Fix regression of always3.1.8A with VHDL target
The patch for pr2516774 exposed a bug which caused always3.1.8A to fail. This patch corrects that. The test to decide when to use a sensitivity list rather than an explicit wait statement wasn't tight enough.
This commit is contained in:
parent
78bc4b5d47
commit
ba8688c0d1
|
|
@ -719,7 +719,8 @@ static int draw_wait(vhdl_procedural *_proc, stmt_container *container,
|
|||
// If this container is the top-level statement (i.e. it is the
|
||||
// first thing inside a process) then we can extract these
|
||||
// events out into the sensitivity list
|
||||
bool is_top_level = proc->get_container()->empty();
|
||||
bool is_top_level = container == proc->get_container()
|
||||
&& container->empty();
|
||||
|
||||
// See if this can be implemented in a more idomatic way before we
|
||||
// fall back on the generic translation
|
||||
|
|
|
|||
Loading…
Reference in New Issue