From mboxrd@z Thu Jan 1 00:00:00 1970 From: Suniel Mahesh Subject: Re: [PATCH] drivers: cpufreq: Fix sysfs duplicate filename creation for platform-device Date: Mon, 18 Sep 2017 22:29:33 +0530 Message-ID: <280ef985-ddb5-bbd5-fb10-d9998225e288@techveda.org> References: <1505586894-20823-1-git-send-email-sunil.m@techveda.org> <20170918011503.GA17030@ubuntu> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20170918011503.GA17030@ubuntu> Content-Language: en-US Sender: linux-kernel-owner@vger.kernel.org To: Viresh Kumar Cc: rjw@rjwysocki.net, linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org, karthik@techveda.org List-Id: linux-pm@vger.kernel.org On Monday 18 September 2017 06:45 AM, Viresh Kumar wrote: > On 17-09-17, 00:04, sunil.m@techveda.org wrote: >> From: Suniel Mahesh >> >> ti-cpufreq.c and cpufreq-dt-platdev.c are registering >> platform device with same name "cpufreq-dt" using >> platform_device_register_*() routines. >> This is leading to build warnings appended below. >> >> This patch does the following: >> (a) Remove platform-device registration routine from >> ti-cpufreq.c, it appears unnecessary as per CPUfreq framework >> kernel Documentation. The ti-cpufreq driver main task is to use >> revision and an efuse value from the SoC and provide the OPP >> framework with supported hardware information. >> Platform-device creation is taken care by cpufreq-dt-platdev.c. >> (b) In case if OPP-v2 is not supported just return without >> registering the platorm-device. >> (c) Rename the goto label >> >> [ 2.370167] ------------[ cut here ]------------ >> [ 2.375087] WARNING: CPU: 0 PID: 1 at fs/sysfs/dir.c:31 sysfs_warn_dup+0x58/0x78 >> [ 2.383112] sysfs: cannot create duplicate filename '/devices/platform/cpufreq-dt' >> [ 2.391219] Modules linked in: >> [ 2.394506] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 4.13.0-next-20170912 #1 >> [ 2.402006] Hardware name: Generic AM33XX (Flattened Device Tree) >> [ 2.408437] [] (unwind_backtrace) from [] (show_stack+0x10/0x14) >> [ 2.416568] [] (show_stack) from [] (dump_stack+0xac/0xe0) >> [ 2.424165] [] (dump_stack) from [] (__warn+0xd8/0x104) >> [ 2.431488] [] (__warn) from [] (warn_slowpath_fmt+0x34/0x44) >> [ 2.439351] [] (warn_slowpath_fmt) from [] (sysfs_warn_dup+0x58/0x78) >> [ 2.447938] [] (sysfs_warn_dup) from [] (sysfs_create_dir_ns+0x80/0x98) >> [ 2.456719] [] (sysfs_create_dir_ns) from [] (kobject_add_internal+0x9c/0x2d4) >> [ 2.466124] [] (kobject_add_internal) from [] (kobject_add+0x4c/0x9c) >> [ 2.474712] [] (kobject_add) from [] (device_add+0xcc/0x57c) >> [ 2.482489] [] (device_add) from [] (platform_device_add+0x100/0x220) >> [ 2.491085] [] (platform_device_add) from [] (platform_device_register_full+0xf4/0x118) >> [ 2.501305] [] (platform_device_register_full) from [] (ti_cpufreq_init+0x150/0x22c) >> [ 2.511253] [] (ti_cpufreq_init) from [] (do_one_initcall+0x3c/0x170) >> [ 2.519838] [] (do_one_initcall) from [] (kernel_init_freeable+0x1fc/0x2c4) >> [ 2.528974] [] (kernel_init_freeable) from [] (kernel_init+0x8/0x110) >> [ 2.537565] [] (kernel_init) from [] (ret_from_fork+0x14/0x3c) >> [ 2.545981] ---[ end trace 2fc00e213c13ab20 ]--- >> [ 2.551051] ------------[ cut here ]------------ >> [ 2.555931] WARNING: CPU: 0 PID: 1 at lib/kobject.c:240 kobject_add_internal+0x254/0x2d4 >> [ 2.564578] kobject_add_internal failed for cpufreq-dt with -EEXIST, don't try to register >> things with the same name in the same directory. >> [ 2.577977] Modules linked in: >> [ 2.581261] CPU: 0 PID: 1 Comm: swapper/0 Tainted: G W 4.13.0-next-20170912 #1 >> [ 2.590013] Hardware name: Generic AM33XX (Flattened Device Tree) >> [ 2.596437] [] (unwind_backtrace) from [] (show_stack+0x10/0x14) >> [ 2.604573] [] (show_stack) from [] (dump_stack+0xac/0xe0) >> [ 2.612172] [] (dump_stack) from [] (__warn+0xd8/0x104) >> [ 2.619494] [] (__warn) from [] (warn_slowpath_fmt+0x34/0x44) >> [ 2.627362] [] (warn_slowpath_fmt) from [] (kobject_add_internal+0x254/0x2d4) >> [ 2.636666] [] (kobject_add_internal) from [] (kobject_add+0x4c/0x9c) >> [ 2.645255] [] (kobject_add) from [] (device_add+0xcc/0x57c) >> [ 2.653027] [] (device_add) from [] (platform_device_add+0x100/0x220) >> [ 2.661615] [] (platform_device_add) from [] (platform_device_register_full+0xf4/0x118) >> [ 2.671833] [] (platform_device_register_full) from [] (ti_cpufreq_init+0x150/0x22c) >> [ 2.681779] [] (ti_cpufreq_init) from [] (do_one_initcall+0x3c/0x170) >> [ 2.690377] [] (do_one_initcall) from [] (kernel_init_freeable+0x1fc/0x2c4) >> [ 2.699510] [] (kernel_init_freeable) from [] (kernel_init+0x8/0x110) >> [ 2.708106] [] (kernel_init) from [] (ret_from_fork+0x14/0x3c) >> [ 2.716217] ---[ end trace 2fc00e213c13ab21 ]--- >> >> Signed-off-by: Suniel Mahesh >> Tested-by: Suniel Mahesh > > You don't need a Tested-by from yourself as it is already assumed after the > Signed-off by. got it, thanks for making me understand in which cases we need to use it. > >> --- >> Note: >> - Patch was compile tested and built(ARCH=arm) on next-20170915. >> - Patch was hardware tested on AM335x SOC based board. >> - cpufreq-dt.c driver got probed and entries related to Dynamic Voltage >> and Frequency scaling were created in >> sysfs @ sys/devices/system/cpu/cpu0/cpufreq. >> - No build/run-time issues reported. >> --- >> drivers/cpufreq/ti-cpufreq.c | 6 ++---- >> 1 file changed, 2 insertions(+), 4 deletions(-) >> >> diff --git a/drivers/cpufreq/ti-cpufreq.c b/drivers/cpufreq/ti-cpufreq.c >> index b29cd33..97c086a 100644 >> --- a/drivers/cpufreq/ti-cpufreq.c >> +++ b/drivers/cpufreq/ti-cpufreq.c >> @@ -224,7 +224,7 @@ static int ti_cpufreq_init(void) >> if (!opp_data->opp_node) { >> dev_info(opp_data->cpu_dev, >> "OPP-v2 not supported, cpufreq-dt will attempt to use legacy tables.\n"); >> - goto register_cpufreq_dt; >> + goto ti_cpufreq_exit; >> } >> >> ret = ti_cpufreq_setup_syscon_register(opp_data); >> @@ -255,9 +255,7 @@ static int ti_cpufreq_init(void) >> >> of_node_put(opp_data->opp_node); >> >> -register_cpufreq_dt: >> - platform_device_register_simple("cpufreq-dt", -1, NULL, 0); >> - > > NACK. > > It was done this way to make sure ti-cpufreq.c does all the platform specific > stuff before the cpufreq-dt driver comes into place. got it, thanks for the explanation. > > If both ti-cpufreq and cpufreq-dt-platform are creating the device, then there > is something else wrong. Did it start failing recently only? Then it might be > due to one of the patches I posted for cpufreq-dt-platdev. Yes, it's happening for the latest mainline rc kernel and next trees because of this commit: "edeec420de24 (cpufreq: dt-platdev: Automatically create cpufreq device with OPP v2)" You need to add that > patch in the fixes tag and add TI SoCs in the blacklist of devices (with opp-v2) > in cpufreq-dt-platform.c. > working on it, thanks for the inputs. Regards Suniel