Re: [PATCH v6 4/8] mm: add a template-based fast path for zone-device page init

[Muchun Song <muchun.song@linux.dev>]

> On Jul 15, 2026, at 14:11, Li Zhe <lizhe.67@bytedance.com> wrote:
> 
> On 7/14/26 4:38 PM, Muchun Song wrote:
>> 
>> 
>> On 2026/7/9 19:25, Li Zhe wrote:
>>> memmap_init_zone_device() repeats nearly identical head-page
>>> initialization for each PFN. Prepare one reusable ZONE_DEVICE head-page
>>> template through the existing slow path, refresh the PFN-dependent
>>> fields in that template before each copy, and memcpy it into each
>>> destination page.
>>> 
>>> The optimized path assigns _refcount through the copied template, so
>>> keep it disabled when the page_ref_set tracepoint is enabled.
>>> 
>>> This patch accelerates head-page initialization. The pfns_per_compound
>>> == 1 case gets the full benefit here, compound tails are handled in the
>>> next patch.
>>> 
>>> Tested in a VM with a 100 GB fsdax namespace device configured with
>>> map=dev on Intel Ice Lake server. This test exercises the nd_pmem rebind
>>> path (pfns_per_compound == 1).
>>> 
>>> Test procedure:
>>> Rebind the nd_pmem driver 30 times and collect the memmap initialization
>>> time from the pr_debug() output of memmap_init_zone_device().
>>> 
>>> Base(v7.2-rc1):
>>>    First binding: 1456 ms
>>>    Average of subsequent rebinds: 244.28 ms
>>> 
>>> With this patch and its prerequisites applied:
>>>    First binding: 1440 ms
>>>    Average of subsequent rebinds: 217.19 ms
>>> 
>>> This reduces the average rebind time from 244.28 ms to 217.19 ms, or
>>> about 11%.
>>> 
>>> Signed-off-by: Li Zhe <lizhe.67@bytedance.com>
>>> ---
>>>   mm/mm_init.c | 59 +++++++++++++++++++++++++++++++++++++++++++++++++++-
>>>   1 file changed, 58 insertions(+), 1 deletion(-)
>>> 
>>> diff --git a/mm/mm_init.c b/mm/mm_init.c
>>> index 5fccfbacf855..170021e182e0 100644
>>> --- a/mm/mm_init.c
>>> +++ b/mm/mm_init.c
>>> @@ -1065,6 +1065,44 @@ static void __ref 
>>> zone_device_page_init_slow(struct page *page,
>>>           set_page_count(page, 0);
>>>   }
>>>   +static inline bool zone_device_page_init_optimization_enabled(void)
>>> +{
>>> +    /*
>>> +     * The template fast path copies a preinitialized struct page 
>>> image.
>>> +     * Skip it when the page_ref_set tracepoint is enabled.
>>> +     */
>>> +    return !page_ref_tracepoint_active(page_ref_set);
>> 
>> I'm not sure why we need to make this. I think we can go ahead and
>> perform this optimization unconditionally.
>> 
>> If we look at why the tracepoint was added in the first place, it's
>> clear that its primary purpose is for debugging, particularly regarding
>> why CMA memory pages cannot be allocated (see commit 95813b8faa0). But
>> at this stage, the struct page hasn't even been initialized yet; we
>> are currently in the middle of initialization. From the perspective of
>> the page being initialized, the user doesn't need to care at all about
>> how the reference count is changing. Therefore, I don't think we need to
>> include these checks. It would only increase the complexity of the code.
>> 
>> I suggest deleting this section of code to keep things simple.
>> 
>> Thanks.
> 
> I would prefer to keep this check.
> 
> When page_ref_set debugging is enabled, preserving the existing per-page
> set_page_count() behavior is still useful, including during
> initialization, because it keeps the debug trace complete and unchanged.

When you say 'useful,' what aspects are you referring to?

> 
> In the normal case, page_ref_set is off by default, so this only affects
> an explicit debugging configuration and leaves the production fast path
> unchanged. The extra logic here is also fairly small and stays local to
> this optimization path.

I know there's no real overhead here. The point I'm actually trying to make
is that since deleting the code has no impact, why not go with the simpler
approach?

Thanks.

> 
> Thanks,
> Zhe
> 
>>> +}
>>> +
>>> +/*
>>> + * 'template' is a reusable page prototype rather than a strictly 
>>> immutable
>>> + * object. Most ZONE_DEVICE fields stay constant across the pages 
>>> covered by
>>> + * the current template, but section bits and page->virtual may 
>>> still depend
>>> + * on the PFN. Refresh those PFN-dependent fields in the template 
>>> before
>>> + * copying it into @page.
>>> + */
>>> +static inline void zone_device_page_update_template(struct page 
>>> *template,
>>> +        unsigned long pfn)
>>> +{
>>> +    set_page_section_from_pfn(template, pfn);
>>> +#ifdef WANT_PAGE_VIRTUAL
>>> +    if (!is_highmem_idx(ZONE_DEVICE))
>>> +        set_page_address(template, __va(pfn << PAGE_SHIFT));
>>> +#endif
>>> +}
>>> +
>>> +static void zone_device_page_init_from_template(struct page *page,
>>> +        unsigned long pfn, struct page *template)
>>> +{
>>> +    /*
>>> +     * 'template' carries the invariant portion of a ZONE_DEVICE struct
>>> +     * page. Update the PFN-dependent fields in place before copying it
>>> +     * to the destination page.
>>> +     */
>>> +    zone_device_page_update_template(template, pfn);
>>> +    memcpy(page, template, sizeof(*page));
>>> +}
>>> +
>>>   /*
>>>    * With compound page geometry and when struct pages are stored in 
>>> ram most
>>>    * tail pages are reused. Consequently, the amount of unique struct 
>>> pages to
>>> @@ -1120,6 +1158,7 @@ void __ref memmap_init_zone_device(struct zone 
>>> *zone,
>>>                      unsigned long nr_pages,
>>>                      struct dev_pagemap *pgmap)
>>>   {
>>> +    bool use_template = zone_device_page_init_optimization_enabled();
>>>       unsigned long pfn, end_pfn = start_pfn + nr_pages;
>>>       struct pglist_data *pgdat = zone->zone_pgdat;
>>>       struct vmem_altmap *altmap = pgmap_altmap(pgmap);
>>> @@ -1127,6 +1166,7 @@ void __ref memmap_init_zone_device(struct zone 
>>> *zone,
>>>       unsigned long zone_idx = zone_idx(zone);
>>>       unsigned long start = jiffies;
>>>       int nid = pgdat->node_id;
>>> +    struct page template;
>>> 
>>>       if (WARN_ON_ONCE(!pgmap || zone_idx != ZONE_DEVICE))
>>>           return;
>>> @@ -1144,7 +1184,24 @@ void __ref memmap_init_zone_device(struct zone 
>>> *zone,
>>>       for (pfn = start_pfn; pfn < end_pfn; pfn += pfns_per_compound) {
>>>           struct page *page = pfn_to_page(pfn);
>>> 
>>> -        zone_device_page_init_slow(page, pfn, zone_idx, nid, pgmap);
>>> +        if (!use_template) {
>>> +            zone_device_page_init_slow(page, pfn, zone_idx,
>>> +                           nid, pgmap);
>>> +        } else if (pfn == start_pfn) {
>>> +            /*
>>> +             * Seed the reusable head-page template from the
>>> +             * first real struct page, because the existing
>>> +             * page-init and pageblock helpers expect a real
>>> +             * memmap entry rather than a stack object.
>>> +             */
>>> +            zone_device_page_init_slow(page, pfn, zone_idx,
>>> +                           nid, pgmap);
>>> +            /* init template page */
>>> +            memcpy(&template, page, sizeof(*page));
>>> +        } else {
>>> +            zone_device_page_init_from_template(page, pfn,
>>> +                                &template);
>>> +        }
>>> 
>>>           if (IS_ALIGNED(pfn, PAGES_PER_SECTION))
>>>               cond_resched();
>>> -- 
>>> 2.20.1