[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: soname bump? (was: GNU Libtasn1 4.11 released)
From: |
Nikos Mavrogiannopoulos |
Subject: |
Re: soname bump? (was: GNU Libtasn1 4.11 released) |
Date: |
Mon, 29 May 2017 06:49:10 +0200 |
On Mon, 2017-05-29 at 06:37 +0200, Nikos Mavrogiannopoulos wrote:
> On Sun, 2017-05-28 at 18:02 +0200, Andreas Metzler wrote:
> > On 2017-05-27 Nikos Mavrogiannopoulos <address@hidden> wrote:
> > [...]
> > > * Noteworthy changes in release 4.11 (released 2017-05-27)
> > > [stable]
> > > - Introduced the ASN1_TIME_ENCODING_ERROR error code to indicate
> > > an invalid encoding in the DER time fields.
> > > - Introduced flag ASN1_DECODE_FLAG_ALLOW_INCORRECT_TIME. This
> > > flag
> > > allows decoding errors in time fields even when in strict DER
> > > mode.
> > > That is introduced in order to allow toleration of invalid
> > > times
> > > in
> > > X.509 certificates (which are common) even though strict DER
> > > adherence
> > > is enforced in other fields.
> > > - Added safety check in asn1_find_node(). That prevents a crash
> > > when a very long variable name is provided by the developer.
> > > Note that this to be exploited requires controlling the ASN.1
> > > definitions used by the developer, i.e., the 'name' parameter
> > > of
> > > asn1_write_value() or asn1_read_value(). The library is
> > > not designed to protect against malicious manipulation of the
> > > developer assigned variable names. Reported by Jakub Jirasek.
> >
> > [...]
> >
> > Hello,
> >
> > this release features a soname bump (libtasn1.so.6 ->
> > libtasn1.so.7).
> > As
> > the changelog does not mention an ABI break I assume this was not
> > done
> > intentionally. Perhaps a typo, bumping LT_CURRENT instead of
> > LT_REVISION?
>
> You are right. I'll make a new release fixing that issue and
> introduce
> an ABI check to avoid such issues in the future.
Thanks, I've released 4.12 addressing only that issue. It seems that
there is already an abi check with abi-compliance-checker as part of
the release process but unfortunately it doesn't include the so-name
version matching as part of its tests. I'm open to suggestions for
that.
regards,
Nikos