Linux kernel mirror (for testing)
git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel
os
linux
docs: net: fix inaccuracies in msg_zerocopy.rst
Replace "sendpage" with "sendfile". Remove comment about
ENOBUFS when the sockopt hasn't been set; experimentation
indicates that this is not true.
Signed-off-by: nick black <dankamongmen@gmail.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://lore.kernel.org/r/Y/gg/EhIIjugLdd3@schwarzgerat.orthanc
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
authored by
nick black
and committed by
Jakub Kicinski
bce90459
766d1bdc
+3
-3
+3
-3
Documentation/networking/msg_zerocopy.rst
+3
-3
Documentation/networking/msg_zerocopy.rst
···
15
15
16
16
Copying large buffers between user process and kernel can be
17
17
expensive. Linux supports various interfaces that eschew copying,
18
-
such as sendpage and splice. The MSG_ZEROCOPY flag extends the
18
+
such as sendfile and splice. The MSG_ZEROCOPY flag extends the
19
19
underlying copy avoidance mechanism to common socket send calls.
20
20
21
21
Copy avoidance is not a free lunch. As implemented, with page pinning,
···
83
83
ret = send(fd, buf, sizeof(buf), MSG_ZEROCOPY);
84
84
85
85
A zerocopy failure will return -1 with errno ENOBUFS. This happens if
86
-
the socket option was not set, the socket exceeds its optmem limit or
87
-
the user exceeds its ulimit on locked pages.
86
+
the socket exceeds its optmem limit or the user exceeds their ulimit on
87
+
locked pages.
88
88
89
89
90
90
Mixing copy avoidance and copying