Re: [PATCH v5 21/25] arm64/mm: Implement new [get_and_]clear_full_ptes() batch APIs

2024-02-13 Thread Mark Rutland
On Tue, Feb 13, 2024 at 04:48:50PM +, Ryan Roberts wrote: > On 13/02/2024 16:43, Mark Rutland wrote: > > On Fri, Feb 02, 2024 at 08:07:52AM +, Ryan Roberts wrote: > >> +static inline void __clear_full_ptes(struct mm_struct *mm, unsigned long > >> addr, > >> +

Re: [PATCH v5 21/25] arm64/mm: Implement new [get_and_]clear_full_ptes() batch APIs

2024-02-13 Thread Ryan Roberts
On 13/02/2024 16:43, Mark Rutland wrote: > On Fri, Feb 02, 2024 at 08:07:52AM +, Ryan Roberts wrote: >> Optimize the contpte implementation to fix some of the >> exit/munmap/dontneed performance regression introduced by the initial >> contpte commit. Subsequent patches will solve it entirely.

Re: [PATCH v5 21/25] arm64/mm: Implement new [get_and_]clear_full_ptes() batch APIs

2024-02-13 Thread Mark Rutland
On Fri, Feb 02, 2024 at 08:07:52AM +, Ryan Roberts wrote: > Optimize the contpte implementation to fix some of the > exit/munmap/dontneed performance regression introduced by the initial > contpte commit. Subsequent patches will solve it entirely. > > During exit(), munmap() or

[PATCH v5 21/25] arm64/mm: Implement new [get_and_]clear_full_ptes() batch APIs

2024-02-02 Thread Ryan Roberts
Optimize the contpte implementation to fix some of the exit/munmap/dontneed performance regression introduced by the initial contpte commit. Subsequent patches will solve it entirely. During exit(), munmap() or madvise(MADV_DONTNEED), mappings must be cleared. Previously this was done 1 PTE at a