Re: [PATCH 1/2] dma-buf: start caching of sg_table objects v2

2019-05-22 Thread Daniel Vetter
On Thu, May 16, 2019 at 08:26:36AM +0200, Daniel Vetter wrote: > On Wed, May 15, 2019 at 11:25 AM Christian König > wrote: > > > > Am 15.05.19 um 10:58 schrieb Daniel Vetter: > > > On Tue, May 14, 2019 at 04:05:14PM +0200, Christian König wrote: > > >> To allow a smooth transition from pinning

Re: [PATCH 1/2] dma-buf: start caching of sg_table objects v2

2019-05-16 Thread Daniel Vetter
On Wed, May 15, 2019 at 11:25 AM Christian König wrote: > > Am 15.05.19 um 10:58 schrieb Daniel Vetter: > > On Tue, May 14, 2019 at 04:05:14PM +0200, Christian König wrote: > >> To allow a smooth transition from pinning buffer objects to dynamic > >> invalidation we first start to cache the

Re: [PATCH 1/2] dma-buf: start caching of sg_table objects v2

2019-05-15 Thread Christian König
Am 15.05.19 um 10:58 schrieb Daniel Vetter: On Tue, May 14, 2019 at 04:05:14PM +0200, Christian König wrote: To allow a smooth transition from pinning buffer objects to dynamic invalidation we first start to cache the sg_table for an attachment. v2: keep closer to the DRM implementation

Re: [PATCH 1/2] dma-buf: start caching of sg_table objects v2

2019-05-15 Thread Daniel Vetter
On Tue, May 14, 2019 at 04:05:14PM +0200, Christian König wrote: > To allow a smooth transition from pinning buffer objects to dynamic > invalidation we first start to cache the sg_table for an attachment. > > v2: keep closer to the DRM implementation > > Signed-off-by: Christian König i915

[PATCH 1/2] dma-buf: start caching of sg_table objects v2

2019-05-14 Thread Christian König
To allow a smooth transition from pinning buffer objects to dynamic invalidation we first start to cache the sg_table for an attachment. v2: keep closer to the DRM implementation Signed-off-by: Christian König --- drivers/dma-buf/dma-buf.c | 27 +--