On Mon, Dec 12, 2016 at 1:59 AM, NeilBrown <neilb@xxxxxxxx> wrote:
> On Sat, Nov 26 2016, Jinpu Wang wrote:
>> [ 810.270860] [<ffffffff813fc851>] blk_prologue_bio+0x91/0xc0
>
> What is this? I cannot find that function in the upstream kernel.
>
> NeilBrown
Hi Neil,
blk_prologue_bio is our internal extension to gather some stats, sorry
not informed before.
It does below, code snip only:
+static void bio_endio_complete_diskstat(struct bio *clone)
+{
+ struct bio *bio = clone->bi_private;
+ struct block_device *bdev = bio->bi_bdev;
+ struct hd_struct *hdp = bdev->bd_part;
+ unsigned long start = clone->bi_start_time;
+ const int rw = !!bio_data_dir(bio);
+ int cpu, err = clone->bi_error;
+
+ bio_put(clone);
+ blk_end_io_acct()
+ bio->bi_error = err;
+ bio_endio(bio);
+}
+
+blk_qc_t blk_prologue_bio(struct request_queue *q, struct bio *bio)
+{
+ struct block_device *bdev = bio->bi_bdev;
+ struct hd_struct *hdp = bdev->bd_part;
+ struct bio *clone;
+
+ if (!hdp->additional_diskstat)
+ return q->custom_make_request_fn(q, bio);
+ clone = bio_clone_fast(bio, GFP_NOWAIT | __GFP_NOWARN, NULL);
+ if (unlikely(!clone))
+ return q->custom_make_request_fn(q, bio);
+ clone->bi_start_time = jiffies;
+ if (hdp->bio_mode_diskstat) {
+ int rw = !!bio_data_dir(bio);
+
+ generic_start_io_acct(rw, bio_sectors(clone), hdp);
+ }
+ clone->bi_end_io = bio_endio_complete_diskstat;
+ clone->bi_private = bio;
+ return q->custom_make_request_fn(q, clone);
+}
ＩＭＨＯ， it seems unrelated, but I will rerun my test without this change.
--
Jinpu Wang
Linux Kernel Developer
ProfitBricks GmbH
Greifswalder Str. 207
D - 10405 Berlin
Tel: +49 30 577 008 042
Fax: +49 30 577 008 299
Email: jinpu.wang@xxxxxxxxxxxxxxxx
URL: https://www.profitbricks.de
Sitz der Gesellschaft: Berlin
Registergericht: Amtsgericht Charlottenburg, HRB 125506 B
Geschäftsführer: Achim Weiss
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
