Merge branch 'main' into fix-batch-ub

Author: ysimonson

.github/workflows/llama-cpp-rs-check.yml

@@ -18,7 +18,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Checkout
-        uses: actions/checkout@08c6903cd8c0fde910a37f88322edcfb5dd907a8
+        uses: actions/checkout@1af3b93b6815bc44a9784bd300feb67ff0d1eeb3
         with:
           submodules: recursive
       - name: Install Compile Deps
@@ -45,7 +45,7 @@ jobs:
         target: [ linux/arm64, linux/amd64 ]
     steps:
       - name: checkout
-        uses: actions/checkout@08c6903cd8c0fde910a37f88322edcfb5dd907a8
+        uses: actions/checkout@1af3b93b6815bc44a9784bd300feb67ff0d1eeb3
       - name: Setup QEMU
         uses: docker/setup-qemu-action@c7c53464625b32c7a7e944ae62b3e17d2b600130
         with:
@@ -63,7 +63,7 @@ jobs:
     runs-on: macos-latest
     steps:
       - name: checkout
-        uses: actions/checkout@08c6903cd8c0fde910a37f88322edcfb5dd907a8
+        uses: actions/checkout@1af3b93b6815bc44a9784bd300feb67ff0d1eeb3
         with:
           submodules: recursive
       - name: Setup Rust
@@ -75,7 +75,7 @@ jobs:
     runs-on: windows-latest
     steps:
       - name: checkout
-        uses: actions/checkout@08c6903cd8c0fde910a37f88322edcfb5dd907a8
+        uses: actions/checkout@1af3b93b6815bc44a9784bd300feb67ff0d1eeb3
         with:
           submodules: recursive
       - name: Setup Rust

.github/workflows/publish-upon-release.yml

@@ -14,7 +14,7 @@ jobs:
     runs-on: ubuntu-latest
 
     steps:
-      - uses: actions/checkout@08c6903cd8c0fde910a37f88322edcfb5dd907a8
+      - uses: actions/checkout@1af3b93b6815bc44a9784bd300feb67ff0d1eeb3
         with:
           submodules: recursive
       - name: Publish crates for llama-cpp-sys-2

.github/workflows/update-llama-cpp.yml

@@ -15,7 +15,7 @@ jobs:
     steps:
       - name: Set date
         run: echo "DATE=$(date -I)" >> $GITHUB_ENV
-      - uses: actions/checkout@08c6903cd8c0fde910a37f88322edcfb5dd907a8
+      - uses: actions/checkout@1af3b93b6815bc44a9784bd300feb67ff0d1eeb3
         name: Checkout latest
         with:
           submodules: recursive

.github/workflows/update-toml-version.yaml

@@ -15,7 +15,7 @@ jobs:
 
     steps:
       - name: Checkout code
-        uses: actions/checkout@08c6903cd8c0fde910a37f88322edcfb5dd907a8
+        uses: actions/checkout@1af3b93b6815bc44a9784bd300feb67ff0d1eeb3
         with:
           submodules: recursive
 

Cargo.lock

@@ -20,9 +20,9 @@ hf-hub = { version = "0.4.3" }
 criterion = "0.5.1"
 pprof = "0.13.0"
 bindgen = "0.72.1"
-cc = "1.2.46"
+cc = "1.2.47"
 anyhow = "1.0.100"
-clap = "4.5.51"
+clap = "4.5.53"
 encoding_rs = "0.8.35"
 tracing-subscriber = { version = "0.3", features = ["json"] }
 

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "embeddings"
-version = "0.1.126"
+version = "0.1.127"
 edition = "2021"
 publish = false
 

examples/embeddings/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "simple"
-version = "0.1.126"
+version = "0.1.127"
 edition = "2021"
 publish = false
 

examples/simple/Cargo.toml

@@ -1,7 +1,7 @@
 [package]
 name = "llama-cpp-2"
 description = "llama.cpp bindings for Rust"
-version = "0.1.126"
+version = "0.1.127"
 edition = "2021"
 license = "MIT OR Apache-2.0"
 repository = "https://github.com/utilityai/llama-cpp-rs"
@@ -10,7 +10,7 @@ repository = "https://github.com/utilityai/llama-cpp-rs"
 
 [dependencies]
 enumflags2 = "0.7.12"
-llama-cpp-sys-2 = { path = "../llama-cpp-sys-2", version = "0.1.126" }
+llama-cpp-sys-2 = { path = "../llama-cpp-sys-2", version = "0.1.127" }
 thiserror = { workspace = true }
 tracing = { workspace = true }
 tracing-core = { workspace = true }
@@ -35,7 +35,7 @@ mtmd = ["llama-cpp-sys-2/mtmd"]
 
 
 [target.'cfg(all(target_os = "macos", any(target_arch = "aarch64", target_arch = "arm64")))'.dependencies]
-llama-cpp-sys-2 = { path = "../llama-cpp-sys-2", version = "0.1.126", features = [
+llama-cpp-sys-2 = { path = "../llama-cpp-sys-2", version = "0.1.127", features = [
     "metal",
 ] }
 

llama-cpp-2/Cargo.toml

@@ -13,7 +13,7 @@
 //!
 //! - `cuda` enables CUDA gpu support.
 //! - `sampler` adds the [`context::sample::sampler`] struct for a more rusty way of sampling.
-use std::ffi::NulError;
+use std::ffi::{c_char, NulError};
 use std::fmt::Debug;
 use std::num::NonZeroI32;
 
@@ -398,7 +398,7 @@ pub struct LlamaBackendDevice {
 pub fn list_llama_ggml_backend_devices() -> Vec<LlamaBackendDevice> {
     let mut devices = Vec::new();
     for i in 0..unsafe { llama_cpp_sys_2::ggml_backend_dev_count() } {
-        fn cstr_to_string(ptr: *const i8) -> String {
+        fn cstr_to_string(ptr: *const c_char) -> String {
             if ptr.is_null() {
                 String::new()
             } else {