Re: [PATCH] backlight: Avoid double fbcon backlight handling
From: joeyli <jlee@suse.com>
Date: 2016-07-07 09:09:42
Also in:
dri-devel, intel-gfx, linux-acpi, linux-omap, linux-renesas-soc, nouveau
On Thu, Jun 30, 2016 at 12:30:56PM +0100, Chris Wilson wrote:
Backlights controlled by i915.ko and only associated with its connectors
and also only associated with the intel_drmfb fbcon, controlled by
i915.ko. In this situation, we already handle adjusting the backlight
when the fbcon is blanked/unblanked and do not require backlight trying
to do the same.
Attempting to register with the fbdev as a client causes lockdep to warn
about a dependency cycle:
[ 18.983763] ===========================
[ 18.983763] [ INFO: possible circular locking dependency detected ]
[ 18.983766] 4.7.0-rc5+ #524 Tainted: G O
[ 18.983767] -------------------------------------------------------
[ 18.983767] kworker/u8:0/6 is trying to acquire lock:
[ 18.983777] (&dev->mode_config.mutex){+.+.+.}, at: [<ffffffff815afde0>] drm_modeset_lock_all+0x40/0x120
[ 18.983777]
but task is already holding lock:
[ 18.983782] ((fb_notifier_list).rwsem){++++.+}, at: [<ffffffff810ac195>] __blocking_notifier_call_chain+0x35/0x70
[ 18.983783]
which lock already depends on the new lock.
[ 18.983783]
the existing dependency chain (in reverse order) is:
[ 18.983785]
-> #1 ((fb_notifier_list).rwsem){++++.+}:
[ 18.983789] [<ffffffff810df611>] lock_acquire+0xb1/0x200
[ 18.983792] [<ffffffff819a55b4>] down_write+0x44/0x80
[ 18.983795] [<ffffffff810abf91>] blocking_notifier_chain_register+0x21/0xb0
[ 18.983798] [<ffffffff814c7448>] fb_register_client+0x18/0x20
[ 18.983800] [<ffffffff814c6c86>] backlight_device_register+0x136/0x260
[ 18.983852] [<ffffffffa0127eb2>] intel_backlight_device_register+0xa2/0x160 [i915]
[ 18.983892] [<ffffffffa00f46be>] intel_connector_register+0xe/0x10 [i915]
[ 18.983932] [<ffffffffa0112bfb>] intel_dp_connector_register+0x1b/0x80 [i915]
[ 18.983936] [<ffffffff8159dfea>] drm_connector_register+0x4a/0x80
[ 18.983938] [<ffffffff8159fe44>] drm_connector_register_all+0x64/0xf0
[ 18.983940] [<ffffffff815a2a64>] drm_modeset_register_all+0x174/0x1c0
[ 18.983942] [<ffffffff81599b72>] drm_dev_register+0xc2/0xd0
[ 18.983976] [<ffffffffa00621d7>] i915_driver_load+0x1547/0x2200 [i915]
[ 18.984010] [<ffffffffa006d80f>] i915_pci_probe+0x4f/0x70 [i915]
[ 18.984014] [<ffffffff814a2135>] local_pci_probe+0x45/0xa0
[ 18.984015] [<ffffffff814a349b>] pci_device_probe+0xdb/0x130
[ 18.984018] [<ffffffff815c07e3>] driver_probe_device+0x223/0x440
[ 18.984020] [<ffffffff815c0ad5>] __driver_attach+0xd5/0x100
[ 18.984022] [<ffffffff815be386>] bus_for_each_dev+0x66/0xa0
[ 18.984023] [<ffffffff815c002e>] driver_attach+0x1e/0x20
[ 18.984025] [<ffffffff815bf9be>] bus_add_driver+0x1ee/0x280
[ 18.984028] [<ffffffff815c1810>] driver_register+0x60/0xe0
[ 18.984030] [<ffffffff814a1a10>] __pci_register_driver+0x60/0x70
[ 18.984063] [<ffffffffa01a905b>] i915_init+0x5b/0x62 [i915]
[ 18.984067] [<ffffffff8100042d>] do_one_initcall+0x3d/0x150
[ 18.984070] [<ffffffff811a935b>] do_init_module+0x5f/0x1d9
[ 18.984073] [<ffffffff81124416>] load_module+0x20e6/0x27e0
[ 18.984075] [<ffffffff81124d63>] SYSC_finit_module+0xc3/0xf0
[ 18.984076] [<ffffffff81124dae>] SyS_finit_module+0xe/0x10
[ 18.984079] [<ffffffff819a83a9>] entry_SYSCALL_64_fastpath+0x1c/0xac
[ 18.984081]
-> #0 (&dev->mode_config.mutex){+.+.+.}:
[ 18.984083] [<ffffffff810df0ac>] __lock_acquire+0x10fc/0x1260
[ 18.984085] [<ffffffff810df611>] lock_acquire+0xb1/0x200
[ 18.984088] [<ffffffff819a3097>] mutex_lock_nested+0x67/0x3c0
[ 18.984090] [<ffffffff815afde0>] drm_modeset_lock_all+0x40/0x120
[ 18.984093] [<ffffffff8158f79b>] drm_fb_helper_restore_fbdev_mode_unlocked+0x2b/0x80
[ 18.984095] [<ffffffff8158f81d>] drm_fb_helper_set_par+0x2d/0x50
[ 18.984134] [<ffffffffa0105f7a>] intel_fbdev_set_par+0x1a/0x60 [i915]
[ 18.984136] [<ffffffff814c13c6>] fbcon_init+0x586/0x610
[ 18.984139] [<ffffffff8154d16a>] visual_init+0xca/0x130
[ 18.984141] [<ffffffff8154e611>] do_bind_con_driver+0x1c1/0x3a0
[ 18.984143] [<ffffffff8154eaf6>] do_take_over_console+0x116/0x180
[ 18.984145] [<ffffffff814bd3a7>] do_fbcon_takeover+0x57/0xb0
[ 18.984147] [<ffffffff814c1e48>] fbcon_event_notify+0x658/0x750
[ 18.984150] [<ffffffff810abcae>] notifier_call_chain+0x3e/0xb0
[ 18.984152] [<ffffffff810ac1ad>] __blocking_notifier_call_chain+0x4d/0x70
[ 18.984154] [<ffffffff810ac1e6>] blocking_notifier_call_chain+0x16/0x20
[ 18.984156] [<ffffffff814c748b>] fb_notifier_call_chain+0x1b/0x20
[ 18.984158] [<ffffffff814c86b1>] register_framebuffer+0x251/0x330
[ 18.984160] [<ffffffff8158fa9f>] drm_fb_helper_initial_config+0x25f/0x3f0
[ 18.984199] [<ffffffffa0106b48>] intel_fbdev_initial_config+0x18/0x30 [i915]
[ 18.984201] [<ffffffff810adfd8>] async_run_entry_fn+0x48/0x150
[ 18.984203] [<ffffffff810a3947>] process_one_work+0x1e7/0x750
[ 18.984205] [<ffffffff810a3efb>] worker_thread+0x4b/0x4f0
[ 18.984207] [<ffffffff810aad4f>] kthread+0xef/0x110
[ 18.984208] [<ffffffff819a85ef>] ret_from_fork+0x1f/0x40
[ 18.984209]
other info that might help us debug this:
[ 18.984210] Possible unsafe locking scenario:
[ 18.984210] CPU0 CPU1
[ 18.984211] ---- ----
[ 18.984212] lock((fb_notifier_list).rwsem);
[ 18.984213] lock(&dev->mode_config.mutex);
[ 18.984215] lock((fb_notifier_list).rwsem);
[ 18.984216] lock(&dev->mode_config.mutex);
[ 18.984216]
*** DEADLOCK ***
[ 18.984217] 6 locks held by kworker/u8:0/6:
[ 18.984220] #0: ("events_unbound"){.+.+.+}, at: [<ffffffff810a38c9>] process_one_work+0x169/0x750
[ 18.984223] #1: ((&entry->work)){+.+.+.}, at: [<ffffffff810a38c9>] process_one_work+0x169/0x750
[ 18.984227] #2: (registration_lock){+.+.+.}, at: [<ffffffff814c8487>] register_framebuffer+0x27/0x330
[ 18.984231] #3: (console_lock){+.+.+.}, at: [<ffffffff814c86ce>] register_framebuffer+0x26e/0x330
[ 18.984234] #4: (&fb_info->lock){+.+.+.}, at: [<ffffffff814c78dd>] lock_fb_info+0x1d/0x40
[ 18.984238] #5: ((fb_notifier_list).rwsem){++++.+}, at: [<ffffffff810ac195>] __blocking_notifier_call_chain+0x35/0x70
[ 18.984238]
stack backtrace:
[ 18.984241] CPU: 2 PID: 6 Comm: kworker/u8:0 Tainted: G O 4.7.0-rc5+ #524
[ 18.984242] Hardware name: Intel Corp. Broxton P/NOTEBOOK, BIOS APLKRVPA.X64.0138.B33.1606250842 06/25/2016
[ 18.984244] Workqueue: events_unbound async_run_entry_fn
[ 18.984248] 0000000000000000 ffff8800758577f0 ffffffff814507a5 ffffffff828b9900
[ 18.984250] ffffffff828b9900 ffff880075857830 ffffffff810dc6fa ffff880075857880
[ 18.984252] ffff88007584d688 0000000000000005 0000000000000006 ffff88007584d6b0
[ 18.984253] Call Trace:
[ 18.984257] [<ffffffff814507a5>] dump_stack+0x67/0x92
[ 18.984259] [<ffffffff810dc6fa>] print_circular_bug+0x1aa/0x200
[ 18.984261] [<ffffffff810df0ac>] __lock_acquire+0x10fc/0x1260
[ 18.984264] [<ffffffff810df611>] lock_acquire+0xb1/0x200
[ 18.984266] [<ffffffff815afde0>] ? drm_modeset_lock_all+0x40/0x120
[ 18.984267] [<ffffffff815afde0>] ? drm_modeset_lock_all+0x40/0x120
[ 18.984269] [<ffffffff819a3097>] mutex_lock_nested+0x67/0x3c0
[ 18.984271] [<ffffffff815afde0>] ? drm_modeset_lock_all+0x40/0x120
[ 18.984273] [<ffffffff810fa85f>] ? rcu_read_lock_sched_held+0x7f/0x90
[ 18.984276] [<ffffffff81208218>] ? kmem_cache_alloc_trace+0x248/0x2b0
[ 18.984277] [<ffffffff815afdc5>] ? drm_modeset_lock_all+0x25/0x120
[ 18.984279] [<ffffffff815afde0>] drm_modeset_lock_all+0x40/0x120
[ 18.984281] [<ffffffff8158f79b>] drm_fb_helper_restore_fbdev_mode_unlocked+0x2b/0x80
[ 18.984283] [<ffffffff8158f81d>] drm_fb_helper_set_par+0x2d/0x50
[ 18.984322] [<ffffffffa0105f7a>] intel_fbdev_set_par+0x1a/0x60 [i915]
[ 18.984324] [<ffffffff814c13c6>] fbcon_init+0x586/0x610
[ 18.984326] [<ffffffff8154d16a>] visual_init+0xca/0x130
[ 18.984328] [<ffffffff8154e611>] do_bind_con_driver+0x1c1/0x3a0
[ 18.984330] [<ffffffff8154eaf6>] do_take_over_console+0x116/0x180
[ 18.984332] [<ffffffff814bd3a7>] do_fbcon_takeover+0x57/0xb0
[ 18.984334] [<ffffffff814c1e48>] fbcon_event_notify+0x658/0x750
[ 18.984336] [<ffffffff810abcae>] notifier_call_chain+0x3e/0xb0
[ 18.984338] [<ffffffff810ac1ad>] __blocking_notifier_call_chain+0x4d/0x70
[ 18.984340] [<ffffffff810ac1e6>] blocking_notifier_call_chain+0x16/0x20
[ 18.984342] [<ffffffff814c748b>] fb_notifier_call_chain+0x1b/0x20
[ 18.984344] [<ffffffff814c86b1>] register_framebuffer+0x251/0x330
[ 18.984347] [<ffffffff815b7e8d>] ? vga_switcheroo_client_fb_set+0x5d/0x70
[ 18.984349] [<ffffffff8158fa9f>] drm_fb_helper_initial_config+0x25f/0x3f0
[ 18.984388] [<ffffffffa0106b48>] intel_fbdev_initial_config+0x18/0x30 [i915]
[ 18.984389] [<ffffffff810adfd8>] async_run_entry_fn+0x48/0x150
[ 18.984391] [<ffffffff810a3947>] process_one_work+0x1e7/0x750
[ 18.984392] [<ffffffff810a38c9>] ? process_one_work+0x169/0x750
[ 18.984394] [<ffffffff810a3efb>] worker_thread+0x4b/0x4f0
[ 18.984396] [<ffffffff810a3eb0>] ? process_one_work+0x750/0x750
[ 18.984398] [<ffffffff810aad4f>] kthread+0xef/0x110
[ 18.984401] [<ffffffff819a85ef>] ret_from_fork+0x1f/0x40
[ 18.984402] [<ffffffff810aac60>] ? kthread_stop+0x2e0/0x2e0
Reported-by: Imre Deak <redacted>
Signed-off-by: Chris Wilson <redacted>
Cc: Imre Deak <redacted>
Cc: Zhang Rui <rui.zhang@intel.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: "Christian König" <christian.koenig@amd.com>
Cc: David Airlie <redacted>
Cc: Patrik Jakobsson <patrik.r.jakobsson@gmail.com>
Cc: Daniel Vetter <redacted>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Ben Skeggs <redacted>
Cc: Tomi Valkeinen <redacted>
Cc: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Cc: "Bruno Prémont" <bonbons@linux-vserver.org>
Cc: Jiri Kosina <jikos@kernel.org>
Cc: Thomas Petazzoni <redacted>
Cc: "Noralf Trønnes" <redacted>
Cc: Jens Frederich <redacted>
Cc: Daniel Drake <redacted>
Cc: Jon Nettleton <redacted>
Cc: Jingoo Han <jingoohan1@gmail.com>
Cc: Lee Jones <redacted>
Cc: Jean-Christophe Plagniol-Villard <redacted>
Cc: Michael Hennerich <michael.hennerich@analog.com>
Cc: Milo Kim <redacted>
Cc: Thierry Reding <redacted>
Cc: Nicolas Ferre <redacted>
Cc: Paul Mackerras <redacted>
Cc: Kukjin Kim <kgene@kernel.org>
Cc: Krzysztof Kozlowski <redacted>
Cc: Antonino Daplas <adaplas@gmail.com>
Cc: linux-acpi@vger.kernel.org
Cc: dri-devel@lists.freedesktop.org
Cc: intel-gfx@lists.freedesktop.org
Cc: nouveau@lists.freedesktop.org
Cc: linux-renesas-soc@vger.kernel.org
Cc: linux-fbdev@vger.kernel.org
Cc: linux-omap@vger.kernel.org
---
drivers/acpi/acpi_video.c | 3 ++-
drivers/gpu/drm/amd/amdgpu/atombios_encoders.c | 3 ++-
drivers/gpu/drm/bridge/parade-ps8622.c | 3 ++-
drivers/gpu/drm/gma500/cdv_device.c | 3 ++-
drivers/gpu/drm/gma500/mdfld_device.c | 3 ++-
drivers/gpu/drm/gma500/oaktrail_device.c | 3 ++-
drivers/gpu/drm/gma500/psb_device.c | 3 ++-
drivers/gpu/drm/i915/intel_panel.c | 3 ++-
drivers/gpu/drm/nouveau/nouveau_backlight.c | 6 ++++--
drivers/gpu/drm/omapdrm/displays/panel-dsi-cm.c | 3 ++-
.../gpu/drm/omapdrm/displays/panel-sony-acx565akm.c | 3 ++-
drivers/gpu/drm/radeon/atombios_encoders.c | 3 ++-
drivers/gpu/drm/radeon/radeon_legacy_encoders.c | 3 ++-
drivers/gpu/drm/shmobile/shmob_drm_backlight.c | 3 ++-
drivers/hid/hid-picolcd_backlight.c | 3 ++-
drivers/macintosh/via-pmu-backlight.c | 3 ++-
drivers/platform/x86/acer-wmi.c | 3 ++-[...snip]
quoted hunk ↗ jump to hunk
diff --git a/drivers/platform/x86/acer-wmi.c b/drivers/platform/x86/acer-wmi.c index 79d64ea00bfb..0b17c5146175 100644 --- a/drivers/platform/x86/acer-wmi.c +++ b/drivers/platform/x86/acer-wmi.c@@ -1419,7 +1419,8 @@ static int acer_backlight_init(struct device *dev) props.type = BACKLIGHT_PLATFORM; props.max_brightness = max_brightness; bd = backlight_device_register("acer-wmi", dev, NULL, &acer_bl_ops, - &props); + &props, + BACKLIGHT_REGISTER_FB_CLIENT); if (IS_ERR(bd)) { pr_err("Could not register Acer backlight device\n"); acer_backlight_device = NULL;
For the part of acer-wmi driver. Review-by: Lee, Chun-Yi [off-list ref] Thanks a lot! Joey Lee