From: Jens Axboe <axboe@kernel.dk>
To: LKML <linux-kernel@vger.kernel.org>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>,
Christian Brauner <brauner@kernel.org>
Subject: [PATCH] iov_iter: streamline iovec/bvec alignment iteration
Date: Tue, 23 Jan 2024 12:49:18 -0700 [thread overview]
Message-ID: <bd2f28ab-78cf-4ae0-a62a-d23a94fb3839@kernel.dk> (raw)
Rewrite iov_iter_aligned_iovec() and iov_iter_aligned_bvec() to be both
easier to read, and also significantly more compact in terms of
generated code. This saves 178 bytes of text on x86-64 for me (with
clang-18) and 136 bytes on arm64 (with gcc-13).
In profiles, also saves a bit of time as well for the same workload:
0.81% -0.18% [kernel.vmlinux] [k] iov_iter_aligned_bvec
0.48% -0.09% [kernel.vmlinux] [k] iov_iter_is_aligned
which is a nice side benefit as well.
diff --git a/lib/iov_iter.c b/lib/iov_iter.c
index e0aa6b440ca5..2fcc47d822e3 100644
--- a/lib/iov_iter.c
+++ b/lib/iov_iter.c
@@ -714,12 +714,11 @@ EXPORT_SYMBOL(iov_iter_discard);
static bool iov_iter_aligned_iovec(const struct iov_iter *i, unsigned addr_mask,
unsigned len_mask)
{
+ const struct iovec *iov = iter_iov(i);
size_t size = i->count;
size_t skip = i->iov_offset;
- unsigned k;
- for (k = 0; k < i->nr_segs; k++, skip = 0) {
- const struct iovec *iov = iter_iov(i) + k;
+ do {
size_t len = iov->iov_len - skip;
if (len > size)
@@ -729,34 +728,36 @@ static bool iov_iter_aligned_iovec(const struct iov_iter *i, unsigned addr_mask,
if ((unsigned long)(iov->iov_base + skip) & addr_mask)
return false;
+ iov++;
size -= len;
- if (!size)
- break;
- }
+ skip = 0;
+ } while (size);
+
return true;
}
static bool iov_iter_aligned_bvec(const struct iov_iter *i, unsigned addr_mask,
unsigned len_mask)
{
- size_t size = i->count;
+ const struct bio_vec *bvec = i->bvec;
unsigned skip = i->iov_offset;
- unsigned k;
+ size_t size = i->count;
- for (k = 0; k < i->nr_segs; k++, skip = 0) {
- size_t len = i->bvec[k].bv_len - skip;
+ do {
+ size_t len = bvec->bv_len;
if (len > size)
len = size;
if (len & len_mask)
return false;
- if ((unsigned long)(i->bvec[k].bv_offset + skip) & addr_mask)
+ if ((unsigned long)(bvec->bv_offset + skip) & addr_mask)
return false;
+ bvec++;
size -= len;
- if (!size)
- break;
- }
+ skip = 0;
+ } while (size);
+
return true;
}
--
Jens Axboe
next reply other threads:[~2024-01-23 19:49 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-01-23 19:49 Jens Axboe [this message]
2024-01-24 12:07 ` [PATCH] iov_iter: streamline iovec/bvec alignment iteration David Laight
2024-01-24 14:49 ` Jens Axboe
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=bd2f28ab-78cf-4ae0-a62a-d23a94fb3839@kernel.dk \
--to=axboe@kernel.dk \
--cc=brauner@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=viro@zeniv.linux.org.uk \
/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