Re: [PATCH v2 1/2] mm: huge_memory: refactor thpsize_shmem_enabled_store() with sysfs_match_string()

[ranxiaokai627]

>On Wed, May 13, 2026 at 09:45:07AM +0000, ranxiaokai627@xxxxxxx wrote:
>> From: Ran Xiaokai <ran.xiaokai@xxxxxxxxxx>
>>
>> Inspired by commit 82d9ff648c6c ("mm: huge_memory: refactor
>> anon_enabled_store() with set_anon_enabled_mode()"), refactor
>> thpsize_shmem_enabled_store() using sysfs_match_string().
>> This eliminates the duplicated spin_lock/unlock(), set/clear_bit(),
>> calls across all branches, reducing code duplication.
>
>You are not mentioning that you're changing the stop/start behaviour. This
>is a fundamental part of the change that should be documented in the commit
>message.

Hi, Lorenzo,
Thanks for your detailed review.

I'll update the commit message to document this:

"Behavioral change:
Call start_stop_khugepaged() only when the mode actually changes.
If unchanged, call set_recommended_min_free_kbytes() to preserve legacy
watermark behavior. This avoids unnecessary khugepaged restarts."

>>
>> Tested with selftests ./run_kselftest.sh -t mm:ksft_thp.sh,
>> all test cases passed.
>>
>> Signed-off-by: Ran Xiaokai <ran.xiaokai@xxxxxxxxxx>
>> ---
>>  mm/shmem.c | 94 +++++++++++++++++++++++++++++-------------------------
>>  1 file changed, 51 insertions(+), 43 deletions(-)
>>
>> diff --git a/mm/shmem.c b/mm/shmem.c
>> index 3b5dc21b323c..60cb10854f11 100644
>> --- a/mm/shmem.c
>> +++ b/mm/shmem.c
>> @@ -5526,6 +5526,29 @@ static ssize_t shmem_enabled_store(struct kobject *kobj,
>>  struct kobj_attribute shmem_enabled_attr = __ATTR_RW(shmem_enabled);
>>  static DEFINE_SPINLOCK(huge_shmem_orders_lock);
>>
>> +enum huge_shmem_enabled_mode {
>> +	HUGE_SHMEM_ENABLED_ALWAYS = 0,
>> +	HUGE_SHMEM_ENABLED_INHERIT,
>> +	HUGE_SHMEM_ENABLED_WITHIN_SIZE,
>> +	HUGE_SHMEM_ENABLED_ADVISE,
>> +	HUGE_SHMEM_ENABLED_NEVER,
>> +};
>> +
>> +static const char * const huge_shmem_enabled_mode_strings[] = {
>> +	[HUGE_SHMEM_ENABLED_ALWAYS]      = "always",
>> +	[HUGE_SHMEM_ENABLED_INHERIT]     = "inherit",
>> +	[HUGE_SHMEM_ENABLED_WITHIN_SIZE] = "within_size",
>> +	[HUGE_SHMEM_ENABLED_ADVISE]      = "advise",
>> +	[HUGE_SHMEM_ENABLED_NEVER]       = "never",
>> +};
>> +
>> +static unsigned long * const huge_shmem_orders_by_mode[] = {
>> +	[HUGE_SHMEM_ENABLED_ALWAYS]      = &huge_shmem_orders_always,
>> +	[HUGE_SHMEM_ENABLED_INHERIT]     = &huge_shmem_orders_inherit,
>> +	[HUGE_SHMEM_ENABLED_WITHIN_SIZE] = &huge_shmem_orders_within_size,
>> +	[HUGE_SHMEM_ENABLED_ADVISE]      = &huge_shmem_orders_madvise,
>> +};
>
>I think it's perfectly cromulent to put this within a function rather than as a
>static value at file scope. See below.
>
>> +
>>  static ssize_t thpsize_shmem_enabled_show(struct kobject *kobj,
>>  					  struct kobj_attribute *attr, char *buf)
>>  {
>> @@ -5551,57 +5574,42 @@ static ssize_t thpsize_shmem_enabled_store(struct kobject *kobj,
>>  					   const char *buf, size_t count)
>>  {
>>  	int order = to_thpsize(kobj)->order;
>> +	int mode, m;
>
>I hate single letter variable names (yes I see set_anon_enabled_mode() has it, I
>should have reviewed against that, oops :).
>
>Please give it its type of enum huge_shmem_enalbed_mode.

int main()
{
    enum huge_shmem_enabled_mode mode = 0;
    const char* type_name = _Generic((mode),
        int: "signed int",
        unsigned int: "unsigned int",
        default: "other"
    );
    printf("enum huge_shmem_enabled_mode underlying type is: %s\n", type_name);
    return 0;
}
 ./a.out
enum huge_shmem_enabled_mode underlying type is: unsigned int

This will cause the following check to always evaluate to false:
if (mode < 0)
	return mode;

So i think we should keep mode as integer.

>>  	ssize_t ret = count;
>> +	int err;
>
>This is silly, we don't need ret and err (the ret is not your fault
>obviously...)
>
>On success, we return count, on failure we return err. Drop the ret here.
>
>(I think Baolin reviewed similarly :)

Agreed, will simplify return logic and return count in v3.

>> +	bool changed = false;
>>
>> -	if (sysfs_streq(buf, "always")) {
>> -		spin_lock(&huge_shmem_orders_lock);
>> -		clear_bit(order, &huge_shmem_orders_inherit);
>> -		clear_bit(order, &huge_shmem_orders_madvise);
>> -		clear_bit(order, &huge_shmem_orders_within_size);
>> -		set_bit(order, &huge_shmem_orders_always);
>> -		spin_unlock(&huge_shmem_orders_lock);
>> -	} else if (sysfs_streq(buf, "inherit")) {
>> -		/* Do not override huge allocation policy with non-PMD sized mTHP */
>> -		if (shmem_huge == SHMEM_HUGE_FORCE && !is_pmd_order(order))
>> -			return -EINVAL;
>> +	mode = sysfs_match_string(huge_shmem_enabled_mode_strings, buf);
>> +	if (mode < 0)
>> +		return -EINVAL;
>
>I don't know why we filter this error, it returns -EINVAL on error anyway, just
>return mode in this case.

yes, will return mode directly, 
but the type of mode needs to remain an integer variable as replied above.

>>
>> -		spin_lock(&huge_shmem_orders_lock);
>> -		clear_bit(order, &huge_shmem_orders_always);
>> -		clear_bit(order, &huge_shmem_orders_madvise);
>> -		clear_bit(order, &huge_shmem_orders_within_size);
>> -		set_bit(order, &huge_shmem_orders_inherit);
>> -		spin_unlock(&huge_shmem_orders_lock);
>> -	} else if (sysfs_streq(buf, "within_size")) {
>> -		spin_lock(&huge_shmem_orders_lock);
>> -		clear_bit(order, &huge_shmem_orders_always);
>> -		clear_bit(order, &huge_shmem_orders_inherit);
>> -		clear_bit(order, &huge_shmem_orders_madvise);
>> -		set_bit(order, &huge_shmem_orders_within_size);
>> -		spin_unlock(&huge_shmem_orders_lock);
>> -	} else if (sysfs_streq(buf, "advise")) {
>> -		spin_lock(&huge_shmem_orders_lock);
>> -		clear_bit(order, &huge_shmem_orders_always);
>> -		clear_bit(order, &huge_shmem_orders_inherit);
>> -		clear_bit(order, &huge_shmem_orders_within_size);
>> -		set_bit(order, &huge_shmem_orders_madvise);
>> -		spin_unlock(&huge_shmem_orders_lock);
>> -	} else if (sysfs_streq(buf, "never")) {
>> -		spin_lock(&huge_shmem_orders_lock);
>> -		clear_bit(order, &huge_shmem_orders_always);
>> -		clear_bit(order, &huge_shmem_orders_inherit);
>> -		clear_bit(order, &huge_shmem_orders_within_size);
>> -		clear_bit(order, &huge_shmem_orders_madvise);
>> -		spin_unlock(&huge_shmem_orders_lock);
>> -	} else {
>> -		ret = -EINVAL;
>> -	}
>> +	/* Do not override huge allocation policy with non-PMD sized mTHP */
>> +	if (mode == HUGE_SHMEM_ENABLED_INHERIT &&
>> +		shmem_huge == SHMEM_HUGE_FORCE && !is_pmd_order(order))
>> +		return -EINVAL;
>>
>> -	if (ret > 0) {
>> -		int err = start_stop_khugepaged();
>> +	spin_lock(&huge_shmem_orders_lock);
>> +	for (m = 0; m < ARRAY_SIZE(huge_shmem_orders_by_mode); m++) {
>> +		if (m == mode)
>> +			changed |= !__test_and_set_bit(order, huge_shmem_orders_by_mode[m]);
>> +		else
>> +			changed |= __test_and_clear_bit(order, huge_shmem_orders_by_mode[m]);
>> +	}
>> +	spin_unlock(&huge_shmem_orders_lock);
>
>You're copy/pasta'ing anon_enabled_store() but not the nicer refactoring that's
>there.
>
>Please split this out like set_anon_enabled_mode() does. And put:
>
>	static unsigned long *enabled_orders[] = {
>		&huge_shmem_orders_always,
>		&huge_shmem_orders_inherit,
>		&huge_shmem_orders_within_size,
>		&huge_shmem_orders_madvise,
>	};
>
>At the start of it like that does also.
>
>Please don't reproduce the single letter var name though :)

Agreed, will extract a set_shmem_enabled_mode() helper with local array.

>> +};
>
>>
>> +	if (changed) {
>> +		err = start_stop_khugepaged();
>>  		if (err)
>>  			ret = err;
>
>
>> +	} else {
>> +		/*
>> +		 * Recalculate watermarks even when the mode hasn't changed
>> +		 * to preserve the legacy behavior, as this is always called
>> +		 * inside start_stop_khugepaged().
>> +		 */
>> +		set_recommended_min_free_kbytes();
>>  	}
>> +
>>  	return ret;
>
>return count;
>
>>  }
>>
>> --
>> 2.25.1
>>
>>
>
>Cheers, Lorenzo