Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Wed, 21 Jul 2021 21:07:08 +0000
From:   Dennis Zhou <dennis@kernel.org>
To:     Yury Norov <yury.norov@gmail.com>
Cc:     linux-kernel@vger.kernel.org, linux-mm@kvack.org,
        linux-mmc@vger.kernel.org,
        Andy Shevchenko <andriy.shevchenko@linux.intel.com>,
        Rasmus Villemoes <linux@rasmusvillemoes.dk>,
        Petr Mladek <pmladek@suse.com>,
        Sergey Senozhatsky <senozhatsky@chromium.org>,
        Tejun Heo <tj@kernel.org>,
        Steven Rostedt <rostedt@goodmis.org>,
        Wolfram Sang <wsa+renesas@sang-engineering.com>,
        Ulf Hansson <ulf.hansson@linaro.org>,
        Christoph Lameter <cl@linux.com>,
        Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [PATCH 1/4] mm/percpu: micro-optimize pcpu_is_populated()
Message-ID: <YPiMUF/sYpjPSADv@google.com>
References: <20210719021755.883182-1-yury.norov@gmail.com>
 <20210719021755.883182-2-yury.norov@gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20210719021755.883182-2-yury.norov@gmail.com>
Precedence: bulk

Hello,

On Sun, Jul 18, 2021 at 07:17:52PM -0700, Yury Norov wrote:
> bitmap_next_clear_region() calls find_next_zero_bit() and find_next_bit()
> sequentially to find a range of clear bits. In case of pcpu_is_populated()
> there's a chance to return earlier if bitmap has all bits set.
> 
> Signed-off-by: Yury Norov <yury.norov@gmail.com>
> ---
>  mm/percpu.c | 15 ++++++++-------
>  1 file changed, 8 insertions(+), 7 deletions(-)
> 
> diff --git a/mm/percpu.c b/mm/percpu.c
> index 7f2e0151c4e2..25461571dcc5 100644
> --- a/mm/percpu.c
> +++ b/mm/percpu.c
> @@ -1071,17 +1071,18 @@ static void pcpu_block_update_hint_free(struct pcpu_chunk *chunk, int bit_off,
>  static bool pcpu_is_populated(struct pcpu_chunk *chunk, int bit_off, int bits,
>  			      int *next_off)
>  {
> -	unsigned int page_start, page_end, rs, re;
> +	unsigned int start, end;
>  
> -	page_start = PFN_DOWN(bit_off * PCPU_MIN_ALLOC_SIZE);
> -	page_end = PFN_UP((bit_off + bits) * PCPU_MIN_ALLOC_SIZE);
> +	start = PFN_DOWN(bit_off * PCPU_MIN_ALLOC_SIZE);
> +	end = PFN_UP((bit_off + bits) * PCPU_MIN_ALLOC_SIZE);
>  
> -	rs = page_start;
> -	bitmap_next_clear_region(chunk->populated, &rs, &re, page_end);
> -	if (rs >= page_end)
> +	start = find_next_zero_bit(chunk->populated, end, start);
> +	if (start >= end)
>  		return true;
>  
> -	*next_off = re * PAGE_SIZE / PCPU_MIN_ALLOC_SIZE;
> +	end = find_next_bit(chunk->populated, end, start + 1);
> +
> +	*next_off = end * PAGE_SIZE / PCPU_MIN_ALLOC_SIZE;
>  	return false;
>  }
>  
> -- 
> 2.30.2
> 

Sorry for the delay.

Acked-by: Dennis Zhou <dennis@kernel.org>

Thanks,
Dennis