Re: [PATCH -v2 -for driver-core-next] ACPI: remove duplicated lines of merging problems with acpi_processor_start

[Thomas Renninger <trenn@suse.de>]

Hi,

this looks like the missing bits, Thanks!
I had a look at Stephan's mergers, but did not realize that (I expect
there) it slip in.
Greg, can you queue this one in your driver-core-next branch, please.

Thanks,

   Thomas

> When checking driver-core tree, found crazying warnings on my setups.
> 
> [  216.025849] calling  acpi_processor_init+0x0/0x81 @ 1
> [  216.045332] ACPI: Requesting acpi_cpufreq
> [  216.047454] Monitor-Mwait will be used to enter C-1 state
> [  216.047912] Monitor-Mwait will be used to enter C-3 state
> [  216.065270] ACPI: acpi_idle registered with cpuidle
> [  216.068241] kobject (ffff8870364a1940): tried to init an initialized object, something is seriously wrong.
> [  216.085287] Pid: 1, comm: swapper/0 Not tainted 3.3.0-rc2-tip-yh-02428-ge663840-dirty #247
> [  216.105041] Call Trace:
> [  216.105192]  [<ffffffff813a9c06>] kobject_init+0x33/0x83
> [  216.124880]  [<ffffffff813aa1f8>] kobject_init_and_add+0x23/0x57
> [  216.125158]  [<ffffffff819f3a08>] cpuidle_add_sysfs+0x49/0x62
> [  216.144850]  [<ffffffff819f2a28>] __cpuidle_register_device+0xe6/0x10e
> [  216.145182]  [<ffffffff819f2ea4>] cpuidle_register_device+0x25/0x4d
> [  216.164912]  [<ffffffff81cb5774>] acpi_processor_power_init+0x13e/0x16c
> [  216.165205]  [<ffffffff81427620>] ? acpi_processor_get_throttling_info+0x128/0x158
> [  216.185012]  [<ffffffff81c68ae5>] acpi_processor_start+0x62/0x11d
> [  216.204861]  [<ffffffff81cb55ff>] acpi_processor_add+0x1b0/0x1e7
> [  216.205144]  [<ffffffff81402a7e>] acpi_device_probe+0x4e/0x11c
> [  216.225063]  [<ffffffff8148f0e7>] really_probe+0x99/0x126
> [  216.225328]  [<ffffffff8148f2a3>] driver_probe_device+0x3b/0x56
> [  216.244846]  [<ffffffff8148f31d>] __driver_attach+0x5f/0x82
> [  216.245101]  [<ffffffff8148f2be>] ? driver_probe_device+0x56/0x56
> [  216.264668]  [<ffffffff8148db80>] bus_for_each_dev+0x5c/0x88
> [  216.264942]  [<ffffffff8148eea7>] driver_attach+0x1e/0x20
> [  216.284639]  [<ffffffff8148eaec>] bus_add_driver+0xca/0x21d
> [  216.284903]  [<ffffffff81095827>] ? local_clock+0xf/0x3c
> [  216.304580]  [<ffffffff82814177>] ? acpi_fan_init+0x18/0x18
> [  216.304849]  [<ffffffff8148f79b>] driver_register+0x91/0xfe
> [  216.324545]  [<ffffffff82814177>] ? acpi_fan_init+0x18/0x18
> [  216.324813]  [<ffffffff81403705>] acpi_bus_register_driver+0x43/0x45
> [  216.344563]  [<ffffffff828141a7>] acpi_processor_init+0x30/0x81
> [  216.344845]  [<ffffffff82814177>] ? acpi_fan_init+0x18/0x18
> [  216.364590]  [<ffffffff810001e7>] do_one_initcall+0x57/0x134
> [  216.364868]  [<ffffffff827e6f8c>] kernel_init+0x146/0x1c0
> [  216.384512]  [<ffffffff81d03aa4>] kernel_thread_helper+0x4/0x10
> [  216.384819]  [<ffffffff81cfbb5d>] ? retint_restore_args+0xe/0xe
> [  216.404578]  [<ffffffff827e6e46>] ? start_kernel+0x3ab/0x3ab
> [  216.424530]  [<ffffffff81d03aa0>] ? gs_change+0xb/0xb
> [  216.424793] ------------[ cut here ]------------
> [  216.425038] WARNING: at fs/sysfs/dir.c:502 sysfs_add_one+0x97/0xab()
> [  216.444480] Hardware name: Sun Fire X4800
> [  216.444668] sysfs: cannot create duplicate filename '/devices/system/cpu/cpu0/cpuidle'
> ...
> 
> It turns out acpi_processor_power_init() get called two time in acpi_processor_add and acpi_processor_start.
> 
> Found several lines are duplicated in those two functions even related commit move them.
> 
> The related patches are ok.  Not sure how it could happen, looks like git problem.
> 
> -v2: add back acpi_processor_load_module(pr) to acpi_processor_load_start
> 
> Signed-off-by: Yinghai Lu <yinghai@kernel.org>
> Cc: Linus Torvalds <torvalds@linux-foundation.org>
> Cc: Thomas Renninger <trenn@suse.de>
> Cc: Len Brown <lenb@kernel.org>
> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> 
> It turns out acpi_processor_power_init() get called two time in acpi_processor_add and acpi_processor_start.
> 
> Found several lines are duplicated in those two functions even related commit move them.
> 
> The related patches are ok.  Not sure how it could happen, looks like git problem.
> 
> -v2: add back acpi_processor_load_module(pr) to acpi_processor_load_start
> 
> Signed-off-by: Yinghai Lu <yinghai@kernel.org>
> Cc: Linus Torvalds <torvalds@linux-foundation.org>
> Cc: Thomas Renninger <trenn@suse.de>
> Cc: Len Brown <lenb@kernel.org>
> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
> 
> ---
>  drivers/acpi/processor_driver.c |   18 +-----------------
>  1 file changed, 1 insertion(+), 17 deletions(-)
> 
> Index: linux-2.6/drivers/acpi/processor_driver.c
> ===================================================================
> --- linux-2.6.orig/drivers/acpi/processor_driver.c
> +++ linux-2.6/drivers/acpi/processor_driver.c
> @@ -474,6 +474,7 @@ static __ref int acpi_processor_start(st
>  
>  #ifdef CONFIG_CPU_FREQ
>  	acpi_processor_ppc_has_changed(pr, 0);
> +	acpi_processor_load_module(pr);
>  #endif
>  	acpi_processor_get_throttling_info(pr);
>  	acpi_processor_get_limit_info(pr);
> @@ -579,23 +580,6 @@ static int __cpuinit acpi_processor_add(
>  		goto err_free_cpumask;
>  	}
>  
> -#ifdef CONFIG_CPU_FREQ
> -	acpi_processor_ppc_has_changed(pr, 0);
> -	acpi_processor_load_module(pr);
> -#endif
> -	acpi_processor_get_throttling_info(pr);
> -	acpi_processor_get_limit_info(pr);
> -
> -	if (!cpuidle_get_driver() || cpuidle_get_driver() == &acpi_idle_driver)
> -		acpi_processor_power_init(pr, device);
> -
> -	pr->cdev = thermal_cooling_device_register("Processor", device,
> -						&processor_cooling_ops);
> -	if (IS_ERR(pr->cdev)) {
> -		result = PTR_ERR(pr->cdev);
> -		goto err_remove_sysfs;
> -	}
> -
>  	/*
>  	 * Do not start hotplugged CPUs now, but when they
>  	 * are onlined the first time
>