2020-06-12 05:33:54

by Geliang Tang

[permalink] [raw]
Subject: [PATCH] mptcp: unify MPTCP_PM_MAX_ADDR and MPTCP_PM_ADDR_MAX

Unify these two duplicate macros into 8.

Signed-off-by: Geliang Tang <[email protected]>
---
net/mptcp/pm_netlink.c | 2 --
net/mptcp/protocol.h | 2 +-
2 files changed, 1 insertion(+), 3 deletions(-)

diff --git a/net/mptcp/pm_netlink.c b/net/mptcp/pm_netlink.c
index b78edf237ba0..b694f13caba8 100644
--- a/net/mptcp/pm_netlink.c
+++ b/net/mptcp/pm_netlink.c
@@ -41,8 +41,6 @@ struct pm_nl_pernet {
unsigned int next_id;
};

-#define MPTCP_PM_ADDR_MAX 8
-
static bool addresses_equal(const struct mptcp_addr_info *a,
struct mptcp_addr_info *b, bool use_port)
{
diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
index 809687d3f410..86d265500cf6 100644
--- a/net/mptcp/protocol.h
+++ b/net/mptcp/protocol.h
@@ -135,7 +135,7 @@ static inline __be32 mptcp_option(u8 subopt, u8 len, u8 nib, u8 field)
((nib & 0xF) << 8) | field);
}

-#define MPTCP_PM_MAX_ADDR 4
+#define MPTCP_PM_ADDR_MAX 8

struct mptcp_addr_info {
sa_family_t family;
--
2.17.1


2020-06-12 17:36:15

by Matthieu Baerts

[permalink] [raw]
Subject: Re: [PATCH] mptcp: unify MPTCP_PM_MAX_ADDR and MPTCP_PM_ADDR_MAX

Hi Geliang,

On 12/06/2020 07:27, Geliang Tang wrote:
> Unify these two duplicate macros into 8.

Thank you for this new patch!

(...)

> diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
> index 809687d3f410..86d265500cf6 100644
> --- a/net/mptcp/protocol.h
> +++ b/net/mptcp/protocol.h
> @@ -135,7 +135,7 @@ static inline __be32 mptcp_option(u8 subopt, u8 len, u8 nib, u8 field)
> ((nib & 0xF) << 8) | field);
> }
>
> -#define MPTCP_PM_MAX_ADDR 4
> +#define MPTCP_PM_ADDR_MAX 8

I think it would be better to drop MPTCP_PM_MAX_ADDR and keep
MPTCP_PM_ADDR_MAX in pm_netlink.c where it is used. Each PM can decide
what's the maximum number of addresses it can support.

MPTCP_PM_MAX_ADDR seems to be a left over from a previous implementation
of a PM that has not been upstreamed but replaced by the Netlink PM later.

Also, please always add "net" or "net-next" prefix in the subject of
your email to help -net maintainers. Do not hesitate to look at the
netdev FAQ for more details.

Here this patch looks like a fix so you should have [PATCH net] and a
"Fixes" tag. I guess for this patch you can use:

Fixes: 1b1c7a0ef7f3 ("mptcp: Add path manager interface")

That's where MPTCP_PM_MAX_ADDR has been introduced. It was already not
used and never used later.

Cheers,
Matt
--
Tessares | Belgium | Hybrid Access Solutions
http://www.tessares.net

2020-06-15 08:50:42

by Geliang Tang

[permalink] [raw]
Subject: Re: [PATCH] mptcp: unify MPTCP_PM_MAX_ADDR and MPTCP_PM_ADDR_MAX

On Fri, Jun 12, 2020 at 07:33:11PM +0200, Matthieu Baerts wrote:
> Hi Geliang,
>
> On 12/06/2020 07:27, Geliang Tang wrote:
> > Unify these two duplicate macros into 8.
>
> Thank you for this new patch!
>
> (...)
>
> > diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h
> > index 809687d3f410..86d265500cf6 100644
> > --- a/net/mptcp/protocol.h
> > +++ b/net/mptcp/protocol.h
> > @@ -135,7 +135,7 @@ static inline __be32 mptcp_option(u8 subopt, u8 len, u8 nib, u8 field)
> > ((nib & 0xF) << 8) | field);
> > }
> > -#define MPTCP_PM_MAX_ADDR 4
> > +#define MPTCP_PM_ADDR_MAX 8
>
> I think it would be better to drop MPTCP_PM_MAX_ADDR and keep
> MPTCP_PM_ADDR_MAX in pm_netlink.c where it is used. Each PM can decide
> what's the maximum number of addresses it can support.
>
> MPTCP_PM_MAX_ADDR seems to be a left over from a previous implementation of
> a PM that has not been upstreamed but replaced by the Netlink PM later.
>
> Also, please always add "net" or "net-next" prefix in the subject of your
> email to help -net maintainers. Do not hesitate to look at the netdev FAQ
> for more details.
>
> Here this patch looks like a fix so you should have [PATCH net] and a
> "Fixes" tag. I guess for this patch you can use:
>
> Fixes: 1b1c7a0ef7f3 ("mptcp: Add path manager interface")
>
> That's where MPTCP_PM_MAX_ADDR has been introduced. It was already not used
> and never used later.
>
> Cheers,
> Matt
> --
> Tessares | Belgium | Hybrid Access Solutions
> http://www.tessares.net

Hi Matt,

Thanks for your reply.
I have already resend patch v2 to you.

-Geliang