[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <004101d21964$3b3d68f0$b1b83ad0$@alibaba-inc.com>
Date: Wed, 28 Sep 2016 16:42:21 +0800
From: "Hillf Danton" <hillf.zj@...baba-inc.com>
To: <tim.c.chen@...ux.intel.com>,
"'Andrew Morton'" <akpm@...ux-foundation.org>
Cc: <dave.hansen@...el.com>, <andi.kleen@...el.com>,
<aaron.lu@...el.com>, <linux-mm@...ck.org>,
<linux-kernel@...r.kernel.org>,
"'Huang Ying'" <ying.huang@...el.com>,
"'Hugh Dickins'" <hughd@...gle.com>,
"'Shaohua Li'" <shli@...nel.org>,
"'Minchan Kim'" <minchan@...nel.org>,
"'Rik van Riel'" <riel@...hat.com>,
"'Andrea Arcangeli'" <aarcange@...hat.com>,
"'Kirill A . Shutemov'" <kirill.shutemov@...ux.intel.com>,
"'Vladimir Davydov'" <vdavydov@...tuozzo.com>,
"'Johannes Weiner'" <hannes@...xchg.org>,
"'Michal Hocko'" <mhocko@...nel.org>
Subject: Re: [PATCH 2/8] mm/swap: Add cluster lock
On Wednesday, September 28, 2016 1:18 AM Tim Chen wrote
>
> @@ -447,8 +505,9 @@ static void scan_swap_map_try_ssd_cluster(struct swap_info_struct *si,
> unsigned long *offset, unsigned long *scan_base)
> {
> struct percpu_cluster *cluster;
> + struct swap_cluster_info *ci;
> bool found_free;
> - unsigned long tmp;
> + unsigned long tmp, max;
>
> new_cluster:
> cluster = this_cpu_ptr(si->percpu_cluster);
> @@ -476,14 +535,21 @@ new_cluster:
> * check if there is still free entry in the cluster
> */
> tmp = cluster->next;
> - while (tmp < si->max && tmp < (cluster_next(&cluster->index) + 1) *
> - SWAPFILE_CLUSTER) {
Currently tmp is checked to be less than both values.
> + max = max_t(unsigned long, si->max,
> + (cluster_next(&cluster->index) + 1) * SWAPFILE_CLUSTER);
> + if (tmp >= max) {
> + cluster_set_null(&cluster->index);
> + goto new_cluster;
> + }
> + ci = lock_cluster(si, tmp);
> + while (tmp < max) {
In this work tmp is checked to be less than the max value.
Semantic change hoped?
> if (!si->swap_map[tmp]) {
> found_free = true;
> break;
> }
> tmp++;
> }
> + unlock_cluster(ci);
> if (!found_free) {
> cluster_set_null(&cluster->index);
> goto new_cluster;
>
thanks
Hillf
Powered by blists - more mailing lists