Instead of allocating a new request when allocating a context, use the
request that initiated the allocation to emit the context
initialisation. This serves two purposes, it makes the initialisation
atomic with first use (simplifying scheduling and our own error
handling). Secondly, it enables us t
Instead of allocating a new request when allocating a context, use the
request that initiated the allocation to emit the context
initialisation. This serves two purposes, it makes the initialisation
atomic with first use (simplifying scheduling and our own error
handling). Secondly, it enables us t
On to, 2016-04-21 at 08:56 +0100, Chris Wilson wrote:
> On Thu, Apr 21, 2016 at 10:47:30AM +0300, Joonas Lahtinen wrote:
> >
> > On to, 2016-04-21 at 08:08 +0100, Chris Wilson wrote:
> > >
> > > On Thu, Apr 21, 2016 at 09:57:03AM +0300, Joonas Lahtinen wrote:
> > > >
> > > >
> > > > On ke, 2016
On Thu, Apr 21, 2016 at 10:47:30AM +0300, Joonas Lahtinen wrote:
> On to, 2016-04-21 at 08:08 +0100, Chris Wilson wrote:
> > On Thu, Apr 21, 2016 at 09:57:03AM +0300, Joonas Lahtinen wrote:
> > >
> > > On ke, 2016-04-20 at 19:42 +0100, Chris Wilson wrote:
> > > >
> > > > + if (!request->ctx
On to, 2016-04-21 at 08:08 +0100, Chris Wilson wrote:
> On Thu, Apr 21, 2016 at 09:57:03AM +0300, Joonas Lahtinen wrote:
> >
> > On ke, 2016-04-20 at 19:42 +0100, Chris Wilson wrote:
> > >
> > > + if (!request->ctx->engine[engine->id].initialised) {
> > > + ret = engine->init_context(requ
On Thu, Apr 21, 2016 at 09:57:03AM +0300, Joonas Lahtinen wrote:
> On ke, 2016-04-20 at 19:42 +0100, Chris Wilson wrote:
> > + if (!request->ctx->engine[engine->id].initialised) {
> > + ret = engine->init_context(request);
> > + if (ret) {
> > + intel_lr_cont
On ke, 2016-04-20 at 19:42 +0100, Chris Wilson wrote:
> Instead of allocating a new request when allocating a context, use the
> request that initiated the allocation to emit the context
> initialisation. This serves two purposes, it makes the initialisation
> atomic with first use (simplifying sch
Instead of allocating a new request when allocating a context, use the
request that initiated the allocation to emit the context
initialisation. This serves two purposes, it makes the initialisation
atomic with first use (simplifying scheduling and our own error
handling). Secondly, it enables us t
