linux-kernel - Re: [PATCH] mm: mmu_gather: remove __tlb_reset

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Tue, 14 May 2019 09:58:16 +0100
From:   Mel Gorman <mgorman@...e.de>
To:     Nadav Amit <namit@...are.com>
Cc:     Will Deacon <will.deacon@....com>,
        Peter Zijlstra <peterz@...radead.org>,
        Yang Shi <yang.shi@...ux.alibaba.com>,
        "jstancek@...hat.com" <jstancek@...hat.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        "stable@...r.kernel.org" <stable@...r.kernel.org>,
        Linux-MM <linux-mm@...ck.org>,
        LKML <linux-kernel@...r.kernel.org>,
        "Aneesh Kumar K . V" <aneesh.kumar@...ux.vnet.ibm.com>,
        Nick Piggin <npiggin@...il.com>,
        Minchan Kim <minchan@...nel.org>
Subject: Re: [PATCH] mm: mmu_gather: remove __tlb_reset_range() for force
 flush

On Mon, May 13, 2019 at 05:06:03PM +0000, Nadav Amit wrote:
> > On May 13, 2019, at 9:37 AM, Will Deacon <will.deacon@....com> wrote:
> > 
> > On Mon, May 13, 2019 at 09:11:38AM +0000, Nadav Amit wrote:
> >>> On May 13, 2019, at 1:36 AM, Peter Zijlstra <peterz@...radead.org> wrote:
> >>> 
> >>> On Thu, May 09, 2019 at 09:21:35PM +0000, Nadav Amit wrote:
> >>> 
> >>>>>>> And we can fix that by having tlb_finish_mmu() sync up. Never let a
> >>>>>>> concurrent tlb_finish_mmu() complete until all concurrenct mmu_gathers
> >>>>>>> have completed.
> >>>>>>> 
> >>>>>>> This should not be too hard to make happen.
> >>>>>> 
> >>>>>> This synchronization sounds much more expensive than what I proposed. But I
> >>>>>> agree that cache-lines that move from one CPU to another might become an
> >>>>>> issue. But I think that the scheme I suggested would minimize this overhead.
> >>>>> 
> >>>>> Well, it would have a lot more unconditional atomic ops. My scheme only
> >>>>> waits when there is actual concurrency.
> >>>> 
> >>>> Well, something has to give. I didn???t think that if the same core does the
> >>>> atomic op it would be too expensive.
> >>> 
> >>> They're still at least 20 cycles a pop, uncontended.
> >>> 
> >>>>> I _think_ something like the below ought to work, but its not even been
> >>>>> near a compiler. The only problem is the unconditional wakeup; we can
> >>>>> play games to avoid that if we want to continue with this.
> >>>>> 
> >>>>> Ideally we'd only do this when there's been actual overlap, but I've not
> >>>>> found a sensible way to detect that.
> >>>>> 
> >>>>> diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
> >>>>> index 4ef4bbe78a1d..b70e35792d29 100644
> >>>>> --- a/include/linux/mm_types.h
> >>>>> +++ b/include/linux/mm_types.h
> >>>>> @@ -590,7 +590,12 @@ static inline void dec_tlb_flush_pending(struct mm_struct *mm)
> >>>>> 	 *
> >>>>> 	 * Therefore we must rely on tlb_flush_*() to guarantee order.
> >>>>> 	 */
> >>>>> -	atomic_dec(&mm->tlb_flush_pending);
> >>>>> +	if (atomic_dec_and_test(&mm->tlb_flush_pending)) {
> >>>>> +		wake_up_var(&mm->tlb_flush_pending);
> >>>>> +	} else {
> >>>>> +		wait_event_var(&mm->tlb_flush_pending,
> >>>>> +			       !atomic_read_acquire(&mm->tlb_flush_pending));
> >>>>> +	}
> >>>>> }
> >>>> 
> >>>> It still seems very expensive to me, at least for certain workloads (e.g.,
> >>>> Apache with multithreaded MPM).
> >>> 
> >>> Is that Apache-MPM workload triggering this lots? Having a known
> >>> benchmark for this stuff is good for when someone has time to play with
> >>> things.
> >> 
> >> Setting Apache2 with mpm_worker causes every request to go through
> >> mmap-writev-munmap flow on every thread. I didn???t run this workload after
> >> the patches that downgrade the mmap_sem to read before the page-table
> >> zapping were introduced. I presume these patches would allow the page-table
> >> zapping to be done concurrently, and therefore would hit this flow.
> > 
> > Hmm, I don't think so: munmap() still has to take the semaphore for write
> > initially, so it will be serialised against other munmap() threads even
> > after they've downgraded afaict.
> > 
> > The initial bug report was about concurrent madvise() vs munmap().
> 
> I guess you are right (and I???m wrong).
> 
> Short search suggests that ebizzy might be affected (a thread by Mel
> Gorman): https://lkml.org/lkml/2015/2/2/493
> 

Glibc has since been fixed to be less munmap/mmap intensive and the
system CPU usage of ebizzy is generally negligible unless configured so
specifically use mmap/munmap instead of malloc/free which is unrealistic
for good application behaviour.

-- 
Mel Gorman
SUSE Labs