Re: [dpdk-dev] [PATCH v7 3/7] spinlock: use wfe to reduce contention on aarch64

2019-10-21 Thread Gavin Hu (Arm Technology China)
; ; Honnappa Nagarahalli > ; Ruifeng Wang (Arm Technology China) > ; Phil Yang (Arm Technology China) > ; Steve Capper > Subject: Re: [dpdk-dev] [PATCH v7 3/7] spinlock: use wfe to reduce > contention on aarch64 > > On Fri, Sep 27, 2019 at 7:43 AM Gavin Hu wrote: > > > >

Re: [dpdk-dev] [PATCH v7 3/7] spinlock: use wfe to reduce contention on aarch64

2019-10-17 Thread Gavin Hu (Arm Technology China)
; ; Honnappa Nagarahalli > ; Ruifeng Wang (Arm Technology China) > ; Phil Yang (Arm Technology China) > ; Steve Capper > Subject: Re: [dpdk-dev] [PATCH v7 3/7] spinlock: use wfe to reduce > contention on aarch64 > > On Fri, Sep 27, 2019 at 7:43 AM Gavin Hu wrote: > > > >

Re: [dpdk-dev] [PATCH v7 3/7] spinlock: use wfe to reduce contention on aarch64

2019-10-17 Thread David Marchand
On Fri, Sep 27, 2019 at 7:43 AM Gavin Hu wrote: > > In acquiring a spinlock, cores repeatedly poll the lock variable. > This is replaced by rte_wait_until_equal API. > > Running the micro benchmarking and the testpmd and l3fwd traffic tests > on ThunderX2, Ampere eMAG80 and Arm N1SDP, everything w

[dpdk-dev] [PATCH v7 3/7] spinlock: use wfe to reduce contention on aarch64

2019-09-26 Thread Gavin Hu
In acquiring a spinlock, cores repeatedly poll the lock variable. This is replaced by rte_wait_until_equal API. Running the micro benchmarking and the testpmd and l3fwd traffic tests on ThunderX2, Ampere eMAG80 and Arm N1SDP, everything went well and no notable performance gain nor degradation was