提交 b262924b 编写于 作者: W Wenwei Tao 提交者: Jens Axboe

lightnvm: fix locking and mempool in rrpc_lun_gc

This patch fix two issues in rrpc_lun_gc

1. prio_list is protected by rrpc_lun's lock not nvm_lun's, so
acquire rlun's lock instead of lun's before operate on the list.

2. we delete block from prio_list before allocating gcb, but gcb
allocation may fail, we end without putting it back to the list,
this makes the block won't get reclaimed in the future. To solve
this issue, delete block after gcb allocation.
Signed-off-by: NWenwei Tao <ww.tao0320@gmail.com>
Signed-off-by: NMatias Bjørling <m@bjorling.me>
Signed-off-by: NJens Axboe <axboe@fb.com>
上级 d0ca798f
......@@ -445,7 +445,7 @@ static void rrpc_lun_gc(struct work_struct *work)
if (nr_blocks_need < rrpc->nr_luns)
nr_blocks_need = rrpc->nr_luns;
spin_lock(&lun->lock);
spin_lock(&rlun->lock);
while (nr_blocks_need > lun->nr_free_blocks &&
!list_empty(&rlun->prio_list)) {
struct rrpc_block *rblock = block_prio_find_max(rlun);
......@@ -454,16 +454,16 @@ static void rrpc_lun_gc(struct work_struct *work)
if (!rblock->nr_invalid_pages)
break;
gcb = mempool_alloc(rrpc->gcb_pool, GFP_ATOMIC);
if (!gcb)
break;
list_del_init(&rblock->prio);
BUG_ON(!block_is_full(rrpc, rblock));
pr_debug("rrpc: selected block '%lu' for GC\n", block->id);
gcb = mempool_alloc(rrpc->gcb_pool, GFP_ATOMIC);
if (!gcb)
break;
gcb->rrpc = rrpc;
gcb->rblk = rblock;
INIT_WORK(&gcb->ws_gc, rrpc_block_gc);
......@@ -472,7 +472,7 @@ static void rrpc_lun_gc(struct work_struct *work)
nr_blocks_need--;
}
spin_unlock(&lun->lock);
spin_unlock(&rlun->lock);
/* TODO: Hint that request queue can be started again */
}
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册
新手
引导
客服 返回
顶部