Acked-by: Gert Doering
Documenation is good :-) - and proper text is even better.
Your patch has been applied to the master branch.
commit 57fe5a263b5a9f48d5c318498352fea7e8ac9908
Author: Simon Rozman
Date: Mon Mar 9 14:17:19 2020 +0100
tapctl: Update documentation
Signed-off-by:
Acked-by: Gert Doering
I won't claim to understand what this all does, but the change matches
the documentation - if a "file name" is returned that has "8.3|long"
format, it will use the "long" part of it, otherwise the string as it is.
Your patch has been applied to the master branch.
commit 3
Acked-by: Gert Doering
It only takes code away, which is now handled outside in the
tap-windows6.msm (tap-windows6/pull/106). Not claiming to understand
any of this, but the code change does not introduce anything new
which looks dangerous - and it's all your code anyway, so you'll know
best :
Hi,
On Fri, Feb 21, 2020 at 11:07:54AM +0100, Arne Schwabe wrote:
> Patch V5: Fix tests with OpenSSL 1.0.2 and libraries missing Chacha
Could you test this with mbedTLS 2.19, please? Maybe it's a local thing,
but the test crashes here, with
[ RUN ] test_check_ncp_ciphers_list
Unsupported c
Hi,
I believe M_ERRNO should handle this…
After the register_ring_buffers() exits, the Win32 error code is left in
GetLastError().
msg(M_ERRNO) picks it from there and formats a human-readable message.
Regards, Simon
From: Lev Stipakov
Sent: Wednesday, February 5, 2020 8:38 PM
T
Your patch has been applied to the master branch.
Taking the ACK from Lev on the v4 patch, verified that v5 has the same
code changes ("diff -uw v4.txt v5.txt") and that the difference is indeed
only in the test module.
Tested on FreeBSD 11.3 / openssl 1.0.2s which does not have chacha20-poly
and
1. We don't need two custom actions to evaluate the system state, do we?
2. FindTUNTAPAdapters was actually broken. It enumerated all existing
network adapters, rather than just the ones we are interested in:
TAP-Windows6 and Wintun.
3. TUNTAPADAPTER and ACTIVETUNTAPADAPTERS were split into
The function and property names that are common to TAP and TUN from
TAP-Windows6 and TUN from Wintun were renamed not to make the now
mainstream TUN sad.
I would have go with just the "adapter". But, wouldn't that cause
confusion when user sees "Deleting adapters" when uninstalling the
OpenVPN?
I
TAP-Windows6 adapters created with tapinstall/devcon.exe have hardware
ID "tap0901", where TAP-Windows6 adapters created with tapctl.exe have
hardware ID "root\\tap0901".
The enumeration of the network adapters have been extended to detect
adapters using a list of acceptable hardware IDs.
Signed-
Signed-off-by: Simon Rozman
---
src/openvpnmsica/openvpnmsica.c | 63 +++--
1 file changed, 37 insertions(+), 26 deletions(-)
diff --git a/src/openvpnmsica/openvpnmsica.c b/src/openvpnmsica/openvpnmsica.c
index 4c186b13..749dce56 100644
--- a/src/openvpnmsica/openvpnm
Interface is not equal to adapter. A quote from Microsoft documentation:
> There is a one-to-one correspondence between the interfaces and
> adapters on a given computer. An interface is an IP-level abstraction,
> whereas an adapter is a datalink-level abstraction.
As tapctl and openvpnmsica are
The MSI packages are switching to TAP-Windows6 and Wintun MSM modules to
install the TAP/TUN driver. The MSM modules have built-in Windows
version detection already.
This commit is now-dead-code clean up with uncrustification.
Signed-off-by: Simon Rozman
---
src/openvpnmsica/openvpnmsica.c | 12
Sequence scripts in temporary files has been discontinued in favor of
much simpler sequence strings passed to individual custom actions.
Pros: no temporary files; less code
Cons: the evaluation phase must make a complete plan what to perform in
each deferred custom action
Signed-off-by: Simon Roz
tap_create_adapter() and tap_list_adapter() no longer default to
"root\tap0901". Defining a default hardware ID value is at the
responsibility of upper layers that process user desires.
Since the tap_list_adapter() no longer defaults the hardware ID to
anything, its behavior was simplified to retu
Signed-off-by: Simon Rozman
---
src/tapctl/tap.c | 287 ---
src/tapctl/tap.h | 28 +
2 files changed, 275 insertions(+), 40 deletions(-)
diff --git a/src/tapctl/tap.c b/src/tapctl/tap.c
index b8249919..576f6740 100644
--- a/src/tapctl/tap.c
+++ b/
The DisplayName field type is Filename in the MSI database. This means
it must be authored as "8.3" filename, or "8.3|long filename".
This issue does not appear until interface names grow longer than 8
characters, or contain characters invalid in 8.3 filename notation.
Signed-off-by: Simon Rozman
Signed-off-by: Simon Rozman
---
src/openvpnmsica/openvpnmsica.c | 73 ++---
1 file changed, 49 insertions(+), 24 deletions(-)
diff --git a/src/openvpnmsica/openvpnmsica.c b/src/openvpnmsica/openvpnmsica.c
index 4223b680..cfbda8da 100644
--- a/src/openvpnmsica/openvpnm
Signed-off-by: Simon Rozman
---
src/tapctl/tap.c | 17 -
src/tapctl/tap.h | 18 +-
2 files changed, 17 insertions(+), 18 deletions(-)
diff --git a/src/tapctl/tap.c b/src/tapctl/tap.c
index 464ce725..b8249919 100644
--- a/src/tapctl/tap.c
+++ b/src/tapctl/tap.c
@@
18 matches
Mail list logo