Re: [igt-dev] [PATCH i-g-t 05/21] wsim/media-bench: i915 balancing

[Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxxxxxxxx>]

On 10/05/2019 14:14, Chris Wilson wrote:
> Quoting Tvrtko Ursulin (2019-05-08 13:10:42)
> > From: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
> >
> > Support i915 virtual engine from gem_wsim (-b i915) and media-bench.pl
> >
> > Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@xxxxxxxxx>
> > ---
> > +       /*
> > +        * Create and configure contexts.
> > +        */
> > +       for (i = 0; i < wrk->nr_ctxs; i += 2) {
> > +               struct ctx *ctx = &wrk->ctx_list[i];
> > +               uint32_t ctx_id, share_vm = 0;
> >   
> > -                       wrk->ctx_list[w->context].id = arg.ctx_id;
> > +               if (ctx->id)
> > +                       continue;
> >   
> > -                       if (flags & GLOBAL_BALANCE) {
> > -                               wrk->ctx_list[w->context].static_vcs = context_vcs_rr;
> > -                               context_vcs_rr ^= 1;
> > -                       } else {
> > -                               wrk->ctx_list[w->context].static_vcs = ctx_vcs;
> > -                               ctx_vcs ^= 1;
> > -                       }
> > +               if (flags & I915) {
>
> vm sharing shouldn't be a i915-balancer only option. For single jobs split
> across multiple contexts, I would expect they will want to share vm.

Could do but I wanted to limit the new features to new features. :) 
Pencil in for later okay?

> > +                       struct drm_i915_gem_context_create_ext_setparam ext = {
> > +                               .base.name = I915_CONTEXT_CREATE_EXT_SETPARAM,
> > +                               .param.param = I915_CONTEXT_PARAM_VM,
> > +                       };
> > +                       struct drm_i915_gem_context_create_ext args = { };
> >   
> > -                       if (wrk->prio) {
> > +                       /* Find existing context to share ppgtt with. */
> > +                       for (j = 0; j < wrk->nr_ctxs; j++) {
> >                                  struct drm_i915_gem_context_param param = {
> > -                                       .ctx_id = arg.ctx_id,
> > -                                       .param = I915_CONTEXT_PARAM_PRIORITY,
> > -                                       .value = wrk->prio,
> > +                                       .param = I915_CONTEXT_PARAM_VM,
> >                                  };
> > -                               gem_context_set_param(fd, &param);
> > +
> > +                               if (!wrk->ctx_list[j].id)
> > +                                       continue;
> > +
> > +                               param.ctx_id = wrk->ctx_list[j].id;
> > +
> > +                               gem_context_get_param(fd, &param);
> > +                               igt_assert(param.value);
> > +
> > +                               share_vm = param.value;
> > +
> > +                               ext.param.value = share_vm;
> > +                               args.flags =
> > +                                   I915_CONTEXT_CREATE_FLAGS_USE_EXTENSIONS;
> > +                               args.extensions = to_user_pointer(&ext);
> > +                               break;
> >                          }
> > +
> > +                       if (!ctx->targets_instance)
> > +                               args.flags |=
> > +                                    I915_CONTEXT_CREATE_FLAGS_SINGLE_TIMELINE;
> > +
> > +                       drmIoctl(fd, DRM_IOCTL_I915_GEM_CONTEXT_CREATE_EXT,
> > +                                &args);
> > +
> > +                       ctx_id = args.ctx_id;
> > +               } else {
> > +                       struct drm_i915_gem_context_create args = {};
> > +
> > +                       drmIoctl(fd, DRM_IOCTL_I915_GEM_CONTEXT_CREATE, &args);
> > +                       ctx_id = args.ctx_id;
> > +               }
> > +
> > +               igt_assert(ctx_id);
> > +               ctx->id = ctx_id;
> > +
> > +               if (flags & GLOBAL_BALANCE) {
> > +                       ctx->static_vcs = context_vcs_rr;
> > +                       context_vcs_rr ^= 1;
> > +               } else {
> > +                       ctx->static_vcs = ctx_vcs;
> > +                       ctx_vcs ^= 1;
> > +               }
> > +
> > +               __ctx_set_prio(ctx_id, wrk->prio);
> > +
> > +               /*
> > +                * Do we need a separate context to satisfy this workloads which
> > +                * both want to target specific engines and be balanced by i915?
> > +                */
> > +               if ((flags & I915) && ctx->wants_balance &&
> > +                   ctx->targets_instance) {
> > +                       struct drm_i915_gem_context_create_ext_setparam ext = {
> > +                               .base.name = I915_CONTEXT_CREATE_EXT_SETPARAM,
> > +                               .param.param = I915_CONTEXT_PARAM_VM,
> > +                               .param.value = share_vm,
> > +                       };
> > +                       struct drm_i915_gem_context_create_ext args = {
> > +                               .extensions = to_user_pointer(&ext),
> > +                               .flags =
> > +                                   I915_CONTEXT_CREATE_FLAGS_USE_EXTENSIONS |
> > +                                   I915_CONTEXT_CREATE_FLAGS_SINGLE_TIMELINE,
> > +                       };
> > +
> > +                       igt_assert(share_vm);
> > +
> > +                       drmIoctl(fd, DRM_IOCTL_I915_GEM_CONTEXT_CREATE_EXT,
> > +                                &args);
> > +
> > +                       igt_assert(args.ctx_id);
> > +                       ctx_id = args.ctx_id;
> > +                       wrk->ctx_list[i + 1].id = args.ctx_id;
> > +
> > +                       __ctx_set_prio(ctx_id, wrk->prio);
> > +               }
> > +
> > +               if (ctx->wants_balance) {
> > +                       I915_DEFINE_CONTEXT_ENGINES_LOAD_BALANCE(load_balance, 2) = {
> > +                               .base.name = I915_CONTEXT_ENGINES_EXT_LOAD_BALANCE,
> > +                               .num_siblings = 2,
> > +                               .engines = {
> > +                                       { .engine_class = I915_ENGINE_CLASS_VIDEO,
> > +                                         .engine_instance = 0 },
> > +                                       { .engine_class = I915_ENGINE_CLASS_VIDEO,
> > +                                         .engine_instance = 1 },
> > +                               },
> > +                       };
> > +                       I915_DEFINE_CONTEXT_PARAM_ENGINES(set_engines, 3) = {
> > +                               .extensions = to_user_pointer(&load_balance),
> > +                               .engines = {
> > +                                       { .engine_class = I915_ENGINE_CLASS_INVALID,
> > +                                         .engine_instance = I915_ENGINE_CLASS_INVALID_NONE },
> > +                                       { .engine_class = I915_ENGINE_CLASS_VIDEO,
> > +                                         .engine_instance = 0 },
> > +                                       { .engine_class = I915_ENGINE_CLASS_VIDEO,
> > +                                         .engine_instance = 1 },
> > +                               },
> > +                       };
> > +
> > +                       struct drm_i915_gem_context_param param = {
> > +                               .ctx_id = ctx_id,
> > +                               .param = I915_CONTEXT_PARAM_ENGINES,
> > +                               .size = sizeof(set_engines),
> > +                               .value = to_user_pointer(&set_engines),
> > +                       };
> > +
> > +                       gem_context_set_param(fd, &param);
> >                  }
>
> if (share_vm)
> 	gem_vm_destroy(share_vm);
>
> Just to drop the local handle as the context has acquired its own
> reference.

Well spotted!

> Other than that, it does what it sets out to do: create a context with
> choice of engines and load balancing amongst them.
>
> Reviewed-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>

Thanks,

Tvrtko

_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx