summaryrefslogtreecommitdiff
path: root/include/sound/atmel-ac97c.h
blob: f2a1cdc3766116d9f8ac4899e3b42c1392983e30 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
/*
 * Driver for the Atmel AC97C controller
 *
 * Copyright (C) 2005-2009 Atmel Corporation
 *
 * This program is free software; you can redistribute it and/or modify it
 * under the terms of the GNU General Public License version 2 as published
 * by the Free Software Foundation.
 */
#ifndef __INCLUDE_SOUND_ATMEL_AC97C_H
#define __INCLUDE_SOUND_ATMEL_AC97C_H

#include <linux/platform_data/dma-dw.h>

#define AC97C_CAPTURE	0x01
#define AC97C_PLAYBACK	0x02
#define AC97C_BOTH	(AC97C_CAPTURE | AC97C_PLAYBACK)

/**
 * struct atmel_ac97c_pdata - board specific AC97C configuration
 * @rx_dws: DMA slave interface to use for sound capture.
 * @tx_dws: DMA slave interface to use for sound playback.
 * @reset_pin: GPIO pin wired to the reset input on the external AC97 codec,
 *             optional to use, set to -ENODEV if not in use. AC97 layer will
 *             try to do a software reset of the external codec anyway.
 *
 * If the user do not want to use a DMA channel for playback or capture, i.e.
 * only one feature is required on the board. The slave for playback or capture
 * can be set to NULL. The AC97C driver will take use of this when setting up
 * the sound streams.
 */
struct ac97c_platform_data {
	struct dw_dma_slave	rx_dws;
	struct dw_dma_slave	tx_dws;
	int			reset_pin;
};

#endif /* __INCLUDE_SOUND_ATMEL_AC97C_H */
t() call in intel_unpin_fb_obj() returns NULL, resulting in an oops immediately afterwards as the (inlined) call to i915_vma_unpin_fence() tries to dereference it. It seems to be some race condition where the object is going away at shutdown time, since both times happened when shutting down the X server. The call chains were different: - VT ioctl(KDSETMODE, KD_TEXT): intel_cleanup_plane_fb+0x5b/0xa0 [i915] drm_atomic_helper_cleanup_planes+0x6f/0x90 [drm_kms_helper] intel_atomic_commit_tail+0x749/0xfe0 [i915] intel_atomic_commit+0x3cb/0x4f0 [i915] drm_atomic_commit+0x4b/0x50 [drm] restore_fbdev_mode+0x14c/0x2a0 [drm_kms_helper] drm_fb_helper_restore_fbdev_mode_unlocked+0x34/0x80 [drm_kms_helper] drm_fb_helper_set_par+0x2d/0x60 [drm_kms_helper] intel_fbdev_set_par+0x18/0x70 [i915] fb_set_var+0x236/0x460 fbcon_blank+0x30f/0x350 do_unblank_screen+0xd2/0x1a0 vt_ioctl+0x507/0x12a0 tty_ioctl+0x355/0xc30 do_vfs_ioctl+0xa3/0x5e0 SyS_ioctl+0x79/0x90 entry_SYSCALL_64_fastpath+0x13/0x94 - i915 unpin_work workqueue: intel_unpin_work_fn+0x58/0x140 [i915] process_one_work+0x1f1/0x480 worker_thread+0x48/0x4d0 kthread+0x101/0x140 and this patch purely papers over the issue by adding a NULL pointer check and a WARN_ON_ONCE() to avoid the oops that would then generally make the machine unresponsive. Other callers of i915_gem_object_to_ggtt() seem to also check for the returned pointer being NULL and warn about it, so this clearly has happened before in other places. [ Reported it originally to the i915 developers on Jan 8, applying the ugly workaround on my own now after triggering the problem for the second time with no feedback. This is likely to be the same bug reported as https://bugs.freedesktop.org/show_bug.cgi?id=98829 https://bugs.freedesktop.org/show_bug.cgi?id=99134 which has a patch for the underlying problem, but it hasn't gotten to me, so I'm applying the workaround. ] Cc: Daniel Vetter <daniel.vetter@intel.com> Cc: Jani Nikula <jani.nikula@linux.intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Imre Deak <imre.deak@intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'sound/soc/codecs/ts3a227e.c')