[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-Id: <169762410085.3624197.15162390059059851386.b4-ty@arm.com>
Date: Wed, 18 Oct 2023 11:15:59 +0100
From: Catalin Marinas <catalin.marinas@....com>
To: Will Deacon <will@...nel.org>, Steven Price <steven.price@....com>,
Peter Collingbourne <pcc@...gle.com>,
Ryan Roberts <ryan.roberts@....com>
Cc: linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] arm64/mm: Hoist synchronization out of set_ptes() loop
On Thu, 05 Oct 2023 15:07:30 +0100, Ryan Roberts wrote:
> set_ptes() sets a physically contiguous block of memory (which all
> belongs to the same folio) to a contiguous block of ptes. The arm64
> implementation of this previously just looped, operating on each
> individual pte. But the __sync_icache_dcache() and mte_sync_tags()
> operations can both be hoisted out of the loop so that they are
> performed once for the contiguous set of pages (which may be less than
> the whole folio). This should result in minor performance gains.
>
> [...]
Applied to arm64 (for-next/misc), thanks!
[1/1] arm64/mm: Hoist synchronization out of set_ptes() loop
https://git.kernel.org/arm64/c/3425cec42c3c
Also pushed the fix to mark 'addr' __always_unused.
--
Catalin
Powered by blists - more mailing lists