Re: [MPTCP][PATCH v2 mptcp-next] Squash to "mptcp: build ADD_ADDR/echo-ADD_ADDR option according pm.add_signal"

[Geliang Tang <geliangtang@gmail.com>]

Hi Mat,

Mat Martineau <mathew.j.martineau@linux.intel.com> 于2021年7月14日周三 上午4:39写道：
>
> On Tue, 13 Jul 2021, Geliang Tang wrote:
>
> > This squash-to patch will conflict with "mptcp: remove
> > MPTCP_ADD_ADDR_IPV6 and MPTCP_ADD_ADDR_PORT":
> >
> > Auto-merging net/mptcp/protocol.h
> > CONFLICT (content): Merge conflict in net/mptcp/protocol.h
> > Auto-merging net/mptcp/pm.c
> > error: could not apply bf1fec79a2bf... mptcp: remove
> > MPTCP_ADD_ADDR_IPV6 and MPTCP_ADD_ADDR_PORT
> >
> > Please fix it like this:
> >
> > <<<<<<< HEAD
> > static inline bool mptcp_pm_should_add_signal_ipv6(struct mptcp_sock *msk)
> > {
> >         return (mptcp_pm_should_add_signal_addr(msk) &&
> > msk->pm.local.family == AF_INET6) ||
> >                 (mptcp_pm_should_add_signal_echo(msk) &&
> > msk->pm.remote.family == AF_INET6);
> > }
> >
> > static inline bool mptcp_pm_should_add_signal_port(struct mptcp_sock *msk)
> > {
> >         return (mptcp_pm_should_add_signal_addr(msk) && msk->pm.local.port) ||
> >                 (mptcp_pm_should_add_signal_echo(msk) && msk->pm.remote.port);
> > }
> >
> > =======
> > >>>>>>> bf1fec79a2bf (mptcp: remove MPTCP_ADD_ADDR_IPV6 and
> > MPTCP_ADD_ADDR_PORT)
> > static inline bool mptcp_pm_should_rm_signal(struct mptcp_sock *msk)
> > {
> >         return READ_ONCE(msk->pm.addr_signal) & BIT(MPTCP_RM_ADDR_SIGNAL);
> > }
> >
> > ->
> >
> > static inline bool mptcp_pm_should_add_signal_ipv6(struct mptcp_sock *msk)
> > {
> >         return (mptcp_pm_should_add_signal_addr(msk) &&
> > msk->pm.local.family == AF_INET6) ||
> >                 (mptcp_pm_should_add_signal_echo(msk) &&
> > msk->pm.remote.family == AF_INET6);
> > }
> >
> > static inline bool mptcp_pm_should_add_signal_port(struct mptcp_sock *msk)
> > {
> >         return (mptcp_pm_should_add_signal_addr(msk) && msk->pm.local.port) ||
> >                 (mptcp_pm_should_add_signal_echo(msk) && msk->pm.remote.port);
> > }
> >
> > static inline bool mptcp_pm_should_rm_signal(struct mptcp_sock *msk)
> > {
> >         return READ_ONCE(msk->pm.addr_signal) & BIT(MPTCP_RM_ADDR_SIGNAL);
> > }
> >
> >
>
> Well, it looks like I gave my Reviewed-by tag too soon. I think it would
> be easier to get the correct results by posting more revisions of the
> patch set - don't want to introduce mistakes when trying to resolve these
> conflicts when applying to mptcp_net-next repo!
>
> Geliang and Yonglong, what do you think about posting v9, possibly with
> co-developed-by tags?
>

Sounds good. I'll send a v9, with my co-developed-by tags, and keep Yonglong
as the author, when the code is ready and the test passes.

Thanks,
-Geliang

>
> --
> Mat Martineau
> Intel