Matthias Aßhauer <mha1993@xxxxxxx> writes:
>>+ switch (variant) {
>>+ case VARIANT_AUTO:
>>+ BUG("VARIANT_AUTO passed to push_ssh_options");
>
> Why do we keep replicating this all over push_ssh_options? The wording
> very much sounds like it's a bug whenever VARIANT_AUTO gets passed to
> push_ssh_options, no exceptions. In that case ...
My reading of the code tells me that it is *not* the case. When the
caller asks to connect only over IPv4 with CONNECT_IPV4 bit set in
the flags, we cannot translate that to an option to underlying SSH
implementation without knowing the variant. As long as the caller
does not specify IPV4/IPV6 or custom port, I think it is OK for the
caller to leave the variant AUTO.
So you probably want a patch along the lines of...
connect.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/connect.c b/connect.c
index b6451ab5e8..b02cf74c9a 100644
--- a/connect.c
+++ b/connect.c
@@ -1118,7 +1118,7 @@ static void push_ssh_options(struct argv_array *args, struct argv_array *env,
if (flags & CONNECT_IPV4) {
switch (variant) {
case VARIANT_AUTO:
- BUG("VARIANT_AUTO passed to push_ssh_options");
+ BUG("VARIANT_AUTO and CONNECT_IPV4 used together");
case VARIANT_SIMPLE:
die(_("ssh variant 'simple' does not support -4"));
case VARIANT_SSH:
