I'm having a somewhat interesting video issue. I'm using the intel
drivers with two monitors. I have two monitors hooked up to the
motherboard, one DVI-D and the other HDMI.
The software I'm using is:
linux-4.3.3 with firmware skl_dmc_ver1.bin compiled into the kernel image.
xf86-video-intel-0340718 Intel Driver
libva-1.6.2
libva-intel-driver-1.6.2
libvdpau-1.1.1
libvdpau-va-gl-0.3.4
libdrm-2.4.66
Mesa-11.0.8
The window manager is xfce-4.12
Visually everything works, however every time I move the cursor from one
monitor to the other, I get in my logs:
WARNING: CPU: 0 PID: 782 at drivers/gpu/drm/i915/intel_pm.c:3668
skl_update_other_pipe_wm+0x1b5/0x1c0()
WARN_ON(!wm_changed)
Modules linked in:
uvcvideo videobuf2_vmalloc videobuf2_memops videobuf2_core
CPU: 0 PID: 782 Comm: Xorg Tainted: G W 4.3.3-lfs-SVN-20160104 #1
Hardware name: Gigabyte Technology Co., Ltd. To be filled by
O.E.M./Z170-HD3P-CF, BIOS F2 07/27/2015
ffffffff81b195f8 ffff880265e43658 ffffffff8142be9f ffff880265e436a0
ffff880265e43690 ffffffff8108de88 ffff880267f81000 ffff880265e437dc
ffff880267f80000 0000000000000000 ffff880265e43728 ffff880265e436f0
Call Trace:
[<ffffffff8142be9f>] dump_stack+0x44/0x55
[<ffffffff8108de88>] warn_slowpath_common+0x78/0xb0
[<ffffffff8108df07>] warn_slowpath_fmt+0x47/0x50
[<ffffffff81553805>] skl_update_other_pipe_wm+0x1b5/0x1c0
[<ffffffff815539a6>] skl_update_wm+0x196/0x6a0
[<ffffffff8158fd40>] ? gen9_read32+0xf0/0x270
[<ffffffff81557e69>] intel_update_watermarks+0x19/0x20
[<ffffffff815ab1e1>] intel_atomic_commit+0x3f1/0x1340
[<ffffffff8153fad0>] ? drm_atomic_check_only+0x210/0x540
[<ffffffff8153eefd>] ? drm_atomic_get_crtc_state+0x2d/0xd0
[<ffffffff8153fe32>] drm_atomic_commit+0x32/0x50
[<ffffffff8151c1ea>] drm_atomic_helper_disable_plane+0xea/0x130
[<ffffffff815307e5>] __setplane_internal+0x235/0x2f0
[<ffffffff815309b6>] drm_mode_cursor_universal+0x116/0x1f0
[<ffffffff81530b0c>] drm_mode_cursor_common+0x7c/0x180
[<ffffffff81185901>] ? set_fd_set+0x21/0x30
[<ffffffff8153478c>] drm_mode_cursor_ioctl+0x3c/0x40
[<ffffffff81526324>] drm_ioctl+0x124/0x4d0
[<ffffffff81534750>] ? drm_mode_setcrtc+0x4c0/0x4c0
[<ffffffff8118860c>] ? dput+0x9c/0x1f0
[<ffffffff81172893>] ? __vfs_read+0x23/0xd0
[<ffffffff81184e00>] do_vfs_ioctl+0x280/0x460
[<ffffffff81097365>] ? recalc_sigpending+0x15/0x50
[<ffffffff8109807d>] ? __set_task_blocked+0x2d/0x70
[<ffffffff81185054>] SyS_ioctl+0x74/0x80
[<ffffffff818fc197>] entry_SYSCALL_64_fastpath+0x12/0x6a
---[ end trace 1fefbff8a5313d63 ]---
I emphasize here that there is no visual problem--only a very full kernel
ring buffer and a large/growing log file.
Since the message references drivers/gpu/drm/i915/intel_pm.c, I looked at
that and saw:
/*
* If we end up re-computing the other pipe WM values, it's
* because it was really needed, so we expect the WM values to
* be different.
*/
WARN_ON(!wm_changed);
So I commented out the WARN_ON line and the messages went away.
I also experimented with the 4.4 kernel, but there is a much bigger
problem there. The DVI-D connected monitor does not seem to be recognized.
I've reported this upstream:
https://software.intel.com/en-us/forums/graphics-driver-bug-reporting/topic/606619
I'll follow up here if I get an answer.
-- Bruce
--
http://lists.linuxfromscratch.org/listinfo/blfs-dev
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
