On Mon, Oct 13, 2025 9:27 AM Matthew Brost wrote: > On Tue, Oct 14, 2025 at 01:11:33AM +0900, Simon Richter wrote: > > It is possible for a BO to exist that is not currently associated with > > a resource, e.g. because it has been evicted. > > > > When devcoredump tries to read the contents of all BOs for dumping, we > > need to expect this as well -- in this case, ENODATA is recorded > > instead of the buffer contents. > > > > Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/6271 > > I think we need a fixes / cc stable but I can add that for you when merging. > > Anyways patch LGTM: > Reviewed-by: Matthew Brost <[email protected]> > > > Signed-off-by: Simon Richter <[email protected]>
It seems this patch is not merged yet. LGTM. Reviewed-by: Shuicheng Lin <[email protected]> > > --- > > drivers/gpu/drm/ttm/ttm_bo_vm.c | 7 +++++++ > > 1 file changed, 7 insertions(+) > > > > diff --git a/drivers/gpu/drm/ttm/ttm_bo_vm.c > > b/drivers/gpu/drm/ttm/ttm_bo_vm.c index b47020fca199..a101ff95b234 > > 100644 > > --- a/drivers/gpu/drm/ttm/ttm_bo_vm.c > > +++ b/drivers/gpu/drm/ttm/ttm_bo_vm.c > > @@ -434,6 +434,11 @@ int ttm_bo_access(struct ttm_buffer_object *bo, > unsigned long offset, > > if (ret) > > return ret; > > > > + if (!bo->resource) { > > + ret = -ENODATA; > > + goto unlock; > > + } > > + > > switch (bo->resource->mem_type) { > > case TTM_PL_SYSTEM: > > fallthrough; > > @@ -448,6 +453,7 @@ int ttm_bo_access(struct ttm_buffer_object *bo, > unsigned long offset, > > ret = -EIO; > > } > > > > +unlock: > > ttm_bo_unreserve(bo); > > > > return ret; > > -- > > 2.47.3 > >
