Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   Daniel Vetter <daniel.vetter@ffwll.ch>
To:     DRI Development <dri-devel@lists.freedesktop.org>
Cc:     Intel Graphics Development <intel-gfx@lists.freedesktop.org>,
        LKML <linux-kernel@vger.kernel.org>, linux-rdma@vger.kernel.org,
        amd-gfx@lists.freedesktop.org,
        Daniel Vetter <daniel.vetter@ffwll.ch>,
        linux-media@vger.kernel.org, linaro-mm-sig@lists.linaro.org,
        Chris Wilson <chris@chris-wilson.co.uk>,
        Maarten Lankhorst <maarten.lankhorst@linux.intel.com>,
        =?UTF-8?q?Christian=20K=C3=B6nig?= <christian.koenig@amd.com>,
        Daniel Vetter <daniel.vetter@intel.com>
Subject: [PATCH 13/18] drm/amdgpu/dc: Stop dma_resv_lock inversion in commit_tail
Date:   Thu,  4 Jun 2020 10:12:19 +0200
Message-Id: <20200604081224.863494-14-daniel.vetter@ffwll.ch>
In-Reply-To: <20200604081224.863494-1-daniel.vetter@ffwll.ch>
References: <20200604081224.863494-1-daniel.vetter@ffwll.ch>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Trying to grab dma_resv_lock while in commit_tail before we've done
all the code that leads to the eventual signalling of the vblank event
(which can be a dma_fence) is deadlock-y. Don't do that.

Here the solution is easy because just grabbing locks to read
something races anyway. We don't need to bother, READ_ONCE is
equivalent. And avoids the locking issue.

Cc: linux-media@vger.kernel.org
Cc: linaro-mm-sig@lists.linaro.org
Cc: linux-rdma@vger.kernel.org
Cc: amd-gfx@lists.freedesktop.org
Cc: intel-gfx@lists.freedesktop.org
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: Christian König <christian.koenig@amd.com>
Signed-off-by: Daniel Vetter <daniel.vetter@intel.com>
---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index c575e7394d03..04c11443b9ca 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -6910,7 +6910,11 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 		 * explicitly on fences instead
 		 * and in general should be called for
 		 * blocking commit to as per framework helpers
+		 *
+		 * Yes, this deadlocks, since you're calling dma_resv_lock in a
+		 * path that leads to a dma_fence_signal(). Don't do that.
 		 */
+#if 0
 		r = amdgpu_bo_reserve(abo, true);
 		if (unlikely(r != 0))
 			DRM_ERROR("failed to reserve buffer before flip\n");
@@ -6920,6 +6924,12 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state,
 		tmz_surface = amdgpu_bo_encrypted(abo);
 
 		amdgpu_bo_unreserve(abo);
+#endif
+		/*
+		 * this races anyway, so READ_ONCE isn't any better or worse
+		 * than the stuff above. Except the stuff above can deadlock.
+		 */
+		tiling_flags = READ_ONCE(abo->tiling_flags);
 
 		fill_dc_plane_info_and_addr(
 			dm->adev, new_plane_state, tiling_flags,
-- 
2.26.2