From patchwork Mon Jun 30 16:53:39 2014 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Rodrigo Vivi X-Patchwork-Id: 4455511 Return-Path: X-Original-To: patchwork-intel-gfx@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork2.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.19.201]) by patchwork2.web.kernel.org (Postfix) with ESMTP id 508E9BEEE9 for ; Mon, 30 Jun 2014 23:52:35 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id D0339203E3 for ; Mon, 30 Jun 2014 23:52:32 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) by mail.kernel.org (Postfix) with ESMTP id 8B579203DC for ; Mon, 30 Jun 2014 23:52:31 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 46D916E416; Mon, 30 Jun 2014 16:52:30 -0700 (PDT) X-Original-To: intel-gfx@lists.freedesktop.org Delivered-To: intel-gfx@lists.freedesktop.org Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by gabe.freedesktop.org (Postfix) with ESMTP id 85CF26E270 for ; Mon, 30 Jun 2014 16:52:27 -0700 (PDT) Received: from orsmga002.jf.intel.com ([10.7.209.21]) by orsmga102.jf.intel.com with ESMTP; 30 Jun 2014 16:46:59 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.01,578,1400050800"; d="scan'208";a="566306962" Received: from di-604.jf.intel.com (HELO rdvivi-hillsboro.jf.intel.com) ([10.7.201.28]) by orsmga002.jf.intel.com with ESMTP; 30 Jun 2014 16:52:26 -0700 From: Rodrigo Vivi To: intel-gfx@lists.freedesktop.org Date: Mon, 30 Jun 2014 09:53:39 -0700 Message-Id: <1404147224-2978-5-git-send-email-rodrigo.vivi@intel.com> X-Mailer: git-send-email 1.9.3 In-Reply-To: <1404147224-2978-1-git-send-email-rodrigo.vivi@intel.com> References: <1404147224-2978-1-git-send-email-rodrigo.vivi@intel.com> MIME-Version: 1.0 Cc: Ben Widawsky , Rodrigo Vivi Subject: [Intel-gfx] [PATCH 05/10] drm/i915: Implement MI decode for gen8 X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" X-Spam-Status: No, score=-2.7 required=5.0 tests=BAYES_00, DATE_IN_PAST_06_12, RCVD_IN_DNSWL_MED, T_RP_MATCHES_RCVD, UNPARSEABLE_RELAY autolearn=unavailable version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP Ipehr just carries Dword 0 and on Gen 8, offsets are located on Dword 2 and 3 of MI_SEMAPHORE_WAIT. This implementation was based on Ben's work and on Ville's suggestion for Ben Cc: Ville Syrjälä Cc: Ben Widawsky Signed-off-by: Rodrigo Vivi Reviewed-by: Ben Widawsky --- drivers/gpu/drm/i915/i915_irq.c | 42 ++++++++++++++++++++++------------------- 1 file changed, 23 insertions(+), 19 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 0217a41..66e2481 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -2784,12 +2784,7 @@ static bool ipehr_is_semaphore_wait(struct drm_device *dev, u32 ipehr) { if (INTEL_INFO(dev)->gen >= 8) { - /* - * FIXME: gen8 semaphore support - currently we don't emit - * semaphores on bdw anyway, but this needs to be addressed when - * we merge that code. - */ - return false; + return (ipehr >> 23) == 0x1c; } else { ipehr &= ~MI_SEMAPHORE_SYNC_MASK; return ipehr == (MI_SEMAPHORE_MBOX | MI_SEMAPHORE_COMPARE | @@ -2798,19 +2793,20 @@ ipehr_is_semaphore_wait(struct drm_device *dev, u32 ipehr) } static struct intel_engine_cs * -semaphore_wait_to_signaller_ring(struct intel_engine_cs *ring, u32 ipehr) +semaphore_wait_to_signaller_ring(struct intel_engine_cs *ring, u32 ipehr, u64 offset) { struct drm_i915_private *dev_priv = ring->dev->dev_private; struct intel_engine_cs *signaller; int i; if (INTEL_INFO(dev_priv->dev)->gen >= 8) { - /* - * FIXME: gen8 semaphore support - currently we don't emit - * semaphores on bdw anyway, but this needs to be addressed when - * we merge that code. - */ - return NULL; + for_each_ring(signaller, dev_priv, i) { + if (ring == signaller) + continue; + + if (offset == signaller->semaphore.signal_ggtt[ring->id]) + return signaller; + } } else { u32 sync_bits = ipehr & MI_SEMAPHORE_SYNC_MASK; @@ -2823,8 +2819,8 @@ semaphore_wait_to_signaller_ring(struct intel_engine_cs *ring, u32 ipehr) } } - DRM_ERROR("No signaller ring found for ring %i, ipehr 0x%08x\n", - ring->id, ipehr); + DRM_ERROR("No signaller ring found for ring %i, ipehr 0x%08x, offset 0x%0%016llx\n", + ring->id, ipehr, offset); return NULL; } @@ -2834,7 +2830,8 @@ semaphore_waits_for(struct intel_engine_cs *ring, u32 *seqno) { struct drm_i915_private *dev_priv = ring->dev->dev_private; u32 cmd, ipehr, head; - int i; + u64 offset = 0; + int i, backwards; ipehr = I915_READ(RING_IPEHR(ring->mmio_base)); if (!ipehr_is_semaphore_wait(ring->dev, ipehr)) @@ -2843,13 +2840,15 @@ semaphore_waits_for(struct intel_engine_cs *ring, u32 *seqno) /* * HEAD is likely pointing to the dword after the actual command, * so scan backwards until we find the MBOX. But limit it to just 3 - * dwords. Note that we don't care about ACTHD here since that might + * or 4 dwords depending on the semaphore wait command size. + * Note that we don't care about ACTHD here since that might * point at at batch, and semaphores are always emitted into the * ringbuffer itself. */ head = I915_READ_HEAD(ring) & HEAD_ADDR; + backwards = (INTEL_INFO(ring->dev)->gen >= 8) ? 5 : 4; - for (i = 4; i; --i) { + for (i = backwards; i; --i) { /* * Be paranoid and presume the hw has gone off into the wild - * our ring is smaller than what the hardware (and hence @@ -2869,7 +2868,12 @@ semaphore_waits_for(struct intel_engine_cs *ring, u32 *seqno) return NULL; *seqno = ioread32(ring->buffer->virtual_start + head + 4) + 1; - return semaphore_wait_to_signaller_ring(ring, ipehr); + if (INTEL_INFO(ring->dev)->gen >= 8) { + offset = ioread32(ring->buffer->virtual_start + head + 12); + offset <<= 32; + offset = ioread32(ring->buffer->virtual_start + head + 8); + } + return semaphore_wait_to_signaller_ring(ring, ipehr, offset); } static int semaphore_passed(struct intel_engine_cs *ring)