Re: [PATCH v4 3/3] zram: implement rw_page operation of zram

[Minchan Kim <minchan@kernel.org>]

Hi karam,

On Wed, Oct 22, 2014 at 04:04:14PM +0900, karam.lee@lge.com wrote:
> From: "karam.lee" <karam.lee@lge.com>
> 
> This patch implements rw_page operation for zram block device.
> 
> I implemented the feature in zram and tested it.
> Test bed was the G2, LG electronic mobile device, whtich has msm8974
> processor and 2GB memory.
> With a memory allocation test program consuming memory, the system
> generates swap.
> And operating time of swap_write_page() was measured.
> 
> --------------------------------------------------
> |             |   operating time   | improvement |
> |             |  (20 runs average) |             |
> --------------------------------------------------
> |with patch   |    1061.15 us      |    +2.4%    |
> --------------------------------------------------
> |without patch|    1087.35 us      |             |
> --------------------------------------------------
> 
> Each test(with paged_io,with BIO) result set shows normal distribution
> and has equal variance.
> I mean the two values are valid result to compare.
> I can say operation with paged I/O(without BIO) is faster 2.4% with
> confidence level 95%.
> 
> Signed-off-by: karam.lee <karam.lee@lge.com>
> ---
>  drivers/block/zram/zram_drv.c |   38 ++++++++++++++++++++++++++++++++++++++
>  1 file changed, 38 insertions(+)
> 
> diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c
> index 4565fdc..8bbd4f2 100644
> --- a/drivers/block/zram/zram_drv.c
> +++ b/drivers/block/zram/zram_drv.c
> @@ -810,8 +810,46 @@ static void zram_slot_free_notify(struct block_device *bdev,
>  	atomic64_inc(&zram->stats.notify_free);
>  }
>  
> +static int zram_rw_page(struct block_device *bdev, sector_t sector,
> +		       struct page *page, int rw)
> +{
> +	int offset, ret;
> +	u32 index;
> +	struct zram *zram;
> +	struct bio_vec bv;
> +
> +	zram = bdev->bd_disk->private_data;
> +	if (!valid_io_request(zram, sector, PAGE_SIZE)) {
> +		atomic64_inc(&zram->stats.invalid_io);
> +		ret = -EINVAL;
> +		goto out;
> +	}
> +
> +	down_read(&zram->init_lock);
> +	if (unlikely(!init_done(zram))) {
> +		ret = -ENOMEM;

Why is it -ENOMEM?
I think EIO is better like bio_io_error.

-- 
Kind regards,
Minchan Kim