From mboxrd@z Thu Jan  1 00:00:00 1970
From: Michael Tremer <michael.tremer@ipfire.org>
To: location@lists.ipfire.org
Subject: Re: Segfault on d5ff39d "tree: Actually delete any deleted nodes"
Date: Mon, 17 Jul 2023 20:09:21 +0100
Message-ID: <35C9DB4A-BAC6-4A2F-BEC5-B41A8ABE7CB3@ipfire.org>
In-Reply-To:
 <CA+sCei3047G9U0J2yojOCVM6xyFiQ28g7=znBXRdit+dhLqxiA@mail.gmail.com>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="===============1842099325158015567=="
List-Id: <location.lists.ipfire.org>

--===============1842099325158015567==
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable

Hello Valters,

Thanks for reviewing my changes. This was indeed a thing that I did observe a=
s well and fixed it right here:

  https://git.ipfire.org/?p=3Dlocation/libloc.git;a=3Dcommitdiff;h=3D0e894966=
750e89d1b83c6ad90f37945d4d349118

The root node got deleted and lots of the code just relied on it being there.=
 Especially when the tree was empty.

-Michael

> On 17 Jul 2023, at 19:35, Valters Jansons <valter.jansons(a)gmail.com> wrot=
e:
>=20
> Hi,
>=20
> I am seeing a segfault during tests (src/test-as, src/test-country and
> src/test-signature) for all Ubuntu LTS versions. This only occurs with
> the latest commit (d5ff39d) applied. Looking at the change, the node
> still seems to be used after dereference, but I am not fully sure of
> the way you have planned out the deletion, therefore I am not offering
> a patch/resolution.
>=20
> From gdb I am seeing:
>=20
> Program received signal SIGSEGV, Segmentation fault.
> loc_network_tree_delete_nodes (tree=3D0x55555555a370) at src/network.c:1045
> 1045            tree->root->deleted =3D 0;
>=20
> Best regards,
> Valters Jansons


--===============1842099325158015567==--