| Message ID | 1455711269-12929-13-git-send-email-janusz.dziedzic@tieto.com |
|---|---|
| State | Changes Requested |
| Headers | show |
On Wed, Feb 17, 2016 at 01:14:15PM +0100, Janusz Dziedzic wrote: > Show udp port when handle INTERFACES command. > diff --git a/wpa_supplicant/ctrl_iface.c b/wpa_supplicant/ctrl_iface.c > @@ -9233,7 +9233,12 @@ static int wpa_supplicant_global_iface_interfaces(struct wpa_global *global, > +#ifdef CONFIG_CTRL_IFACE_UDP > + res = os_snprintf(pos, end - pos, "%s ctrl_iface=%s\n", > + wpa_s->ifname, wpa_s->conf->ctrl_interface); > +#else > res = os_snprintf(pos, end - pos, "%s\n", wpa_s->ifname); > +#endif Wouldn't this break all existing expectations on INTERFACES command output format? It does not look good to return different values based on which control interface backend is used.
On 20 February 2016 at 16:26, Jouni Malinen <j@w1.fi> wrote: > On Wed, Feb 17, 2016 at 01:14:15PM +0100, Janusz Dziedzic wrote: >> Show udp port when handle INTERFACES command. > >> diff --git a/wpa_supplicant/ctrl_iface.c b/wpa_supplicant/ctrl_iface.c >> @@ -9233,7 +9233,12 @@ static int wpa_supplicant_global_iface_interfaces(struct wpa_global *global, >> +#ifdef CONFIG_CTRL_IFACE_UDP >> + res = os_snprintf(pos, end - pos, "%s ctrl_iface=%s\n", >> + wpa_s->ifname, wpa_s->conf->ctrl_interface); >> +#else >> res = os_snprintf(pos, end - pos, "%s\n", wpa_s->ifname); >> +#endif > > Wouldn't this break all existing expectations on INTERFACES command > output format? It does not look good to return different values based on > which control interface backend is used. > Added this to not break hwsim command parsing. In next version I can remove this #ifdef and fix hwsim parsing, or we can introduce new command here? > -- > Jouni Malinen PGP id EFC895FA
On 21 February 2016 at 13:48, Janusz Dziedzic <janusz.dziedzic@tieto.com> wrote: > On 20 February 2016 at 16:26, Jouni Malinen <j@w1.fi> wrote: >> On Wed, Feb 17, 2016 at 01:14:15PM +0100, Janusz Dziedzic wrote: >>> Show udp port when handle INTERFACES command. >> >>> diff --git a/wpa_supplicant/ctrl_iface.c b/wpa_supplicant/ctrl_iface.c >>> @@ -9233,7 +9233,12 @@ static int wpa_supplicant_global_iface_interfaces(struct wpa_global *global, >>> +#ifdef CONFIG_CTRL_IFACE_UDP >>> + res = os_snprintf(pos, end - pos, "%s ctrl_iface=%s\n", >>> + wpa_s->ifname, wpa_s->conf->ctrl_interface); >>> +#else >>> res = os_snprintf(pos, end - pos, "%s\n", wpa_s->ifname); >>> +#endif >> >> Wouldn't this break all existing expectations on INTERFACES command >> output format? It does not look good to return different values based on >> which control interface backend is used. >> > Added this to not break hwsim command parsing. In next version I can > remove this #ifdef and fix hwsim parsing, or we can introduce new > command here? > Maybe we should introduce: <INTERFACES param=ctrl> command (or something similar)? So, <INTERFACES> will work as before? >> -- >> Jouni Malinen PGP id EFC895FA
diff --git a/wpa_supplicant/ctrl_iface.c b/wpa_supplicant/ctrl_iface.c index 6829aa0..0bd5eca 100644 --- a/wpa_supplicant/ctrl_iface.c +++ b/wpa_supplicant/ctrl_iface.c @@ -9233,7 +9233,12 @@ static int wpa_supplicant_global_iface_interfaces(struct wpa_global *global, end = buf + len; while (wpa_s) { +#ifdef CONFIG_CTRL_IFACE_UDP + res = os_snprintf(pos, end - pos, "%s ctrl_iface=%s\n", + wpa_s->ifname, wpa_s->conf->ctrl_interface); +#else res = os_snprintf(pos, end - pos, "%s\n", wpa_s->ifname); +#endif if (os_snprintf_error(end - pos, res)) { *pos = '\0'; break;
Show udp port when handle INTERFACES command. Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com> --- wpa_supplicant/ctrl_iface.c | 5 +++++ 1 file changed, 5 insertions(+)