Re: [mm/page_alloc] f26b3fa046: netperf.Throughput_Mbps -18.0% regression

["ying.huang@intel.com" <ying.huang@intel.com>]

On Wed, 2022-05-11 at 19:04 +0800, Aaron Lu wrote:
> On Wed, May 11, 2022 at 09:58:23AM +0800, ying.huang@intel.com wrote:
> > On Tue, 2022-05-10 at 11:05 -0700, Linus Torvalds wrote:
> > > [ Adding locking people in case they have any input ]
> > > 
> > > On Mon, May 9, 2022 at 11:23 PM ying.huang@intel.com
> > > <ying.huang@intel.com> wrote:
> > > > 
> > > > > 
> > > > > Can you point me to the regression report? I would like to take a look,
> > > > > thanks.
> > > > 
> > > > https://lore.kernel.org/all/1425108604.10337.84.camel@linux.intel.com/
> > > 
> > > Hmm.
> > > 
> > > That explanation looks believable, except that our qspinlocks
> > > shouldn't be spinning on the lock itself, but spinning on the mcs node
> > > it inserts into the lock.
> > 
> > The referenced regression report is very old (in Feb 2015 for 3.16-
> > 3.17).  The ticket spinlock was still used at that time.  I believe that
> > things become much better after we used qspinlock.  We can test that.
> 
> 'will-it-scale/page_fault1 process mode' can greatly stress both zone
> lock and LRU lock when nr_process = nr_cpu with thp disabled. So I run
> it to see if it still makes a difference with qspinlock.
> https://github.com/antonblanchard/will-it-scale/blob/master/tests/page_fault1.c
> 
> The result on an Icelake 2 sockets server with a total of 48cores/96cpus:
> 
> tbox_group/testcase/rootfs/kconfig/compiler/nr_task/mode/test/thp_enabled/cpufreq_governor/ucode:
>   lkp-icl-2sp4/will-it-scale/debian-10.4-x86_64-20200603.cgz/x86_64-rhel-8.3/gcc-11/100%/process/page_fault1/never/performance/0xd000331
> 
> commit:
>   v5.18-rc4
>   731a704c0d8760cfd641af4bf57167d8c68f9b99
> 
>        v5.18-rc4 731a704c0d8760cfd641af4bf57
> ---------------- ---------------------------
>          %stddev     %change         %stddev
> 	     \          |                \
>   12323894           -26.0%    9125299        will-it-scale.128.processes
> 
>      22.33 ±  4%     -22.3        0.00        perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.folio_lruvec_lock_irqsave.release_pages.tlb_flush_mmu
>       9.80            -9.2        0.57 ±  3%  perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock_irqsave.folio_lruvec_lock_irqsave.__pagevec_lru_add.folio_add_lru
>      36.25            +6.7       42.94        perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.rmqueue_bulk.rmqueue.get_page_from_freelist
>       4.28 ± 10%     +34.6       38.93        perf-profile.calltrace.cycles-pp.native_queued_spin_lock_slowpath._raw_spin_lock.free_pcppages_bulk.free_unref_page_list.release_pages
>      75.05            +7.8       82.83        perf-profile.self.cycles-pp.native_queued_spin_lock_slowpath
> 
> commit 731a704c0d8760cfd641af4bf57 moves zone's lock back to above
> free_area by reverting commit a368ab67aa55("mm: move zone lock to a
> different cache line than order-0 free page lists") on top of v5.18-rc4.
> 
> The interpretation of the above result is: after the revert, performance
> dropped 26%, zone lock increased 41% from 40% to 81%, the overall lock
> contention increased 7.8% from 75% to 82.83%. So it appears it still
> makes a difference with qspinlock.

The performance impact is larger than what I have thought before.  The
test may be too stressing to be realistic?  Can you run the test with
less process number to check?

Best Regards,
Huang, Ying