Increase unix socket send buffer to at least 64KB

Some systems (e.g. OpenBSD) have a rather small default socket send buffer
size of 4KB.  The result is that sending requests with a largish payload
requires serveral writev(2) system calls.  Make sure the socket send buffer
is at least 64KB such that we're likely to succeed with a single system
call for most requests.  A similar change was made to the xtrans code
some time ago.

Signed-off-by: Mark Kettenis <kettenis@openbsd.org>
Reviewed-by: Matthieu Herrb <matthieu@herrb.eu>
This commit is contained in:
Mark Kettenis 2016-01-23 17:29:32 +01:00 committed by Christian Linhart
parent b3516102b4
commit 095353ff1a

View File

@ -428,6 +428,8 @@ static int _xcb_open_unix(char *protocol, const char *file)
{
int fd;
struct sockaddr_un addr;
socklen_t len = sizeof(int);
int val;
if (protocol && strcmp("unix",protocol))
return -1;
@ -440,6 +442,11 @@ static int _xcb_open_unix(char *protocol, const char *file)
fd = _xcb_socket(AF_UNIX, SOCK_STREAM, 0);
if(fd == -1)
return -1;
if(getsockopt(fd, SOL_SOCKET, SO_SNDBUF, &val, &len) == 0 && val < 64 * 1024)
{
val = 64 * 1024;
setsockopt(fd, SOL_SOCKET, SO_SNDBUF, &val, sizeof(int));
}
if(connect(fd, (struct sockaddr *) &addr, sizeof(addr)) == -1) {
close(fd);
return -1;