[Intel-gfx] [Regression] WARNING: drivers/gpu/drm/i915/i915_gem.c:4525 i915_gem_free_object

[Steven Rostedt]

On Mon, Feb 23, 2015 at 11:12:39PM +0300, Andrey Skvortsov wrote:
> 
> This warning is moved from linux-next to v4.0-rc1 now. After system boot is just a black screen.
> I ssh'ed into the machine and saved the log. I attached updated dmesg.log with drm.debug=6. Hopefully it helps. 
> If you need any other debug information, traces, core dump or something else. Feel free to ask.
> 

I'm hitting this exact same warning on one of my test boxes.

 [drm] Initialized i915 1.6.0 20150130 for 0000:00:02.0 on minor 0
 ------------[ cut here ]------------ 
 WARNING: CPU: 7 PID: 116 at /home/rostedt/work/git/linux-trace.git/drivers/gpu/drm/i915/i915_gem.c:4525 i915_gem_free_object+0x2de/0x300 [i915]()
 WARN_ON(obj->frontbuffer_bits)
 Modules linked in: i915 i2c_algo_bit e1000e drm_kms_helper ptp crc32c_intel drm pps_core i2c_core video sunrpc
 CPU: 7 PID: 116 Comm: kworker/u16:5 Not tainted 4.0.0-rc1-test #270
 Hardware name: Hewlett-Packard HP Compaq Pro 6300 SFF/339A, BIOS K01 v02.05 05/07/2012
 Workqueue: events_unbound async_run_entry_fn
  00000000000011ad ffff88001fa837e8 ffffffff816cc3cc 000000000000d4d4
  ffff88001fa83838 ffff88001fa83828 ffffffff8107b977 ffff8800d4fd0000
  ffff88001f2b5800 ffff8800d4f40040 ffff8800d4f40000 ffff8800d4f400c0
 Call Trace:
  [<ffffffff816cc3cc>] dump_stack+0x4c/0x6e
  [<ffffffff8107b977>] warn_slowpath_common+0x97/0xe0
  [<ffffffff8107ba76>] warn_slowpath_fmt+0x46/0x50
  [<ffffffffa01b513e>] i915_gem_free_object+0x2de/0x300 [i915]
  [<ffffffffa0079797>] drm_gem_object_free+0x27/0x30 [drm]
  [<ffffffffa01ebf55>] intel_user_framebuffer_destroy+0x75/0xa0 [i915]
  [<ffffffffa0084b1b>] drm_framebuffer_free+0x5b/0x70 [drm]
  [<ffffffffa00858f9>] drm_framebuffer_unreference+0x39/0x70 [drm]
  [<ffffffffa00859fc>] drm_mode_set_config_internal+0xcc/0x110 [drm]
  [<ffffffffa013dca8>] restore_fbdev_mode+0xc8/0xf0 [drm_kms_helper]
  [<ffffffffa013f780>] drm_fb_helper_restore_fbdev_mode_unlocked+0x30/0x80 [drm_kms_helper]
  [<ffffffffa013f7f4>] drm_fb_helper_set_par+0x24/0x50 [drm_kms_helper]
  [<ffffffffa020bada>] intel_fbdev_set_par+0x1a/0x60 [i915]
  [<ffffffff8139f9c4>] fbcon_init+0x514/0x5a0
  [<ffffffff8140f39c>] visual_init+0xbc/0x120
  [<ffffffff81411aea>] do_bind_con_driver+0x16a/0x320
  [<ffffffff81411dc4>] do_take_over_console+0x124/0x1d0
  [<ffffffff8139de6b>] do_fbcon_takeover+0x5b/0xb0
  [<ffffffff813a2bd2>] fbcon_event_notify+0x672/0x800
  [<ffffffff810997dd>] notifier_call_chain+0x4d/0x70
  [<ffffffff810999c8>] __blocking_notifier_call_chain+0x58/0x80
  [<ffffffff81099a06>] blocking_notifier_call_chain+0x16/0x20
  [<ffffffff813a970b>] fb_notifier_call_chain+0x1b/0x20
  [<ffffffff813ab96b>] register_framebuffer+0x1eb/0x320
  [<ffffffffa013fa34>] drm_fb_helper_initial_config+0x214/0x3a0 [drm_kms_helper]
  [<ffffffffa020c64b>] intel_fbdev_initial_config+0x1b/0x20 [i915]
  [<ffffffff8109b40c>] async_run_entry_fn+0x4c/0x160
  [<ffffffff81092bab>] process_one_work+0x14b/0x4a0
  [<ffffffff8109336b>] worker_thread+0x11b/0x480
  [<ffffffff81093250>] ? rescuer_thread+0x310/0x310
  [<ffffffff81098e89>] kthread+0xc9/0xe0
  [<ffffffff81090000>] ? wq_unbind_fn+0xa0/0x100
  [<ffffffff81098dc0>] ? flush_kthread_worker+0x90/0x90
  [<ffffffff816d3aac>] ret_from_fork+0x7c/0xb0
  [<ffffffff81098dc0>] ? flush_kthread_worker+0x90/0x90
 ---[ end trace 507288b4d65403f3 ]---

It's not connected to a monitor so I don't know if it has a blank screen or not.
But because of the warning, it makes my tests fail (ktest fails or call traces
like this).

I bisected it down to this commit:

commit 455a68086d1dfb801ad7c867d5ca0ed0e0f758b0
Author: Gustavo Padovan <gustavo.padovan at collabora.co.uk>
Date:   Mon Dec 1 15:40:11 2014 -0800

    drm/i915: remove intel_pipe_set_base() (v4)


-- Steve