Date:	Mon, 07 Apr 2014 11:44:50 +0530
From:	Raghavendra K T <raghavendra.kt@...ux.vnet.ibm.com>
To:	Waiman Long <Waiman.Long@...com>
CC:	Thomas Gleixner <tglx@...utronix.de>,
	Ingo Molnar <mingo@...hat.com>,
	"H. Peter Anvin" <hpa@...or.com>,
	Peter Zijlstra <peterz@...radead.org>,
	linux-arch@...r.kernel.org, x86@...nel.org,
	linux-kernel@...r.kernel.org,
	virtualization@...ts.linux-foundation.org,
	xen-devel@...ts.xenproject.org, kvm@...r.kernel.org,
	Paolo Bonzini <paolo.bonzini@...il.com>,
	Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>,
	"Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>,
	Rik van Riel <riel@...hat.com>,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	David Vrabel <david.vrabel@...rix.com>,
	Oleg Nesterov <oleg@...hat.com>,
	Gleb Natapov <gleb@...hat.com>,
	Aswin Chandramouleeswaran <aswin@...com>,
	Scott J Norton <scott.norton@...com>,
	Chegu Vinod <chegu_vinod@...com>
Subject: Re: [PATCH v8 00/10] qspinlock: a 4-byte queue spinlock with PV support

On 04/02/2014 06:57 PM, Waiman Long wrote:
> N.B. Sorry for the duplicate. This patch series were resent as the
>       original one was rejected by the vger.kernel.org list server
>       due to long header. There is no change in content.
>
> v7->v8:
>    - Remove one unneeded atomic operation from the slowpath, thus
>      improving performance.
>    - Simplify some of the codes and add more comments.
>    - Test for X86_FEATURE_HYPERVISOR CPU feature bit to enable/disable
>      unfair lock.
>    - Reduce unfair lock slowpath lock stealing frequency depending
>      on its distance from the queue head.
>    - Add performance data for IvyBridge-EX CPU.
>
> v6->v7:
>    - Remove an atomic operation from the 2-task contending code
>    - Shorten the names of some macros
>    - Make the queue waiter to attempt to steal lock when unfair lock is
>      enabled.
>    - Remove lock holder kick from the PV code and fix a race condition
>    - Run the unfair lock & PV code on overcommitted KVM guests to collect
>      performance data.
>
> v5->v6:
>   - Change the optimized 2-task contending code to make it fairer at the
>     expense of a bit of performance.
>   - Add a patch to support unfair queue spinlock for Xen.
>   - Modify the PV qspinlock code to follow what was done in the PV
>     ticketlock.
>   - Add performance data for the unfair lock as well as the PV
>     support code.
>
> v4->v5:
>   - Move the optimized 2-task contending code to the generic file to
>     enable more architectures to use it without code duplication.
>   - Address some of the style-related comments by PeterZ.
>   - Allow the use of unfair queue spinlock in a real para-virtualized
>     execution environment.
>   - Add para-virtualization support to the qspinlock code by ensuring
>     that the lock holder and queue head stay alive as much as possible.
>
> v3->v4:
>   - Remove debugging code and fix a configuration error
>   - Simplify the qspinlock structure and streamline the code to make it
>     perform a bit better
>   - Add an x86 version of asm/qspinlock.h for holding x86 specific
>     optimization.
>   - Add an optimized x86 code path for 2 contending tasks to improve
>     low contention performance.
>
> v2->v3:
>   - Simplify the code by using numerous mode only without an unfair option.
>   - Use the latest smp_load_acquire()/smp_store_release() barriers.
>   - Move the queue spinlock code to kernel/locking.
>   - Make the use of queue spinlock the default for x86-64 without user
>     configuration.
>   - Additional performance tuning.
>
> v1->v2:
>   - Add some more comments to document what the code does.
>   - Add a numerous CPU mode to support >= 16K CPUs
>   - Add a configuration option to allow lock stealing which can further
>     improve performance in many cases.
>   - Enable wakeup of queue head CPU at unlock time for non-numerous
>     CPU mode.
>
> This patch set has 3 different sections:
>   1) Patches 1-4: Introduces a queue-based spinlock implementation that
>      can replace the default ticket spinlock without increasing the
>      size of the spinlock data structure. As a result, critical kernel
>      data structures that embed spinlock won't increase in size and
>      break data alignments.
>   2) Patches 5-6: Enables the use of unfair queue spinlock in a
>      para-virtualized execution environment. This can resolve some
>      of the locking related performance issues due to the fact that
>      the next CPU to get the lock may have been scheduled out for a
>      period of time.
>   3) Patches 7-10: Enable qspinlock para-virtualization support
>      by halting the waiting CPUs after spinning for a certain amount of
>      time. The unlock code will detect the a sleeping waiter and wake it
>      up. This is essentially the same logic as the PV ticketlock code.
>
> The queue spinlock has slightly better performance than the ticket
> spinlock in uncontended case. Its performance can be much better
> with moderate to heavy contention.  This patch has the potential of
> improving the performance of all the workloads that have moderate to
> heavy spinlock contention.
>
> The queue spinlock is especially suitable for NUMA machines with at
> least 2 sockets, though noticeable performance benefit probably won't
> show up in machines with less than 4 sockets.
>
> The purpose of this patch set is not to solve any particular spinlock
> contention problems. Those need to be solved by refactoring the code
> to make more efficient use of the lock or finer granularity ones. The
> main purpose is to make the lock contention problems more tolerable
> until someone can spend the time and effort to fix them.
>
> To illustrate the performance benefit of the queue spinlock, the
> ebizzy benchmark was run with the -m option in two different computers:
>
>    Test machine		ticket-lock		queue-lock
>    ------------		-----------		----------
>    4-socket 40-core	2316 rec/s		2899 rec/s
>    Westmere-EX (HT off)
>    2-socket 12-core	2130 rec/s		2176 rec/s
>    Westmere-EP (HT on)
>

I tested the v7,v8 of qspinlock with unfair config on kvm guest.
I was curious about unfair locks performance in undercommit cases. 
(overcommit case is expected to perform well)

But I am seeing hang in overcommit cases. Gdb showed that many vcpus
are halted and there was no progress. Suspecting the problem /race with
halting, I removed the halt() part of kvm_hibernate().  I am yet to take 
a closer look at the code on halt() related changes.

Patch series with that change gave around 20% improvement for dbench 2x 
and 30% improvement for ebizzy 2x cases. (1x has no significant loss/gain).


--
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