Date:   Fri, 27 May 2022 10:53:36 -0400
From:   Peter Xu <peterx@...hat.com>
To:     Ingo Molnar <mingo@...nel.org>
Cc:     linux-kernel@...r.kernel.org, linux-mm@...ck.org,
        Richard Henderson <rth@...ddle.net>,
        David Hildenbrand <david@...hat.com>,
        Matt Turner <mattst88@...il.com>,
        Albert Ou <aou@...s.berkeley.edu>,
        Michal Simek <monstr@...str.eu>,
        Russell King <linux@...linux.org.uk>,
        Ivan Kokshaysky <ink@...assic.park.msu.ru>,
        linux-riscv@...ts.infradead.org,
        Alexander Gordeev <agordeev@...ux.ibm.com>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        Jonas Bonn <jonas@...thpole.se>, Will Deacon <will@...nel.org>,
        "James E . J . Bottomley" <James.Bottomley@...senpartnership.com>,
        "H . Peter Anvin" <hpa@...or.com>,
        Andrea Arcangeli <aarcange@...hat.com>,
        openrisc@...ts.librecores.org, linux-s390@...r.kernel.org,
        Ingo Molnar <mingo@...hat.com>,
        linux-m68k@...ts.linux-m68k.org,
        Palmer Dabbelt <palmer@...belt.com>,
        Heiko Carstens <hca@...ux.ibm.com>,
        Chris Zankel <chris@...kel.net>,
        Peter Zijlstra <peterz@...radead.org>,
        Alistair Popple <apopple@...dia.com>,
        linux-csky@...r.kernel.org, linux-hexagon@...r.kernel.org,
        Vlastimil Babka <vbabka@...e.cz>,
        Thomas Gleixner <tglx@...utronix.de>,
        sparclinux@...r.kernel.org,
        Christian Borntraeger <borntraeger@...ux.ibm.com>,
        Stafford Horne <shorne@...il.com>,
        Michael Ellerman <mpe@...erman.id.au>, x86@...nel.org,
        Thomas Bogendoerfer <tsbogend@...ha.franken.de>,
        Paul Mackerras <paulus@...ba.org>,
        linux-arm-kernel@...ts.infradead.org,
        Sven Schnelle <svens@...ux.ibm.com>,
        Benjamin Herrenschmidt <benh@...nel.crashing.org>,
        linux-xtensa@...ux-xtensa.org, Nicholas Piggin <npiggin@...il.com>,
        linux-sh@...r.kernel.org, Vasily Gorbik <gor@...ux.ibm.com>,
        Borislav Petkov <bp@...en8.de>, linux-mips@...r.kernel.org,
        Max Filippov <jcmvbkbc@...il.com>,
        Helge Deller <deller@....de>, Vineet Gupta <vgupta@...nel.org>,
        Al Viro <viro@...iv.linux.org.uk>,
        Paul Walmsley <paul.walmsley@...ive.com>,
        Johannes Weiner <hannes@...xchg.org>,
        Anton Ivanov <anton.ivanov@...bridgegreys.com>,
        Catalin Marinas <catalin.marinas@....com>,
        linux-um@...ts.infradead.org, linux-alpha@...r.kernel.org,
        Johannes Berg <johannes@...solutions.net>,
        linux-ia64@...r.kernel.org,
        Geert Uytterhoeven <geert@...ux-m68k.org>,
        Dinh Nguyen <dinguyen@...nel.org>, Guo Ren <guoren@...nel.org>,
        linux-snps-arc@...ts.infradead.org,
        Hugh Dickins <hughd@...gle.com>, Rich Felker <dalias@...c.org>,
        Andy Lutomirski <luto@...nel.org>,
        Richard Weinberger <richard@....at>,
        linuxppc-dev@...ts.ozlabs.org, Brian Cain <bcain@...cinc.com>,
        Yoshinori Sato <ysato@...rs.sourceforge.jp>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Stefan Kristiansson <stefan.kristiansson@...nalahti.fi>,
        linux-parisc@...r.kernel.org,
        "David S . Miller" <davem@...emloft.net>
Subject: Re: [PATCH v3] mm: Avoid unnecessary page fault retires on shared
 memory types

On Fri, May 27, 2022 at 12:46:31PM +0200, Ingo Molnar wrote:
> 
> * Peter Xu <peterx@...hat.com> wrote:
> 
> > This patch provides a ~12% perf boost on my aarch64 test VM with a simple
> > program sequentially dirtying 400MB shmem file being mmap()ed and these are
> > the time it needs:
> >
> >   Before: 650.980 ms (+-1.94%)
> >   After:  569.396 ms (+-1.38%)
> 
> Nice!
> 
> >  arch/x86/mm/fault.c           |  4 ++++
> 
> Reviewed-by: Ingo Molnar <mingo@...nel.org>
> 
> Minor comment typo:
> 
> > +		/*
> > +		 * We should do the same as VM_FAULT_RETRY, but let's not
> > +		 * return -EBUSY since that's not reflecting the reality on
> > +		 * what has happened - we've just fully completed a page
> > +		 * fault, with the mmap lock released.  Use -EAGAIN to show
> > +		 * that we want to take the mmap lock _again_.
> > +		 */
> 
> s/reflecting the reality on what has happened
>  /reflecting the reality of what has happened

Will fix.

> 
> >  	ret = handle_mm_fault(vma, address, fault_flags, NULL);
> > +
> > +	if (ret & VM_FAULT_COMPLETED) {
> > +		/*
> > +		 * NOTE: it's a pity that we need to retake the lock here
> > +		 * to pair with the unlock() in the callers. Ideally we
> > +		 * could tell the callers so they do not need to unlock.
> > +		 */
> > +		mmap_read_lock(mm);
> > +		*unlocked = true;
> > +		return 0;
> 
> Indeed that's a pity - I guess more performance could be gained here, 
> especially in highly parallel threaded workloads?

Yes I think so.

The patch avoids the page fault retry, including the mmap lock/unlock side.
Now if we retake the lock for fixup_user_fault() we still safe time for
pgtable walks but the lock overhead will be somehow kept, just with smaller
critical sections.

Some fixup_user_fault() callers won't be affected as long as unlocked==NULL
is passed - e.g. the futex code path (fault_in_user_writeable).  After all
they never needed to retake the lock before/after this patch.

It's about the other callers, and they may need some more touch-ups case by
case.  Examples are follow_fault_pfn() in vfio and hva_to_pfn_remapped() in
KVM: both of them returns -EAGAIN when *unlocked==true.  We need to teach
them to know "*unlocked==true" does not necessarily mean a retry attempt.

I think I can look into them if this patch can be accepted as a follow up.

Thanks for taking a look!

-- 
Peter Xu

Powered by blists - more mailing lists