Re: [PATCH v4 29/33] x86/mm: try VMA lock-based page fault handling first

From: Suren Baghdasaryan <surenb@google.com>
To: "Holger Hoffstätte" <holger@applied-asynchrony.com>
Cc: Jiri Slaby <jirislaby@kernel.org>,
	akpm@linux-foundation.org, michel@lespinasse.org,
	 jglisse@google.com, mhocko@suse.com, vbabka@suse.cz,
	hannes@cmpxchg.org,  mgorman@techsingularity.net,
	dave@stgolabs.net, willy@infradead.org,  liam.howlett@oracle.com,
	peterz@infradead.org, ldufour@linux.ibm.com,  paulmck@kernel.org,
	mingo@redhat.com, will@kernel.org, luto@kernel.org,
	 songliubraving@fb.com, peterx@redhat.com, david@redhat.com,
	 dhowells@redhat.com, hughd@google.com, bigeasy@linutronix.de,
	 kent.overstreet@linux.dev, punit.agrawal@bytedance.com,
	lstoakes@gmail.com,  peterjung1337@gmail.com,
	rientjes@google.com, chriscli@google.com,
	 axelrasmussen@google.com, joelaf@google.com, minchan@google.com,
	 rppt@kernel.org, jannh@google.com, shakeelb@google.com,
	tatashin@google.com,  edumazet@google.com, gthelen@google.com,
	linux-mm <linux-mm@kvack.org>
Subject: Re: [PATCH v4 29/33] x86/mm: try VMA lock-based page fault handling first
Date: Thu, 6 Jul 2023 21:40:00 -0700	[thread overview]
Message-ID: <CAJuCfpGQzcPzMQppa=vENyc1NX263W6TE4qGiO8hLzvZp+pZHA@mail.gmail.com> (raw)
In-Reply-To: <CAJuCfpHeU3esC47Kg_g0rOXFDiGe-Q8JEJ+kP96mRJq9rLsdpA@mail.gmail.com>

On Thu, Jul 6, 2023 at 7:23 PM Suren Baghdasaryan <surenb@google.com> wrote:
>
> On Thu, Jul 6, 2023 at 9:11 AM Suren Baghdasaryan <surenb@google.com> wrote:
> >
> > On Thu, Jul 6, 2023 at 7:27 AM Holger Hoffstätte
> > <holger@applied-asynchrony.com> wrote:
> > >
> > > On 2023-07-06 00:55, Suren Baghdasaryan wrote:
> > > > On Wed, Jul 5, 2023 at 3:37 PM Holger Hoffstätte
> > > > <holger@applied-asynchrony.com> wrote:
> > > >>> Jiri, Holger, would you be able to try
> > > >>> https://lore.kernel.org/all/20230705171213.2843068-2-surenb@google.com/
> > > >>> and see if your issues still exist?
> > > >>
> > > >> Just in time! Not 2 minutes ago I finished rebuilding 6.4.2 + the last version of
> > > >> your patches on a second machine (old Intel Sandy Bridge workstation) to be my
> > > >> crash test dummy. I removed the BROKEN dependency in mm/Kconfig, manually set
> > > >> PER_VMA_LOCK=y and ... it seems to work?! Boots fine, Firefox seems to work
> > > >> (but no exhaustive tests yet). I will also rerun a few reboot laps, just to
> > > >> exercise this a bit harder and see if something comes up.
> > > >>
> > > >> Tomorrow I'll also try again on my Zen2 Thinkpad and will report back.
> > > >>
> > > >> Fingers crossed!
> > > >
> > > > Thanks! This is promising.
> > >
> > > Indeed it was, and still is. :)
> > >
> > > This morning I wrangled 6.4.2 + v4 of the patches into all my machines,
> > > enabled PER_VMA_LOCK=y, removed BROKEN and so far everything has been humming
> > > along just fine. One machine has been compiling for several hours without issue.
> > > My Zen2 thinkpad - which was previously really unhappy with enabled PER_VMA_LOCK -
> > > has booted a few times without hiccup, and Firefox has been happily roaming the
> > > interwebs for several hours as well. \o/
> >
> > This is great! Thanks so much for verifying!
> > Andrew, if it's not too late, maybe we can drop the BROKEN dependency
> > now that we have this confirmation?
>
> Liam pointed me to another possible issue with per-VMA locks due to
> the recent changes in stack expansion locking rules. So, it's probably
> safer to keep it marked BROKEN. I'll post a simple fix for that
> shortly.

The fix along with another less critical one is posted at
https://lore.kernel.org/all/20230707043211.3682710-1-surenb@google.com/

>
> >
> > >
> > > cheers
> > > Holger