*   XLA reduction emitter is deterministic when the environment variable
        `TF_DETERMINISTIC_OPS` is set to "true" or "1". This extends
        deterministic `tf.nn.bias_add` back-prop functionality (and therefore
        also deterministic back-prop of bias-addition in Keras layers) to
        include when XLA JIT compilation is enabled.
    *   Fix problem, when running on a CUDA GPU and when either environment

func cryptBlocks(c code, key, dst, src *byte, length int) TEXT ·cryptBlocks(SB),NOSPLIT,$0-40 MOVD key+8(FP), R1 MOVD dst+16(FP), R2 MOVD src+24(FP), R4 MOVD length+32(FP), R5 MOVD c+0(FP), R0 loop: KM R2, R4 // cipher message (KM) BVS loop // branch back if interrupted XOR R0, R0 RET // func cryptBlocksChain(c code, iv, key, dst, src *byte, length int) TEXT ·cryptBlocksChain(SB),NOSPLIT,$48-48 LA params-48(SP), R1 MOVD iv+8(FP), R8 MOVD key+16(FP), R9 MVC $16, 0(R8), 0(R1) // move iv into params MVC...

func cryptBlocks(c code, key, dst, src *byte, length int) TEXT ·cryptBlocks(SB),NOSPLIT,$0-40 MOVD key+8(FP), R1 MOVD dst+16(FP), R2 MOVD src+24(FP), R4 MOVD length+32(FP), R5 MOVD c+0(FP), R0 loop: KM R2, R4 // cipher message (KM) BVS loop // branch back if interrupted XOR R0, R0 RET // func cryptBlocksChain(c code, iv, key, dst, src *byte, length int) TEXT ·cryptBlocksChain(SB),NOSPLIT,$48-48 LA params-48(SP), R1 MOVD iv+8(FP), R8 MOVD key+16(FP), R9 MVC $16, 0(R8), 0(R1) // move iv into params MVC...

### Internal

* 🔧 Add sponsor Porter. PR [#10051](https://github.com/tiangolo/fastapi/pull/10051) by [@tiangolo](https://github.com/tiangolo).
* 🔧 Update sponsors, add Jina back as bronze sponsor. PR [#10050](https://github.com/tiangolo/fastapi/pull/10050) by [@tiangolo](https://github.com/tiangolo).