg(dev->dev, "unaligned va_address 0x%LX, offset_in_bo 0x%LX, or map_size 0x%LX\n", + args->va_address, args->offset_in_bo, args->map_size); + return -EINVAL; + } + + if (args->map_size == 0 || + args->va_address + args->map_size < args->va_address || + args->offset_in_bo + args->map_size < args->offset_in_bo) { + dev_dbg(dev->dev, "invalid map_size 0x%LX (va_address 0x%LX, offset_in_bo 0x%LX)\n", + args->map_size, args->va_address, args->offset_in_bo); + return -EINVAL; + } + if (args->va_address < AMDGPU_VA_RESERVED_SIZE) { dev_dbg(dev->dev, "va_address 0x%LX is in reserved area 0x%LX\n", diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index b9441ab457ea7..6307baaa136cf 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -1501,15 +1501,9 @@ int amdgpu_vm_bo_replace_map(struct amdgpu_device *adev, uint64_t eaddr; int r; - /* validate the parameters */ - if (saddr & ~PAGE_MASK || offset & ~PAGE_MASK || - size == 0 || size & ~PAGE_MASK) - return -EINVAL; - /* make sure object fit at this offset */ eaddr = saddr + size - 1; - if (saddr >= eaddr || - (bo && offset + size > amdgpu_bo_size(bo)) || + if ((bo && offset + size > amdgpu_bo_size(bo)) || (eaddr >= adev->vm_manager.max_pfn << AMDGPU_GPU_PAGE_SHIFT)) return -EINVAL;
Validate drm_amdgpu_gem_va addrs in amdgpu_gem_va_ioctl. amdgpu_vm_bo_replace_map no longer needs to validate (and its validations were insufficient either). amdgpu_vm_bo_map has internal users and its validations are kept. This is motivated by OOB access in amdgpu_vm_update_range when offset_in_bo+map_size overflows. Userspace (radeonsi and radv) seems fine as well. v2: keep the validations in amdgpu_vm_bo_map Fixes: 9f7eb5367d00 ("drm/amdgpu: actually use the VM map parameters") Signed-off-by: Chia-I Wu <olvaffe@gmail.com> --- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 15 +++++++++++++++ drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 8 +------- 2 files changed, 16 insertions(+), 7 deletions(-)