public inbox for io-uring@vger.kernel.org
 help / color / mirror / Atom feed
From: Pavel Begunkov <asml.silence@gmail.com>
To: io-uring@vger.kernel.org
Cc: asml.silence@gmail.com, Nitesh Shetty <nj.shetty@samsung.com>
Subject: [PATCH 3/4] io_uring/rsrc: refactor io_import_fixed
Date: Thu, 17 Apr 2025 10:32:33 +0100	[thread overview]
Message-ID: <2d5107fed24f8b23245ef2ede9a5a7f7c426df61.1744882081.git.asml.silence@gmail.com> (raw)
In-Reply-To: <cover.1744882081.git.asml.silence@gmail.com>

io_import_fixed is a mess. Even though we know the final len of the
iterator, we still assign offset + len and do some magic after to
correct for that.

Do offset calculation first and finalise it with iov_iter_bvec at the
end.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
---
 io_uring/rsrc.c | 24 +++++++-----------------
 1 file changed, 7 insertions(+), 17 deletions(-)

diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c
index fddde8ffe81e..5cf854318b1d 100644
--- a/io_uring/rsrc.c
+++ b/io_uring/rsrc.c
@@ -1037,6 +1037,7 @@ static int io_import_fixed(int ddir, struct iov_iter *iter,
 			   u64 buf_addr, size_t len)
 {
 	const struct bio_vec *bvec;
+	unsigned nr_segs;
 	size_t offset;
 	int ret;
 
@@ -1056,8 +1057,6 @@ static int io_import_fixed(int ddir, struct iov_iter *iter,
 		return 0;
 	}
 
-	iov_iter_bvec(iter, ddir, imu->bvec, imu->nr_bvecs, offset + len);
-
 	/*
 	 * Don't use iov_iter_advance() here, as it's really slow for
 	 * using the latter parts of a big fixed buffer - it iterates
@@ -1067,30 +1066,21 @@ static int io_import_fixed(int ddir, struct iov_iter *iter,
 	 * 1) it's a BVEC iter, we set it up
 	 * 2) all bvecs are the same in size, except potentially the
 	 *    first and last bvec
-	 *
-	 * So just find our index, and adjust the iterator afterwards.
-	 * If the offset is within the first bvec (or the whole first
-	 * bvec, just use iov_iter_advance(). This makes it easier
-	 * since we can just skip the first segment, which may not
-	 * be folio_size aligned.
 	 */
 	bvec = imu->bvec;
-	if (offset < bvec->bv_len) {
-		iter->count -= offset;
-		iter->iov_offset = offset;
-	} else {
+	if (offset >= bvec->bv_len) {
 		unsigned long seg_skip;
 
 		/* skip first vec */
 		offset -= bvec->bv_len;
 		seg_skip = 1 + (offset >> imu->folio_shift);
-
-		iter->bvec += seg_skip;
-		iter->nr_segs -= seg_skip;
-		iter->count -= bvec->bv_len + offset;
-		iter->iov_offset = offset & ((1UL << imu->folio_shift) - 1);
+		bvec += seg_skip;
+		offset &= (1UL << imu->folio_shift) - 1;
 	}
 
+	nr_segs = imu->nr_bvecs - (bvec - imu->bvec);
+	iov_iter_bvec(iter, ddir, bvec, nr_segs, len);
+	iter->iov_offset = offset;
 	return 0;
 }
 
-- 
2.48.1


  parent reply	other threads:[~2025-04-17  9:31 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-04-17  9:32 [PATCH 0/4] io_import_fixed cleanups and optimisation Pavel Begunkov
2025-04-17  9:32 ` [PATCH 1/4] io_uring/rsrc: don't skip offset calculation Pavel Begunkov
2025-04-17  9:32 ` [PATCH 2/4] io_uring/rsrc: separate kbuf offset adjustments Pavel Begunkov
2025-04-17  9:32 ` Pavel Begunkov [this message]
2025-04-17  9:32 ` [PATCH 4/4] io_uring/rsrc: send exact nr_segs for fixed buffer Pavel Begunkov
2025-04-17  9:34   ` Pavel Begunkov
     [not found]     ` <CGME20250417103133epcas5p32c1e004e7f8a5135c4c7e3662b087470@epcas5p3.samsung.com>
2025-04-17 10:23       ` Nitesh Shetty
2025-04-17 11:50         ` Nitesh Shetty
2025-04-17 12:56           ` Pavel Begunkov
2025-04-17 13:41             ` Jens Axboe
2025-04-17 13:57               ` Jens Axboe
2025-04-17 14:02                 ` Pavel Begunkov
2025-04-17 14:05                   ` Jens Axboe
2025-04-17 12:31 ` [PATCH 0/4] io_import_fixed cleanups and optimisation Jens Axboe
     [not found] ` <CGME20250417142334epcas5p36df55874d21c896115d92e505f9793fd@epcas5p3.samsung.com>
2025-04-17 14:15   ` Nitesh Shetty

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=2d5107fed24f8b23245ef2ede9a5a7f7c426df61.1744882081.git.asml.silence@gmail.com \
    --to=asml.silence@gmail.com \
    --cc=io-uring@vger.kernel.org \
    --cc=nj.shetty@samsung.com \
    /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