Question on UAPI for fences

2014-09-15 Thread Daniel Vetter
On Sun, Sep 14, 2014 at 12:36:43PM +0200, Christian K?nig wrote: > Yeah, right. Providing the fd to reassign to a fence would indeed reduce the > create/close overhead. > > But it would still be more overhead than for example a simple on demand > growing ring buffer which then uses 64bit sequence

Question on UAPI for fences

2014-09-14 Thread Christian König
Yeah, right. Providing the fd to reassign to a fence would indeed reduce the create/close overhead. But it would still be more overhead than for example a simple on demand growing ring buffer which then uses 64bit sequence numbers in userspace to refer to a fence in the kernel. Apart from

Question on UAPI for fences

2014-09-14 Thread Marek Olšák
BTW, we can recycle fences in userspace just like we recycle buffers. That should make the create/close overhead non-existent. Marek On Sat, Sep 13, 2014 at 2:25 PM, Christian K?nig wrote: >> Doing such combining and cleaning up fds as soon as they have been passed >> on should keep each

Question on UAPI for fences

2014-09-13 Thread Christian König
> Doing such combining and cleaning up fds as soon as they have been > passed on should keep each application's fd usage fairly small. Yeah, but this is exactly what we wanted to avoid internally because of the IOCTL overhead. And thinking more about it for our driver internal use we will

Question on UAPI for fences

2014-09-12 Thread Christian König
> As Daniel said using fd is most likely the way we want to do it but this > remains vague. Separating the discussion if it should be an fd or not. Using an fd sounds fine to me in general, but I have some concerns as well. For example what was the maximum number of opened FDs per process again?

Question on UAPI for fences

2014-09-12 Thread Christian König
Am 12.09.2014 um 17:48 schrieb Jerome Glisse: > On Fri, Sep 12, 2014 at 05:42:57PM +0200, Christian K?nig wrote: >> Am 12.09.2014 um 17:33 schrieb Jerome Glisse: >>> On Fri, Sep 12, 2014 at 11:25:12AM -0400, Alex Deucher wrote: On Fri, Sep 12, 2014 at 10:50 AM, Jerome Glisse wrote:

Question on UAPI for fences

2014-09-12 Thread Christian König
Am 12.09.2014 um 17:33 schrieb Jerome Glisse: > On Fri, Sep 12, 2014 at 11:25:12AM -0400, Alex Deucher wrote: >> On Fri, Sep 12, 2014 at 10:50 AM, Jerome Glisse >> wrote: >>> On Fri, Sep 12, 2014 at 04:43:44PM +0200, Daniel Vetter wrote: On Fri, Sep 12, 2014 at 4:09 PM, Daniel Vetter

Question on UAPI for fences

2014-09-12 Thread John Harrison
On Fri, Sep 12, 2014 at 05:58:09PM +0200, Christian K?nig wrote: > pass in a list of fences to wait for before beginning a command submission. The Android implementation has a mechanism for combining multiple sync points into a brand new single sync pt. Thus APIs only ever need to take in a

Question on UAPI for fences

2014-09-12 Thread Daniel Vetter
On Fri, Sep 12, 2014 at 10:50:49AM -0400, Jerome Glisse wrote: > On Fri, Sep 12, 2014 at 04:43:44PM +0200, Daniel Vetter wrote: > > On Fri, Sep 12, 2014 at 4:09 PM, Daniel Vetter wrote: > > > On Fri, Sep 12, 2014 at 03:23:22PM +0200, Christian K?nig wrote: > > >> Hello everyone, > > >> > > >> to

Question on UAPI for fences

2014-09-12 Thread Daniel Vetter
On Fri, Sep 12, 2014 at 4:09 PM, Daniel Vetter wrote: > On Fri, Sep 12, 2014 at 03:23:22PM +0200, Christian K?nig wrote: >> Hello everyone, >> >> to allow concurrent buffer access by different engines beyond the multiple >> readers/single writer model that we currently use in radeon and other >>

Question on UAPI for fences

2014-09-12 Thread Daniel Vetter
On Fri, Sep 12, 2014 at 03:23:22PM +0200, Christian K?nig wrote: > Hello everyone, > > to allow concurrent buffer access by different engines beyond the multiple > readers/single writer model that we currently use in radeon and other > drivers we need some kind of synchonization object exposed to

Question on UAPI for fences

2014-09-12 Thread Christian König
Hello everyone, to allow concurrent buffer access by different engines beyond the multiple readers/single writer model that we currently use in radeon and other drivers we need some kind of synchonization object exposed to userspace. My initial patch set for this used (or rather abused) zero

Question on UAPI for fences

2014-09-12 Thread Jerome Glisse
On Fri, Sep 12, 2014 at 05:58:09PM +0200, Christian K?nig wrote: > Am 12.09.2014 um 17:48 schrieb Jerome Glisse: > >On Fri, Sep 12, 2014 at 05:42:57PM +0200, Christian K?nig wrote: > >>Am 12.09.2014 um 17:33 schrieb Jerome Glisse: > >>>On Fri, Sep 12, 2014 at 11:25:12AM -0400, Alex Deucher wrote:

Question on UAPI for fences

2014-09-12 Thread Jerome Glisse
On Fri, Sep 12, 2014 at 05:42:57PM +0200, Christian K?nig wrote: > Am 12.09.2014 um 17:33 schrieb Jerome Glisse: > >On Fri, Sep 12, 2014 at 11:25:12AM -0400, Alex Deucher wrote: > >>On Fri, Sep 12, 2014 at 10:50 AM, Jerome Glisse > >>wrote: > >>>On Fri, Sep 12, 2014 at 04:43:44PM +0200, Daniel

Question on UAPI for fences

2014-09-12 Thread Alex Deucher
On Fri, Sep 12, 2014 at 11:33 AM, Jerome Glisse wrote: > On Fri, Sep 12, 2014 at 11:25:12AM -0400, Alex Deucher wrote: >> On Fri, Sep 12, 2014 at 10:50 AM, Jerome Glisse >> wrote: >> > On Fri, Sep 12, 2014 at 04:43:44PM +0200, Daniel Vetter wrote: >> >> On Fri, Sep 12, 2014 at 4:09 PM, Daniel

Question on UAPI for fences

2014-09-12 Thread Jerome Glisse
On Fri, Sep 12, 2014 at 11:25:12AM -0400, Alex Deucher wrote: > On Fri, Sep 12, 2014 at 10:50 AM, Jerome Glisse wrote: > > On Fri, Sep 12, 2014 at 04:43:44PM +0200, Daniel Vetter wrote: > >> On Fri, Sep 12, 2014 at 4:09 PM, Daniel Vetter wrote: > >> > On Fri, Sep 12, 2014 at 03:23:22PM +0200,

Question on UAPI for fences

2014-09-12 Thread Alex Deucher
On Fri, Sep 12, 2014 at 10:50 AM, Jerome Glisse wrote: > On Fri, Sep 12, 2014 at 04:43:44PM +0200, Daniel Vetter wrote: >> On Fri, Sep 12, 2014 at 4:09 PM, Daniel Vetter wrote: >> > On Fri, Sep 12, 2014 at 03:23:22PM +0200, Christian K?nig wrote: >> >> Hello everyone, >> >> >> >> to allow

Question on UAPI for fences

2014-09-12 Thread Jerome Glisse
On Fri, Sep 12, 2014 at 04:43:44PM +0200, Daniel Vetter wrote: > On Fri, Sep 12, 2014 at 4:09 PM, Daniel Vetter wrote: > > On Fri, Sep 12, 2014 at 03:23:22PM +0200, Christian K?nig wrote: > >> Hello everyone, > >> > >> to allow concurrent buffer access by different engines beyond the multiple >

Question on UAPI for fences

2014-09-12 Thread Jesse Barnes
On Fri, 12 Sep 2014 18:08:23 +0200 Christian K?nig wrote: > > As Daniel said using fd is most likely the way we want to do it but this > > remains vague. > Separating the discussion if it should be an fd or not. Using an fd > sounds fine to me in general, but I have some concerns as well. > >