Date:	Tue, 20 Oct 2015 14:55:35 -0400
From:	Waiman Long <waiman.long@....com>
To:	Ling Ma <ling.ma.program@...il.com>
CC:	Peter Zijlstra <peterz@...radead.org>, mingo@...hat.com,
	linux-kernel@...r.kernel.org, Ma Ling <ling.ml@...baba-inc.com>
Subject: Re: [RFC PATCH] qspinlock: Improve performance by reducing load instruction
 rollback

On 10/19/2015 11:12 PM, Ling Ma wrote:
> 2015-10-20 1:18 GMT+08:00 Waiman Long<waiman.long@....com>:
>> On 10/18/2015 10:27 PM, ling.ma.program@...il.com wrote:
>>> From: Ma Ling<ling.ml@...baba-inc.com>
>>>
>>> All load instructions can run speculatively but they have to follow
>>> memory order rule in multiple cores as below:
>>> _x = _y = 0
>>>
>>> Processor 0                             Processor 1
>>>
>>> mov r1, [ _y]  //M1                     mov [ _x], 1  //M3
>>> mov r2, [ _x]  //M2                     mov [ _y], 1  //M4
>>>
>>> If r1 = 1, r2 must be 1
>>>
>>> In order to guarantee above rule, although Processor 0 execute
>>> M1 and M2 instruction out of order, they are kept in ROB,
>>> when load buffer for _x in Processor 0 received the update
>>> message from Processor 1, Processor 0 need to roll back
>>> from M2 instruction, which will flush the whole pipeline,
>>> the latency is over the penalty from branch prediction miss.
>>>
>>> In this patch we use lock cmpxchg instruction to force load
>>> instructions to be serialization, the destination operand
>>> receives a write cycle without regard to the result of
>>> the comparison, which can help us to reduce the penalty
>>> from load instruction roll back.
>>>
>>> Our experiment indicates the performance can be improved by 10%~15%
>>> for 2 and 3 threads cases, the conflicts from lock cache line
>>> spend them most of the time.
>>
>> What kind of performance test were you running? With the right timing, it is
>> possible that you see some performance gain. However, if the lock hold time
>> is longer so that a fair number of cmpxchg instructions have to be executed
>> before it can get the lock, you may see a performance degradation especially
>> if the lock holder needs to access the lock cacheline.
>>
>> In general, we try to avoid this kind of cmpxchg loop unless we are sure
>> that at most a few iterations of the loop may happen.
> Waiman,
>
> The machine is Haswell (2699 V3, COD off, HT on, 2 sockets)
> (we have sent test module in separate email)
>
>
>
> A.      Data is located with lock in one cache line On 2 threads cases
> (only write struct member data_a)
>
>   1.       Load version test 5 times, the cost time is below:
>
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 103904620
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 104351876
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 118599784
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 103064024
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 103389696
>
> Totally cost time is 533310000
>
> 2.       Lock cmpxchg version test 5 times, the cost time is below:
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 67081220
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 97640708
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 96439612
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 66699296
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 96464800
>
>
>
> Totally cost time is 424325636
>
>
>
> Above data shows lock cmpxchg is better about average 25% (533310000/424325636)
>
>
>
> B.      Data is located with lock in different cache line On 2 threads
> cases(only write struct member data_b)
>
>
>
> 1.       Load version test 5 times, the cost time is below:
>
>
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 174266128
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 205053924
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 160165124
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 173241552
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 205765008
>
> Totally cost time is 918491736
>
>
>
> 2.       Lock cmpxchg version test 5 times, the cost time is below:
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 113410044
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 116293104
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 116064256
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 189320876
>
> [root@...alhost spinlock]# insmod dummy.ko; rmmod dummy;dmesg -c
>
>
>
>   all cost time is 123735352
>
> Totally cost time is 658823632
>
>
>
> Above data shows lock cmpxchg is better about average 39%  (918491736/658823632)
>
>

I did see some performance improvement when I used your test program on 
a Haswell-EX system. It seems like the use of cmpxchg has forced the 
changed memory values to be visible to other processors earlier. I also 
ran your test on an older machine with Westmere-EX processors. This 
time, I didn't see any performance improvement. In fact, your change 
actually make it a tiny bit slower. So the benefit of your patch can be 
highly processor sensitive.

As other architectures like ARM & AA64 are going to adopt qspinlock in 
the near future, we will also need to make sure that it won't cause a 
regression there. So I don't see your patch has a big chance of being 
merged upstream unless you can provide a real world workload that can 
benefit from your patch. Even then, proving that it won't cause 
regression in other processors or architectures can be tedious.

Cheers,
Longman
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists