[6/9] cpu/SMT: Allow enabling partial SMT states via sysfs

Add support to the /sys/devices/system/cpu/smt/control interface for
enabling a specified number of SMT threads per core, including partial
SMT states where not all threads are brought online.

The current interface accepts "on" and "off", to enable either 1 or all
SMT threads per core.

This commit allows writing an integer, between 1 and the number of SMT
threads supported by the machine. Writing 1 is a synonym for "off", 2 or
more enables SMT with the specified number of threads.

When reading the file, if all threads are online "on" is returned, to
avoid changing behaviour for existing users. If some other number of
threads is online then the integer value is returned.

There is a hook which allows arch code to control how many threads per
core are supported. To retain the existing behaviour, the x86 hook only
supports 1 thread or all threads.

Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
---
 .../ABI/testing/sysfs-devices-system-cpu      |  1 +
 kernel/cpu.c                                  | 39 ++++++++++++++++---
 2 files changed, 34 insertions(+), 6 deletions(-)

On Thu, May 25 2023 at 01:56, Michael Ellerman wrote:
> There is a hook which allows arch code to control how many threads per

Can you please write out architecture in changelogs and comments?

I know 'arch' is commonly used but while my brain parser tolerates
'arch_' prefixes it raises an exception on 'arch' in prose as 'arch' is
a regular word with a completely different meaning. Changelogs and
comments are not space constraint.

> @@ -2505,20 +2505,38 @@ __store_smt_control(struct device *dev, struct device_attribute *attr,
>  	if (cpu_smt_control == CPU_SMT_NOT_SUPPORTED)
>  		return -ENODEV;
>  
> -	if (sysfs_streq(buf, "on"))
> +	if (sysfs_streq(buf, "on")) {
>  		ctrlval = CPU_SMT_ENABLED;
> -	else if (sysfs_streq(buf, "off"))
> +		num_threads = cpu_smt_max_threads;
> +	} else if (sysfs_streq(buf, "off")) {
>  		ctrlval = CPU_SMT_DISABLED;
> -	else if (sysfs_streq(buf, "forceoff"))
> +		num_threads = 1;
> +	} else if (sysfs_streq(buf, "forceoff")) {
>  		ctrlval = CPU_SMT_FORCE_DISABLED;
> -	else
> +		num_threads = 1;
> +	} else if (kstrtoint(buf, 10, &num_threads) == 0) {
> +		if (num_threads == 1)
> +			ctrlval = CPU_SMT_DISABLED;
> +		else if (num_threads > 1 && topology_smt_threads_supported(num_threads))
> +			ctrlval = CPU_SMT_ENABLED;
> +		else
> +			return -EINVAL;
> +	} else {
>  		return -EINVAL;
> +	}
>  
>  	ret = lock_device_hotplug_sysfs();
>  	if (ret)
>  		return ret;
>  
> -	if (ctrlval != cpu_smt_control) {
> +	orig_threads = cpu_smt_num_threads;
> +	cpu_smt_num_threads = num_threads;
> +
> +	if (num_threads > orig_threads) {
> +		ret = cpuhp_smt_enable();
> +	} else if (num_threads < orig_threads) {
> +		ret = cpuhp_smt_disable(ctrlval);
> +	} else if (ctrlval != cpu_smt_control) {
>  		switch (ctrlval) {
>  		case CPU_SMT_ENABLED:
>  			ret = cpuhp_smt_enable();

This switch case does not make sense anymore.

The only situation which reaches this is when the control value goes
from CPU_SMT_DISABLED to CPU_SMT_FORCE_DISABLED because that's not
changing the number of threads.

So something like this is completely sufficient:

	if (num_threads > orig_threads)
		ret = cpuhp_smt_enable();
        else if (num_threads < orig_threads || ctrval == CPU_SMT_FORCE_DISABLED)
		ret = cpuhp_smt_disable(ctrlval);

No?

Thanks,

        tglx

On Sat, Jun 10 2023 at 22:09, Thomas Gleixner wrote:

> On Thu, May 25 2023 at 01:56, Michael Ellerman wrote:
>> There is a hook which allows arch code to control how many threads per
>
> Can you please write out architecture in changelogs and comments?
>
> I know 'arch' is commonly used but while my brain parser tolerates
> 'arch_' prefixes it raises an exception on 'arch' in prose as 'arch' is
> a regular word with a completely different meaning. Changelogs and
> comments are not space constraint.
>
>> @@ -2505,20 +2505,38 @@ __store_smt_control(struct device *dev, struct device_attribute *attr,
>>  	if (cpu_smt_control == CPU_SMT_NOT_SUPPORTED)
>>  		return -ENODEV;
>>  
>> -	if (sysfs_streq(buf, "on"))
>> +	if (sysfs_streq(buf, "on")) {
>>  		ctrlval = CPU_SMT_ENABLED;
>> -	else if (sysfs_streq(buf, "off"))
>> +		num_threads = cpu_smt_max_threads;
>> +	} else if (sysfs_streq(buf, "off")) {
>>  		ctrlval = CPU_SMT_DISABLED;
>> -	else if (sysfs_streq(buf, "forceoff"))
>> +		num_threads = 1;
>> +	} else if (sysfs_streq(buf, "forceoff")) {
>>  		ctrlval = CPU_SMT_FORCE_DISABLED;
>> -	else
>> +		num_threads = 1;
>> +	} else if (kstrtoint(buf, 10, &num_threads) == 0) {
>> +		if (num_threads == 1)
>> +			ctrlval = CPU_SMT_DISABLED;
>> +		else if (num_threads > 1 && topology_smt_threads_supported(num_threads))

Why does this not simply check cpu_smt_max_threads?

		else if (num_threads > 1 && num_threads <= cpu_smt_max_threads)

cpu_smt_max_threads should have been established already, no?

Thanks,

        tglx