Am 25.11.21 um 10:31 schrieb Maarten Lankhorst:
[SNIP]
diff --git a/include/linux/dma-resv.h b/include/linux/dma-resv.h index eebf04325b34..e0558429a5ee 100644 --- a/include/linux/dma-resv.h +++ b/include/linux/dma-resv.h @@ -458,6 +458,7 @@ void dma_resv_fini(struct dma_resv *obj); int dma_resv_reserve_shared(struct dma_resv *obj, unsigned int num_fences); void dma_resv_add_shared_fence(struct dma_resv *obj, struct dma_fence *fence); void dma_resv_add_excl_fence(struct dma_resv *obj, struct dma_fence *fence); +void dma_resv_prune_unlocked(struct dma_resv *obj); int dma_resv_get_fences(struct dma_resv *obj, struct dma_fence **pfence_excl, unsigned *pshared_count, struct dma_fence ***pshared); int dma_resv_copy_fences(struct dma_resv *dst, struct dma_resv *src);
I don't mind adding a dma_resv_prune for locked case, but I don't think unlocked would have benefits.
Furthermore, I'm trying to remove the unlocked versions from i915. Could this be a prereq patch instead?
https://patchwork.freedesktop.org/patch/460722/?series=96115&rev=1
Yeah, that works for me as well.
I was on the edge of dropping that from TTM as well since this is really just abusing the interface to save a few bytes of memory.
Feel free to add an Acked-by: Christian König christian.koenig@amd.com to the i915 patch if it helps to get that committed.
Regards, Christian.
~Maarten
~Maarten