emo/busybox
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

libiproute: use if_nametoindex

Browse Source 
Saves 87 bytes.  Assuming, of course, all platforms have it.

Signed-off-by: Ron Yorston <rmy@pobox.com>
Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
This commit is contained in:
Ron Yorston 2015-10-21 16:57:25 +01:00 committed by Denys Vlasenko
parent 6d777b75ed
commit 8814431aca
1 changed files with 1 additions and 24 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
networking/libiproute/ll_map.c
View File

@ -136,7 +136,6 @@ unsigned FAST_FUNC ll_index_to_flags(int idx)
int FAST_FUNC xll_name_to_index(const char *name)
{
int ret = 0;
int sock_fd;
/* caching is not warranted - no users which repeatedly call it */
#ifdef UNUSED
@ -164,30 +163,8 @@ int FAST_FUNC xll_name_to_index(const char *name)
}
}
}
/* We have not found the interface in our cache, but the kernel
* may still know about it. One reason is that we may be using
* module on-demand loading, which means that the kernel will
* load the module and make the interface exist only when
* we explicitely request it (check for dev_load() in net/core/dev.c).
* I can think of other similar scenario, but they are less common...
* Jean II */
#endif
sock_fd = socket(AF_INET, SOCK_DGRAM, 0);
if (sock_fd >= 0) {
struct ifreq ifr;
int tmp;
strncpy_IFNAMSIZ(ifr.ifr_name, name);
ifr.ifr_ifindex = -1;
tmp = ioctl(sock_fd, SIOCGIFINDEX, &ifr);
close(sock_fd);
if (tmp >= 0)
/* In theory, we should redump the interface list
* to update our cache, this is left as an exercise
* to the reader... Jean II */
ret = ifr.ifr_ifindex;
}
ret = if_nametoindex(name);
/* out:*/
if (ret <= 0)
bb_error_msg_and_die("can't find device '%s'", name);