Upgrade minor rust version (Fixes rust build compilation cache) (#2617)

* Upgrade minor rust version (Fixes rust build compilation cache)

* Black

Dockerfile

 # Rust builder
-FROM lukemathwalker/cargo-chef:latest-rust-1.80 AS chef
+FROM lukemathwalker/cargo-chef:latest-rust-1.80.1 AS chef
 WORKDIR /usr/src
 
 ARG CARGO_REGISTRIES_CRATES_IO_PROTOCOL=sparse
@@ -32,6 +32,7 @@ RUN cargo chef cook --profile release-opt --recipe-path recipe.json
 ARG GIT_SHA
 ARG DOCKER_LABEL
 
+COPY Cargo.lock Cargo.lock
 COPY Cargo.toml Cargo.toml
 COPY rust-toolchain.toml rust-toolchain.toml
 COPY proto proto
@@ -39,7 +40,7 @@ COPY benchmark benchmark
 COPY router router
 COPY backends backends
 COPY launcher launcher
-RUN cargo build --profile release-opt
+RUN cargo build --profile release-opt --frozen
 
 # Python builder
 # Adapted from: https://github.com/pytorch/pytorch/blob/master/Dockerfile

Dockerfile_amd

 # Rust builder
-FROM lukemathwalker/cargo-chef:latest-rust-1.80 AS chef
+FROM lukemathwalker/cargo-chef:latest-rust-1.80.1 AS chef
 WORKDIR /usr/src
 
 ARG CARGO_REGISTRIES_CRATES_IO_PROTOCOL=sparse
@@ -31,6 +31,7 @@ RUN cargo chef cook --profile release-opt --recipe-path recipe.json
 ARG GIT_SHA
 ARG DOCKER_LABEL
 
+COPY Cargo.lock Cargo.lock
 COPY Cargo.toml Cargo.toml
 COPY rust-toolchain.toml rust-toolchain.toml
 COPY proto proto
@@ -38,7 +39,7 @@ COPY benchmark benchmark
 COPY router router
 COPY backends backends
 COPY launcher launcher
-RUN cargo build --profile release-opt
+RUN cargo build --profile release-opt --frozen
 
 # Text Generation Inference base image for RoCm
 FROM rocm/dev-ubuntu-22.04:6.2 AS base

Dockerfile_intel

 ARG PLATFORM=xpu
 
-FROM lukemathwalker/cargo-chef:latest-rust-1.80 AS chef
+FROM lukemathwalker/cargo-chef:latest-rust-1.80.1 AS chef
 WORKDIR /usr/src
 
 ARG CARGO_REGISTRIES_CRATES_IO_PROTOCOL=sparse
@@ -32,6 +32,7 @@ RUN cargo chef cook --profile release-opt --recipe-path recipe.json
 ARG GIT_SHA
 ARG DOCKER_LABEL
 
+COPY Cargo.lock Cargo.lock
 COPY Cargo.toml Cargo.toml
 COPY rust-toolchain.toml rust-toolchain.toml
 COPY proto proto
@@ -39,7 +40,7 @@ COPY benchmark benchmark
 COPY router router
 COPY backends backends
 COPY launcher launcher
-RUN cargo build --profile release-opt
+RUN cargo build --profile release-opt --frozen
 
 
 # Text Generation Inference base image for Intel
@@ -52,7 +53,7 @@ ARG MAMBA_VERSION=23.1.0-1
 ARG PYTHON_VERSION='3.11.10'
 # Automatically set by buildx
 ARG TARGETPLATFORM
-ENV PATH /opt/conda/bin:$PATH
+ENV PATH=/opt/conda/bin:$PATH
 
 # TGI seem to require libssl.so.1.1 instead of libssl.so.3 so we can't use ubuntu 22.04. Ubuntu 20.04 has python==3.8, and TGI requires python>=3.9, hence the need for miniconda.
 # Install mamba

rust-toolchain.toml

 [toolchain]
 # Released on: June 13, 2024
 # https://releases.rs/docs/1.79.0/
-channel = "1.80.0"
+channel = "1.80.1"
 components = ["rustfmt", "clippy"]

server/text_generation_server/layers/attention/kv_cache.py

@@ -24,10 +24,8 @@ class KVCache:
     ):
         """Construct the key-value cache for a layer."""
 
-        if (
-            dtype == torch.float8_e5m2
-            and (ATTENTION != "flashinfer"
-            or SYSTEM != "cuda")
+        if dtype == torch.float8_e5m2 and (
+            ATTENTION != "flashinfer" or SYSTEM != "cuda"
         ):
             raise ValueError(
                 "float8_e5m2 KV cache is currently only supported for flashinfer on CUDA"