| Message ID | 20190712080314.21018-1-chris@chris-wilson.co.uk (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [1/2] dma-buf: Expand reservation_list to fill allocation | expand |
On 2019-07-12 10:03 a.m., Chris Wilson wrote: > Since kmalloc() will round up the allocation to the next slab size or > page, it will normally return a pointer to a memory block bigger than we > asked for. We can query for the actual size of the allocated block using > ksize() and expand our variable size reservation_list to take advantage > of that extra space. > > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> > Cc: Christian König <christian.koenig@amd.com> > Cc: Michel Dänzer <michel.daenzer@amd.com> > --- > drivers/dma-buf/reservation.c | 6 ++++-- > 1 file changed, 4 insertions(+), 2 deletions(-) > > diff --git a/drivers/dma-buf/reservation.c b/drivers/dma-buf/reservation.c > index a6ac2b3a0185..80ecc1283d15 100644 > --- a/drivers/dma-buf/reservation.c > +++ b/drivers/dma-buf/reservation.c > @@ -153,7 +153,9 @@ int reservation_object_reserve_shared(struct reservation_object *obj, > RCU_INIT_POINTER(new->shared[j++], fence); > } > new->shared_count = j; > - new->shared_max = max; > + new->shared_max = > + (ksize(new) - offsetof(typeof(*new), shared)) / > + sizeof(*new->shared); > > preempt_disable(); > write_seqcount_begin(&obj->seq); > @@ -169,7 +171,7 @@ int reservation_object_reserve_shared(struct reservation_object *obj, > return 0; > > /* Drop the references to the signaled fences */ > - for (i = k; i < new->shared_max; ++i) { > + for (i = k; i < max; ++i) { > struct dma_fence *fence; > > fence = rcu_dereference_protected(new->shared[i], > Nice, TIL about ksize(), wonder where else that could be used. Reviewed-by: Michel Dänzer <michel.daenzer@amd.com> P.S. According to scripts/get_maintainer.pl , this series should be sent to more recipients for review.
Am 12.07.19 um 10:03 schrieb Chris Wilson: > Since kmalloc() will round up the allocation to the next slab size or > page, it will normally return a pointer to a memory block bigger than we > asked for. We can query for the actual size of the allocated block using > ksize() and expand our variable size reservation_list to take advantage > of that extra space. > > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> > Cc: Christian König <christian.koenig@amd.com> > Cc: Michel Dänzer <michel.daenzer@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> BTW: I was wondering if we shouldn't replace the reservation_object_list with a dma_fence_chain. That would costs us a bit more memory and is slightly slower on querying the fence in the container. But it would be much faster on adding new fences and massively simplifies waiting or returning all fences currently in the container. Christian. > --- > drivers/dma-buf/reservation.c | 6 ++++-- > 1 file changed, 4 insertions(+), 2 deletions(-) > > diff --git a/drivers/dma-buf/reservation.c b/drivers/dma-buf/reservation.c > index a6ac2b3a0185..80ecc1283d15 100644 > --- a/drivers/dma-buf/reservation.c > +++ b/drivers/dma-buf/reservation.c > @@ -153,7 +153,9 @@ int reservation_object_reserve_shared(struct reservation_object *obj, > RCU_INIT_POINTER(new->shared[j++], fence); > } > new->shared_count = j; > - new->shared_max = max; > + new->shared_max = > + (ksize(new) - offsetof(typeof(*new), shared)) / > + sizeof(*new->shared); > > preempt_disable(); > write_seqcount_begin(&obj->seq); > @@ -169,7 +171,7 @@ int reservation_object_reserve_shared(struct reservation_object *obj, > return 0; > > /* Drop the references to the signaled fences */ > - for (i = k; i < new->shared_max; ++i) { > + for (i = k; i < max; ++i) { > struct dma_fence *fence; > > fence = rcu_dereference_protected(new->shared[i],
Quoting Koenig, Christian (2019-07-14 08:37:47) > Am 12.07.19 um 10:03 schrieb Chris Wilson: > > Since kmalloc() will round up the allocation to the next slab size or > > page, it will normally return a pointer to a memory block bigger than we > > asked for. We can query for the actual size of the allocated block using > > ksize() and expand our variable size reservation_list to take advantage > > of that extra space. > > > > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> > > Cc: Christian König <christian.koenig@amd.com> > > Cc: Michel Dänzer <michel.daenzer@amd.com> > > Reviewed-by: Christian König <christian.koenig@amd.com> > > BTW: I was wondering if we shouldn't replace the reservation_object_list > with a dma_fence_chain. I thought the dma_fence_chain tracked points (naturally ordered) along a singe timeline, whereas the reservation list tracked parallel timelines. Seems like a semantic mismatch? (Making lookup slower would not be pleasant, tbh, both waiting on and updating are an issue with the severe amount of reservation_objects we currently process per execbuf.) -Chris
Quoting Chris Wilson (2019-07-12 09:03:13) > Since kmalloc() will round up the allocation to the next slab size or > page, it will normally return a pointer to a memory block bigger than we > asked for. We can query for the actual size of the allocated block using > ksize() and expand our variable size reservation_list to take advantage > of that extra space. > > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> > Cc: Christian König <christian.koenig@amd.com> > Cc: Michel Dänzer <michel.daenzer@amd.com> Pushed to drm-misc-next, thanks for the reviews! Anyone feel brave enough for the second? :) -Chris
diff --git a/drivers/dma-buf/reservation.c b/drivers/dma-buf/reservation.c index a6ac2b3a0185..80ecc1283d15 100644 --- a/drivers/dma-buf/reservation.c +++ b/drivers/dma-buf/reservation.c @@ -153,7 +153,9 @@ int reservation_object_reserve_shared(struct reservation_object *obj, RCU_INIT_POINTER(new->shared[j++], fence); } new->shared_count = j; - new->shared_max = max; + new->shared_max = + (ksize(new) - offsetof(typeof(*new), shared)) / + sizeof(*new->shared); preempt_disable(); write_seqcount_begin(&obj->seq); @@ -169,7 +171,7 @@ int reservation_object_reserve_shared(struct reservation_object *obj, return 0; /* Drop the references to the signaled fences */ - for (i = k; i < new->shared_max; ++i) { + for (i = k; i < max; ++i) { struct dma_fence *fence; fence = rcu_dereference_protected(new->shared[i],
Since kmalloc() will round up the allocation to the next slab size or page, it will normally return a pointer to a memory block bigger than we asked for. We can query for the actual size of the allocated block using ksize() and expand our variable size reservation_list to take advantage of that extra space. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Christian König <christian.koenig@amd.com> Cc: Michel Dänzer <michel.daenzer@amd.com> --- drivers/dma-buf/reservation.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)