$1, t0, t1, y0 EXTR $1, t1, t2, y1 EXTR $1, t2, t3, y2 EXTR $1, t3, hlp0, y3 STy(y3out) LDx(x1in) LDy(s) CALL p256MulInternal<>(SB) STy(s) p256MulBy2Inline STx(tmp) LDx(m) CALL p256SqrInternal<>(SB) LDx(tmp) CALL p256SubInternal<>(SB) STx(x3out) LDy(s) CALL p256SubInternal<>(SB) LDy(m) CALL p256MulInternal<>(SB) LDx(y3out) CALL p256SubInternal<>(SB) STx(y3out) RET /* ---------- #undef y2in #undef x3out #undef y3out #undef z3out #define y2in(off) (off + 32)(b_ptr) #define x3out(off) (off)(b_ptr) #define...

### Bug Fixes and Other Changes

* `tf.py_function` and `tf.numpy_function` can now be used as function decorators for clearer code:
   ```
   @tf.py_function(Tout=tf.float32)
   def my_fun(x):
     print("This always executes eagerly.")
     return x+1
   ```

* `tf.lite`
    * Strided_Slice now supports `UINT32`.

* `tf.config.experimental.enable_tensor_float_32_execution`