llama.swiftui : initial bench functionality

examples/llama.swiftui/llama.cpp.swift/LibLlama.swift

@@ -6,6 +6,22 @@ enum LlamaError: Error {
     case couldNotInitializeContext
 }
 
+func llama_batch_clear(_ batch: inout llama_batch) {
+    batch.n_tokens = 0
+}
+
+func llama_batch_add(_ batch: inout llama_batch, _ id: llama_token, _ pos: llama_pos, _ seq_ids: [llama_seq_id], _ logits: Bool) {
+    batch.token   [Int(batch.n_tokens)] = id
+    batch.pos     [Int(batch.n_tokens)] = pos
+    batch.n_seq_id[Int(batch.n_tokens)] = Int32(seq_ids.count)
+    for i in 0..<seq_ids.count {
+        batch.seq_id[Int(batch.n_tokens)]![Int(i)] = seq_ids[i]
+    }
+    batch.logits  [Int(batch.n_tokens)] = logits ? 1 : 0
+
+    batch.n_tokens += 1
+}
+
 actor LlamaContext {
     private var model: OpaquePointer
     private var context: OpaquePointer
@@ -16,6 +32,7 @@ actor LlamaContext {
 
     var n_len: Int32 = 512
     var n_cur: Int32 = 0
+
     var n_decode: Int32 = 0
 
     init(model: OpaquePointer, context: OpaquePointer) {
@@ -27,12 +44,13 @@ actor LlamaContext {
     }
 
     deinit {
+        llama_batch_free(batch)
         llama_free(context)
         llama_free_model(model)
         llama_backend_free()
     }
 
-    static func createContext(path: String) throws -> LlamaContext {
+    static func create_context(path: String) throws -> LlamaContext {
         llama_backend_init(false)
         let model_params = llama_model_default_params()
 
@@ -41,11 +59,15 @@ actor LlamaContext {
             print("Could not load model at \(path)")
             throw LlamaError.couldNotInitializeContext
         }
+
+        let n_threads = max(1, min(8, ProcessInfo.processInfo.processorCount - 2))
+        print("Using \(n_threads) threads")
+
         var ctx_params = llama_context_default_params()
-        ctx_params.seed = 1234
+        ctx_params.seed  = 1234
         ctx_params.n_ctx = 2048
-        ctx_params.n_threads = 8
-        ctx_params.n_threads_batch = 8
+        ctx_params.n_threads       = UInt32(n_threads)
+        ctx_params.n_threads_batch = UInt32(n_threads)
 
         let context = llama_new_context_with_model(model, ctx_params)
         guard let context else {
@@ -56,6 +78,26 @@ actor LlamaContext {
         return LlamaContext(model: model, context: context)
     }
 
+    func model_info() -> String {
+        let result = UnsafeMutablePointer<Int8>.allocate(capacity: 256)
+        result.initialize(repeating: Int8(0), count: 256)
+        defer {
+            result.deallocate()
+        }
+
+        // TODO: this is probably very stupid way to get the string from C
+
+        let nChars = llama_model_desc(model, result, 256)
+        let bufferPointer = UnsafeBufferPointer(start: result, count: Int(nChars))
+
+        var SwiftString = ""
+        for char in bufferPointer {
+            SwiftString.append(Character(UnicodeScalar(UInt8(char))))
+        }
+
+        return SwiftString
+    }
+
     func get_n_tokens() -> Int32 {
         return batch.n_tokens;
     }
@@ -79,16 +121,11 @@ actor LlamaContext {
             print(String(cString: token_to_piece(token: id) + [0]))
         }
 
-        // batch = llama_batch_init(512, 0) // done in init()
-        batch.n_tokens = Int32(tokens_list.count)
+        llama_batch_clear(&batch)
 
-        for i1 in 0..<batch.n_tokens {
+        for i1 in 0..<tokens_list.count {
             let i = Int(i1)
-            batch.token[i] = tokens_list[i]
-            batch.pos[i] = i1
-            batch.n_seq_id[Int(i)] = 1
-            batch.seq_id[Int(i)]![0] = 0
-            batch.logits[i] = 0
+            llama_batch_add(&batch, tokens_list[i], Int32(i), [0], false)
         }
         batch.logits[Int(batch.n_tokens) - 1] = 1 // true
 
@@ -141,18 +178,11 @@ actor LlamaContext {
         print(new_token_str)
         // tokens_list.append(new_token_id)
 
-        batch.n_tokens = 0
-
-        batch.token[Int(batch.n_tokens)] = new_token_id
-        batch.pos[Int(batch.n_tokens)] = n_cur
-        batch.n_seq_id[Int(batch.n_tokens)] = 1
-        batch.seq_id[Int(batch.n_tokens)]![0] = 0
-        batch.logits[Int(batch.n_tokens)] = 1 // true
-        batch.n_tokens += 1
+        llama_batch_clear(&batch)
+        llama_batch_add(&batch, new_token_id, n_cur, [0], true)
 
         n_decode += 1
-
-        n_cur += 1
+        n_cur    += 1
 
         if llama_decode(context, batch) != 0 {
             print("failed to evaluate llama!")
@@ -161,8 +191,60 @@ actor LlamaContext {
         return new_token_str
     }
 
-    func bench() -> String{
-        return "bench not implemented"
+    func bench() -> String {
+        let pp = 512
+        let tg = 128
+        let pl = 1
+
+        // bench prompt processing
+
+        llama_batch_clear(&batch)
+
+        let n_tokens = pp
+
+        for i in 0..<n_tokens {
+            llama_batch_add(&batch, 0, Int32(i), [0], false)
+        }
+        batch.logits[Int(batch.n_tokens) - 1] = 1 // true
+
+        llama_kv_cache_clear(context)
+
+        let t_pp_start = ggml_time_us()
+
+        if llama_decode(context, batch) != 0 {
+            print("llama_decode() failed during prompt")
+        }
+
+        let t_pp_end = ggml_time_us()
+
+        // bench text generation
+
+        llama_kv_cache_clear(context)
+
+        let t_tg_start = ggml_time_us()
+
+        for i in 0..<tg {
+            llama_batch_clear(&batch)
+
+            for j in 0..<pl {
+                llama_batch_add(&batch, 0, Int32(i), [Int32(j)], true)
+            }
+
+            if llama_decode(context, batch) != 0 {
+                print("llama_decode() failed during text generation")
+            }
+        }
+
+        let t_tg_end = ggml_time_us()
+
+        let t_pp = Double(t_pp_end - t_pp_start) / 1000000.0
+        let t_tg = Double(t_tg_end - t_tg_start) / 1000000.0
+
+        let speed_pp = Double(pp)    / t_pp
+        let speed_tg = Double(pl*tg) / t_tg
+
+        return String(format: "PP 512 speed: %7.2f t/s\n", speed_pp) +
+               String(format: "TG 128 speed: %7.2f t/s\n", speed_tg)
     }
 
     func clear() {

examples/llama.swiftui/llama.swiftui/Models/LlamaState.swift

@@ -6,7 +6,7 @@ class LlamaState: ObservableObject {
 
     private var llamaContext: LlamaContext?
     private var modelUrl: URL? {
-        Bundle.main.url(forResource: "ggml-model-q8_0", withExtension: "gguf", subdirectory: "models")
+        Bundle.main.url(forResource: "ggml-model", withExtension: "gguf", subdirectory: "models")
         // Bundle.main.url(forResource: "llama-2-7b-chat", withExtension: "Q2_K.gguf", subdirectory: "models")
     }
     init() {
@@ -20,7 +20,7 @@ class LlamaState: ObservableObject {
     private func loadModel() throws {
         messageLog += "Loading model...\n"
         if let modelUrl {
-            llamaContext = try LlamaContext.createContext(path: modelUrl.path())
+            llamaContext = try LlamaContext.create_context(path: modelUrl.path())
             messageLog += "Loaded model \(modelUrl.lastPathComponent)\n"
         } else {
             messageLog += "Could not locate model\n"
@@ -49,7 +49,10 @@ class LlamaState: ObservableObject {
             return
         }
 
+        messageLog += "Model info: "
+        messageLog += await llamaContext.model_info() + "\n"
         messageLog += "Running benchmark...\n"
+        await llamaContext.bench() // heat up
         let result = await llamaContext.bench()
         messageLog += "\(result)"
     }

examples/llama.swiftui/llama.swiftui/UI/ContentView.swift

@@ -53,6 +53,6 @@ struct ContentView: View {
     }
 }
 
-#Preview {
-    ContentView()
-}
+//#Preview {
+//    ContentView()
+//}