Whatever this does, it only affecst Windows builds, and both GHA and
Lev agree that it still works...
Your patch has been applied to the master branch.
commit 5e94e8de4bfaf6637124947a3489710b591e5e26
Author: Frank Lichtenheld
Date: Tue Jun 20 15:53:09 2023 +0200
CMake: Add /Brepro to MSVC
Whatever it does... GHA and Lev agree that it's fine.
(One thing I noticed is that the "msbuild" builds now also run the
cmocka tests - this seems to be a new thing, and very welcome!)
I have extended the commit message to include the cache thing.
Your patch has been applied to the master branch
This patch touches a large number of files, but all the source code
changes are trivial (remove the "config-msvc.h" include).
The larger change "get rid of the evil MSVC xml files" is very
welcome :-) - one of the changes confuses me, though - this recent
"tapctl.exe.manifest" inclusion in EXTRA_
Acked-by: Gert Doering
Verified by building a tarball on 2.6, cert_data.h is now included.
Your patch has been applied to the release/2.6 branch.
commit 5eb84eb19749eb7e12f3becb36d6aaae89457b5b
Author: Frank Lichtenheld
Date: Wed Jun 21 14:58:42 2023 +0200
unit_tests: Add missing cert_d
I have tested this by pushing to GHA (as there are quite significant
changes to build.yaml) - it seems to still pass, which is good :-)
I do wonder why all unit tests now produce differently-named .exe
files (packet_id_testdriver.exe -> test_packet_id.exe etc)? Is this
a side effect of a new cmoc
I understand that this is work in progress, does not change anything,
and is not feature-complete either (but due to shared authorship, this
is a separate commit). It is not packaged into tarballs either.
According to one of Frank's latest mails, this patch series does not
work with 2.6 - so I'm
I have not tested this beyond "does it still compile?", but a quick
glance at the changes and the definitions agrees with Antonio's ACK.
Your patch has been applied to the master and release/2.6 branches.
commit 330bef679544b6a22d16a800c898927a785d74fc (master)
commit fa434643534d506c834df3eb1dbb
On Tue, Jun 27, 2023 at 11:40:30AM +0300, Lev Stipakov wrote:
> It would be nice to see the reason for the version bump in the commit message.
We could add something like: "Uses improved binary cache handling."
Not sure that helps much ;)
> Anyway, GHA is happy with this.
>
> Acked-by: Lev Stipa
It would be nice to see the reason for the version bump in the commit message.
Anyway, GHA is happy with this.
Acked-by: Lev Stipakov
ti 20. kesäk. 2023 klo 16.54 Frank Lichtenheld (fr...@lichtenheld.com)
kirjoitti:
>
> Change-Id: I4183edacc0295e9671c586cfcd77cf687015a22c
> Signed-off-by: Frank
Hi,
On 26/06/2023 15:09, Sergey Korolev via Openvpn-devel wrote:
Avoid compilation warnings on 32 bit platforms.
dco_linux.c: In function 'dco_update_peer_stat':
dco_linux.c:830:26: error: format '%lu' expects argument of type
'long unsigned int', but argument 4 has type 'counter_type'
{aka 'lo
Hi,
Interesting that MSFT hasn't been able to document this flag for almost 4 years:
https://developercommunity.visualstudio.com/t/document-brepro-flag/731265
I checked that the flag is indeed passed to the linker command line but haven't
tested that binaries are indeed identical.
Acked-by: Lev
11 matches
Mail list logo
