Created attachment 114723 [details] dmesg with drm.debug info System Environment: ----------------------------------------------------- Regression: Yes, Non-working platforms: PNV SNB IVB BYT ==Kernel== -------------------------------------------------- commit 07e5c9b5061ad65c85285c0bb9ecd5626ed301d0 Author: Daniel Vetter <daniel.vetter@ffwll.ch> Date: Fri Mar 27 15:29:51 2015 +0100 drm-intel-nightly: 2015y-03m-27d-14h-29m-07s UTC integration manifest ==Bug detailed description== Igt/kms_setmode subcases clone-exclusive-crtc and clone-single-crtc causes time out and call trace on SNB mobile, it does not exists on SNB desktop, By bisected, shows the first bad commit is 0b9018. commit 0b9018793939970aad49e4e0ed21f0fd0db74e4a Author: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com> AuthorDate: Fri Mar 20 16:18:08 2015 +0200 Commit: Daniel Vetter <daniel.vetter@ffwll.ch> CommitDate: Thu Mar 26 17:49:52 2015 +0100 drm/i915: Don't use encoder->new_crtc in intel_modeset_pipe_config() Move towards atomic by using the legacy modeset's drm_atomic_state instead. v2: Move call to drm_atomic_add_affected_connectors() to intel_modeset_compute_config(). (Daniel) Output: [ 127.901701] Call Trace: [ 127.901722] [<ffffffffa00d26e8>] ? intel_modeset_compute_config+0x5b8/0xd51 [ i915] [ 127.901760] [<ffffffffa00d4cb9>] ? intel_set_mode+0x39/0x91 [i915] [ 127.901792] [<ffffffffa00d5df8>] ? intel_crtc_set_config+0xb8a/0xc55 [i915] [ 127.901821] [<ffffffff81791d97>] ? printk+0x48/0x4d [ 127.901848] [<ffffffffa000c1e7>] ? drm_mode_set_config_internal+0x4e/0xd2 [dr m] [ 127.901882] [<ffffffffa000febf>] ? drm_mode_setcrtc+0x3fe/0x4f4 [drm] [ 127.901913] [<ffffffffa00047d4>] ? drm_ioctl+0x344/0x3b3 [drm] [ 127.901941] [<ffffffffa000fac1>] ? drm_mode_setplane+0x1dc/0x1dc [drm] [ 127.901969] [<ffffffff8111da12>] ? do_vfs_ioctl+0x360/0x424 [ 127.901992] [<ffffffff810a2634>] ? __audit_syscall_entry+0xb3/0xd3 [ 127.902018] [<ffffffff8100d41f>] ? syscall_trace_enter_phase1+0x11a/0x123 [ 127.902045] [<ffffffff8111db1f>] ? SyS_ioctl+0x49/0x7a [ 127.902066] [<ffffffff8179ae0e>] ? int_check_syscall_exit_work+0x34/0x3d [ 127.902092] [<ffffffff8179abf2>] ? system_call_fastpath+0x12/0x17 [ 127.902116] Code: c7 44 24 08 18 00 00 00 e9 68 fc ff ff 48 63 44 24 20 49 8b 56 38 4c 8b 2c c2 4d 85 ed 0f 84 c6 00 00 00 49 8b 56 40 48 8b 04 c2 <48> 39 58 0 8 0f 85 b4 00 00 00 f6 05 58 de f5 ff 04 41 8b 87 34 [ 127.902280] RIP [<ffffffffa00d28b3>] intel_modeset_compute_config+0x783/0xd51 [i915] [ 127.902320] RSP <ffff8800b095fb78> [ 127.902334] CR2: 0000000000000008 [ 127.908820] ---[ end trace 77b9184de44fe47f ]--- Reproduce steps: ---------------------------- 1, ./kms_setmode --run-subtest invalid-clone-single-crtc
[ 127.900863] BUG: unable to handle kernel NULL pointer dereference at 0000000000000008 [ 127.900901] IP: [<ffffffffa00d28b3>] intel_modeset_compute_config+0x783/0xd51 [i915] Ideally you would translate that address back to the source line.
Could you please test if the following patch fixes the issue? http://lists.freedesktop.org/archives/intel-gfx/2015-March/063389.html
This issue does not exists on the latest kernel(nightly-ff2304). verified it.
Closing old verified.
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.