linux-xfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Dave Chinner <david@fromorbit.com>
To: linux-xfs@vger.kernel.org
Subject: [PATCH 11/16] xfs: introduce xlog_write_full()
Date: Wed,  9 Mar 2022 16:29:32 +1100	[thread overview]
Message-ID: <20220309052937.2696447-12-david@fromorbit.com> (raw)
In-Reply-To: <20220309052937.2696447-1-david@fromorbit.com>

From: Dave Chinner <dchinner@redhat.com>

Introduce an optimised version of xlog_write() that is used when the
entire write will fit in a single iclog. This greatly simplifies the
implementation of writing a log vector chain into an iclog, and sets
the ground work for a much more understandable xlog_write()
implementation.

This incorporates some factoring and simplifications proposed by
Christoph Hellwig.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Chandan Babu R <chandan.babu@oracle.com>
---
 fs/xfs/xfs_log.c | 69 +++++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 66 insertions(+), 3 deletions(-)

diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index f26c85dbc765..6d93b2c96262 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -2224,6 +2224,58 @@ xlog_print_trans(
 	}
 }
 
+static inline void
+xlog_write_iovec(
+	struct xlog_in_core	*iclog,
+	uint32_t		*log_offset,
+	void			*data,
+	uint32_t		write_len,
+	int			*bytes_left,
+	uint32_t		*record_cnt,
+	uint32_t		*data_cnt)
+{
+	ASSERT(*log_offset % sizeof(int32_t) == 0);
+	ASSERT(write_len % sizeof(int32_t) == 0);
+
+	memcpy(iclog->ic_datap + *log_offset, data, write_len);
+	*log_offset += write_len;
+	*bytes_left -= write_len;
+	(*record_cnt)++;
+	*data_cnt += write_len;
+}
+
+/*
+ * Write log vectors into a single iclog which is guaranteed by the caller
+ * to have enough space to write the entire log vector into.
+ */
+static void
+xlog_write_full(
+	struct xfs_log_vec	*lv,
+	struct xlog_ticket	*ticket,
+	struct xlog_in_core	*iclog,
+	uint32_t		*log_offset,
+	uint32_t		*len,
+	uint32_t		*record_cnt,
+	uint32_t		*data_cnt)
+{
+	int			index;
+
+	ASSERT(*log_offset + *len <= iclog->ic_size);
+
+	/*
+	 * Ordered log vectors have no regions to write so this
+	 * loop will naturally skip them.
+	 */
+	for (index = 0; index < lv->lv_niovecs; index++) {
+		struct xfs_log_iovec	*reg = &lv->lv_iovecp[index];
+		struct xlog_op_header	*ophdr = reg->i_addr;
+
+		ophdr->oh_tid = cpu_to_be32(ticket->t_tid);
+		xlog_write_iovec(iclog, log_offset, reg->i_addr,
+				reg->i_len, len, record_cnt, data_cnt);
+	}
+}
+
 static xlog_op_header_t *
 xlog_write_setup_ophdr(
 	struct xlog_op_header	*ophdr,
@@ -2388,8 +2440,8 @@ xlog_write(
 	int			partial_copy = 0;
 	int			partial_copy_len = 0;
 	int			contwr = 0;
-	int			record_cnt = 0;
-	int			data_cnt = 0;
+	uint32_t		record_cnt = 0;
+	uint32_t		data_cnt = 0;
 	int			error = 0;
 
 	if (ticket->t_curr_res < 0) {
@@ -2409,7 +2461,6 @@ xlog_write(
 			return error;
 
 		ASSERT(log_offset <= iclog->ic_size - 1);
-		ptr = iclog->ic_datap + log_offset;
 
 		/*
 		 * If we have a context pointer, pass it the first iclog we are
@@ -2421,10 +2472,22 @@ xlog_write(
 			ctx = NULL;
 		}
 
+		/* If this is a single iclog write, go fast... */
+		if (!contwr && lv == log_vector) {
+			while (lv) {
+				xlog_write_full(lv, ticket, iclog, &log_offset,
+						 &len, &record_cnt, &data_cnt);
+				lv = lv->lv_next;
+			}
+			data_cnt = 0;
+			break;
+		}
+
 		/*
 		 * This loop writes out as many regions as can fit in the amount
 		 * of space which was allocated by xlog_state_get_iclog_space().
 		 */
+		ptr = iclog->ic_datap + log_offset;
 		while (lv && (!lv->lv_niovecs || index < lv->lv_niovecs)) {
 			struct xfs_log_iovec	*reg;
 			struct xlog_op_header	*ophdr;
-- 
2.33.0


  parent reply	other threads:[~2022-03-09  5:29 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-03-09  5:29 [PATCH 00/16 v8] xfs: rework xlog_write() Dave Chinner
2022-03-09  5:29 ` [PATCH 01/16] xfs: factor out the CIL transaction header building Dave Chinner
2022-03-09  5:29 ` [PATCH 02/16] xfs: only CIL pushes require a start record Dave Chinner
2022-03-09  5:29 ` [PATCH 03/16] xfs: embed the xlog_op_header in the unmount record Dave Chinner
2022-03-09  5:29 ` [PATCH 04/16] xfs: embed the xlog_op_header in the commit record Dave Chinner
2022-03-09  5:29 ` [PATCH 05/16] xfs: log tickets don't need log client id Dave Chinner
2022-03-09  5:29 ` [PATCH 06/16] xfs: move log iovec alignment to preparation function Dave Chinner
2022-03-09  5:29 ` [PATCH 07/16] xfs: reserve space and initialise xlog_op_header in item formatting Dave Chinner
2022-03-09  5:29 ` [PATCH 08/16] xfs: log ticket region debug is largely useless Dave Chinner
2022-03-09  5:29 ` [PATCH 09/16] xfs: pass lv chain length into xlog_write() Dave Chinner
2022-03-09  5:29 ` [PATCH 10/16] xfs: change the type of ic_datap Dave Chinner
2022-03-09  5:29 ` Dave Chinner [this message]
2022-03-09  5:29 ` [PATCH 12/16] xfs: introduce xlog_write_partial() Dave Chinner
2022-03-09  5:29 ` [PATCH 13/16] xfs: remove xlog_verify_dest_ptr Dave Chinner
2022-03-09  5:29 ` [PATCH 14/16] xfs: xlog_write() no longer needs contwr state Dave Chinner
2022-03-09  5:29 ` [PATCH 15/16] xfs: xlog_write() doesn't need optype anymore Dave Chinner
2022-03-09  5:29 ` [PATCH 16/16] xfs: CIL context doesn't need to count iovecs Dave Chinner
  -- strict thread matches above, loose matches on Subject: below --
2021-11-18 23:13 [PATCH 00/16 v7] xfs: rework xlog_write() Dave Chinner
2021-11-18 23:13 ` [PATCH 11/16] xfs: introduce xlog_write_full() Dave Chinner
2021-11-22 11:40   ` Chandan Babu R
2021-11-09  1:50 [PATCH 00/16 v6] xfs: rework xlog_write() Dave Chinner
2021-11-09  1:50 ` [PATCH 11/16] xfs: introduce xlog_write_full() Dave Chinner
2021-11-11  8:12   ` Christoph Hellwig
2021-11-17  4:56   ` Darrick J. Wong

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220309052937.2696447-12-david@fromorbit.com \
    --to=david@fromorbit.com \
    --cc=linux-xfs@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).