Messages in this thread Patch in this message | | | From | Chao Yu <> | Subject | [PATCH v2 4/6] f2fs: reduce cmd_lock coverage in __issue_discard_cmd | Date | Wed, 4 Oct 2017 09:08:35 +0800 |
| |
From: Chao Yu <yuchao0@huawei.com>
__submit_discard_cmd may lead long latency due to exhaustion of I/O request resource in block layer, so issuing all discard under cmd_lock may lead to hangtask, in order to avoid that, let's reduce it's coverage.
Signed-off-by: Chao Yu <yuchao0@huawei.com> --- v2: - rebase on last codes of Jaegeuk's dev-test branch. fs/f2fs/segment.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-)
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index e8fcc1eb4e0a..ba433828c170 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -1158,14 +1158,14 @@ static int __issue_discard_cmd(struct f2fs_sb_info *sbi, int i, iter = 0, issued = 0; bool io_interrupted = false; - mutex_lock(&dcc->cmd_lock); - f2fs_bug_on(sbi, - !__check_rb_tree_consistence(sbi, &dcc->root)); - blk_start_plug(&plug); for (i = MAX_PLIST_NUM - 1; i >= 0; i--) { if (i + 1 < dpolicy->granularity) break; pend_list = &dcc->pend_list[i]; + + mutex_lock(&dcc->cmd_lock); + f2fs_bug_on(sbi, !__check_rb_tree_consistence(sbi, &dcc->root)); + blk_start_plug(&plug); list_for_each_entry_safe(dc, tmp, pend_list, list) { f2fs_bug_on(sbi, dc->state != D_PREP); @@ -1179,12 +1179,14 @@ static int __issue_discard_cmd(struct f2fs_sb_info *sbi, issued++; skip: if (++iter >= dpolicy->max_requests) - goto out; + break; } + blk_finish_plug(&plug); + mutex_unlock(&dcc->cmd_lock); + + if (iter >= dpolicy->max_requests) + break; } -out: - blk_finish_plug(&plug); - mutex_unlock(&dcc->cmd_lock); if (!issued && io_interrupted) issued = -1; -- 2.14.1.145.gb3622a4ee
| |