Re: [PATCH v3 3/3] arm64: mm: enable per pmd page table lock

2019-03-11 Thread Yu Zhao
On Mon, Mar 11, 2019 at 12:12:28PM +, Mark Rutland wrote: > Hi, > > On Sat, Mar 09, 2019 at 06:19:06PM -0700, Yu Zhao wrote: > > Switch from per mm_struct to per pmd page table lock by enabling > > ARCH_ENABLE_SPLIT_PMD_PTLOCK. This provides better granularity for > > large system. > > > >

Re: [PATCH v3 3/3] arm64: mm: enable per pmd page table lock

2019-03-11 Thread Yu Zhao
On Mon, Mar 11, 2019 at 01:58:27PM +0530, Anshuman Khandual wrote: > On 03/10/2019 06:49 AM, Yu Zhao wrote: > > Switch from per mm_struct to per pmd page table lock by enabling > > ARCH_ENABLE_SPLIT_PMD_PTLOCK. This provides better granularity for > > large system. > > > > I'm not sure if there

Re: [PATCH v3 3/3] arm64: mm: enable per pmd page table lock

2019-03-11 Thread Anshuman Khandual
On 03/11/2019 05:42 PM, Mark Rutland wrote: > Hi, > > On Sat, Mar 09, 2019 at 06:19:06PM -0700, Yu Zhao wrote: >> Switch from per mm_struct to per pmd page table lock by enabling >> ARCH_ENABLE_SPLIT_PMD_PTLOCK. This provides better granularity for >> large system. >> >> I'm not sure if there is

Re: [PATCH v3 3/3] arm64: mm: enable per pmd page table lock

2019-03-11 Thread Mark Rutland
Hi, On Sat, Mar 09, 2019 at 06:19:06PM -0700, Yu Zhao wrote: > Switch from per mm_struct to per pmd page table lock by enabling > ARCH_ENABLE_SPLIT_PMD_PTLOCK. This provides better granularity for > large system. > > I'm not sure if there is contention on mm->page_table_lock. Given > the option

Re: [PATCH v3 3/3] arm64: mm: enable per pmd page table lock

2019-03-11 Thread Anshuman Khandual
On 03/10/2019 06:49 AM, Yu Zhao wrote: > Switch from per mm_struct to per pmd page table lock by enabling > ARCH_ENABLE_SPLIT_PMD_PTLOCK. This provides better granularity for > large system. > > I'm not sure if there is contention on mm->page_table_lock. Given > the option comes at no cost (apart

[PATCH v3 3/3] arm64: mm: enable per pmd page table lock

2019-03-09 Thread Yu Zhao
Switch from per mm_struct to per pmd page table lock by enabling ARCH_ENABLE_SPLIT_PMD_PTLOCK. This provides better granularity for large system. I'm not sure if there is contention on mm->page_table_lock. Given the option comes at no cost (apart from initializing more spin locks), why not enable