Messages in this thread | | | Subject | Re: [1/2] cpufreq: dt-platdev: Automatically create cpufreq device with OPP v2 | From | Dave Gerlach <> | Date | Mon, 11 Sep 2017 09:18:46 -0500 |
| |
Hi, On 09/11/2017 07:13 AM, Keerthy wrote: > > > On Tuesday 29 August 2017 03:41 PM, Viresh Kumar wrote: >> On 21-08-17, 15:17, Simon Horman wrote: >>> Sorry, I seem to have accidently sent this email as Virish rather than >>> myself. I will try again. >> >> No issues. I see that Rafael has already applied my patches, you can send >> additional patches over that to make sure everything is clear. > > Viresh, > > I see below warning with this patch on am57xx-beagle > > [ 3.545247] WARNING: CPU: 1 PID: 1 at fs/sysfs/dir.c:31 > sysfs_warn_dup+0x58/0x78 > [ 3.553099] sysfs: cannot create duplicate filename > '/devices/platform/cpufreq-dt' > [ 3.561126] Modules linked in: > [ 3.564402] CPU: 1 PID: 1 Comm: swapper/0 Tainted: G W > 4.13.0-next-20170911-12690-ga31cc45-dirty #7 > [ 3.575071] Hardware name: Generic DRA74X (Flattened Device Tree) > [ 3.581484] [<c0110a88>] (unwind_backtrace) from [<c010cae4>] > (show_stack+0x10/0x14) > [ 3.589620] [<c010cae4>] (show_stack) from [<c0828184>] > (dump_stack+0xac/0xe0) > [ 3.597209] [<c0828184>] (dump_stack) from [<c01374d0>] > (__warn+0xd8/0x104) > [ 3.604532] [<c01374d0>] (__warn) from [<c0137530>] > (warn_slowpath_fmt+0x34/0x44) > [ 3.612390] [<c0137530>] (warn_slowpath_fmt) from [<c0345b30>] > (sysfs_warn_dup+0x58/0x78) > [ 3.620988] [<c0345b30>] (sysfs_warn_dup) from [<c0345c18>] > (sysfs_create_dir_ns+0x80/0x98) > [ 3.629773] [<c0345c18>] (sysfs_create_dir_ns) from [<c082c974>] > (kobject_add_internal+0x9c/0x2d4) > [ 3.639177] [<c082c974>] (kobject_add_internal) from [<c082cbf8>] > (kobject_add+0x4c/0x9c) > [ 3.647764] [<c082cbf8>] (kobject_add) from [<c05805a8>] > (device_add+0xcc/0x57c) > [ 3.655542] [<c05805a8>] (device_add) from [<c0584d38>] > (platform_device_add+0x100/0x220) > [ 3.664138] [<c0584d38>] (platform_device_add) from [<c058576c>] > (platform_device_register_full+0xf4/0x118) > [ 3.674378] [<c058576c>] (platform_device_register_full) from > [<c0670514>] (ti_cpufreq_init+0x150/0x22c) > [ 3.684326] [<c0670514>] (ti_cpufreq_init) from [<c0101df4>] > (do_one_initcall+0x3c/0x170) > [ 3.692916] [<c0101df4>] (do_one_initcall) from [<c0c00eb4>] > (kernel_init_freeable+0x1fc/0x2c4) > [ 3.702049] [<c0c00eb4>] (kernel_init_freeable) from [<c083c0cc>] > (kernel_init+0x8/0x110) > [ 3.710639] [<c083c0cc>] (kernel_init) from [<c0107d78>] > (ret_from_fork+0x14/0x3c) > [ 3.718680] ---[ end trace 33482508cbb50156 ]--- > [ 3.720850] ata1: SATA link down (SStatus 0 SControl 300) > [ 3.729278] ------------[ cut here ]------------ > [ 3.734147] WARNING: CPU: 1 PID: 1 at lib/kobject.c:240 > kobject_add_internal+0x254/0x2d4 > [ 3.742709] kobject_add_internal failed for cpufreq-dt with -EEXIST, > don't try to register things with the same name in the same directory. > [ 3.755900] Modules linked in: > [ 3.759239] CPU: 1 PID: 1 Comm: swapper/0 Tainted: G W > 4.13.0-next-20170911-12690-ga31cc45-dirty #7 > [ 3.769907] Hardware name: Generic DRA74X (Flattened Device Tree) > [ 3.776321] [<c0110a88>] (unwind_backtrace) from [<c010cae4>] > (show_stack+0x10/0x14) > [ 3.784454] [<c010cae4>] (show_stack) from [<c0828184>] > (dump_stack+0xac/0xe0) > [ 3.792043] [<c0828184>] (dump_stack) from [<c01374d0>] > (__warn+0xd8/0x104) > [ 3.799364] [<c01374d0>] (__warn) from [<c0137530>] > (warn_slowpath_fmt+0x34/0x44) > [ 3.807227] [<c0137530>] (warn_slowpath_fmt) from [<c082cb2c>] > (kobject_add_internal+0x254/0x2d4) > [ 3.816552] [<c082cb2c>] (kobject_add_internal) from [<c082cbf8>] > (kobject_add+0x4c/0x9c) > [ 3.825139] [<c082cbf8>] (kobject_add) from [<c05805a8>] > (device_add+0xcc/0x57c) > [ 3.832913] [<c05805a8>] (device_add) from [<c0584d38>] > (platform_device_add+0x100/0x220) > [ 3.841504] [<c0584d38>] (platform_device_add) from [<c058576c>] > (platform_device_register_full+0xf4/0x118) > [ 3.851729] [<c058576c>] (platform_device_register_full) from > [<c0670514>] (ti_cpufreq_init+0x150/0x22c) > [ 3.861686] [<c0670514>] (ti_cpufreq_init) from [<c0101df4>] > (do_one_initcall+0x3c/0x170) > [ 3.870273] [<c0101df4>] (do_one_initcall) from [<c0c00eb4>] > (kernel_init_freeable+0x1fc/0x2c4) > [ 3.879412] [<c0c00eb4>] (kernel_init_freeable) from [<c083c0cc>] > (kernel_init+0x8/0x110) > [ 3.888002] [<c083c0cc>] (kernel_init) from [<c0107d78>] > (ret_from_fork+0x14/0x3c) > [ 3.896035] ---[ end trace 33482508cbb50157 ]--- > > I believe in two places platform_device_register_* calls are made with > the same name 'cpufreq-dt' > > First place is: cpufreq_dt_platdev_init and the other i believe is > ti_cpufreq_init. > > Dave, > > Should the ti-cpufreq driver also be calling > platform_device_register_simple?
Yes, ti-cpufreq registers the cpufreq-dt platdev in order to ensure the ti-cpufreq driver probes first and provides the necessary opp-supported-hw for cpufreq-dt. This applies to am335x, am437x, dra7xx, and am57xx. I suppose these platforms will need to be added to the 'blacklist' to prevent the cpufreq-dt platdev from being created automatically, unless there is a better way to handle this dependency...
Regards, Dave
> > Regards, > Keerthy > >>
| |