Hi, On Tue, Dec 29, 2015 at 06:03:04PM +0200, Lev Stipakov wrote: > * Use adapter name instead of index on WinXP - sadly XP does not support > indexes > * Write Windows version to log > * Send it with peer-info as IV_PLAT_VER
Overall, I'm fine with the patch - thanks a lot.
I've built release/2.3 + patch on ubuntu 14.04 and tested on XP SP3, and
it nicely worked - sending IV_PLAT_VER=5.1 ("WIN_XP"), and using interface
name again.
A few small details (half of them already discussed on IRC):
- please send IV_PLAT_VER only if the client requested this (push-peer-info)
(this is slightly more sensitive information, comparable to library
versions, and when introducing library versions in IV_SSL, we decided
to not send such information by default)
- and please send a patch of the windows version bits for master ("all but
the changes against route.c and tun.c" :) )
- src/compat/compat-versionhelpers.h needs to be listed "somewhere"
(src/compat/Makefile.am libcompat_la_SOURCES=... maybe?) so that
"make dist" will include it into the generated .tar.gz
> #elif defined (WIN32)
>
> - struct buffer out = alloc_buf_gc (64, &gc);
> - buf_printf (&out, "interface=%d", tt->adapter_index );
> - device = buf_bptr(&out);
> + if (win32_version_info() != WIN_XP)
> + {
> + struct buffer out = alloc_buf_gc (64, &gc);
> + buf_printf (&out, "interface=%d", tt->adapter_index );
> + device = buf_bptr(&out);
> + }
the indentation on these is off - they are indented with a full <tab>
but our funny 2.3 convention requires *six* spaces here... (sorry for
being a pain... trying to be consistent)
Same thing for delete_route_ipv6().
Functionally, this all looks perfectly fine (I only tested on XP, but
as the rest is just existing code, it will "obviously" work for the
non-XP case).
[..]
> +int
> +win32_version_info()
> +{
> + if (!IsWindowsXPOrGreater())
> + {
> + msg (M_FATAL, "Error: Windows version must be XP or greater.");
> + }
> +
> + if (!IsWindowsVistaOrGreater())
> + {
> + return WIN_XP;
> + }
I'm not sure if this is the canonical best version to deal with it or
not, but I don't particularily *want* to get myself involved into this.
If someone else thinks this needs improvement (like, working Win10
detection :-) ) I'm all open to accept additional patches on top of it,
but to solve our immediate needs, this is good enough for me and does
not *need* changes.
thanks, and waiting for v2 :-)
gert
--
USENET is *not* the non-clickable part of WWW!
//www.muc.de/~gert/
Gert Doering - Munich, Germany [email protected]
fax: +49-89-35655025 [email protected]
signature.asc
Description: PGP signature
