From patchwork Mon Apr 9 11:14:13 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Chris Wilson X-Patchwork-Id: 10331123 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id E7E0B6022E for ; Mon, 9 Apr 2018 11:15:32 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id DB5E128AC6 for ; Mon, 9 Apr 2018 11:15:32 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id CE47A28AFB; Mon, 9 Apr 2018 11:15:32 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-5.2 required=2.0 tests=BAYES_00, MAILING_LIST_MULTI, RCVD_IN_DNSWL_MED autolearn=ham version=3.3.1 Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher DHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.wl.linuxfoundation.org (Postfix) with ESMTPS id 4FFD828AC6 for ; Mon, 9 Apr 2018 11:15:32 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id D1FB06E1BA; Mon, 9 Apr 2018 11:15:31 +0000 (UTC) X-Original-To: intel-gfx@lists.freedesktop.org Delivered-To: intel-gfx@lists.freedesktop.org Received: from fireflyinternet.com (mail.fireflyinternet.com [109.228.58.192]) by gabe.freedesktop.org (Postfix) with ESMTPS id 4C2C189199 for ; Mon, 9 Apr 2018 11:15:28 +0000 (UTC) X-Default-Received-SPF: pass (skip=forwardok (res=PASS)) x-ip-name=78.156.65.138; Received: from haswell.alporthouse.com (unverified [78.156.65.138]) by fireflyinternet.com (Firefly Internet (M1)) with ESMTP id 11308945-1500050 for multiple; Mon, 09 Apr 2018 12:15:18 +0100 Received: by haswell.alporthouse.com (sSMTP sendmail emulation); Mon, 09 Apr 2018 12:15:18 +0100 From: Chris Wilson To: intel-gfx@lists.freedesktop.org Date: Mon, 9 Apr 2018 12:14:13 +0100 Message-Id: <20180409111413.6352-19-chris@chris-wilson.co.uk> X-Mailer: git-send-email 2.17.0 In-Reply-To: <20180409111413.6352-1-chris@chris-wilson.co.uk> References: <20180409111413.6352-1-chris@chris-wilson.co.uk> X-Originating-IP: 78.156.65.138 X-Country: code=GB country="United Kingdom" ip=78.156.65.138 Subject: [Intel-gfx] [PATCH 18/18] drm/i915: Allow user control over preempt timeout on their important context X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , MIME-Version: 1.0 Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" X-Virus-Scanned: ClamAV using ClamSMTP One usecase would be to couple in via EGL_NV_context_priority_realtime in userspace to provide some QoS guarantees in conjunction with setting the highest priority. Signed-off-by: Chris Wilson --- drivers/gpu/drm/i915/i915_gem_context.c | 22 ++++++ drivers/gpu/drm/i915/i915_gem_context.h | 13 ++++ drivers/gpu/drm/i915/i915_request.c | 8 +- drivers/gpu/drm/i915/intel_lrc.c | 2 +- drivers/gpu/drm/i915/selftests/intel_lrc.c | 85 ++++++++++++++++++++++ include/uapi/drm/i915_drm.h | 12 +++ 6 files changed, 139 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c index 5cfac0255758..932ca1082b26 100644 --- a/drivers/gpu/drm/i915/i915_gem_context.c +++ b/drivers/gpu/drm/i915/i915_gem_context.c @@ -749,6 +749,15 @@ int i915_gem_context_getparam_ioctl(struct drm_device *dev, void *data, case I915_CONTEXT_PARAM_PRIORITY: args->value = ctx->priority; break; + case I915_CONTEXT_PARAM_PREEMPT_TIMEOUT: + if (!(to_i915(dev)->caps.scheduler & I915_SCHEDULER_CAP_PREEMPTION)) + ret = -ENODEV; + else if (args->size) + ret = -EINVAL; + else + args->value = ctx->preempt_timeout; + break; + default: ret = -EINVAL; break; @@ -824,6 +833,19 @@ int i915_gem_context_setparam_ioctl(struct drm_device *dev, void *data, } break; + case I915_CONTEXT_PARAM_PREEMPT_TIMEOUT: + if (args->size) + ret = -EINVAL; + else if (args->value > U32_MAX) + ret = -EINVAL; + else if (!(to_i915(dev)->caps.scheduler & I915_SCHEDULER_CAP_PREEMPTION)) + ret = -ENODEV; + else if (args->value && !capable(CAP_SYS_ADMIN)) + ret = -EPERM; + else + ctx->preempt_timeout = args->value; + break; + default: ret = -EINVAL; break; diff --git a/drivers/gpu/drm/i915/i915_gem_context.h b/drivers/gpu/drm/i915/i915_gem_context.h index 7854262ddfd9..1fc7181edd2d 100644 --- a/drivers/gpu/drm/i915/i915_gem_context.h +++ b/drivers/gpu/drm/i915/i915_gem_context.h @@ -150,6 +150,19 @@ struct i915_gem_context { */ int priority; + /** + * @preempt_timeout: QoS guarantee for the high priority context + * + * Some clients need a guarantee that they will start executing + * within a certain window, even at the expense of others. This entails + * that if a preemption request is not honoured by the active context + * within the timeout, we will reset the GPU to evict the hog and + * run the high priority context instead. + * + * Timeout is stored in nanoseconds. + */ + u32 preempt_timeout; + /** ggtt_offset_bias: placement restriction for context objects */ u32 ggtt_offset_bias; diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c index ddffffb829ef..b9dfd49a173e 100644 --- a/drivers/gpu/drm/i915/i915_request.c +++ b/drivers/gpu/drm/i915/i915_request.c @@ -1061,8 +1061,12 @@ void __i915_request_add(struct i915_request *request, bool flush_caches) * run at the earliest possible convenience. */ rcu_read_lock(); - if (engine->schedule) - engine->schedule(request, request->ctx->priority, 0); + if (engine->schedule) { + unsigned int timeout = request->ctx->preempt_timeout; + int priority = request->ctx->priority; + + engine->schedule(request, priority, timeout); + } rcu_read_unlock(); local_bh_disable(); diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index 877340afb63d..18cc83520e95 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -1232,7 +1232,7 @@ static void execlists_submit_request(struct i915_request *request) spin_lock_irqsave(&engine->timeline->lock, flags); queue_request(engine, &request->priotree, rq_prio(request)); - submit_queue(engine, rq_prio(request), 0); + submit_queue(engine, rq_prio(request), request->ctx->preempt_timeout); GEM_BUG_ON(!engine->execlists.first); GEM_BUG_ON(list_empty(&request->priotree.link)); diff --git a/drivers/gpu/drm/i915/selftests/intel_lrc.c b/drivers/gpu/drm/i915/selftests/intel_lrc.c index 5e4d6d07fff5..4fac9c552595 100644 --- a/drivers/gpu/drm/i915/selftests/intel_lrc.c +++ b/drivers/gpu/drm/i915/selftests/intel_lrc.c @@ -780,6 +780,90 @@ static int live_late_preempt_timeout(void *arg) goto err_ctx_lo; } +static int live_context_preempt_timeout(void *arg) +{ + struct drm_i915_private *i915 = arg; + struct i915_gem_context *ctx_hi, *ctx_lo; + struct spinner spin_hi, spin_lo; + struct intel_engine_cs *engine; + enum intel_engine_id id; + int err = -ENOMEM; + + if (!HAS_LOGICAL_RING_PREEMPTION(i915)) + return 0; + + mutex_lock(&i915->drm.struct_mutex); + + if (spinner_init(&spin_hi, i915)) + goto err_unlock; + + if (spinner_init(&spin_lo, i915)) + goto err_spin_hi; + + ctx_hi = kernel_context(i915); + if (!ctx_hi) + goto err_spin_lo; + ctx_hi->priority = I915_CONTEXT_MAX_USER_PRIORITY; + ctx_hi->preempt_timeout = 50 * 1000; /* 50us */ + + ctx_lo = kernel_context(i915); + if (!ctx_lo) + goto err_ctx_hi; + ctx_lo->priority = I915_CONTEXT_MIN_USER_PRIORITY; + + for_each_engine(engine, i915, id) { + struct i915_request *rq; + + rq = spinner_create_request(&spin_lo, ctx_lo, engine, MI_NOOP); + if (IS_ERR(rq)) { + err = PTR_ERR(rq); + goto err_ctx_lo; + } + + i915_request_add(rq); + if (!wait_for_spinner(&spin_lo, rq)) { + i915_gem_set_wedged(i915); + err = -EIO; + goto err_ctx_lo; + } + + rq = spinner_create_request(&spin_hi, ctx_hi, engine, MI_NOOP); + if (IS_ERR(rq)) { + spinner_end(&spin_lo); + err = PTR_ERR(rq); + goto err_ctx_lo; + } + + i915_request_add(rq); + if (!wait_for_spinner(&spin_hi, rq)) { + i915_gem_set_wedged(i915); + err = -EIO; + goto err_ctx_lo; + } + + spinner_end(&spin_hi); + spinner_end(&spin_lo); + if (flush_test(i915, I915_WAIT_LOCKED)) { + err = -EIO; + goto err_ctx_lo; + } + } + + err = 0; +err_ctx_lo: + kernel_context_close(ctx_lo); +err_ctx_hi: + kernel_context_close(ctx_hi); +err_spin_lo: + spinner_fini(&spin_lo); +err_spin_hi: + spinner_fini(&spin_hi); +err_unlock: + flush_test(i915, I915_WAIT_LOCKED); + mutex_unlock(&i915->drm.struct_mutex); + return err; +} + int intel_execlists_live_selftests(struct drm_i915_private *i915) { static const struct i915_subtest tests[] = { @@ -789,6 +873,7 @@ int intel_execlists_live_selftests(struct drm_i915_private *i915) SUBTEST(live_preempt_timeout), SUBTEST(live_preempt_reset), SUBTEST(live_late_preempt_timeout), + SUBTEST(live_context_preempt_timeout), }; return i915_subtests(tests, i915); } diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h index 7f5634ce8e88..853e0c7e0e85 100644 --- a/include/uapi/drm/i915_drm.h +++ b/include/uapi/drm/i915_drm.h @@ -1456,6 +1456,18 @@ struct drm_i915_gem_context_param { #define I915_CONTEXT_MAX_USER_PRIORITY 1023 /* inclusive */ #define I915_CONTEXT_DEFAULT_PRIORITY 0 #define I915_CONTEXT_MIN_USER_PRIORITY -1023 /* inclusive */ + +/* + * I915_CONTEXT_PARAM_PREEMPT_TIMEOUT: + * + * Preemption timeout give in nanoseconds. + * + * Only allowed for privileged clients (CAP_SYS_ADMIN), this property allows + * the preempting context to kick out a GPU hog using a GPU reset if they do + * not honour our preemption request in time. + */ +#define I915_CONTEXT_PARAM_PREEMPT_TIMEOUT 0x7 + __u64 value; };