Hi all,
I'm trying to define some header fields of type FT_UINT64 which have non-zero bitmasks (they all sit inside the same 64-bit word). These fields are sign extended when displayed, which is not what I expected. The sign extension happens in epan/proto.c, function proto_tree_set_uint64().
Comparing the 32-bit and 64-bit, and signed/unsigned variants of this function, it seems to me that this is a bug, possibly caused by forgetting to remove the sign extension code when proto_tree_set_int64 was added in commit e2bcee6a. Am I right, or is this behavior intended?
Thanks,
Jens.