On Mon, Nov 09, 2015 at 03:21:11PM -0500, Brian Foster wrote:
> Each log recovery pass walks from the tail block to the head block and
> processes records appropriately based on the associated log pass type.
> There are various failure conditions that can occur through this
> sequence, such as I/O errors, CRC errors, etc. Log torn write detection
> will perform CRC verification near the head of the log to detect torn
> writes and trim torn records from the log appropriately.
>
> As it is, xlog_do_recovery_pass() only returns an error code in the
> event of CRC failure, which isn't enough information to trim the head of
> the log. Update xlog_do_recovery_pass() to optionally return the start
> block of the associated record when an error occurs. This patch contains
> no functional changes.
>
> Signed-off-by: Brian Foster <bfoster@xxxxxxxxxx>
> ---
> fs/xfs/xfs_log_recover.c | 19 +++++++++++++++----
> 1 file changed, 15 insertions(+), 4 deletions(-)
>
> diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
> index c2bf307..2ef0880 100644
> --- a/fs/xfs/xfs_log_recover.c
> +++ b/fs/xfs/xfs_log_recover.c
> @@ -4239,10 +4239,12 @@ xlog_do_recovery_pass(
> struct xlog *log,
> xfs_daddr_t head_blk,
> xfs_daddr_t tail_blk,
> - int pass)
> + int pass,
> + xfs_daddr_t *first_bad) /* out: first bad log rec */
> {
> xlog_rec_header_t *rhead;
> xfs_daddr_t blk_no;
> + xfs_daddr_t rhead_blk;
> char *offset;
> xfs_buf_t *hbp, *dbp;
> int error = 0, h_size, h_len;
> @@ -4251,6 +4253,7 @@ xlog_do_recovery_pass(
> struct hlist_head rhash[XLOG_RHASH_SIZE];
>
> ASSERT(head_blk != tail_blk);
> + rhead_blk = 0;
>
> /*
> * Read the header of the tail block and get the iclog buffer size from
> @@ -4325,7 +4328,7 @@ xlog_do_recovery_pass(
> }
>
> memset(rhash, 0, sizeof(rhash));
> - blk_no = tail_blk;
> + blk_no = rhead_blk = tail_blk;
> if (tail_blk > head_blk) {
> /*
> * Perform recovery around the end of the physical log.
> @@ -4436,7 +4439,9 @@ xlog_do_recovery_pass(
> pass);
> if (error)
> goto bread_err2;
> +
> blk_no += bblks;
> + rhead_blk = blk_no;
> }
>
> ASSERT(blk_no >= log->l_logBBsize);
There's a rewind of blk_no (not shown in this patch) between the above
loop and the subsequent loop to handle wrapping around the end of the
log. We need to update rhead_blk there as well. Otherwise, if the first
record processed in the following loop is bad, rhead_blk can still point
beyond the end of the log and thus throw off all of the recovery bits
that follow.
Brian
> @@ -4464,13 +4469,19 @@ xlog_do_recovery_pass(
> error = xlog_recover_process(log, rhash, rhead, offset, pass);
> if (error)
> goto bread_err2;
> +
> blk_no += bblks + hblks;
> + rhead_blk = blk_no;
> }
>
> bread_err2:
> xlog_put_bp(dbp);
> bread_err1:
> xlog_put_bp(hbp);
> +
> + if (error && first_bad)
> + *first_bad = rhead_blk;
> +
> return error;
> }
>
> @@ -4508,7 +4519,7 @@ xlog_do_log_recovery(
> INIT_LIST_HEAD(&log->l_buf_cancel_table[i]);
>
> error = xlog_do_recovery_pass(log, head_blk, tail_blk,
> - XLOG_RECOVER_PASS1);
> + XLOG_RECOVER_PASS1, NULL);
> if (error != 0) {
> kmem_free(log->l_buf_cancel_table);
> log->l_buf_cancel_table = NULL;
> @@ -4519,7 +4530,7 @@ xlog_do_log_recovery(
> * When it is complete free the table of buf cancel items.
> */
> error = xlog_do_recovery_pass(log, head_blk, tail_blk,
> - XLOG_RECOVER_PASS2);
> + XLOG_RECOVER_PASS2, NULL);
> #ifdef DEBUG
> if (!error) {
> int i;
> --
> 2.1.0
>
> _______________________________________________
> xfs mailing list
> xfs@xxxxxxxxxxx
> http://oss.sgi.com/mailman/listinfo/xfs
_______________________________________________
xfs mailing list
xfs@xxxxxxxxxxx
http://oss.sgi.com/mailman/listinfo/xfs
