From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-bl2on0071.outbound.protection.outlook.com ([65.55.169.71]:64048 "EHLO na01-bl2-obe.outbound.protection.outlook.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751359AbcDGNVn (ORCPT ); Thu, 7 Apr 2016 09:21:43 -0400 Subject: Re: Fwd: Please revert wrong kernel OOPS patch in -69 ! To: Taras Prokopenko , , Luis Henriques , , Kamal Mostafa , Sasha Levin , Jiri Slaby , Greg Kroah-Hartman References: From: =?UTF-8?Q?Nicolai_H=c3=a4hnle?= Message-ID: <57065B31.5010503@amd.com> Date: Thu, 7 Apr 2016 08:05:53 -0500 MIME-Version: 1.0 In-Reply-To: Content-Type: multipart/mixed; boundary="------------070600010001060007000200" Sender: stable-owner@vger.kernel.org List-ID: --------------070600010001060007000200 Content-Type: text/plain; charset="utf-8"; format=flowed Content-Transfer-Encoding: 7bit Hmm, I was under the impression from an email I got that this was already reverted from 3.16. Clearly, communication is hard. Let me re-iterate, to be sure: kernels 3.17 and older 1. need to revert the backport of upstream commit f6ff4f67cdf8455d0a4226eeeaf5af17c37d05eb (I think most have done that already) and 2. should apply the corrected backport which I am attaching yet again (seems like when I sent this patch around two weeks ago, it wasn't picked up). Thanks, Nicolai On 07.04.2016 04:15, Taras Prokopenko wrote: > > ---------- Forwarded message ---------- > From: *Taras Prokopenko* > > Date: 2016-04-07 12:04 GMT+03:00 > Subject: Please revert wrong kernel OOPS patch in -69 ! > To: luis.henriques@canonical.com > Cc: jslaby@suse.cz > > > Please test your changes! Lost 1/2 a day to figure out whats causing my > pc to hang. > > OopsText: > BUG: unable to handle kernel NULL pointer dereference at 0000000000000008 > IP: [] radeon_fence_ref+0xd/0x50 [radeon] > PGD 0 > Oops: 0002 [#1] SMP > > > https://lists.ubuntu.com/archives/kernel-team/2016-February/072148.html > > Damn > --------------070600010001060007000200 Content-Type: text/x-patch; name="0001-drm-radeon-hold-reference-to-fences-in-radeon_sa_bo_.patch" Content-Transfer-Encoding: 8bit Content-Disposition: attachment; filename*0="0001-drm-radeon-hold-reference-to-fences-in-radeon_sa_bo_.pa"; filename*1="tch" >>From ad94965f69c2681832f64473d28c23ae71b6e52f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Nicolai=20H=C3=A4hnle?= Date: Tue, 15 Mar 2016 12:56:45 -0500 Subject: [PATCH] drm/radeon: hold reference to fences in radeon_sa_bo_new (3.17 and older) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit [Backport of upstream commit f6ff4f67cdf8455d0a4226eeeaf5af17c37d05eb, with an additional NULL pointer guard that is required for kernels 3.17 and older. To be precise, any kernel that does *not* have commit 954605ca3 "drm/radeon: use common fence implementation for fences, v4" requires this additional NULL pointer guard.] An arbitrary amount of time can pass between spin_unlock and radeon_fence_wait_any, so we need to ensure that nobody frees the fences from under us. Based on the analogous fix for amdgpu. Signed-off-by: Nicolai Hähnle Reviewed-by: Christian König (v1 + fix) Tested-by: Lutz Euler Cc: stable@vger.kernel.org --- drivers/gpu/drm/radeon/radeon_sa.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/gpu/drm/radeon/radeon_sa.c b/drivers/gpu/drm/radeon/radeon_sa.c index f0bac68..8962411 100644 --- a/drivers/gpu/drm/radeon/radeon_sa.c +++ b/drivers/gpu/drm/radeon/radeon_sa.c @@ -349,8 +349,15 @@ int radeon_sa_bo_new(struct radeon_device *rdev, /* see if we can skip over some allocations */ } while (radeon_sa_bo_next_hole(sa_manager, fences, tries)); + for (i = 0; i < RADEON_NUM_RINGS; ++i) { + if (fences[i]) + radeon_fence_ref(fences[i]); + } + spin_unlock(&sa_manager->wq.lock); r = radeon_fence_wait_any(rdev, fences, false); + for (i = 0; i < RADEON_NUM_RINGS; ++i) + radeon_fence_unref(&fences[i]); spin_lock(&sa_manager->wq.lock); /* if we have nothing to wait for block */ if (r == -ENOENT && block) { -- 2.5.0 --------------070600010001060007000200--