Re: [PATCH] mm: increase totalram_pages on freeing to buddy system

[Wei Yang <richard.weiyang@gmail.com>]

On Thu, Jun 06, 2024 at 09:14:25AM +0200, David Hildenbrand wrote:
>On 06.06.24 00:44, Wei Yang wrote:
>> On Mon, Jun 03, 2024 at 10:43:51PM +0200, David Hildenbrand wrote:
>> > On 03.06.24 22:01, Wei Yang wrote:
>> > > On Mon, Jun 03, 2024 at 10:55:10AM +0200, David Hildenbrand wrote:
>> > > > On 02.06.24 02:58, Wei Yang wrote:
>> > > > > On Sat, Jun 01, 2024 at 06:15:33PM +0200, David Hildenbrand wrote:
>> > > > > > On 01.06.24 17:32, David Hildenbrand wrote:
>> > > > > > > On 01.06.24 15:34, Wei Yang wrote:
>> > > > > > > > Total memory represents pages managed by buddy system.
>> > > > > > > 
>> > > > > > > No, that's managed pages.
>> > > > > > > 
>> > > > > > > > After the
>> > > > > > > > introduction of DEFERRED_STRUCT_PAGE_INIT, it may count the pages before
>> > > > > > > > being managed.
>> > > > > > > > 
>> > > > > > > 
>> > > > > > > I recall one reason that is done, so other subsystem know the total
>> > > > > > > memory size even before deferred init is done.
>> > > > > > > 
>> > > > > > > > free_low_memory_core_early() returns number of pages for all free pages,
>> > > > > > > > even at this moment only early initialized pages are freed to buddy
>> > > > > > > > system. This means the total memory at this moment is not correct.
>> > > > > > > > 
>> > > > > > > > Let's increase it when pages are freed to buddy system.
>> > > > > > > 
>> > > > > > > I'm missing the "why", and the very first sentence of this patch is wrong.
>> > > > > > 
>> > > > > > Correction: your statement was correct :) That's why
>> > > > > > adjust_managed_page_count() adjusts that as well.
>> > > > > > 
>> > > > > > __free_pages_core() only adjusts managed page count, because it assumes
>> > > > > > totalram has already been adjusted early during boot.
>> > > > > > 
>> > > > > > The reason we have this split for now, I think, is because of subsystems that
>> > > > > > call totalram_pages() during init.
>> > > > > > 
>> > > > > > So the "why" question remains, because this change has the potential to break
>> > > > > > other stuff.
>> > > > > > 
>> > > > > 
>> > > > > Thanks, I didn't notice this.
>> > > > 
>> > > > I think having your cleanup would be very nice, as I have patches in the
>> > > > works that would benefit from being able to move the totalram update from
>> > > > memory hotplug code to __free_pages_core().
>> > > > 
>> > > 
>> > > I got the same feeling.
>> > > 
>> > > > We'd have to make sure that no code relies on totalram being sane/fixed
>> > > > during boot for the initial memory. I think right now we might have such
>> > > > code.
>> > > > 
>> > > 
>> > > One concern is totalram would change when hotplug is enabled. That sounds
>> > > those codes should do some re-calculation after totalram changes?
>> > 
>> > We don't have such code in place -- there were discussions regarding that
>> > recently.
>> > 
>> > It would be reasonable to take a look at all totalram_pages() users and
>> > determine if they could be affected by deferring updating it.
>> > 
>> > At least page_alloc_init_late()->deferred_init_memmap() happens before
>> > do_basic_setup()->do_initcalls(), which is good.
>> > 
>> 
>> But deferred_init_memmap() will spawn threads to do the work. I am afraid
>> do_initcalls() won't wait for the completion of defer_init? Do I miss
>> something?
>
>Don't we wait for them to finish?
>
>/* Block until all are initialised */
>wait_for_completion(&pgdat_init_all_done_comp);

You are right. I missed this.

So we still need to wait for mm fully initialized and then to initialise
others. I thought everything would start in parallel.

>
>-- 
>Cheers,
>
>David / dhildenb

-- 
Wei Yang
Help you, Help me