jit: Adjust back-patch of f90b4a84 to 12 and 13.

While back-patching f90b4a84, I missed that branches before
REL_14_STABLE did some (accidental?) type punning in a function
parameter, and failed to adjust these two branches accordingly.  That
didn't seem to cause a problem for newer LLVM versions or non-debug
builds, but older debug builds would fail a type cross-check assertion.
Fix by supplying the correct function argument type.  In REL_14_STABLE
the same change was made by commit df99ddc7.

Per build farm animal xenodermus, which runs a debug build of LLVM 6
with jit_above_cost=0.

Discussion: https://postgr.es/m/CA%2BhUKGLQ38rgZ3bvNHXPRjsWFAg3pa%3Dtnpeq0osa%2B%3DmiFD5jAw%40mail.gmail.com

diff --git a/src/backend/jit/llvm/llvmjit_expr.c b/src/backend/jit/llvm/llvmjit_expr.c
index d84fbba7cc5a2b9a98afe32007c1919a396b49c0..c2e367f00dba16df458f5608bba2202f9f921c5f 100644 (file)
--- a/src/backend/jit/llvm/llvmjit_expr.c
+++ b/src/backend/jit/llvm/llvmjit_expr.c
@@ -1126,7 +1126,7 @@ llvm_compile_expr(ExprState *state)
                                         llvm_pg_var_type("TypeExecEvalSubroutine"));
 
                    v_params[0] = v_state;
-                   v_params[1] = l_ptr_const(op, l_ptr(TypeSizeT));
+                   v_params[1] = l_ptr_const(op, l_ptr(StructExprEvalStep));
                    v_params[2] = v_econtext;
                    l_call(b,
                           LLVMGetFunctionType(ExecEvalSubroutineTemplate),