Fixed the rest of the build issues

crates/llama-cpp-bindings/build.rs

@@ -33,7 +33,7 @@ fn main() {
         println!("cargo:rustc-link-lib=cublasLt");
     }
     if cfg!(feature = "rocm") {
-        let amd_gpu_targets: Vec<&str> = vec![
+        let amd_gpu_default_targets: Vec<&str> = vec![
             "gfx803",
             "gfx900",
             "gfx906:xnack-",
@@ -51,6 +51,8 @@ fn main() {
             "gfx1102",
             "gfx1103",
         ];
+        let amd_gpu_targets =
+            env::var("AMDGPU_TARGETS").unwrap_or(amd_gpu_default_targets.join(";"));
 
         let rocm_root = env::var("ROCM_ROOT").unwrap_or("/opt/rocm".to_string());
         config.define("LLAMA_HIPBLAS", "ON");
@@ -59,7 +61,7 @@ fn main() {
             "CMAKE_CXX_COMPILER",
             format!("{}/llvm/bin/clang++", rocm_root),
         );
-        config.define("AMDGPU_TARGETS", amd_gpu_targets.join(";"));
+        config.define("AMDGPU_TARGETS", amd_gpu_targets);
         println!("cargo:rustc-link-arg=-Wl,--copy-dt-needed-entries");
         println!("cargo:rustc-link-search=native={}/hip/lib", rocm_root);
         println!("cargo:rustc-link-search=native={}/rocblas/lib", rocm_root);

crates/llama-cpp-bindings/src/llama.rs

@@ -83,7 +83,7 @@ impl LlamaServiceImpl {
         };
 
         for ffi::StepOutput { request_id, text } in result {
-            let mut stopped = false;
+            let mut stopped;
             let LlamaRunningRequest { tx, stop_condition } =
                 self.requests.get_mut(&request_id).unwrap();
 

crates/tabby/src/services/health.rs

@@ -22,7 +22,7 @@ pub struct HealthState {
     cpu_info: String,
     cpu_count: usize,
     accelerators: Vec<Accelerator>,
-    #[deprecated(note = "Use the more generic gpu_devices instead")]
+    #[deprecated(note = "Please use the more generic accelerators instead")]
     cuda_devices: Vec<String>,
     version: Version,
 }
@@ -103,7 +103,7 @@ pub fn read_rocm_devices() -> Result<Vec<Accelerator>> {
             Regex::new(r"(?m)^  Device Type: +([a-zA-Z0-9-]+) *$").unwrap();
     }
 
-    let cmd_res = Command::new("rocminfon").output()?;
+    let cmd_res = Command::new("rocminfo").output()?;
     let output = from_utf8(cmd_res.stdout.as_slice())?;
     let agent_outputs = output.split("Agent ").skip(1);
     let mut rocm_devices = vec![];

ee/tabby-webserver/src/api.rs

@@ -6,7 +6,7 @@ use tabby_common::api::{
 };
 use tokio_tungstenite::connect_async;
 
-use crate::{
+pub use crate::{
     schema::worker::{RegisterWorkerError, Worker, WorkerKind},
     websocket::WebSocketTransport,
 };

ee/tabby-webserver/src/lib.rs

@@ -26,7 +26,7 @@ use schema::{
 };
 use service::create_service_locator;
 use tabby_common::api::{
-    accelerator::Accelerator,
+    accelerator::{Accelerator, DeviceType},
     code::{CodeSearch, SearchResponse},
     event::RawEventLogger,
 };
@@ -159,6 +159,15 @@ impl Hub for Arc<HubImpl> {
         let addr = format!("http://{}:{}", self.conn.ip(), port);
         *worker_addr = addr.clone();
 
+        let mut cuda_devices = vec![];
+
+        for accelerator in &accelerators {
+            if accelerator.device_type == DeviceType::Cuda {
+                cuda_devices.push(accelerator.display_name.clone())
+            }
+        }
+
+        #[allow(deprecated)]
         let worker = Worker {
             name,
             kind,
@@ -168,6 +177,7 @@ impl Hub for Arc<HubImpl> {
             cpu_info,
             cpu_count,
             accelerators,
+            cuda_devices,
         };
         self.ctx.worker().register_worker(worker).await
     }

ee/tabby-webserver/src/schema/worker.rs

@@ -23,6 +23,8 @@ pub struct Worker {
     pub cpu_info: String,
     pub cpu_count: i32,
     pub accelerators: Vec<Accelerator>,
+    #[deprecated(note = "Please use the more generic accelerators instead")]
+    pub cuda_devices: Vec<String>,
 }
 
 #[derive(Serialize, Deserialize, Error, Debug)]

ee/tabby-webserver/src/service/worker.rs

@@ -82,6 +82,7 @@ mod tests {
     }
 
     fn make_worker(addr: &str) -> Worker {
+        #[allow(deprecated)]
         Worker {
             name: "Fake worker".to_owned(),
             kind: WorkerKind::Chat,
@@ -91,6 +92,7 @@ mod tests {
             cpu_info: "Fake CPU".to_owned(),
             cpu_count: 32,
             accelerators: vec![],
+            cuda_devices: vec![],
         }
     }
 }