* [PATCH] drm/i915/lvds: Only act on lid notify when the device is on
@ 2011-04-21 22:08 Alex Williamson
2011-04-22 5:58 ` Chris Wilson
2011-04-24 10:05 ` Chris Wilson
0 siblings, 2 replies; 7+ messages in thread
From: Alex Williamson @ 2011-04-21 22:08 UTC (permalink / raw)
To: chris; +Cc: dri-devel
If we're using vga switcheroo, the device may be turned off
and poking at it can cause an oops.
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
---
drivers/gpu/drm/i915/intel_lvds.c | 3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index a562bd2..67cb076 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -539,6 +539,9 @@ static int intel_lid_notify(struct notifier_block *nb, unsigned long val,
struct drm_device *dev = dev_priv->dev;
struct drm_connector *connector = dev_priv->int_lvds_connector;
+ if (dev->switch_power_state != DRM_SWITCH_POWER_ON)
+ return NOTIFY_OK;
+
/*
* check and update the status of LVDS connector after receiving
* the LID nofication event.
^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [PATCH] drm/i915/lvds: Only act on lid notify when the device is on
2011-04-21 22:08 [PATCH] drm/i915/lvds: Only act on lid notify when the device is on Alex Williamson
@ 2011-04-22 5:58 ` Chris Wilson
2011-04-22 6:13 ` Dave Airlie
2011-04-23 14:42 ` Alex Williamson
2011-04-24 10:05 ` Chris Wilson
1 sibling, 2 replies; 7+ messages in thread
From: Chris Wilson @ 2011-04-22 5:58 UTC (permalink / raw)
To: Alex Williamson; +Cc: dri-devel
On Thu, 21 Apr 2011 16:08:14 -0600, Alex Williamson <alex.williamson@redhat.com> wrote:
> If we're using vga switcheroo, the device may be turned off
> and poking at it can cause an oops.
I'm unfamiliar with what actually happens here after it is turned off. Can
you post an OOPS for an example?
-Chris
--
Chris Wilson, Intel Open Source Technology Centre
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] drm/i915/lvds: Only act on lid notify when the device is on
2011-04-22 5:58 ` Chris Wilson
@ 2011-04-22 6:13 ` Dave Airlie
2011-04-22 6:26 ` Chris Wilson
2011-04-23 14:42 ` Alex Williamson
1 sibling, 1 reply; 7+ messages in thread
From: Dave Airlie @ 2011-04-22 6:13 UTC (permalink / raw)
To: Chris Wilson; +Cc: dri-devel
On Fri, Apr 22, 2011 at 3:58 PM, Chris Wilson <chris@chris-wilson.co.uk> wrote:
> On Thu, 21 Apr 2011 16:08:14 -0600, Alex Williamson <alex.williamson@redhat.com> wrote:
>> If we're using vga switcheroo, the device may be turned off
>> and poking at it can cause an oops.
>
> I'm unfamiliar with what actually happens here after it is turned off. Can
> you post an OOPS for an example?
The device will be in D3, so I'm guessing we read back some register
and get 0xFFFFFFFF,
and later fail.
Dave.
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] drm/i915/lvds: Only act on lid notify when the device is on
2011-04-22 6:13 ` Dave Airlie
@ 2011-04-22 6:26 ` Chris Wilson
0 siblings, 0 replies; 7+ messages in thread
From: Chris Wilson @ 2011-04-22 6:26 UTC (permalink / raw)
To: Dave Airlie; +Cc: dri-devel
On Fri, 22 Apr 2011 16:13:27 +1000, Dave Airlie <airlied@gmail.com> wrote:
> On Fri, Apr 22, 2011 at 3:58 PM, Chris Wilson <chris@chris-wilson.co.uk> wrote:
> > On Thu, 21 Apr 2011 16:08:14 -0600, Alex Williamson <alex.williamson@redhat.com> wrote:
> >> If we're using vga switcheroo, the device may be turned off
> >> and poking at it can cause an oops.
> >
> > I'm unfamiliar with what actually happens here after it is turned off. Can
> > you post an OOPS for an example?
>
> The device will be in D3, so I'm guessing we read back some register
> and get 0xFFFFFFFF,
> and later fail.
That's sounds reasonable. A bit of explanation does wonders if I (or
anyone else reading through the history *or* code) ever encounter
similar bug reports.
Is anyone else concerned by the acpi_lid_notifier never being decoupled
when the module is unloaded? And the randomness of int_lvds_connector?
I love our code, every time you read it you find another bug.
-Chris
--
Chris Wilson, Intel Open Source Technology Centre
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] drm/i915/lvds: Only act on lid notify when the device is on
2011-04-22 5:58 ` Chris Wilson
2011-04-22 6:13 ` Dave Airlie
@ 2011-04-23 14:42 ` Alex Williamson
1 sibling, 0 replies; 7+ messages in thread
From: Alex Williamson @ 2011-04-23 14:42 UTC (permalink / raw)
To: Chris Wilson; +Cc: dri-devel
On Fri, 2011-04-22 at 06:58 +0100, Chris Wilson wrote:
> On Thu, 21 Apr 2011 16:08:14 -0600, Alex Williamson <alex.williamson@redhat.com> wrote:
> > If we're using vga switcheroo, the device may be turned off
> > and poking at it can cause an oops.
>
> I'm unfamiliar with what actually happens here after it is turned off. Can
> you post an OOPS for an example?
See below. Thanks,
Alex
[ 941.024073] BUG: unable to handle kernel NULL pointer dereference at 0000000000000230
[ 941.024598] IP: [<ffffffffa0098415>] intel_dp_link_down+0x133/0x16f [i915]
[ 941.025026] PGD 0
[ 941.025026] Oops: 0000 [#1] SMP
[ 941.025026] last sysfs file: /sys/devices/system/cpu/cpu1/cache/index2/shared_cpu_map
[ 941.025026] CPU 0
[ 941.025026] Modules linked in: netconsole configfs ftdi_sio usbserial ebtable_nat ebtables ipt_MASQUERADE iptable_nat nf_nat bridge stp llc sunrpc cpufreq_ondemand acpi_cpufreq freq_table mperf ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 ip6table_filter ip6_tables ipv6 kvm_intel kvm uinput nouveau asus_switcheroo ttm snd_hda_codec_hdmi arc4 snd_hda_codec_realtek ecb snd_hda_intel snd_hda_codec iwlagn mac80211 snd_hwdep snd_seq snd_seq_device snd_pcm snd_timer snd soundcore cfg80211 uvcvideo videodev btusb bluetooth v4l2_compat_ioctl32 asus_laptop pcspkr serio_raw sparse_keymap joydev atl1c rfkill iTCO_wdt iTCO_vendor_support snd_page_alloc wmi microcode i915 drm_kms_helper drm i2c_algo_bit i2c_core video [last unloaded: configfs]
[ 941.025026]
[ 941.025026] Pid: 1907, comm: kworker/0:2 Tainted: G W 2.6.39-rc4+ #1 ASUSTeK Computer Inc. UL30VT /UL30VT
[ 941.025026] RIP: 0010:[<ffffffffa0098415>] [<ffffffffa0098415>] intel_dp_link_down+0x133/0x16f [i915]
[ 941.025026] RSP: 0018:ffff880136de1990 EFLAGS: 00010286
[ 941.025026] RAX: 00000000ffffffff RBX: ffff880135dabc00 RCX: 0000000000002052
[ 941.025026] RDX: 0000000000000000 RSI: ffff880136de1980 RDI: 0000000000000001
[ 941.025026] RBP: ffff880136de19c0 R08: ffff880135db3800 R09: 0000000000000000
[ 941.025026] R10: ffffffff81ca5ee0 R11: 0000000000000040 R12: ffff880133a10000
[ 941.025026] R13: 0000000000000000 R14: ffff880135db3800 R15: 0000000000000000
[ 941.025026] FS: 0000000000000000(0000) GS:ffff88013fc00000(0000) knlGS:0000000000000000
[ 941.025026] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[ 941.025026] CR2: 0000000000000230 CR3: 0000000001a03000 CR4: 00000000000406f0
[ 941.025026] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 941.025026] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[ 941.025026] Process kworker/0:2 (pid: 1907, threadinfo ffff880136de0000, task ffff88013646ae00)
[ 941.025026] Stack:
[ 941.025026] 0000000000000003 ffff880135dabc00 ffff880135db3800 0000000000000003
[ 941.025026] 00000000ffffffff ffff880135e0b030 ffff880136de19f0 ffffffffa0098a9b
[ 941.025026] ffff880135e0b000 ffff880135db3800 ffff880135dabc00 ffff880135db3c01
[ 941.025026] Call Trace:
[ 941.025026] [<ffffffffa0098a9b>] intel_dp_dpms+0x4b/0xc6 [i915]
[ 941.025026] [<ffffffffa0066271>] drm_encoder_disable+0x21/0x23 [drm_kms_helper]
[ 941.025026] [<ffffffffa0066ab7>] drm_crtc_helper_set_mode+0x1db/0x3ea [drm_kms_helper]
[ 941.025026] [<ffffffff8107b952>] ? arch_local_irq_save+0x18/0x1e
[ 941.025026] [<ffffffffa0066d23>] drm_helper_resume_force_mode+0x5d/0x122 [drm_kms_helper]
[ 941.025026] [<ffffffff8147510e>] ? _cond_resched+0xe/0x22
[ 941.025026] [<ffffffffa009525d>] intel_lid_notify+0x8d/0xa2 [i915]
[ 941.025026] [<ffffffff8147993b>] notifier_call_chain+0x37/0x63
[ 941.025026] [<ffffffff8106dc6d>] __blocking_notifier_call_chain+0x4b/0x60
[ 941.025026] [<ffffffff8106dc96>] blocking_notifier_call_chain+0x14/0x16
[ 941.025026] [<ffffffff81298be8>] acpi_lid_send_state+0x86/0xaf
[ 941.025026] [<ffffffff81298c6a>] acpi_button_notify+0x3b/0xab
[ 941.025026] [<ffffffff81272951>] ? acpi_os_wait_events_complete+0x23/0x23
[ 941.025026] [<ffffffff8127594d>] acpi_device_notify+0x19/0x1b
[ 941.025026] [<ffffffff81281e4f>] acpi_ev_notify_dispatch+0x67/0x7e
[ 941.025026] [<ffffffff8127297a>] acpi_os_execute_deferred+0x29/0x36
[ 941.025026] [<ffffffff8106482f>] process_one_work+0x18f/0x28a
[ 941.025026] [<ffffffff81065a9e>] worker_thread+0xfd/0x181
[ 941.025026] [<ffffffff810659a1>] ? manage_workers.clone.16+0x172/0x172
[ 941.025026] [<ffffffff81069076>] kthread+0x82/0x8a
[ 941.025026] [<ffffffff8147dee4>] kernel_thread_helper+0x4/0x10
[ 941.025026] [<ffffffff81068ff4>] ? kthread_worker_fn+0x14b/0x14b
[ 941.025026] [<ffffffff8147dee0>] ? gs_change+0x13/0x13
[ 941.025026] Code: 8b 7c 24 18 e8 3c f3 ff ff a9 00 00 00 40 74 2a 4c 8b 7b 30 8b 73 60 49 8d 7c 24 18 41 81 e5 ff ff ff bf 44 89 ea e8 e1 f2 ff ff
[ 941.025026] 8b b7 30 02 00 00 4c 89 f7 e8 b7 42 ff ff 8b 73 60 44 89 ea
[ 941.025026] RIP [<ffffffffa0098415>] intel_dp_link_down+0x133/0x16f [i915]
[ 941.025026] RSP <ffff880136de1990>
[ 941.025026] CR2: 0000000000000230
[ 941.064863] ---[ end trace 704c59211af95870 ]---
[ 941.065579] BUG: unable to handle kernel paging request at fffffffffffffff8
[ 941.066242] IP: [<ffffffff810692d1>] kthread_data+0x10/0x16
[ 941.066251] PGD 1a05067 PUD 1a06067 PMD 0
[ 941.066257] Oops: 0000 [#2] SMP
[ 941.066261] last sysfs file: /sys/devices/system/cpu/cpu1/cache/index2/shared_cpu_map
[ 941.066265] CPU 0
[ 941.066267] Modules linked in: netconsole configfs ftdi_sio usbserial ebtable_nat ebtables ipt_MASQUERADE iptable_nat nf_nat bridge stp llc sunrpc cpufreq_ondemand acpi_cpufreq freq_table mperf ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 ip6table_filter ip6_tables ipv6 kvm_intel kvm uinput nouveau asus_switcheroo ttm snd_hda_codec_hdmi arc4 snd_hda_codec_realtek ecb snd_hda_intel snd_hda_codec iwlagn mac80211 snd_hwdep snd_seq snd_seq_device snd_pcm snd_timer snd soundcore cfg80211 uvcvideo videodev btusb bluetooth v4l2_compat_ioctl32 asus_laptop pcspkr serio_raw sparse_keymap joydev atl1c rfkill iTCO_wdt iTCO_vendor_support snd_page_alloc wmi microcode i915 drm_kms_helper drm i2c_algo_bit i2c_core video [last unloaded: configfs]
[ 941.066330]
[ 941.066335] Pid: 1907, comm: kworker/0:2 Tainted: G D W 2.6.39-rc4+ #1 ASUSTeK Computer Inc. UL30VT /UL30VT
[ 941.066342] RIP: 0010:[<ffffffff810692d1>] [<ffffffff810692d1>] kthread_data+0x10/0x16
[ 941.066350] RSP: 0018:ffff880136de1568 EFLAGS: 00010096
[ 941.066354] RAX: 0000000000000000 RBX: ffff88013fc13a80 RCX: ffff88013646ae00
[ 941.066358] RDX: 0000000000008eb0 RSI: 0000000000000000 RDI: ffff88013646ae00
[ 941.066362] RBP: ffff880136de1568 R08: ffffffff8104c5ab R09: 00000000fffffff8
[ 941.066367] R10: ffffffff81a29580 R11: ffff88013646ae00 R12: 0000000000000000
[ 941.066371] R13: ffff880136de1698 R14: 0000000000000000 R15: ffff88013646b1b8
[ 941.066376] FS: 0000000000000000(0000) GS:ffff88013fc00000(0000) knlGS:0000000000000000
[ 941.066380] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[ 941.066384] CR2: fffffffffffffff8 CR3: 0000000001a03000 CR4: 00000000000406f0
[ 941.066388] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 941.066393] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[ 941.066398] Process kworker/0:2 (pid: 1907, threadinfo ffff880136de0000, task ffff88013646ae00)
[ 941.066401] Stack:
[ 941.066403] ffff880136de1598 ffffffff81065def ffffffff81a29580 ffff88013fc13a80
[ 941.066409] 0000000000000000 ffff880136de1698 ffff880136de1658 ffffffff81474a02
[ 941.066415] ffffffff81a29580 0000000000000246 0000000000013a80 0000000000013a80
[ 941.066420] Call Trace:
[ 941.066428] [<ffffffff81065def>] wq_worker_sleeping+0x1a/0x83
[ 941.066434] [<ffffffff81474a02>] schedule+0x199/0x6b6
[ 941.066441] [<ffffffff8106d614>] ? switch_task_namespaces+0x4b/0x64
[ 941.066448] [<ffffffff81051fc1>] do_exit+0x714/0x716
[ 941.066455] [<ffffffff814777e7>] oops_end+0xbf/0xc7
[ 941.066461] [<ffffffff8103253c>] no_context+0x1f9/0x208
[ 941.066467] [<ffffffff810326dd>] __bad_area_nosemaphore+0x192/0x1b5
[ 941.066473] [<ffffffff81041b95>] ? update_stats_wait_end.clone.74+0x6a/0xac
[ 941.066479] [<ffffffff81032713>] bad_area_nosemaphore+0x13/0x15
[ 941.066484] [<ffffffff81479731>] do_page_fault+0x187/0x35a
[ 941.066489] [<ffffffff81040ea8>] ? mmdrop+0x1a/0x2a
[ 941.066495] [<ffffffff8104130c>] ? finish_task_switch+0x89/0xa8
[ 941.066502] [<ffffffff8103cc9e>] ? need_resched+0x23/0x2d
[ 941.066504] [<ffffffff8107b952>] ? arch_local_irq_save+0x18/0x1e
[ 941.066504] [<ffffffff814765d2>] ? _raw_spin_lock_irqsave+0x12/0x2f
[ 941.066504] [<ffffffff8105b388>] ? lock_timer_base.clone.20+0x2b/0x50
[ 941.066504] [<ffffffff81476627>] ? _raw_spin_unlock_irqrestore+0x17/0x19
[ 941.066504] [<ffffffff8105b424>] ? try_to_del_timer_sync+0x77/0x85
[ 941.066504] [<ffffffff81476cd5>] page_fault+0x25/0x30
[ 941.066504] [<ffffffffa0098415>] ? intel_dp_link_down+0x133/0x16f [i915]
[ 941.066504] [<ffffffffa0098a9b>] intel_dp_dpms+0x4b/0xc6 [i915]
[ 941.066504] [<ffffffffa0066271>] drm_encoder_disable+0x21/0x23 [drm_kms_helper]
[ 941.066504] [<ffffffffa0066ab7>] drm_crtc_helper_set_mode+0x1db/0x3ea [drm_kms_helper]
[ 941.066504] [<ffffffff8107b952>] ? arch_local_irq_save+0x18/0x1e
[ 941.066504] [<ffffffffa0066d23>] drm_helper_resume_force_mode+0x5d/0x122 [drm_kms_helper]
[ 941.066504] [<ffffffff8147510e>] ? _cond_resched+0xe/0x22
[ 941.066504] [<ffffffffa009525d>] intel_lid_notify+0x8d/0xa2 [i915]
[ 941.066504] [<ffffffff8147993b>] notifier_call_chain+0x37/0x63
[ 941.066504] [<ffffffff8106dc6d>] __blocking_notifier_call_chain+0x4b/0x60
[ 941.066504] [<ffffffff8106dc96>] blocking_notifier_call_chain+0x14/0x16
[ 941.066504] [<ffffffff81298be8>] acpi_lid_send_state+0x86/0xaf
[ 941.066504] [<ffffffff81298c6a>] acpi_button_notify+0x3b/0xab
[ 941.066504] [<ffffffff81272951>] ? acpi_os_wait_events_complete+0x23/0x23
[ 941.066504] [<ffffffff8127594d>] acpi_device_notify+0x19/0x1b
[ 941.066504] [<ffffffff81281e4f>] acpi_ev_notify_dispatch+0x67/0x7e
[ 941.066504] [<ffffffff8127297a>] acpi_os_execute_deferred+0x29/0x36
[ 941.066504] [<ffffffff8106482f>] process_one_work+0x18f/0x28a
[ 941.066504] [<ffffffff81065a9e>] worker_thread+0xfd/0x181
[ 941.066504] [<ffffffff810659a1>] ? manage_workers.clone.16+0x172/0x172
[ 941.066504] [<ffffffff81069076>] kthread+0x82/0x8a
[ 941.066504] [<ffffffff8147dee4>] kernel_thread_helper+0x4/0x10
[ 941.066504] [<ffffffff81068ff4>] ? kthread_worker_fn+0x14b/0x14b
[ 941.066504] [<ffffffff8147dee0>] ? gs_change+0x13/0x13
[ 941.066504] Code: 5c 60 81 e8 0a 7c fd ff 48 8b 85 68 ff ff ff 48 81 c4 a0 00 00 00 5b 41 5c c9 c3 55 48 89 e5 0f 1f 44 00 00 48 8b 87 60 03 00 00
[ 941.066504] 8b 40 f8 c9 c3 55 48 89 e5 0f 1f 44 00 00 48 3b 3d 19 e5 c3
[ 941.066504] RIP [<ffffffff810692d1>] kthread_data+0x10/0x16
[ 941.066504] RSP <ffff880136de1568>
[ 941.066504] CR2: fffffffffffffff8
[ 941.066504] ---[ end trace 704c59211af95871 ]---
[ 941.066504] Fixing recursive fault but reboot is needed!
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] drm/i915/lvds: Only act on lid notify when the device is on
2011-04-21 22:08 [PATCH] drm/i915/lvds: Only act on lid notify when the device is on Alex Williamson
2011-04-22 5:58 ` Chris Wilson
@ 2011-04-24 10:05 ` Chris Wilson
2011-04-24 13:26 ` Alex Williamson
1 sibling, 1 reply; 7+ messages in thread
From: Chris Wilson @ 2011-04-24 10:05 UTC (permalink / raw)
To: Alex Williamson; +Cc: dri-devel
On Thu, 21 Apr 2011 16:08:14 -0600, Alex Williamson <alex.williamson@redhat.com> wrote:
> If we're using vga switcheroo, the device may be turned off
> and poking at it can cause an oops.
I added just a couple more details to the changelog and added to my -fixes
queue.
The question for the audience is: what is the difference between using the
PCI power state and the drm->power_switch_state?
Thanks,
-Chris
--
Chris Wilson, Intel Open Source Technology Centre
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] drm/i915/lvds: Only act on lid notify when the device is on
2011-04-24 10:05 ` Chris Wilson
@ 2011-04-24 13:26 ` Alex Williamson
0 siblings, 0 replies; 7+ messages in thread
From: Alex Williamson @ 2011-04-24 13:26 UTC (permalink / raw)
To: Chris Wilson; +Cc: dri-devel
On Sun, 2011-04-24 at 11:05 +0100, Chris Wilson wrote:
> On Thu, 21 Apr 2011 16:08:14 -0600, Alex Williamson <alex.williamson@redhat.com> wrote:
> > If we're using vga switcheroo, the device may be turned off
> > and poking at it can cause an oops.
>
> I added just a couple more details to the changelog and added to my -fixes
> queue.
>
> The question for the audience is: what is the difference between using the
> PCI power state and the drm->power_switch_state?
Seems possible we could use either. I picked the latter because
i915_switcheroo_set_state() seems to be keeping it in sync with power
state and has the "changing" state, which might help reduce races.
Thanks,
Alex
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2011-04-24 13:26 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-04-21 22:08 [PATCH] drm/i915/lvds: Only act on lid notify when the device is on Alex Williamson
2011-04-22 5:58 ` Chris Wilson
2011-04-22 6:13 ` Dave Airlie
2011-04-22 6:26 ` Chris Wilson
2011-04-23 14:42 ` Alex Williamson
2011-04-24 10:05 ` Chris Wilson
2011-04-24 13:26 ` Alex Williamson
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.