Quoting Manasi Navare (2017-10-26 22:52:00)
> During modeset cleanup on driver unload we may have a pending
> hotplug work. This needs to be canceled early during the teardown
> so that it does not fire after we have freed the connector.
> We do this after drm_kms_helper_poll_fini(dev) since this might
> trigger modeset retry work due to link retrain and before
> intel_fbdev_fini() since this work requires the lock from fbdev.
>
> If this is not done we may see something like:
> DEBUG_LOCKS_WARN_ON(mutex_is_locked(lock))
> ------------[ cut here ]------------
> WARNING: CPU: 4 PID: 5010 at kernel/locking/mutex-debug.c:103 mutex_destroy+0x4e/0x60
> Modules linked in: i915(-) snd_hda_codec_hdmi snd_hda_codec_realtek snd_hda_codec_generic snd_hda_codec snd_hwdep snd_hda_core snd_pcm vgem ax88179_178
> +a usbnet mii x86_pkg_temp_thermal intel_powerclamp coretemp crct10dif_pclmul crc32_pclmul ghash_clmulni_intel e1000e ptp pps_core prime_numbers i2c_hid
> +[last unloaded: snd_hda_intel]
> CPU: 4 PID: 5010 Comm: drv_module_relo Tainted: G U 4.14.0-rc3-CI-CI_DRM_3186+ #1
> Hardware name: Intel Corporation CoffeeLake Client Platform/CoffeeLake S UDIMM RVP, BIOS CNLSFWX1.R00.X104.A03.1709140524 09/14/2017
> task: ffff8803c827aa40 task.stack: ffffc90000520000
> RIP: 0010:mutex_destroy+0x4e/0x60
> RSP: 0018:ffffc90000523d58 EFLAGS: 00010292
> RAX: 000000000000002a RBX: ffff88044fbef648 RCX: 0000000000000000
> RDX: 0000000080000001 RSI: 0000000000000001 RDI: ffffffff810f0cf0
> RBP: ffffc90000523d60 R08: 0000000000000001 R09: 0000000000000001
> R10: 000000000f21cb81 R11: 0000000000000000 R12: ffff88044f71efc8
> R13: ffffffffa02b3d20 R14: ffffffffa02b3d90 R15: ffff880459b29308
> FS: 00007f5df4d6e8c0(0000) GS:ffff88045d300000(0000) knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 000055ec51f00a18 CR3: 0000000451782006 CR4: 00000000003606e0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> Call Trace:
> drm_fb_helper_fini+0xd9/0x130
> intel_fbdev_destroy+0x12/0x60 [i915]
> intel_fbdev_fini+0x28/0x30 [i915]
> intel_modeset_cleanup+0x45/0xa0 [i915]
> i915_driver_unload+0x92/0x180 [i915]
> i915_pci_remove+0x19/0x30 [i915]
> i915_driver_unload+0x92/0x180 [i915]
> i915_pci_remove+0x19/0x30 [i915]
> pci_device_remove+0x39/0xb0
> device_release_driver_internal+0x15d/0x220
> driver_detach+0x40/0x80
> bus_remove_driver+0x58/0xd0
> driver_unregister+0x2c/0x40
> pci_unregister_driver+0x36/0xb0
> i915_exit+0x1a/0x8b [i915]
> SyS_delete_module+0x18c/0x1e0
> entry_SYSCALL_64_fastpath+0x1c/0xb1
> RIP: 0033:0x7f5df3286287
> RSP: 002b:00007fff8e107cc8 EFLAGS: 00000246 ORIG_RAX: 00000000000000b0
> RAX: ffffffffffffffda RBX: ffffffff81493a03 RCX: 00007f5df3286287
> RDX: 0000000000000001 RSI: 0000000000000800 RDI: 0000564c7be02e48
> RBP: ffffc90000523f88 R08: 0000000000000000 R09: 0000000000000080
> R10: 00007f5df4d6e8c0 R11: 0000000000000246 R12: 0000000000000000
> R13: 00007fff8e107eb0 R14: 0000000000000000 R15: 0000000000000000
> Or a GPF like:
>
> general protection fault: 0000 [#1] PREEMPT SMP
> Modules linked in: i915(-) snd_hda_codec_hdmi snd_hda_codec_realtek snd_hda_codec_generic snd_hda_codec snd_hwdep snd_hda_core snd_pcm vgem ax88179_178
> +a usbnet mii x86_pkg_temp_thermal intel_powerclamp coretemp crct10dif_pclmul crc32_pclmul ghash_clmulni_intel e1000e ptp pps_core prime_numbers i2c_hid
> +[last unloaded: snd_hda_intel]
> CPU: 0 PID: 82 Comm: kworker/0:1 Tainted: G U W 4.14.0-rc3-CI-CI_DRM_3186+ #1
> Hardware name: Intel Corporation CoffeeLake Client Platform/CoffeeLake S UDIMM RVP, BIOS CNLSFWX1.R00.X104.A03.1709140524 09/14/2017
> Workqueue: events intel_dp_modeset_retry_work_fn [i915]
> task: ffff88045a5caa40 task.stack: ffffc90000378000
> RIP: 0010:drm_setup_crtcs+0x143/0xbf0
> RSP: 0018:ffffc9000037bd20 EFLAGS: 00010202
> RAX: 6b6b6b6b6b6b6b6b RBX: 0000000000000002 RCX: 0000000000000001
> RDX: 0000000000000001 RSI: 0000000000000780 RDI: 00000000ffffffff
> RBP: ffffc9000037bdb8 R08: 0000000000000001 R09: 0000000000000001
> R10: 0000000000000780 R11: 0000000000000000 R12: 0000000000000002
> R13: ffff88044fbef4e8 R14: 0000000000000780 R15: 0000000000000438
> FS: 0000000000000000(0000) GS:ffff88045d200000(0000) knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 000055ec51ee5168 CR3: 000000044c89d003 CR4: 00000000003606f0
> Call Trace:
> drm_fb_helper_hotplug_event.part.18+0x7e/0xc0
> drm_fb_helper_hotplug_event+0x1a/0x20
> intel_fbdev_output_poll_changed+0x1a/0x20 [i915]
> drm_kms_helper_hotplug_event+0x27/0x30
> intel_dp_modeset_retry_work_fn+0x77/0x80 [i915]
> process_one_work+0x233/0x660
> worker_thread+0x206/0x3b0
> kthread+0x152/0x190
> ? process_one_work+0x660/0x660
> ? kthread_create_on_node+0x40/0x40
> ret_from_fork+0x27/0x40
> Code: 06 00 00 45 8b 45 20 31 db 45 31 e4 45 85 c0 0f 8e 91 06 00 00 44 8b 75 94 44 8b 7d 90 49 8b 45 28 49 63 d4 44 89 f6 41 83 c4 01 <48> 8b 04 d0 44
> +89 fa 48 8b 38 48 8b 87 a8 01 00 00 ff 50 20 01
> RIP: drm_setup_crtcs+0x143/0xbf0 RSP: ffffc9000037bd20
> ---[ end trace 08901ff1a77d30c7 ]---
>
> v2:
> * Rename it to intel_hpd_poll_fini() and call drm_kms_helper_fini() inside it
> as the first step before cancel work (Chris Wilson)
> * Add GPF trace in commit message and make the function static (Maarten Lankhorst)
>
> Suggested-by: Maarten Lankhorst <maarten.lankhorst@xxxxxxxxxxxxxxx>
> Suggested-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
> Fixes: 9301397a63b3 ("drm/i915: Implement Link Rate fallback on Link training failure")
> Cc: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
> Cc: Tony Cheng <tony.cheng@xxxxxxx>
> Cc: Harry Wentland <Harry.wentland@xxxxxxx>
> Cc: Jani Nikula <jani.nikula@xxxxxxxxxxxxxxx>
> Cc: Daniel Vetter <daniel.vetter@xxxxxxxxx>
> Cc: Ville Syrjala <ville.syrjala@xxxxxxxxxxxxxxx>
> Cc: Manasi Navare <manasi.d.navare@xxxxxxxxx>
> Cc: Maarten Lankhorst <maarten.lankhorst@xxxxxxxxxxxxxxx>
> Signed-off-by: Manasi Navare <manasi.d.navare@xxxxxxxxx>
> ---
> drivers/gpu/drm/i915/intel_display.c | 21 ++++++++++++++++++++-
> 1 file changed, 20 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
> index 0e493a1..1ffe7a4 100644
> --- a/drivers/gpu/drm/i915/intel_display.c
> +++ b/drivers/gpu/drm/i915/intel_display.c
> @@ -15200,6 +15200,23 @@ void intel_connector_unregister(struct drm_connector *connector)
> intel_panel_destroy_backlight(connector);
> }
>
> +static void intel_hpd_poll_fini(struct drm_device *dev)
> +{
> + struct intel_connector *connector;
> + struct drm_connector_list_iter conn_iter;
> +
> + /* Disable Polling */
> + drm_kms_helper_poll_fini(dev);
> +
> + /* kill the modeset retry work after killing the polling */
> + drm_connector_list_iter_begin(dev, &conn_iter);
> + for_each_intel_connector_iter(connector, &conn_iter) {
> + if (connector->modeset_retry_work.func)
> + cancel_work_sync(&connector->modeset_retry_work);
> + }
> + drm_connector_list_iter_end(&conn_iter);
> +}
Reviewed-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
