Re: [PATCH v2 3/3] drivers: platform: goldfish: goldfish_sync: add a driver

2019-01-08 Thread Greg KH
On Tue, Jan 08, 2019 at 04:48:02PM -0800, Roman Kiryanov wrote: > > Also, why is this driver needed at all? Why can't you use the "normal" > > drm sync api interface? Why write a custom ioctl driver just for this > > one kernel interface? > > This driver allows us to talk to host's sync api. Th

Re: [PATCH v2 3/3] drivers: platform: goldfish: goldfish_sync: add a driver

2019-01-08 Thread Roman Kiryanov
> You have a number of WARN_ON() calls in here that are not needed, please > remove. Thank you for noticing, I will remove them. > Also, why is this driver needed at all? Why can't you use the "normal" > drm sync api interface? Why write a custom ioctl driver just for this > one kernel interfac

Re: [PATCH v2 3/3] drivers: platform: goldfish: goldfish_sync: add a driver

2019-01-08 Thread Greg KH
On Mon, Jan 07, 2019 at 10:50:39AM -0800, r...@google.com wrote: > From: Roman Kiryanov > > The Goldfish sync driver is designed to provide a interface > between the underlying host's sync device and the kernel's > fence sync framework. > > Signed-off-by: Roman Kiryanov > --- > Changes in v2: >

[PATCH v2 3/3] drivers: platform: goldfish: goldfish_sync: add a driver

2019-01-07 Thread rkir
From: Roman Kiryanov The Goldfish sync driver is designed to provide a interface between the underlying host's sync device and the kernel's fence sync framework. Signed-off-by: Roman Kiryanov --- Changes in v2: - Added a missing include (mod_devicetable.h). - Put in one batch with goldfish_ad