Re: [PATCH v2] blk-throtl: optimize IOPS throttle for large IO scenarios

[Tejun Heo]

Hello,

On Wed, Jul 28, 2021 at 05:01:41PM +0800, brookxu wrote:
> diff --git a/block/blk-merge.c b/block/blk-merge.c
> index a11b3b5..86ff943 100644
> --- a/block/blk-merge.c
> +++ b/block/blk-merge.c
> @@ -348,6 +348,8 @@ void __blk_queue_split(struct bio **bio, unsigned int *nr_segs)
>  		trace_block_split(split, (*bio)->bi_iter.bi_sector);
>  		submit_bio_noacct(*bio);
>  		*bio = split;
> +
> +		blk_throtl_recharge_bio(*bio);

Can you rename this blk_throtl_charge_bio_split()?

> @@ -524,6 +537,11 @@ static struct blkg_policy_data *throtl_pd_alloc(gfp_t gfp,
>  	tg->idletime_threshold = DFL_IDLE_THRESHOLD;
>  	tg->idletime_threshold_conf = DFL_IDLE_THRESHOLD;
>  
> +	atomic_set(&tg->io_split_cnt[0], 0);
> +	atomic_set(&tg->io_split_cnt[1], 0);
> +	atomic_set(&tg->last_io_split_cnt[0], 0);
> +	atomic_set(&tg->last_io_split_cnt[1], 0);

We likely don't need these. pd's zeroed on allocation.

> @@ -877,10 +900,19 @@ static inline void throtl_trim_slice(struct throtl_grp *tg, bool rw)
>  	else
>  		tg->bytes_disp[rw] = 0;
>  
> -	if (tg->io_disp[rw] >= io_trim)
> +	if (tg_io_disp(tg, rw) >= io_trim) {

Instead of checking this in multiple places, would it be simpler to transfer
the atomic counters to the existing counters whenever we enter blk-throtl
and leave the rest of the code as-is?

Thanks.

-- 
tejun