Fast kernels for rotary embeddings. (#1928)

* Fast kernels for rotary embeddings.

* Add a test for the fast CPU kernel.

* Rope cuda bindings.

* Cuda kernel.

* Metal kernel (part 1).

* Cuda kernels.

* Finish the metal kernel.

* Use the new kernels in the quantized example.

* Fix warning.

2 files changed, 64 insertions, 0 deletions

diff --git a/candle-metal-kernels/src/lib.rs b/candle-metal-kernels/src/lib.rs
index e17365a0..e83814a8 100644
--- a/candle-metal-kernels/src/lib.rs
+++ b/candle-metal-kernels/src/lib.rs
@@ -809,6 +809,47 @@ pub fn call_rms_norm(
 }
 
 #[allow(clippy::too_many_arguments)]
+pub fn call_rope_i(
+    device: &Device,
+    command_buffer: &CommandBufferRef,
+    kernels: &Kernels,
+    kernel_name: &'static str,
+    bh: usize,
+    td: usize,
+    src: &Buffer,
+    src_offset: usize,
+    cos: &Buffer,
+    cos_offset: usize,
+    sin: &Buffer,
+    sin_offset: usize,
+    output: &Buffer,
+) -> Result<(), MetalKernelError> {
+    let pipeline = kernels.load_pipeline(device, Source::Reduce, kernel_name)?;
+    let encoder = command_buffer.new_compute_command_encoder();
+    encoder.set_compute_pipeline_state(&pipeline);
+
+    set_params!(
+        encoder,
+        (
+            bh,
+            td,
+            (src, src_offset),
+            (cos, cos_offset),
+            (sin, sin_offset),
+            output
+        )
+    );
+    let (thread_group_count, thread_group_size) = linear_split(&pipeline, (bh * td) / 2);
+    encoder.use_resource(src, metal::MTLResourceUsage::Read);
+    encoder.use_resource(cos, metal::MTLResourceUsage::Read);
+    encoder.use_resource(sin, metal::MTLResourceUsage::Read);
+    encoder.use_resource(output, metal::MTLResourceUsage::Write);
+    encoder.dispatch_thread_groups(thread_group_count, thread_group_size);
+    encoder.end_encoding();
+    Ok(())
+}
+
+#[allow(clippy::too_many_arguments)]
 pub fn call_affine(
     device: &Device,
     command_buffer: &CommandBufferRef,
diff --git a/candle-metal-kernels/src/reduce.metal b/candle-metal-kernels/src/reduce.metal
index 3c3cbc14..fa980dea 100644
--- a/candle-metal-kernels/src/reduce.metal
+++ b/candle-metal-kernels/src/reduce.metal
@@ -313,6 +313,26 @@ kernel void NAME(
     }                                                                             \
 }                                                                                 \
 
+#define ROPEI(FN_NAME, TYPENAME) \
+kernel void FN_NAME( \
+    constant size_t &bh, \
+    constant size_t &td, \
+    device const TYPENAME *src,  \
+    device const TYPENAME *cos,  \
+    device const TYPENAME *sin,  \
+    device TYPENAME *dst, \
+    uint tid [[ thread_position_in_grid ]] \
+) { \
+    if (2 * tid >= bh * td) { \
+        return; \
+    } \
+    size_t rope_idx = tid % (td / 2); \
+    TYPENAME c = cos[rope_idx]; \
+    TYPENAME s = sin[rope_idx]; \
+    dst[2 * tid] = src[2 * tid] * c - src[2 * tid + 1] * s; \
+    dst[2 * tid + 1] = src[2 * tid] * s + src[2 * tid + 1] * c; \
+}\
+
 REDUCE(x + y, fast_sum_f32_strided, float, 0)
 REDUCE(x + y, fast_sum_u32_strided, uint, 0)
 REDUCE(x + y, fast_sum_f16_strided, half, 0)
@@ -341,6 +361,8 @@ SOFTMAX(softmax_f32, float)
 SOFTMAX(softmax_f16, half)
 RMSNORM(rmsnorm_f32, float)
 RMSNORM(rmsnorm_f16, half)
+ROPEI(rope_i_f32, float)
+ROPEI(rope_i_f16, half)
 
 #if __METAL_VERSION__ >= 220
 REDUCE(x + y, fast_sum_i64_strided, int64_t, 0)
@@ -359,4 +381,5 @@ ARGMIN(fast_argmin_bf16, bfloat, HUGE_VALBF)
 ARGMAX(fast_argmax_bf16, bfloat, -HUGE_VALBF)
 SOFTMAX(softmax_bf16, bfloat)
 RMSNORM(rmsnorm_bf16, bfloat)
+ROPEI(rope_i_bf16, bfloat)
 #endif