Re: + mm-percpu-drop-unused-max_upa-in-pcpu_build_alloc_info.patch added to mm-new branch

[Dennis Zhou <dennis@kernel.org>]

Hi Andrew,

On Tue, Dec 16, 2025 at 12:42:00PM -0800, Andrew Morton wrote:
> 
> The patch titled
>      Subject: mm/percpu: drop unused max_upa in pcpu_build_alloc_info()
> has been added to the -mm mm-new branch.  Its filename is
>      mm-percpu-drop-unused-max_upa-in-pcpu_build_alloc_info.patch
> 
> This patch will shortly appear at
>      https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-percpu-drop-unused-max_upa-in-pcpu_build_alloc_info.patch
> 
> This patch will later appear in the mm-new branch at
>     git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
> 
> Note, mm-new is a provisional staging ground for work-in-progress
> patches, and acceptance into mm-new is a notification for others take
> notice and to finish up reviews.  Please do not hesitate to respond to
> review feedback and post updated versions to replace or incrementally
> fixup patches in mm-new.
> 
> The mm-new branch of mm.git is not included in linux-next
> 
> Before you just go and hit "reply", please:
>    a) Consider who else should be cc'ed
>    b) Prefer to cc a suitable mailing list as well
>    c) Ideally: find the original patch on the mailing list and do a
>       reply-to-all to that, adding suitable additional cc's
> 
> *** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
> 
> The -mm tree is included into linux-next via various
> branches at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
> and is updated there most days
> 
> ------------------------------------------------------
> From: WanLi Niu <niuwl1@chinatelecom.cn>
> Subject: mm/percpu: drop unused max_upa in pcpu_build_alloc_info()
> Date: Mon, 8 Dec 2025 17:26:27 +0800
> 
> pcpu_build_alloc_info() only needs the initial upa as the starting point
> for the downward search.  Storing it in max_upa and reusing it adds no
> value.  Iterate from upa directly to remove the redundant local variable
> without changing behavior, improving readability.
> 
> Link: https://lkml.kernel.org/r/20251208092627.5603-1-niuwl1@chinatelecom.cn
> Signed-off-by: WanLi Niu <niuwl1@chinatelecom.cn>
> Co-developed-by: hlleng <a909204013@gmail.com>
> Signed-off-by: hlleng <a909204013@gmail.com>
> Cc: Christoph Lameter (Ampere) <cl@gentwo.org>
> Cc: Dennis Zhou <dennis@kernel.org>
> Cc: Tejun Heo <tj@kernel.org>
> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
> ---
> 
>  mm/percpu.c |    7 +++----
>  1 file changed, 3 insertions(+), 4 deletions(-)
> 
> --- a/mm/percpu.c~mm-percpu-drop-unused-max_upa-in-pcpu_build_alloc_info
> +++ a/mm/percpu.c
> @@ -2800,7 +2800,7 @@ static struct pcpu_alloc_info * __init _
>  	const size_t static_size = __per_cpu_end - __per_cpu_start;
>  	int nr_groups = 1, nr_units = 0;
>  	size_t size_sum, min_unit_size, alloc_size;
> -	int upa, max_upa, best_upa;	/* units_per_alloc */
> +	int upa, best_upa;	/* units_per_alloc */
>  	int last_allocs, group, unit;
>  	unsigned int cpu, tcpu;
>  	struct pcpu_alloc_info *ai;
> @@ -2817,7 +2817,7 @@ static struct pcpu_alloc_info * __init _
>  	dyn_size = size_sum - static_size - reserved_size;
>  
>  	/*
> -	 * Determine min_unit_size, alloc_size and max_upa such that
> +	 * Determine min_unit_size, alloc_size such that
>  	 * alloc_size is multiple of atom_size and is the smallest
>  	 * which can accommodate 4k aligned segments which are equal to
>  	 * or larger than min_unit_size.
> @@ -2829,7 +2829,6 @@ static struct pcpu_alloc_info * __init _
>  	upa = alloc_size / min_unit_size;
>  	while (alloc_size % upa || (offset_in_page(alloc_size / upa)))
>  		upa--;
> -	max_upa = upa;
>  
>  	cpumask_copy(&mask, cpu_possible_mask);
>  
> @@ -2860,7 +2859,7 @@ static struct pcpu_alloc_info * __init _
>  	 */
>  	last_allocs = INT_MAX;
>  	best_upa = 0;
> -	for (upa = max_upa; upa; upa--) {
> +	for (; upa; upa--) {
>  		int allocs = 0, wasted = 0;
>  
>  		if (alloc_size % upa || (offset_in_page(alloc_size / upa)))
> _
> 
> Patches currently in -mm which might be from niuwl1@chinatelecom.cn are
> 
> mm-percpu-drop-unused-max_upa-in-pcpu_build_alloc_info.patch
> 

I'd prefer if we dropped this because it's part of init code and makes
it clear max_upa is the outcome of the code above and the start for the
code below.

Separately, I just sent you a respin of the percpu double free check
[1].

[1] https://lore.kernel.org/lkml/20251220002737.84100-1-dennis@kernel.org/


Thanks,
Dennis