Re: [GIT PULL] MM updates for 6.4-rc1

From: Suren Baghdasaryan <surenb@google.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	mm-commits@vger.kernel.org, linux-mm@kvack.org,
	linux-kernel@vger.kernel.org,
	Lorenzo Stoakes <lstoakes@gmail.com>,
	Vlastimil Babka <vbabka@suse.cz>
Subject: Re: [GIT PULL] MM updates for 6.4-rc1
Date: Fri, 28 Apr 2023 09:03:27 -0700	[thread overview]
Message-ID: <CAJuCfpHZSEjG2nti5iWbPQjWgXcHZ95hFS2FKwaPC-sZPdSznw@mail.gmail.com> (raw)
In-Reply-To: <CAHk-=wi5=LfeCMv5ZPAtzWhJyW=y1rVJFTRBfgcqzoUNk5NRDA@mail.gmail.com>

On Thu, Apr 27, 2023 at 8:03 PM Linus Torvalds
<torvalds@linux-foundation.org> wrote:
>
> On Thu, Apr 27, 2023 at 3:33 PM Andrew Morton <akpm@linux-foundation.org> wrote:
> >
> > - Suren Baghdasaryan has improved mmap_lock scalability by switching to
> >   per-VMA locking.
>
> .. related to this, we have that new PER_VMA_LOCK_STATS config entry,
> which is 'default y' and has basically no help test.
>
> Please don't do that.
>
> I don't think any of the VM debugging stuff should likely be 'default
> y' in the first place, but if they are, they should most definitely
> have a good *explanation* for why they should be on for a normal user.

Understood.
I wanted these stats enabled by default to be able to identify
possible pathological cases and to also let users disable them if they
can't tolerate even a small overhead in the pagefault path. Should I
document this reasoning for the config option?

>
>                       Linus