Hello,
Joan Lledó, le sam. 22 juin 2019 12:00:02 +0200, a ecrit:
> * lwip/port/netif/ifcommon.c:
> * Changing flags for a device (e.g. by inetutils-ifconfig) now
> takes the big lock to ensure the stack is not doing anything else.
Applied, thanks!
* lwip/port/netif/ifcommon.c:
* Changing flags for a device (e.g. by inetutils-ifconfig) now
takes the big lock to ensure the stack is not doing anything else.
---
lwip/port/netif/ifcommon.c | 60 ++
1 file changed, 55 insertions(+), 5 deletion
Hi,
I updated this patch, now it uses a new function I wrote for lwip
which calls any callback inside a thread-safe context.
Regards.
Joan Lledó, le ven. 03 mai 2019 13:37:59 +0200, a ecrit:
> Missatge de Samuel Thibault del dia dv., 3
> de maig 2019 a les 0:02:
> > In C, there is no need to explicit the cast from void*.
>
> I'm doing a similar cast on lwip-util.c:256. Should I remove it too?
Yes, there is no need to clutter t
Missatge de Samuel Thibault del dia dv., 3
de maig 2019 a les 0:02:
> In C, there is no need to explicit the cast from void*.
I'm doing a similar cast on lwip-util.c:256. Should I remove it too?
> Errno is a per-thread variable, you can not propagate the error this
> way. You can for instance ad
Joan Lledó, le jeu. 02 mai 2019 11:55:02 +0200, a ecrit:
> + uint16_t oldflags;
> + struct if_change_flags_args *args = (struct if_change_flags_args *) arg;
In C, there is no need to explicit the cast from void*.
> + ifc = netif_get_state (args->netif);
> +
> + if (ifc == NULL)
> +{
> +
* lwip/port/netif/ifcommon.c:
* Changing flags for a device (e.g. by inetutils-ifconfig) now
takes the big lock to ensure the stack is not doing anything else.
---
lwip/port/netif/ifcommon.c | 61 +-
1 file changed, 54 insertions(+), 7 deletion
