From: Thomas Petazzoni Date: Sat, 25 Apr 2020 16:13:18 +0200 Subject: [Buildroot] [PATCH 1/1] package/libtorrent: annotate CVE-2009-1760 and CVE-2016-5301 In-Reply-To: <20200301200049.211453-1-fontaine.fabrice@gmail.com> References: <20200301200049.211453-1-fontaine.fabrice@gmail.com> Message-ID: <20200425161318.00ee0b62@windsurf.home> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net On Sun, 1 Mar 2020 21:00:49 +0100 Fabrice Fontaine wrote: > CVE-2009-1760 and CVE-2016-5301 are misclassified (by our CVE tracker) > as affecting libtorrent, while in fact they affect libtorrent-rasterbar. > > Signed-off-by: Fabrice Fontaine > --- > package/libtorrent/libtorrent.mk | 4 ++++ > 1 file changed, 4 insertions(+) > > diff --git a/package/libtorrent/libtorrent.mk b/package/libtorrent/libtorrent.mk > index c8310cab65..17c6f92ab4 100644 > --- a/package/libtorrent/libtorrent.mk > +++ b/package/libtorrent/libtorrent.mk > @@ -14,6 +14,10 @@ LIBTORRENT_INSTALL_STAGING = YES > LIBTORRENT_LICENSE = GPL-2.0 > LIBTORRENT_LICENSE_FILES = COPYING > > +# CVE-2009-1760 and CVE-2016-5301 are misclassified (by our CVE tracker) as > +# affecting libtorrent, while in fact they affect libtorrent-rasterbar. > +LIBTORRENT_IGNORE_CVES += CVE-2009-1760 CVE-2016-5301 Here as well, I don't think this is the proper solution. You're just papering over the actual problem, not solving it. CVE-2009-1760 is described like this: "affects" : { "vendor" : { "vendor_data" : [ { "vendor_name" : "rasterbar_software", "product" : { "product_data" : [ { "product_name" : "libtorrent", So indeed, since we currently only match on "product_name", we believe this CVE is for libtorrent, while it is for libtorrent-rasterbar. The proper way to solve this is to have better CPE matching logic. Your patch is just removing the CVEs from libtorrent, but not making them appear for libtorrent-rasterbar (even though they probably wouldn't appear as they are quite old). Let's work together on a proper CPE matching logic instead. Thanks, Thomas -- Thomas Petazzoni, CTO, Bootlin Embedded Linux and Kernel engineering https://bootlin.com