I have not tested this on a recent-enough OpenBSD or OpenSSL 1.0.x
(no time to build or find such a system) but I have tested on FreeBSD
with 1.1.1l - unsurprisingly (the #ifdef say so) no difference.
I'm a bit unsure if we need this for 2.5 - it's "long term compat"
and not very intrusive, but
I have not tested this on a recent-enough OpenBSD or OpenSSL 1.0.x
(no time to build or find such a system) but I have tested on FreeBSD
with 1.1.1l - unsurprisingly (the #ifdef say so) no difference.
I'm a bit unsure if we need this for 2.5 - it's "long term compat"
and not very intrusive, but
> I'm a bit unsure if we need this for 2.5 - it's "long term compat"
> and not very intrusive, but on the other hand, not too many people
> seem to care about LibreSSL.
OpenBSD has packaged 2.5.7 for snapshots, so they must already have a
workaround.
___
Hi,
On Tue, Aug 23, 2022 at 10:21:30AM +, Maximilian Fillinger wrote:
> > I'm a bit unsure if we need this for 2.5 - it's "long term compat"
> > and not very intrusive, but on the other hand, not too many people
> > seem to care about LibreSSL.
>
> OpenBSD has packaged 2.5.7 for snapshots, so
Am 23.08.22 um 20:55 schrieb Gert Doering:
Hi,
On Tue, Aug 23, 2022 at 10:21:30AM +, Maximilian Fillinger wrote:
I'm a bit unsure if we need this for 2.5 - it's "long term compat"
and not very intrusive, but on the other hand, not too many people
seem to care about LibreSSL.
OpenBSD has p
> but they think the revamped OpenSSL 3.0 way of calculating the TLS1 PRF
> might actually not be in 2.5 yet, so they do not need a patch for that.
In 2.5, openssl_compat.h also doesn't try to define X509_OBJECT_free(), so
there's nothing to backport there.
_
