CUDA: add bf16 and f32 support to cublas_mul_mat_batched (#14361)

* CUDA: add bf16 and f32 support to cublas_mul_mat_batched

* Review: add type traits and make function more generic

* Review: make check more explicit, add back comments, and fix formatting

* Review: fix formatting, remove useless type conversion, fix naming for bools

ggml/src/ggml-cuda/convert.cu

@@ -728,3 +728,25 @@ to_fp16_nc_cuda_t ggml_get_to_fp16_nc_cuda(ggml_type type) {
             return nullptr;
     }
 }
+
+to_bf16_nc_cuda_t ggml_get_to_bf16_nc_cuda(ggml_type type) {
+    switch (type) {
+        case GGML_TYPE_F32:
+            return convert_unary_cuda<float, nv_bfloat16>;
+        case GGML_TYPE_F16:
+            return convert_unary_cuda<half, nv_bfloat16>;
+        default:
+            return nullptr;
+    }
+}
+
+to_fp32_nc_cuda_t ggml_get_to_fp32_nc_cuda(ggml_type type) {
+    switch (type) {
+        case GGML_TYPE_F16:
+            return convert_unary_cuda<half, float>;
+        case GGML_TYPE_BF16:
+            return convert_unary_cuda<nv_bfloat16, float>;
+        default:
+            return nullptr;
+    }
+}

ggml/src/ggml-cuda/convert.cuh

@@ -22,5 +22,10 @@ using to_t_nc_cuda_t = void (*)(const void * x, T * y,
     int64_t ne00, int64_t ne01, int64_t ne02, int64_t ne03,
     int64_t s01, int64_t s02, int64_t s03, cudaStream_t stream);
 
+typedef to_t_nc_cuda_t<float> to_fp32_nc_cuda_t;
 typedef to_t_nc_cuda_t<half> to_fp16_nc_cuda_t;
+typedef to_t_nc_cuda_t<nv_bfloat16> to_bf16_nc_cuda_t;
+
+to_fp32_nc_cuda_t ggml_get_to_fp32_nc_cuda(ggml_type type);
 to_fp16_nc_cuda_t ggml_get_to_fp16_nc_cuda(ggml_type type);
+to_bf16_nc_cuda_t ggml_get_to_bf16_nc_cuda(ggml_type type);

ggml/src/ggml-cuda/ggml-cuda.cu

@@ -1749,7 +1749,7 @@ static void ggml_cuda_op_mul_mat(
 }
 
 static __global__ void k_compute_batched_ptrs(
-        const half * src0_as_f16, const half * src1_as_f16, char * dst,
+        const void * src0_as_f16, const void * src1_as_f16, char * dst,
         const void ** ptrs_src, void ** ptrs_dst,
         int64_t ne12, int64_t ne13,
         int64_t ne23,
@@ -1772,83 +1772,131 @@ static __global__ void k_compute_batched_ptrs(
     ptrs_dst[0*ne23 + i12 + i13*ne12] = (      char *)         dst + i12*nbd2 + i13*nbd3;
 }
 
-static void ggml_cuda_mul_mat_batched_cublas(ggml_backend_cuda_context & ctx, const ggml_tensor * src0, const ggml_tensor * src1, ggml_tensor * dst) {
+// Type traits for mapping ggml types to CUDA/cuBLAS types
+template<ggml_type T>
+struct batched_mul_mat_traits;
+
+template<>
+struct batched_mul_mat_traits<GGML_TYPE_F32> {
+    using cuda_type = float;
+    static inline const cublasComputeType_t compute_type = CUBLAS_COMPUTE_32F;
+    static inline const cudaDataType_t data_type = CUDA_R_32F;
+    static inline const ggml_type ggml_type_val = GGML_TYPE_F32;
+    static inline const float alpha = 1.0f;
+    static inline const float beta = 0.0f;
+    static inline const void* get_alpha() { static const float val = alpha; return &val; }
+    static inline const void* get_beta() { static const float val = beta; return &val; }
+    static inline auto get_nc_converter(ggml_type src_type) { return ggml_get_to_fp32_nc_cuda(src_type); }
+};
+
+template<>
+struct batched_mul_mat_traits<GGML_TYPE_BF16> {
+    using cuda_type = nv_bfloat16;
+    static inline const cublasComputeType_t compute_type = CUBLAS_COMPUTE_32F;
+    static inline const cudaDataType_t data_type = CUDA_R_16BF;
+    static inline const ggml_type ggml_type_val = GGML_TYPE_BF16;
+    static inline const float alpha = 1.0f;
+    static inline const float beta = 0.0f;
+    static inline const void* get_alpha() { static const float val = alpha; return &val; }
+    static inline const void* get_beta() { static const float val = beta; return &val; }
+    static inline auto get_nc_converter(ggml_type src_type) { return ggml_get_to_bf16_nc_cuda(src_type); }
+};
+
+template<>
+struct batched_mul_mat_traits<GGML_TYPE_F16> {
+    using cuda_type = half;
+    static inline const cublasComputeType_t compute_type = CUBLAS_COMPUTE_16F;
+    static inline const cudaDataType_t data_type = CUDA_R_16F;
+    static inline const ggml_type ggml_type_val = GGML_TYPE_F16;
+    static inline const half alpha = 1.0;
+    static inline const half beta = 0.0;
+    static inline const void* get_alpha() { static const half val = alpha; return &val; }
+    static inline const void* get_beta() { static const half val = beta; return &val; }
+    static inline auto get_nc_converter(ggml_type src_type) { return ggml_get_to_fp16_nc_cuda(src_type); }
+};
+
+template<ggml_type src0_type>
+static void ggml_cuda_mul_mat_batched_cublas_impl(ggml_backend_cuda_context & ctx, const ggml_tensor * src0, const ggml_tensor * src1, ggml_tensor * dst) {
+    using traits = batched_mul_mat_traits<src0_type>;
+    using cuda_t = typename traits::cuda_type;
+
     GGML_ASSERT(!ggml_is_transposed(src0));
     GGML_ASSERT(!ggml_is_transposed(src1));
-
     GGML_ASSERT(!ggml_backend_buft_is_cuda_split(src0->buffer->buft));
-    GGML_ASSERT(src0->type == GGML_TYPE_F16);
+    GGML_ASSERT(src0->type == src0_type);
+    GGML_ASSERT(ggml_is_contiguous(dst));
 
     // Byte offsets and tensor dimensions are currently used in an inconsistent way for dst.
     // As long as dst is contiguous this does not matter though.
-    GGML_ASSERT(ggml_is_contiguous(dst));
 
     GGML_TENSOR_BINARY_OP_LOCALS
 
     const int64_t ne_dst = ggml_nelements(dst);
-
     cudaStream_t main_stream = ctx.stream();
-
     CUBLAS_CHECK(cublasSetStream(ctx.cublas_handle(), main_stream));
 
-    const half * src0_f16 = (const half *) src0->data;
     float * dst_ddf = (float *) dst->data;
-
-    const half * src1_f16 = (const half *) src1->data;
     const size_t ts_src1 = ggml_type_size(src1->type);
     GGML_ASSERT(nb10 == ts_src1);
     int64_t s11 = nb11 / ts_src1;
     int64_t s12 = nb12 / ts_src1;
     int64_t s13 = nb13 / ts_src1;
-    ggml_cuda_pool_alloc<half> src1_f16_alloc(ctx.pool());
 
-    // convert src1 to fp16
+    const cuda_t * src0_ptr = nullptr;
-    if (src1->type != GGML_TYPE_F16) {
+    const cuda_t * src1_ptr = nullptr;
-        const to_fp16_nc_cuda_t to_fp16_cuda = ggml_get_to_fp16_nc_cuda(src1->type);
+
+    ggml_cuda_pool_alloc<cuda_t> src0_alloc(ctx.pool());
+    ggml_cuda_pool_alloc<cuda_t> src1_alloc(ctx.pool());
+
+    // Handle src0
+    src0_ptr = (const cuda_t *) src0->data;
+
+    // Handle src1 - convert if necessary
+    if (src1->type == src0_type) {
+        src1_ptr = (const cuda_t *) src1->data;
+    } else {
+        // Convert src1 to target type using traits conversion functions
         const int64_t ne_src1 = ggml_nelements(src1);
-        src1_f16_alloc.alloc(ne_src1);
+        src1_alloc.alloc(ne_src1);
-        GGML_ASSERT(to_fp16_cuda != nullptr);
 
-        to_fp16_cuda(src1_f16, src1_f16_alloc.get(), ne10, ne11, ne12, ne13, s11, s12, s13, main_stream);
+        const auto convert_func = traits::get_nc_converter(src1->type);
-
+        GGML_ASSERT(convert_func != nullptr);
-        src1_f16 = src1_f16_alloc.get();
+        convert_func(src1->data, src1_alloc.get(), ne10, ne11, ne12, ne13, s11, s12, s13, main_stream);
+        src1_ptr = src1_alloc.get();
         s11 = ne10;
         s12 = ne11*s11;
         s13 = ne12*s12;
     }
 
-    ggml_cuda_pool_alloc<half> dst_f16(ctx.pool());
+    // Setup destination buffer
+    ggml_cuda_pool_alloc<cuda_t> dst_temp(ctx.pool());
     char * dst_t;
-
-    cublasComputeType_t cu_compute_type = CUBLAS_COMPUTE_16F;
-    cudaDataType_t      cu_data_type    = CUDA_R_16F;
-
-    // dst strides
     size_t nbd2 = dst->nb[2];
     size_t nbd3 = dst->nb[3];
 
-    const half  alpha_f16 = 1.0f;
+    cublasComputeType_t cu_compute_type = traits::compute_type;
-    const half  beta_f16  = 0.0f;
+    cudaDataType_t cu_data_type = traits::data_type;
-
+    cudaDataType_t cu_data_type_a = traits::data_type;
+    cudaDataType_t cu_data_type_b = traits::data_type;
+    const void * alpha = traits::get_alpha();
+    const void * beta = traits::get_beta();
     const float alpha_f32 = 1.0f;
-    const float beta_f32  = 0.0f;
+    const float beta_f32 = 0.0f;
-
-    const void * alpha = &alpha_f16;
-    const void * beta  = &beta_f16;
 
     if (dst->op_params[0] == GGML_PREC_DEFAULT) {
-        dst_t = (char *) dst_f16.alloc(ne_dst);
+        if constexpr (src0_type == GGML_TYPE_F32) {
-
+            dst_t = (char *) dst_ddf;  // Direct F32 output
-        nbd2 /= sizeof(float) / sizeof(half);
+        } else {
-        nbd3 /= sizeof(float) / sizeof(half);
+            dst_t = (char *) dst_temp.alloc(ne_dst);
+            nbd2 /= sizeof(float) / sizeof(cuda_t);
+            nbd3 /= sizeof(float) / sizeof(cuda_t);
+        }
     } else {
         dst_t = (char *) dst_ddf;
-
         cu_compute_type = CUBLAS_COMPUTE_32F;
-        cu_data_type    = CUDA_R_32F;
+        cu_data_type = CUDA_R_32F;
-
         alpha = &alpha_f32;
-        beta  = &beta_f32;
+        beta = &beta_f32;
     }
 
     int id = ggml_cuda_get_device();
@@ -1856,7 +1904,7 @@ static void ggml_cuda_mul_mat_batched_cublas(ggml_backend_cuda_context & ctx, co
     if (GGML_CUDA_CC_IS_CDNA(cc) || GGML_CUDA_CC_IS_RDNA4(cc)) {
         cu_compute_type = CUBLAS_COMPUTE_32F;
         alpha = &alpha_f32;
-        beta  = &beta_f32;
+        beta = &beta_f32;
     }
 
     GGML_ASSERT(ne12 % ne02 == 0);
@@ -1866,35 +1914,15 @@ static void ggml_cuda_mul_mat_batched_cublas(ggml_backend_cuda_context & ctx, co
     const int64_t r2 = ne12/ne02;
     const int64_t r3 = ne13/ne03;
 
-#if 0
-    // use cublasGemmEx
-    {
-        for (int i13 = 0; i13 < ne13; ++i13) {
-            for (int i12 = 0; i12 < ne12; ++i12) {
-                int i03 = i13 / r3;
-                int i02 = i12 / r2;
-
-                CUBLAS_CHECK(
-                cublasGemmEx(ctx.cublas_handle(), CUBLAS_OP_T, CUBLAS_OP_N,
-                    ne01, ne11, ne10,
-                    alpha, (const char *) src0_f16 + i03*nb03 + i02*nb02, CUDA_R_16F,   nb01/sizeof(half),
-                                          src1_f16 + i13*s13  + i12*s12,  CUDA_R_16F,   s11,
-                    beta,  (      char *)    dst_t + i13*nbd3 + i12*nbd2, cu_data_type, ne0,
-                    cu_compute_type,
-                    CUBLAS_GEMM_DEFAULT_TENSOR_OP));
-            }
-        }
-    }
-#else
     if (r2 == 1 && r3 == 1 && ggml_is_contiguous_2(src0) && ggml_is_contiguous_2(src1)) {
         // there is no broadcast and src0, src1 are contiguous across dims 2, 3
         // use cublasGemmStridedBatchedEx
         CUBLAS_CHECK(
         cublasGemmStridedBatchedEx(ctx.cublas_handle(), CUBLAS_OP_T, CUBLAS_OP_N,
                 ne01, ne11, ne10,
-                alpha, src0_f16, CUDA_R_16F,   nb01/nb00, nb02/nb00, // strideA
+                alpha, src0_ptr, cu_data_type_a, nb01/nb00, nb02/nb00, // strideA
-                       src1_f16, CUDA_R_16F,   s11,       s12,       // strideB
+                       src1_ptr, cu_data_type_b, s11,       s12,       // strideB
-                beta,     dst_t, cu_data_type, ne0,       ne1*ne0,   // strideC
+                beta,     dst_t, cu_data_type,   ne0,       ne1*ne0,   // strideC
                 ne12*ne13,
                 cu_compute_type,
                 CUBLAS_GEMM_DEFAULT_TENSOR_OP));
@@ -1905,34 +1933,55 @@ static void ggml_cuda_mul_mat_batched_cublas(ggml_backend_cuda_context & ctx, co
         ggml_cuda_pool_alloc<const void *> ptrs_src(ctx.pool(), 2*ne23);
         ggml_cuda_pool_alloc<      void *> ptrs_dst(ctx.pool(), 1*ne23);
 
+        size_t src1_stride_size = sizeof(cuda_t);
+
         dim3 block_dims(ne13, ne12);
         k_compute_batched_ptrs<<<1, block_dims, 0, main_stream>>>(
-                src0_f16, src1_f16, dst_t,
+                src0_ptr, src1_ptr, dst_t,
                 ptrs_src.get(), ptrs_dst.get(),
                 ne12, ne13,
                 ne23,
                 nb02, nb03,
-                src1->type == GGML_TYPE_F16 ? nb12 : s12*sizeof(half),
+                (src1->type == src0_type) ? nb12 : s12*src1_stride_size,
-                src1->type == GGML_TYPE_F16 ? nb13 : s13*sizeof(half),
+                (src1->type == src0_type) ? nb13 : s13*src1_stride_size,
                 nbd2, nbd3,
                 r2, r3);
+
         CUDA_CHECK(cudaGetLastError());
 
         CUBLAS_CHECK(
         cublasGemmBatchedEx(ctx.cublas_handle(), CUBLAS_OP_T, CUBLAS_OP_N,
                 ne01, ne11, ne10,
-                alpha, (const void **) (ptrs_src.get() + 0*ne23), CUDA_R_16F,   nb01/nb00,
+                alpha, (const void **) (ptrs_src.get() + 0*ne23), cu_data_type_a, nb01/nb00,
-                       (const void **) (ptrs_src.get() + 1*ne23), CUDA_R_16F,   s11,
+                       (const void **) (ptrs_src.get() + 1*ne23), cu_data_type_b, s11,
-                beta,  (      void **) (ptrs_dst.get() + 0*ne23), cu_data_type, ne0,
+                beta,  (      void **) (ptrs_dst.get() + 0*ne23), cu_data_type,   ne0,
                 ne23,
                 cu_compute_type,
                 CUBLAS_GEMM_DEFAULT_TENSOR_OP));
     }
-#endif
 
-    if (dst->op_params[0] == GGML_PREC_DEFAULT && cu_data_type == CUDA_R_16F) {
+    // Convert output back to F32 if needed
-        const to_fp32_cuda_t to_fp32_cuda = ggml_get_to_fp32_cuda(GGML_TYPE_F16);
+    if (dst->op_params[0] == GGML_PREC_DEFAULT && cu_data_type != CUDA_R_32F) {
-        to_fp32_cuda(dst_f16.get(), dst_ddf, ne_dst, main_stream);
+        const to_fp32_cuda_t to_fp32_cuda = ggml_get_to_fp32_cuda(traits::ggml_type_val);
+        to_fp32_cuda(dst_temp.get(), dst_ddf, ne_dst, main_stream);
+    }
+}
+
+static void ggml_cuda_mul_mat_batched_cublas(ggml_backend_cuda_context & ctx, const ggml_tensor * src0, const ggml_tensor * src1, ggml_tensor * dst) {
+    GGML_ASSERT(src0->type == GGML_TYPE_F16 || src0->type == GGML_TYPE_BF16 || src0->type == GGML_TYPE_F32);
+
+    switch (src0->type) {
+        case GGML_TYPE_F32:
+            ggml_cuda_mul_mat_batched_cublas_impl<GGML_TYPE_F32>(ctx, src0, src1, dst);
+            break;
+        case GGML_TYPE_BF16:
+            ggml_cuda_mul_mat_batched_cublas_impl<GGML_TYPE_BF16>(ctx, src0, src1, dst);
+            break;
+        case GGML_TYPE_F16:
+            ggml_cuda_mul_mat_batched_cublas_impl<GGML_TYPE_F16>(ctx, src0, src1, dst);
+            break;
+        default:
+            GGML_ABORT("Unsupported type");
     }
 }
 
@@ -1984,6 +2033,12 @@ static void ggml_cuda_mul_mat(ggml_backend_cuda_context & ctx, const ggml_tensor
     //printf("src0 is contiguous %d, transposed %d, type = %s, name = %s\n", ggml_is_contiguous(src0), ggml_is_transposed(src0), ggml_type_name(src0->type), src0->name);
     //printf("src1 is contiguous %d, transposed %d, type = %s, name = %s\n", ggml_is_contiguous(src1), ggml_is_transposed(src1), ggml_type_name(src1->type), src1->name);
 
+    //TODO update for generic tensor parallelism
+    const int cc                     = ggml_cuda_info().devices[ggml_cuda_get_device()].cc;
+    bool use_batched_cublas_f16  = src0->type == GGML_TYPE_F16 && (src1->type == GGML_TYPE_F16 || !any_gpus_with_slow_fp16);
+    bool use_batched_cublas_bf16 = src0->type == GGML_TYPE_BF16 && bf16_mma_hardware_available(cc);
+    bool use_batched_cublas_f32  = src0->type == GGML_TYPE_F32;
+
     if (!split && use_mul_mat_vec) {
         // the custom F16 vector kernel can be used over batched cuBLAS GEMM
         // but this is only faster for GPUs without tensor cores or with a thin src0 matrix (particularly KQV in attention)
@@ -1992,8 +2047,8 @@ static void ggml_cuda_mul_mat(ggml_backend_cuda_context & ctx, const ggml_tensor
         ggml_cuda_mul_mat_vec_q(ctx, src0, src1, nullptr, dst);
     } else if (!split && use_mul_mat_q) {
         ggml_cuda_mul_mat_q(ctx, src0, src1, nullptr, dst);
-    } else if (!split && src0->type == GGML_TYPE_F16 && (src1->type == GGML_TYPE_F16 || !any_gpus_with_slow_fp16) &&
+    } else if (!split && (use_batched_cublas_f16 || use_batched_cublas_bf16 || use_batched_cublas_f32)
-            !ggml_is_transposed(src0) && !ggml_is_transposed(src1) && src1->ne[2]*src1->ne[3] > 1) {
+        && !ggml_is_transposed(src0) && !ggml_is_transposed(src1) && src1->ne[2]*src1->ne[3] > 1) {
         // general KQ + KQV multi-batch without FlashAttention
         ggml_cuda_mul_mat_batched_cublas(ctx, src0, src1, dst);
     } else if (use_mul_mat_vec) {

tests/test-backend-ops.cpp

@@ -4425,8 +4425,10 @@ static std::vector<std::unique_ptr<test_case>> make_test_cases_eval() {
         for (auto nr : {1,4}) {
             for (uint32_t m = 0; m < 2; ++m) {
                 for (uint32_t k = 0; k < 2; ++k) {
-                    test_cases.emplace_back(new test_mul_mat(GGML_TYPE_F16, GGML_TYPE_F32, 1056 + m, 1, 128 + k,  {bs,  1}, {nr, 1}, {0, 2, 1, 3}));
+                    for (ggml_type type: {GGML_TYPE_F16, GGML_TYPE_BF16, GGML_TYPE_F32}) {
-                    test_cases.emplace_back(new test_mul_mat(GGML_TYPE_F16, GGML_TYPE_F32, 128 + m,  1, 1056 + k, {bs,  1}, {nr, 1}, {0, 1, 2, 3}, true));
+                        test_cases.emplace_back(new test_mul_mat(type, GGML_TYPE_F32, 1056 + m, 1, 128 + k,  {bs,  1}, {nr, 1}, {0, 2, 1, 3}));
+                        test_cases.emplace_back(new test_mul_mat(type, GGML_TYPE_F32, 128 + m,  1, 1056 + k, {bs,  1}, {nr, 1}, {0, 1, 2, 3}, true));
+                    }
                 }
             }
         }