From: Weinan Li Date: Mon, 19 Jun 2017 00:49:17 +0000 (+0800) Subject: drm/i915/gvt: remove scheduler_mutex in per-engine workload_thread X-Git-Url: https://git.stricted.de/?a=commitdiff_plain;h=4cc74389a551dc95fce72d58c11e55a93b6ecd19;p=GitHub%2Fmoto-9609%2Fandroid_kernel_motorola_exynos9610.git drm/i915/gvt: remove scheduler_mutex in per-engine workload_thread For the vGPU workloads, now GVT-g use per vGPU scheduler, the per-ring work_thread only pick workload belongs to the current vGPU. And with time slice based scheduler, it waits all the engines become idle before do vGPU switch. So we can run free dispatch in per-ring work_thread, different ring running in different 'vGPU' won't happen. For the workloads between vGPU and Host, this scheduler_mutex can't block host to dispatch workload into other ring engines. Here remove this mutex since it impacts the performance when applications use more than 1 ring engines in 1 vgpu. ring0 running in vGPU1, ring1 running in Host. Will happen. ring0 running in vGPU1, ring1 running in vGPU2. Won't happen. Signed-off-by: Weinan Li Signed-off-by: Ping Gao Signed-off-by: Zhenyu Wang --- diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c index 488fdea348a9..5aeba13a5de4 100644 --- a/drivers/gpu/drm/i915/gvt/scheduler.c +++ b/drivers/gpu/drm/i915/gvt/scheduler.c @@ -464,8 +464,6 @@ struct workload_thread_param { int ring_id; }; -static DEFINE_MUTEX(scheduler_mutex); - static int workload_thread(void *priv) { struct workload_thread_param *p = (struct workload_thread_param *)priv; @@ -497,8 +495,6 @@ static int workload_thread(void *priv) if (!workload) break; - mutex_lock(&scheduler_mutex); - gvt_dbg_sched("ring id %d next workload %p vgpu %d\n", workload->ring_id, workload, workload->vgpu->id); @@ -537,9 +533,6 @@ complete: FORCEWAKE_ALL); intel_runtime_pm_put(gvt->dev_priv); - - mutex_unlock(&scheduler_mutex); - } return 0; }