Messages in this thread | | | Subject | Re: [Patch v3 0/3] iommu: reduce spinlock contention on fast path | From | Qian Cai <> | Date | Fri, 6 Dec 2019 16:57:57 -0500 |
| |
> On Dec 6, 2019, at 4:38 PM, Cong Wang <xiyou.wangcong@gmail.com> wrote: > > This patchset contains three small optimizations for the global spinlock > contention in IOVA cache. Our memcache perf test shows this reduced its > p999 latency down by 45% on AMD when IOMMU is enabled.
Can you at least have a changelog compared to previous versions?
> > Cong Wang (3): > iommu: avoid unnecessary magazine allocations > iommu: optimize iova_magazine_free_pfns() > iommu: avoid taking iova_rbtree_lock twice > --- > drivers/iommu/iova.c | 75 ++++++++++++++++++++++++++------------------ > 1 file changed, 45 insertions(+), 30 deletions(-) > > -- > 2.21.0 > > _______________________________________________ > iommu mailing list > iommu@lists.linux-foundation.org > https://lists.linuxfoundation.org/mailman/listinfo/iommu
| |