From patchwork Thu Feb 28 16:01:33 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Mika Kuoppala X-Patchwork-Id: 10833389 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 6BEC21390 for ; Thu, 28 Feb 2019 16:02:13 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 5C7642F388 for ; Thu, 28 Feb 2019 16:02:12 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 515172F391; Thu, 28 Feb 2019 16:02:12 +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 F11DF2F39D for ; Thu, 28 Feb 2019 16:02:11 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 90B566E20B; Thu, 28 Feb 2019 16:02:09 +0000 (UTC) X-Original-To: intel-gfx@lists.freedesktop.org Delivered-To: intel-gfx@lists.freedesktop.org Received: from mga01.intel.com (mga01.intel.com [192.55.52.88]) by gabe.freedesktop.org (Postfix) with ESMTPS id 0303E6E20C for ; Thu, 28 Feb 2019 16:02:07 +0000 (UTC) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga004.jf.intel.com ([10.7.209.38]) by fmsmga101.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 28 Feb 2019 08:02:07 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.58,423,1544515200"; d="scan'208";a="278610732" Received: from rosetta.fi.intel.com ([10.237.72.186]) by orsmga004.jf.intel.com with ESMTP; 28 Feb 2019 08:02:05 -0800 Received: by rosetta.fi.intel.com (Postfix, from userid 1000) id 6633D840B51; Thu, 28 Feb 2019 18:01:33 +0200 (EET) From: Mika Kuoppala To: intel-gfx@lists.freedesktop.org Date: Thu, 28 Feb 2019 18:01:33 +0200 Message-Id: <20190228160133.819-3-mika.kuoppala@linux.intel.com> X-Mailer: git-send-email 2.17.1 In-Reply-To: <20190228160133.819-1-mika.kuoppala@linux.intel.com> References: <20190228160133.819-1-mika.kuoppala@linux.intel.com> Subject: [Intel-gfx] [PATCH 3/3] drm/i915: Disable PSMI idle messaging when stopping ring 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 Hardware cannot be in a middle of idle flow messaging when we pull the plug from ringbuffer. Disable idle messaging before we do so to avoid potential deadlock on engine initialization and reset. v2: INVALID_MMIO_REG, unconditional enable (Chris) v3: comment (Chris), bspec reference References: bspec/11751 Cc: Chris Wilson Acked-by: Chris Wilson Signed-off-by: Mika Kuoppala --- drivers/gpu/drm/i915/intel_engine_cs.c | 64 +++++++++++++++++++++++++- 1 file changed, 63 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c b/drivers/gpu/drm/i915/intel_engine_cs.c index 2a94b92cfcd3..10cb6b22fe92 100644 --- a/drivers/gpu/drm/i915/intel_engine_cs.c +++ b/drivers/gpu/drm/i915/intel_engine_cs.c @@ -764,7 +764,7 @@ int intel_engine_init_common(struct intel_engine_cs *engine) /** * intel_engines_cleanup_common - cleans up the engine state created by - * the common initiailizers. + * the common initializers. * @engine: Engine to cleanup. * * This cleans up everything created by the common helpers. @@ -865,6 +865,63 @@ void intel_engine_cancel_stop_cs(struct intel_engine_cs *engine) _MASKED_BIT_DISABLE(STOP_RING)); } +static i915_reg_t get_idle_poll_reg(const struct intel_engine_cs *engine) +{ + if (engine->id != RCS) + return INVALID_MMIO_REG; + + if (IS_GEN(engine->i915, 9)) + return GEN9_RCS_FE_FSM2; + + if (IS_GEN(engine->i915, 8)) + return GEN6_RCS_PWR_FSM; + + return INVALID_MMIO_REG; +} + +static void disable_idle_messaging(struct intel_engine_cs *engine) +{ + struct drm_i915_private *dev_priv = engine->i915; + i915_reg_t poll_reg; + + poll_reg = get_idle_poll_reg(engine); + if (!i915_mmio_reg_valid(poll_reg)) + return; + + GEM_DEBUG_WARN_ON(I915_READ_FW(GEN6_RC_SLEEP_PSMI_CONTROL) & + GEN6_PSMI_SLEEP_MSG_DISABLE); + /* + * Hardware must not be in middle of idle flow signalling + * when the RING_CTL is zeroed. In order to prevent this + * we disable the messaging temporarily. + */ + I915_WRITE_FW(GEN6_RC_SLEEP_PSMI_CONTROL, + _MASKED_BIT_ENABLE(GEN6_PSMI_SLEEP_MSG_DISABLE)); + + /* + * Poll the magic bits to ensure that engine state machine + * is in valid state for stopping the ring. + */ + if (__intel_wait_for_register_fw(dev_priv, poll_reg, + 0x7f, 0x30, + 5000, 0, + NULL)) + DRM_DEBUG_DRIVER("psmi idle msg poll timeout\n"); +} + +static void enable_idle_messaging(struct intel_engine_cs *engine) +{ + struct drm_i915_private *dev_priv = engine->i915; + i915_reg_t poll_reg; + + poll_reg = get_idle_poll_reg(engine); + if (!i915_mmio_reg_valid(poll_reg)) + return; + + I915_WRITE_FW(GEN6_RC_SLEEP_PSMI_CONTROL, + _MASKED_BIT_DISABLE(GEN6_PSMI_SLEEP_MSG_DISABLE)); +} + int intel_engine_stop_ring(struct intel_engine_cs *engine) { struct drm_i915_private *dev_priv = engine->i915; @@ -880,9 +937,14 @@ int intel_engine_stop_ring(struct intel_engine_cs *engine) I915_WRITE_FW(RING_TAIL(base), 0); POSTING_READ_FW(RING_TAIL(base)); + /* Idle messaging needs to be off during ring disable */ + disable_idle_messaging(engine); + /* The ring must be empty before it is disabled */ I915_WRITE_FW(RING_CTL(base), 0); + enable_idle_messaging(engine); + /* Check acts as a post */ if (I915_READ_FW(RING_HEAD(base))) { GEM_TRACE("%s: ring head [%x] not parked\n",