*   Options returned by `tf.data.Dataset.options()` are no longer mutable.
    *   tf.data input pipelines can now be executed in debug mode, which
        disables any asynchrony, parallelism, or non-determinism and forces
        Python execution (as opposed to trace-compiled graph execution) of
        user-defined functions passed into transformations such as `map`. The
        debug mode can be enabled through

        dep -->> dep: Handle exception, can't change response. E.g. close DB session.
    end
```

The new execution flow can be found in the docs: [Execution of dependencies with `yield`](https://fastapi.tiangolo.com/tutorial/dependencies/dependencies-with-yield/#execution-of-dependencies-with-yield).

### Features

encryption // streams to be done in parallel, which is achieved by interleaving // vcipher instructions from each stream. This is also referred to as // stitching, and provides significant performance improvements. // Some macros are defined which enable execution for big or little // endian as well as different ISA targets. //func (g *gcmAsm) counterCrypt(out, in []byte, counter *[gcmBlockSize]byte, key[gcmBlockSize]uint32) //func counterCryptASM(xr, out, in, counter, key) TEXT ·counterCryptASM(SB), NOSPLIT,...

encryption // streams to be done in parallel, which is achieved by interleaving // vcipher instructions from each stream. This is also referred to as // stitching, and provides significant performance improvements. // Some macros are defined which enable execution for big or little // endian as well as different ISA targets. //func (g *gcmAsm) counterCrypt(out, in []byte, counter *[gcmBlockSize]byte, key[gcmBlockSize]uint32) //func counterCryptASM(xr, out, in, counter, key) TEXT ·counterCryptASM(SB), NOSPLIT,...