memfill v2 now with ARM and x86 implementations

From: Matthew Wilcox <willy@infradead.org>
To: Minchan Kim <minchan@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	linux-kernel@vger.kernel.org, sergey.senozhatsky@gmail.com,
	iamjoonsoo.kim@lge.com, ngupta@vflare.org,
	zhouxianrong@huawei.com, zhouxiyu@huawei.com,
	weidu.du@huawei.com, zhangshiming5@huawei.com,
	Mi.Sophia.Wang@huawei.com, won.ho.park@huawei.com
Subject: memfill v2 now with ARM and x86 implementations
Date: Sat, 11 Mar 2017 06:56:40 -0800	[thread overview]
Message-ID: <20170311145640.GB1860@bombadil.infradead.org> (raw)
In-Reply-To: <1486307804-27903-1-git-send-email-minchan@kernel.org>

On Mon, Feb 06, 2017 at 12:16:44AM +0900, Minchan Kim wrote:
> +static inline void zram_fill_page(char *ptr, unsigned long len,
> +					unsigned long value)
> +{
> +	int i;
> +	unsigned long *page = (unsigned long *)ptr;
> +
> +	WARN_ON_ONCE(!IS_ALIGNED(len, sizeof(unsigned long)));
> +
> +	if (likely(value == 0)) {
> +		memset(ptr, 0, len);
> +	} else {
> +		for (i = 0; i < len / sizeof(*page); i++)
> +			page[i] = value;
> +	}
> +}

I've hacked up memset32/memset64 for both ARM and x86 here:

http://git.infradead.org/users/willy/linux-dax.git/shortlog/refs/heads/memfill

Can you do some performance testing and see if it makes a difference?

At this point, I'd probably ask for the first 5 patches in that git
branch to be included, and leave out memfill and the shoddy testsuite.

I haven't actually tested either asm implementation ... only the
C fallback.