From: "its_Giaan (Code Review)"
Date: Fri, 24 Jan 2025 11:48:44 +0000
In-Reply-To:
References:
Reply-To: gianmarco@...2726..., arne-openvpn@...1227...,
openvpn-devel@lists.sourceforge.net, frank@...2641...,
gert@...1296..., a@...2181...
Message-ID: <325f40cf54aecda355caa586eac4ab1dc5e05d14-EmailReviewComments-HTML@...2715...>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Content-Disposition: inline
Content-Type: multipart/alternative; boundary="a6RmCWhHOHA="; charset=UTF-8
Subject: [Openvpn-devel] [S] Change in openvpn[master]: mroute: adapt to new
protocol handling and hashing improvements
List-Id:
List-Unsubscribe: ,
List-Archive:
List-Post:
List-Help:
List-Subscribe: ,
Cc: cron2 , plaisthos , flichtenheld , openvpn-devel
--a6RmCWhHOHA=
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Attention is currently required from: cron2, flichtenheld, ordex, plaisthos.
its_Giaan has posted comments on this change. ( http://gerrit.openvpn.net/c=
/openvpn/+/765?usp=3Demail )
Change subject: mroute: adapt to new protocol handling and hashing improvem=
ents
......................................................................
Patch Set 15:
(7 comments)
File src/openvpn/forward.c:
http://gerrit.openvpn.net/c/openvpn/+/765/comment/7d7d6798_7dba137c :
PS14, Line 1132: for (int i =3D 0; i < c->c1.link_sockets_num; i++)
> I assume this hunk is needed, but it's sort of "outside the scope of mrou=
te handling" (so should eit [=E2=80=A6]
This is still about client context, so no need to loop and no need to touch=
this again.
http://gerrit.openvpn.net/c/openvpn/+/765/comment/4c321545_89555b0c :
PS14, Line 1135: /* on the instance context we have only on=
e socket, so just check the first one */
> whatever the outcome is, this comment is no longer correct :-)
Done
File src/openvpn/mroute.h:
http://gerrit.openvpn.net/c/openvpn/+/765/comment/9bb3a8db_d8b619b4 :
PS14, Line 225: return (uint32_t) a->len + 2;
> if you change the start of the to-be-hashed block, but all the existing d=
ata should still be hashed, [=E2=80=A6]
Done
File src/openvpn/mudp.c:
http://gerrit.openvpn.net/c/openvpn/+/765/comment/ff1ec800_dab6a655 :
PS14, Line 196: m->local.proto =3D real.proto;
> This is a bit confusing. I can see the assignment to `real. [=E2=80=A6]
That's a leftover, I see that m->local is used only on virtual addresses co=
ntext so no need to copy the real proto and since it's CLEAR() no need to s=
et the proto to 0.
File src/openvpn/multi.c:
http://gerrit.openvpn.net/c/openvpn/+/765/comment/f9fda981_a1023801 :
PS14, Line 1156: struct mroute_addr *add=
r,
> this looks wrong. A lookup function should not modify its arguments. [=
=E2=80=A6]
Done
http://gerrit.openvpn.net/c/openvpn/+/765/comment/285a38d0_93d7ac29 :
PS14, Line 3356:
> see comment above - now you're actually setting `proto` in the caller, ev=
en if I do not agree this i [=E2=80=A6]
Done
http://gerrit.openvpn.net/c/openvpn/+/765/comment/9738102a_e71eb117 :
PS14, Line 3561: dest.proto =3D src.proto;
> same thing -> `mroute_extract_addr_from_packet()`
Done
--
To view, visit http://gerrit.openvpn.net/c/openvpn/+/765?usp=3Demail
To unsubscribe, or for help writing mail filters, visit http://gerrit.openv=
pn.net/settings
Gerrit-Project: openvpn
Gerrit-Branch: master
Gerrit-Change-Id: Ic66eccb5058fe9c0fae64d8e2ca88728068a92ab
Gerrit-Change-Number: 765
Gerrit-PatchSet: 15
Gerrit-Owner: its_Giaan
Gerrit-Reviewer: cron2
Gerrit-Reviewer: flichtenheld
Gerrit-Reviewer: plaisthos
Gerrit-CC: openvpn-devel
Gerrit-CC: ordex
Gerrit-Attention: plaisthos
Gerrit-Attention: cron2
Gerrit-Attention: flichtenheld
Gerrit-Attention: ordex
Gerrit-Comment-Date: Fri, 24 Jan 2025 11:48:44 +0000
Gerrit-HasComments: Yes
Gerrit-Has-Labels: No
Comment-In-Reply-To: cron2
Gerrit-MessageType: comment
--a6RmCWhHOHA=
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
This is a bit confusing. I can see t=
he assignment to `real. [=E2=80=A6]
That's a leftover, I see that m->=
;local is used only on virtual addresses context so no need to copy the rea=
l proto and since it's CLEAR() no need to set the proto to 0.