From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andrzej Hajda Subject: Re: [BUG] blocked task after exynos_drm_init Date: Tue, 18 Nov 2014 12:04:20 +0100 Message-ID: <546B27B4.4040808@samsung.com> References: <1415264811.27478.4.camel@AMDC1943> <546B228E.6030500@samsung.com> <546B2508.8070409@samsung.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from mailout4.w1.samsung.com ([210.118.77.14]:13509 "EHLO mailout4.w1.samsung.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752006AbaKRLEZ (ORCPT ); Tue, 18 Nov 2014 06:04:25 -0500 In-reply-to: <546B2508.8070409@samsung.com> Sender: linux-samsung-soc-owner@vger.kernel.org List-Id: linux-samsung-soc@vger.kernel.org To: Inki Dae Cc: linux-samsung-soc@vger.kernel.org, linux-arm-kernel@lists.infradead.org, dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org On 11/18/2014 11:52 AM, Inki Dae wrote: > On 2014=EB=85=84 11=EC=9B=94 18=EC=9D=BC 19:42, Andrzej Hajda wrote: >> On 11/06/2014 10:06 AM, Krzysztof Kozlowski wrote: >>> Hi, >>> >>> On last next (next-20141104, next-20141105) booting locks after >>> initializing Exynos DRM (Trats2 board): >>> >>> [ 2.028283] [drm] Initialized drm 1.1.0 20060810 >>> [ 240.505795] INFO: task swapper/0:1 blocked for more than 120 sec= onds. >>> [ 240.510825] Not tainted 3.18.0-rc3-next-20141105 #794 >>> [ 240.516418] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" d= isables this message. >>> [ 240.524173] swapper/0 D c052534c 0 1 0 0x0000= 0000 >>> [ 240.530527] [] (__schedule) from [] (schedul= e_preempt_disabled+0x14/0x20) >>> [ 240.539030] [] (schedule_preempt_disabled) from [] (mutex_lock_nested+0x1c4/0x464) >>> [ 240.548320] [] (mutex_lock_nested) from [] (= __driver_attach+0x48/0x98) >>> [ 240.556562] [] (__driver_attach) from [] (bu= s_for_each_dev+0x54/0x88) >>> [ 240.564717] [] (bus_for_each_dev) from [] (b= us_add_driver+0xe4/0x200) >>> [ 240.572876] [] (bus_add_driver) from [] (dri= ver_register+0x78/0xf4) >>> [ 240.580864] [] (driver_register) from [] (ex= ynos_drm_platform_probe+0x34/0x234) >>> [ 240.589890] [] (exynos_drm_platform_probe) from [] (platform_drv_probe+0x48/0xa4) >>> [ 240.599090] [] (platform_drv_probe) from [] = (driver_probe_device+0x13c/0x37c) >>> [ 240.607940] [] (driver_probe_device) from []= (__driver_attach+0x94/0x98) >>> [ 240.616360] [] (__driver_attach) from [] (bu= s_for_each_dev+0x54/0x88) >>> [ 240.624517] [] (bus_for_each_dev) from [] (b= us_add_driver+0xe4/0x200) >>> [ 240.632679] [] (bus_add_driver) from [] (dri= ver_register+0x78/0xf4) >>> [ 240.640667] [] (driver_register) from [] (ex= ynos_drm_init+0x70/0xa0) >>> [ 240.648739] [] (exynos_drm_init) from [] (do= _one_initcall+0xac/0x1f0) >>> [ 240.656914] [] (do_one_initcall) from [] (ke= rnel_init_freeable+0x10c/0x1d8) >>> [ 240.665591] [] (kernel_init_freeable) from [= ] (kernel_init+0x8/0xec) >>> [ 240.673661] [] (kernel_init) from [] (ret_fr= om_fork+0x14/0x2c) >>> [ 240.681196] 3 locks held by swapper/0/1: >>> [ 240.685091] #0: (&dev->mutex){......}, at: [] __driv= er_attach+0x48/0x98 >>> [ 240.692732] #1: (&dev->mutex){......}, at: [] __driv= er_attach+0x58/0x98 >>> [ 240.700367] #2: (&dev->mutex){......}, at: [] __driv= er_attach+0x48/0x98 >> >> >> This is caused by patch moving platform devices to >> /sys/devices/platform[1]. Since this patch registering platform >> drivers/devices in probe of platform device causes deadlocks. I gues= s >> now all driver registration should be moved to exynos_drm_init and i= t >> seems better location for it IMHO. >=20 > Thanks. It might be a chance that we could separate sub drivers of > Exynos drm into independent modules so that they can be called > independently because if we move them to exynos_drm_init then the > deferred probe wouldn't work correctly. =46or full separation of sub-drivers component matching code should be fixed first. Now it usually works because components are probed before exynos_drm_match_add and this is because components are probed during sub-driver registrations. Regards Andrzej >=20 > Thanks, > Inki Dae >=20 >> >> Regards >> Andrzej >> >> [1]: http://www.spinics.net/lists/devicetree/msg56101.html >> >> >> >>> >>> Full dmesg and config attached. >>> >>> Best regards, >>> Krzysztof >>> >>> >>> >>> _______________________________________________ >>> dri-devel mailing list >>> dri-devel@lists.freedesktop.org >>> http://lists.freedesktop.org/mailman/listinfo/dri-devel >>> >> >> >> -- >> To unsubscribe from this list: send the line "unsubscribe linux-sams= ung-soc" in >> the body of a message to majordomo@vger.kernel.org >> More majordomo info at http://vger.kernel.org/majordomo-info.html >> >=20 > _______________________________________________ > dri-devel mailing list > dri-devel@lists.freedesktop.org > http://lists.freedesktop.org/mailman/listinfo/dri-devel >=20 From mboxrd@z Thu Jan 1 00:00:00 1970 From: a.hajda@samsung.com (Andrzej Hajda) Date: Tue, 18 Nov 2014 12:04:20 +0100 Subject: [BUG] blocked task after exynos_drm_init In-Reply-To: <546B2508.8070409@samsung.com> References: <1415264811.27478.4.camel@AMDC1943> <546B228E.6030500@samsung.com> <546B2508.8070409@samsung.com> Message-ID: <546B27B4.4040808@samsung.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On 11/18/2014 11:52 AM, Inki Dae wrote: > On 2014? 11? 18? 19:42, Andrzej Hajda wrote: >> On 11/06/2014 10:06 AM, Krzysztof Kozlowski wrote: >>> Hi, >>> >>> On last next (next-20141104, next-20141105) booting locks after >>> initializing Exynos DRM (Trats2 board): >>> >>> [ 2.028283] [drm] Initialized drm 1.1.0 20060810 >>> [ 240.505795] INFO: task swapper/0:1 blocked for more than 120 seconds. >>> [ 240.510825] Not tainted 3.18.0-rc3-next-20141105 #794 >>> [ 240.516418] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. >>> [ 240.524173] swapper/0 D c052534c 0 1 0 0x00000000 >>> [ 240.530527] [] (__schedule) from [] (schedule_preempt_disabled+0x14/0x20) >>> [ 240.539030] [] (schedule_preempt_disabled) from [] (mutex_lock_nested+0x1c4/0x464) >>> [ 240.548320] [] (mutex_lock_nested) from [] (__driver_attach+0x48/0x98) >>> [ 240.556562] [] (__driver_attach) from [] (bus_for_each_dev+0x54/0x88) >>> [ 240.564717] [] (bus_for_each_dev) from [] (bus_add_driver+0xe4/0x200) >>> [ 240.572876] [] (bus_add_driver) from [] (driver_register+0x78/0xf4) >>> [ 240.580864] [] (driver_register) from [] (exynos_drm_platform_probe+0x34/0x234) >>> [ 240.589890] [] (exynos_drm_platform_probe) from [] (platform_drv_probe+0x48/0xa4) >>> [ 240.599090] [] (platform_drv_probe) from [] (driver_probe_device+0x13c/0x37c) >>> [ 240.607940] [] (driver_probe_device) from [] (__driver_attach+0x94/0x98) >>> [ 240.616360] [] (__driver_attach) from [] (bus_for_each_dev+0x54/0x88) >>> [ 240.624517] [] (bus_for_each_dev) from [] (bus_add_driver+0xe4/0x200) >>> [ 240.632679] [] (bus_add_driver) from [] (driver_register+0x78/0xf4) >>> [ 240.640667] [] (driver_register) from [] (exynos_drm_init+0x70/0xa0) >>> [ 240.648739] [] (exynos_drm_init) from [] (do_one_initcall+0xac/0x1f0) >>> [ 240.656914] [] (do_one_initcall) from [] (kernel_init_freeable+0x10c/0x1d8) >>> [ 240.665591] [] (kernel_init_freeable) from [] (kernel_init+0x8/0xec) >>> [ 240.673661] [] (kernel_init) from [] (ret_from_fork+0x14/0x2c) >>> [ 240.681196] 3 locks held by swapper/0/1: >>> [ 240.685091] #0: (&dev->mutex){......}, at: [] __driver_attach+0x48/0x98 >>> [ 240.692732] #1: (&dev->mutex){......}, at: [] __driver_attach+0x58/0x98 >>> [ 240.700367] #2: (&dev->mutex){......}, at: [] __driver_attach+0x48/0x98 >> >> >> This is caused by patch moving platform devices to >> /sys/devices/platform[1]. Since this patch registering platform >> drivers/devices in probe of platform device causes deadlocks. I guess >> now all driver registration should be moved to exynos_drm_init and it >> seems better location for it IMHO. > > Thanks. It might be a chance that we could separate sub drivers of > Exynos drm into independent modules so that they can be called > independently because if we move them to exynos_drm_init then the > deferred probe wouldn't work correctly. For full separation of sub-drivers component matching code should be fixed first. Now it usually works because components are probed before exynos_drm_match_add and this is because components are probed during sub-driver registrations. Regards Andrzej > > Thanks, > Inki Dae > >> >> Regards >> Andrzej >> >> [1]: http://www.spinics.net/lists/devicetree/msg56101.html >> >> >> >>> >>> Full dmesg and config attached. >>> >>> Best regards, >>> Krzysztof >>> >>> >>> >>> _______________________________________________ >>> dri-devel mailing list >>> dri-devel at lists.freedesktop.org >>> http://lists.freedesktop.org/mailman/listinfo/dri-devel >>> >> >> >> -- >> To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in >> the body of a message to majordomo at vger.kernel.org >> More majordomo info at http://vger.kernel.org/majordomo-info.html >> > > _______________________________________________ > dri-devel mailing list > dri-devel at lists.freedesktop.org > http://lists.freedesktop.org/mailman/listinfo/dri-devel >