CUDA: quantized KV support for FA vec (llama/7527)

* CUDA: quantized KV support for FA vec

* try CI fix

* fix commented-out kernel variants

* add q8_0 q4_0 tests

* fix nwarps > batch size

* split fattn compile via extern templates

* fix flake8

* fix metal tests

* fix cmake

* make generate_cu_files.py executable

* add autogenerated .cu files

* fix AMD

* error if type_v != FP16 and not flash_attn

* remove obsolete code

ggml-cuda.cu

@@ -2905,10 +2905,14 @@ GGML_CALL static bool ggml_backend_cuda_supports_op(ggml_backend_t backend, cons
 #if defined(GGML_USE_HIPBLAS) && defined(__HIP_PLATFORM_AMD__)
             return op->src[0]->ne[0] == 64 || op->src[0]->ne[0] == 128;
 #else
-            if (op->src[0]->ne[0] == 64 || op->src[0]->ne[0] == 128) {
+            if (op->src[0]->ne[0] == 128) {
                 return true;
             }
-            return ggml_cuda_info().devices[cuda_ctx->device].cc >= CC_VOLTA;
+            if (op->src[0]->ne[0] ==  64 && op->src[1]->type == GGML_TYPE_F16) {
+                return true;
+            }
+            return ggml_cuda_info().devices[cuda_ctx->device].cc >= CC_VOLTA &&
+                op->src[1]->type == GGML_TYPE_F16 && op->src[2]->type == GGML_TYPE_F16;
 #endif // defined(GGML_USE_HIPBLAS) && defined(__HIP_PLATFORM_AMD__)
         default:
             return false;

ggml-cuda/fattn-common.cuh

@@ -1,4 +1,7 @@
+#pragma once
+
 #include "common.cuh"
+#include "vecdotq.cuh"
 
 #include <cstdint>
 
@@ -34,11 +37,523 @@ typedef void (* fattn_kernel_t)(
         const int nb11,
         const int nb12,
         const int nb13,
+        const int nb21,
+        const int nb22,
+        const int nb23,
         const int ne0,
         const int ne1,
         const int ne2,
         const int ne3);
 
+typedef half (*vec_dot_KQ_f16_t)(
+    const char * __restrict__ K_c, const void * __restrict__ Q_v, const int * __restrict__ Q_q8 , const void * __restrict__ Q_ds);
+typedef float (*vec_dot_KQ_f32_t)(
+    const char * __restrict__ K_c, const void * __restrict__ Q_v, const int * __restrict__ Q_q8 , const void * __restrict__ Q_ds);
+
+template<typename T, int D>
+static __device__ __forceinline__ T vec_dot_fattn_vec_KQ_q4_0(
+    const char * __restrict__ K_c, const void * __restrict__ Q_v, const int * __restrict__ Q_q8, const void * __restrict__ Q_ds_v) {
+#if __CUDA_ARCH__ > MIN_CC_DP4A
+
+    const block_q4_0 * K_q4_0 = (const block_q4_0 *) K_c;
+    GGML_UNUSED(Q_v);
+
+    half sum = 0.0f;
+
+#pragma unroll
+    for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) {
+        const int k_KQ = k_KQ_0 + threadIdx.x;
+
+        const int ib    = k_KQ /  QI8_1;
+        const int iqs4  = k_KQ %  QI4_0;
+        const int shift = k_KQ & (QI8_1/2);
+
+        const int v = (get_int_from_uint8(K_q4_0[ib].qs, iqs4) >> shift) & 0x0F0F0F0F;
+        const int u = Q_q8[k_KQ_0/WARP_SIZE];
+
+        const int sumi = __dp4a(v, u, 0);
+
+#if FP16_AVAILABLE
+        if (std::is_same<T, half>::value) {
+            const half2  * Q_ds = (const half2  *) Q_ds_v;
+
+            const half2 sum2 = __half2half2(K_q4_0[ib].d) * Q_ds[k_KQ_0/WARP_SIZE];
+            sum += (T) (((half) sumi)*__low2half(sum2) - __high2half(sum2) /* *8/QI8_1 == 1 */);
+        } else
+#endif // FP16_AVAILABLE
+        {
+            const float2 * Q_ds = (const float2 *) Q_ds_v;
+
+            sum += (T) (__half2float(K_q4_0[ib].d) * (sumi*Q_ds[k_KQ_0/WARP_SIZE].x - (8/QI8_1)*Q_ds[k_KQ_0/WARP_SIZE].y));
+        }
+    }
+
+    return sum;
+#else
+    GGML_UNUSED(K_c);
+    GGML_UNUSED(Q_v);
+    GGML_UNUSED(Q_q8);
+    GGML_UNUSED(Q_ds_v);
+    NO_DEVICE_CODE;
+#endif  // __CUDA_ARCH__ > MIN_CC_DP4A
+}
+
+template<typename T, int D>
+static __device__ __forceinline__ T vec_dot_fattn_vec_KQ_q4_1(
+    const char * __restrict__ K_c, const void * __restrict__ Q_v, const int * __restrict__ Q_q8, const void * __restrict__ Q_ds_v) {
+#if __CUDA_ARCH__ > MIN_CC_DP4A
+
+    const block_q4_1 * K_q4_1 = (const block_q4_1 *) K_c;
+    GGML_UNUSED(Q_v);
+
+    T sum = 0.0f;
+
+#pragma unroll
+    for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) {
+        const int k_KQ = k_KQ_0 + threadIdx.x;
+
+        const int ib    = k_KQ /  QI8_1;
+        const int iqs4  = k_KQ %  QI4_1;
+        const int shift = k_KQ & (QI8_1/2);
+
+        const int v = (get_int_from_uint8_aligned(K_q4_1[ib].qs, iqs4) >> shift) & 0x0F0F0F0F;
+        const int u = Q_q8[k_KQ_0/WARP_SIZE];
+
+        const int sumi = __dp4a(v, u, 0);
+
+#if FP16_AVAILABLE
+        if (std::is_same<T, half>::value) {
+            const half2  * Q_ds = (const half2  *) Q_ds_v;
+
+            const half2 d4d8_m4s8 = K_q4_1[ib].dm * Q_ds[k_KQ_0/WARP_SIZE];
+            const half2 sumid4d8_m4s8scaled = d4d8_m4s8 * make_half2(sumi, 1.0f/QI8_1);
+            sum += (T) (__low2half(sumid4d8_m4s8scaled) + __high2half(sumid4d8_m4s8scaled));
+        } else
+#endif // FP16_AVAILABLE
+        {
+            const float2 * Q_ds = (const float2 *) Q_ds_v;
+
+            const float sumid4d8   =  __low2float(K_q4_1[ib].dm)*Q_ds[k_KQ_0/WARP_SIZE].x * sumi;
+            const float m4s8scaled = __high2float(K_q4_1[ib].dm)*Q_ds[k_KQ_0/WARP_SIZE].y / QI8_1;
+
+            sum += (T) (sumid4d8 + m4s8scaled);
+        }
+    }
+
+    return sum;
+#else
+    GGML_UNUSED(K_c);
+    GGML_UNUSED(Q_v);
+    GGML_UNUSED(Q_q8);
+    GGML_UNUSED(Q_ds_v);
+    NO_DEVICE_CODE;
+#endif  // __CUDA_ARCH__ > MIN_CC_DP4A
+}
+
+template<typename T, int D>
+static __device__ __forceinline__ T vec_dot_fattn_vec_KQ_q5_0(
+    const char * __restrict__ K_c, const void * __restrict__ Q_v, const int * __restrict__ Q_q8, const void * __restrict__ Q_ds_v) {
+#if __CUDA_ARCH__ > MIN_CC_DP4A
+
+    const block_q5_0 * K_q5_0 = (const block_q5_0 *) K_c;
+    GGML_UNUSED(Q_v);
+
+    T sum = 0.0f;
+
+#pragma unroll
+    for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) {
+        const int k_KQ = k_KQ_0 + threadIdx.x;
+
+        const int ib    = k_KQ /  QI8_1;
+        const int iqs4  = k_KQ %  QI5_0;
+        const int iqs8  = k_KQ %  QI8_1;
+        const int shift = k_KQ & (QI8_1/2);
+
+        int v = (get_int_from_uint8(K_q5_0[ib].qs, iqs4) >> shift) & 0x0F0F0F0F;
+        const int vh = get_int_from_uint8(K_q5_0[ib].qh, 0) >> (iqs8 * QI5_0);
+        v |= (vh <<  4) & 0x00000010; // 0 ->  4
+        v |= (vh << 11) & 0x00001000; // 1 -> 12
+        v |= (vh << 18) & 0x00100000; // 2 -> 20
+        v |= (vh << 25) & 0x10000000; // 3 -> 28
+
+        const int u = Q_q8[k_KQ_0/WARP_SIZE];
+
+        const int sumi = __dp4a(v, u, 0);
+
+#if FP16_AVAILABLE
+        if (std::is_same<T, half>::value) {
+            const half2  * Q_ds = (const half2  *) Q_ds_v;
+
+            const half2 sum2 = __half2half2(K_q5_0[ib].d) * Q_ds[k_KQ_0/WARP_SIZE];
+            sum += (T) (((half) sumi)*__low2half(sum2) - __high2half(sum2)*__float2half(2.0f)) /* *16/QI8_1 == 2 */;
+        } else
+#endif // FP16_AVAILABLE
+        {
+            const float2 * Q_ds = (const float2 *) Q_ds_v;
+
+            sum += (T) (__half2float(K_q5_0[ib].d) * (sumi*Q_ds[k_KQ_0/WARP_SIZE].x - (16/QI8_1)*Q_ds[k_KQ_0/WARP_SIZE].y));
+        }
+    }
+
+    return sum;
+#else
+    GGML_UNUSED(K_c);
+    GGML_UNUSED(Q_v);
+    GGML_UNUSED(Q_q8);
+    GGML_UNUSED(Q_ds_v);
+    NO_DEVICE_CODE;
+#endif  // __CUDA_ARCH__ > MIN_CC_DP4A
+}
+
+template<typename T, int D>
+static __device__ __forceinline__ T vec_dot_fattn_vec_KQ_q5_1(
+    const char * __restrict__ K_c, const void * __restrict__ Q_v, const int * __restrict__ Q_q8, const void * __restrict__ Q_ds_v) {
+#if __CUDA_ARCH__ > MIN_CC_DP4A
+
+    const block_q5_1 * K_q5_1 = (const block_q5_1 *) K_c;
+    GGML_UNUSED(Q_v);
+
+    T sum = 0.0f;
+
+#pragma unroll
+    for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) {
+        const int k_KQ = k_KQ_0 + threadIdx.x;
+
+        const int ib    = k_KQ /  QI8_1;
+        const int iqs4  = k_KQ %  QI5_1;
+        const int iqs8  = k_KQ %  QI8_1;
+        const int shift = k_KQ & (QI8_1/2);
+
+        int v = (get_int_from_uint8(K_q5_1[ib].qs, iqs4) >> shift) & 0x0F0F0F0F;
+        const int vh = get_int_from_uint8(K_q5_1[ib].qh, 0) >> (iqs8 * QI5_1);
+        v |= (vh <<  4) & 0x00000010; // 0 ->  4
+        v |= (vh << 11) & 0x00001000; // 1 -> 12
+        v |= (vh << 18) & 0x00100000; // 2 -> 20
+        v |= (vh << 25) & 0x10000000; // 3 -> 28
+
+        const int u = Q_q8[k_KQ_0/WARP_SIZE];
+
+        const int sumi = __dp4a(v, u, 0);
+
+#if FP16_AVAILABLE
+        if (std::is_same<T, half>::value) {
+            const half2  * Q_ds = (const half2  *) Q_ds_v;
+
+            const half2 d5d8_m5s8 = K_q5_1[ib].dm * Q_ds[k_KQ_0/WARP_SIZE];
+            const half2 sumid5d8_m5s8scaled = d5d8_m5s8 * make_half2(sumi, 1.0f/QI8_1);
+            sum += (T) (__low2half(sumid5d8_m5s8scaled) + __high2half(sumid5d8_m5s8scaled));
+        } else
+#endif // FP16_AVAILABLE
+        {
+            const float2 * Q_ds = (const float2 *) Q_ds_v;
+
+            const float sumid5d8   =  __low2float(K_q5_1[ib].dm)*Q_ds[k_KQ_0/WARP_SIZE].x * sumi;
+            const float m5s8scaled = __high2float(K_q5_1[ib].dm)*Q_ds[k_KQ_0/WARP_SIZE].y / QI8_1;
+
+            sum += (T) (sumid5d8 + m5s8scaled);
+        }
+    }
+
+    return sum;
+#else
+    GGML_UNUSED(K_c);
+    GGML_UNUSED(Q_v);
+    GGML_UNUSED(Q_q8);
+    GGML_UNUSED(Q_ds_v);
+    NO_DEVICE_CODE;
+#endif  // __CUDA_ARCH__ > MIN_CC_DP4A
+}
+
+template <typename T, int D>
+static __device__ __forceinline__ T vec_dot_fattn_vec_KQ_q8_0(
+    const char * __restrict__ K_c, const void * __restrict__ Q_v, const int * __restrict__ Q_q8, const void * __restrict__ Q_ds_v) {
+#if __CUDA_ARCH__ > MIN_CC_DP4A
+
+    const block_q8_0 * K_q8_0 = (const block_q8_0 *) K_c;
+    GGML_UNUSED(Q_v);
+
+    T sum = 0.0f;
+
+#pragma unroll
+    for (int k_KQ_0 = 0; k_KQ_0 < D/sizeof(int); k_KQ_0 += WARP_SIZE) {
+        const int k_KQ = k_KQ_0 + threadIdx.x;
+
+        const int ib  = k_KQ / QI8_0;
+        const int iqs = k_KQ % QI8_0;
+
+        const int v = get_int_from_int8(K_q8_0[ib].qs, iqs);
+
+        T Q_d;
+        if (std::is_same<T, half>::value) {
+            const half2  * Q_ds = (const half2  *) Q_ds_v;
+            Q_d = __low2half(Q_ds[k_KQ_0/WARP_SIZE]);
+        } else {
+            const float2 * Q_ds = (const float2 *) Q_ds_v;
+            Q_d = Q_ds[k_KQ_0/WARP_SIZE].x;
+        }
+
+        sum += vec_dot_q8_0_q8_1_impl<T, 1>(&v, &Q_q8[k_KQ_0/WARP_SIZE], K_q8_0[ib].d, Q_d);
+    }
+
+    return sum;
+#else
+    GGML_UNUSED(K_c);
+    GGML_UNUSED(Q_v);
+    GGML_UNUSED(Q_q8);
+    GGML_UNUSED(Q_ds_v);
+    NO_DEVICE_CODE;
+#endif  // __CUDA_ARCH__ > MIN_CC_DP4A
+}
+
+template <typename T, int D>
+static __device__ __forceinline__ T vec_dot_fattn_vec_KQ_f16(
+    const char * __restrict__ K_c, const void * __restrict__ Q_v, const int * __restrict__ Q_q8 , const void * __restrict__ Q_ds_v) {
+
+    const half2 * K_h2 = (const half2 *) K_c;
+    GGML_UNUSED(Q_q8);
+    GGML_UNUSED(Q_ds_v);
+
+#if FP16_AVAILABLE
+    if (std::is_same<T, half>::value) {
+        const half2 * Q_h2 = (const half2 *) Q_v;
+
+        half2 sum2 = make_half2(0.0f, 0.0f);
+
+#pragma unroll
+        for (int k_KQ_0 = 0; k_KQ_0 < D/2; k_KQ_0 += WARP_SIZE) {
+            const int k_KQ = k_KQ_0 + threadIdx.x;
+
+            const half2 K_ik = K_h2[k_KQ];
+            sum2 += K_ik * Q_h2[k_KQ_0/WARP_SIZE];
+        }
+
+        return __low2half(sum2) + __high2half(sum2);
+    }
+#endif // FP16_AVAILABLE
+
+    const float2 * Q_f2 = (const float2 *) Q_v;
+
+    float sum = 0.0f;
+
+#pragma unroll
+    for (int k_KQ_0 = 0; k_KQ_0 < D/2; k_KQ_0 += WARP_SIZE) {
+        const int k_KQ = k_KQ_0 + threadIdx.x;
+
+        const half2 K_ik = K_h2[k_KQ];
+        sum +=  __low2float(K_ik) * Q_f2[k_KQ_0/WARP_SIZE].x;
+        sum += __high2float(K_ik) * Q_f2[k_KQ_0/WARP_SIZE].y;
+    }
+
+    return sum;
+}
+
+template <typename Tds>
+static __device__ __forceinline__ void quantize_q8_1_to_shared(
+    const float * __restrict__ x, const float scale, int * __restrict__ yq32, void * __restrict__ yds) {
+
+    float vals[sizeof(int)] = {0.0f};
+#pragma unroll
+    for (int l = 0; l < sizeof(int); ++l) {
+        vals[l] = scale * x[4*threadIdx.x + l];
+    }
+
+    float amax = fabsf(vals[0]);
+    float sum  = vals[0];
+#pragma unroll
+    for (int l = 1; l < sizeof(int); ++l) {
+        amax = fmaxf(amax, fabsf(vals[l]));
+        sum += vals[l];
+    }
+#pragma unroll
+    for (int mask = QI8_1/2; mask > 0; mask >>= 1) {
+        amax = fmaxf(amax, __shfl_xor_sync(0xFFFFFFFF, amax, mask, 32));
+        sum +=             __shfl_xor_sync(0xFFFFFFFF, sum,  mask, 32);
+    }
+
+    const float d = amax / 127;
+    int q32 = 0;
+    int8_t * q8 = (int8_t *) &q32;
+
+    if (d != 0.0f) {
+#pragma unroll
+        for (int l = 0; l < sizeof(int); ++l) {
+            q8[l] = roundf(vals[l] / d);
+        }
+    }
+
+    yq32[threadIdx.x] = q32;
+    if (threadIdx.x % QI8_1 == 0) {
+        if (std::is_same<Tds, half2>::value) {
+            ((half2  *) yds)[threadIdx.x/QI8_1] =  make_half2(d, sum);
+        } else {
+            ((float2 *) yds)[threadIdx.x/QI8_1] = make_float2(d, sum);
+        }
+    }
+}
+
+typedef half  (*dequantize_1_f16_t)(const void *, const int64_t);
+typedef float (*dequantize_1_f32_t)(const void *, const int64_t);
+
+template <typename T>
+static __device__ __forceinline__ T dequantize_1_q4_0(const void * __restrict__ vx, const int64_t i) {
+    const block_q4_0 * x = (const block_q4_0 *) vx;
+
+    const int64_t ib    =  i          /  QK4_0;
+    const int     iqs   =  i          % (QK4_0/2);
+    const int     shift = (i % QK4_0) / (QK4_0/2);
+
+    const T   d  = x[ib].d;
+    const int q0 = x[ib].qs[iqs];
+    const int q  = ((q0 >> (4*shift)) & 0x0F) - 8;
+
+#if FP16_AVAILABLE
+    if (std::is_same<T, half>::value) {
+        return ((half) d)*((half) q);
+    }
+#endif // FP16_AVAILABLE
+
+    return ((float) d)*((float) q);
+}
+
+template <typename T>
+static __device__ __forceinline__ T dequantize_1_q4_1(const void * __restrict__ vx, const int64_t i) {
+    const block_q4_1 * x = (const block_q4_1 *) vx;
+
+    const int64_t ib    =  i          /  QK4_1;
+    const int     iqs   =  i          % (QK4_1/2);
+    const int     shift = (i % QK4_1) / (QK4_1/2);
+
+    const half2 dm = x[ib].dm;
+    const int   q0 = x[ib].qs[iqs];
+    const int   q  = ((q0 >> (4*shift)) & 0x0F);
+
+#if FP16_AVAILABLE
+    if (std::is_same<T, half>::value) {
+        return __low2half(dm)*((half) q) + __high2half(dm);
+    }
+#endif // FP16_AVAILABLE
+
+    return __low2float(dm)*((float) q) + __high2float(dm);
+}
+
+template <typename T>
+static __device__ __forceinline__ T dequantize_1_q5_0(const void * __restrict__ vx, const int64_t i) {
+    const block_q5_0 * x = (const block_q5_0 *) vx;
+
+    const int64_t ib    =  i          /  QK5_0;
+    const int     idq   =  i          %  QK5_0;
+    const int     iqs   =  i          % (QK5_0/2);
+    const int     shift = (i % QK5_0) / (QK5_0/2);
+
+    const T   d   = x[ib].d;
+    const int ql0 = x[ib].qs[iqs];
+    const int qh0 = get_int_from_uint8(x[ib].qh, 0);
+    const int ql  = ((ql0 >> (4*shift)) & 0x0F);
+    const int qh  = ((qh0 >> idq) << 4) & 0x10;
+    const int q   = (ql | qh) - 16;
+
+#if FP16_AVAILABLE
+    if (std::is_same<T, half>::value) {
+        return ((half) d)*((half) q);
+    }
+#endif // FP16_AVAILABLE
+
+    return ((float) d)*((float) q);
+}
+
+template <typename T>
+static __device__ __forceinline__ T dequantize_1_q5_1(const void * __restrict__ vx, const int64_t i) {
+    const block_q5_1 * x = (const block_q5_1 *) vx;
+
+    const int64_t ib    =  i          /  QK5_1;
+    const int     idq   =  i          %  QK5_1;
+    const int     iqs   =  i          % (QK5_1/2);
+    const int     shift = (i % QK5_1) / (QK5_1/2);
+
+    const half2 dm  = x[ib].dm;
+    const int   ql0 = x[ib].qs[iqs];
+    const int   qh0 = get_int_from_uint8_aligned(x[ib].qh, 0);
+    const int   ql  = ((ql0 >> (4*shift)) & 0x0F);
+    const int   qh  = ((qh0 >> idq) << 4) & 0x10;
+    const int   q   = (ql | qh);
+
+#if FP16_AVAILABLE
+    if (std::is_same<T, half>::value) {
+        return __low2half(dm)*((half) q) + __high2half(dm);
+    }
+#endif // FP16_AVAILABLE
+
+    return __low2float(dm)*((float) q) + __high2float(dm);
+}
+
+template <typename T>
+static __device__ __forceinline__ T dequantize_1_q8_0(const void * __restrict__ vx, const int64_t i) {
+    const block_q8_0 * x = (const block_q8_0 *) vx;
+
+    const int64_t ib  = i / QK8_0;
+    const int     iqs = i % QK8_0;
+
+    const T   d = x[ib].d;
+    const int q = x[ib].qs[iqs];
+
+#if FP16_AVAILABLE
+    if (std::is_same<T, half>::value) {
+        return ((half) d)*((half) q);
+    }
+#endif // FP16_AVAILABLE
+
+    return ((float) d)*((float) q);
+}
+
+template <typename T>
+static __device__ __forceinline__ T dequantize_1_f16(const void * __restrict__ vx, const int64_t i) {
+    const half * x = (const half *) vx;
+
+    return x[i];
+}
+
+template <int D>
+constexpr __device__ vec_dot_KQ_f16_t get_vec_dot_KQ_f16(ggml_type type_K) {
+    return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0<half, D> :
+        type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1<half, D> :
+        type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0<half, D> :
+        type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1<half, D> :
+        type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0<half, D> :
+        type_K == GGML_TYPE_F16 ? vec_dot_fattn_vec_KQ_f16<half, D> :
+        nullptr;
+}
+
+template <int D>
+constexpr __device__ vec_dot_KQ_f32_t get_vec_dot_KQ_f32(ggml_type type_K) {
+    return type_K == GGML_TYPE_Q4_0 ? vec_dot_fattn_vec_KQ_q4_0<float, D> :
+        type_K == GGML_TYPE_Q4_1 ? vec_dot_fattn_vec_KQ_q4_1<float, D> :
+        type_K == GGML_TYPE_Q5_0 ? vec_dot_fattn_vec_KQ_q5_0<float, D> :
+        type_K == GGML_TYPE_Q5_1 ? vec_dot_fattn_vec_KQ_q5_1<float, D> :
+        type_K == GGML_TYPE_Q8_0 ? vec_dot_fattn_vec_KQ_q8_0<float, D> :
+        type_K == GGML_TYPE_F16 ? vec_dot_fattn_vec_KQ_f16<float, D> :
+        nullptr;
+}
+
+constexpr __device__ dequantize_1_f16_t get_dequantize_1_f16(ggml_type type_V) {
+    return type_V == GGML_TYPE_Q4_0 ? dequantize_1_q4_0<half> :
+        type_V == GGML_TYPE_Q4_1 ? dequantize_1_q4_1<half> :
+        type_V == GGML_TYPE_Q5_0 ? dequantize_1_q5_0<half> :
+        type_V == GGML_TYPE_Q5_1 ? dequantize_1_q5_1<half> :
+        type_V == GGML_TYPE_Q8_0 ? dequantize_1_q8_0<half> :
+        type_V == GGML_TYPE_F16 ? dequantize_1_f16<half> :
+        nullptr;
+}
+
+constexpr __device__ dequantize_1_f32_t get_dequantize_1_f32(ggml_type type_V) {
+    return type_V == GGML_TYPE_Q4_0 ? dequantize_1_q4_0<float> :
+        type_V == GGML_TYPE_Q4_1 ? dequantize_1_q4_1<float> :
+        type_V == GGML_TYPE_Q5_0 ? dequantize_1_q5_0<float> :
+        type_V == GGML_TYPE_Q5_1 ? dequantize_1_q5_1<float> :
+        type_V == GGML_TYPE_Q8_0 ? dequantize_1_q8_0<float> :
+        type_V == GGML_TYPE_F16 ? dequantize_1_f16<float> :
+        nullptr;
+}
+
 template<int D, int parallel_blocks> // D == head size
 #if !(defined(GGML_USE_HIPBLAS) && defined(__HIP_PLATFORM_AMD__))
 __launch_bounds__(D, 1)
@@ -83,6 +598,27 @@ static __global__ void flash_attn_combine_results(
     dst[blockIdx.y*D + tid] = VKQ_numerator / VKQ_denominator;
 }
 
+static void on_no_fattn_vec_case(const int D) {
+    if (D == 64) {
+        fprintf(stderr, "Unsupported KV type combination for head_size 64.\n");
+        fprintf(stderr, "By default only f16 KV cache is supported.\n");
+        fprintf(stderr, "Compile with LLAMA_CUDA_FA_ALL_QUANTS for V cache quantization support.\n");
+        GGML_ASSERT(false);
+    } else if (D == 128) {
+        fprintf(stderr, "Unsupported KV type combination for head_size 128.\n");
+        fprintf(stderr, "Supported combinations:\n");
+        fprintf(stderr, "  - K == q4_0, V == q4_0,  4.50 BPV\n");
+        fprintf(stderr, "  - K == q8_0, V == q8_0,  8.50 BPV\n");
+        fprintf(stderr, "  - K == f16,  V == f16,  16.00 BPV\n");
+        fprintf(stderr, "Compile with LLAMA_CUDA_FA_ALL_QUANTS for all combinations of q4_0, q4_1, q5_0, q5_1, q8_0, and f16.\n");
+        GGML_ASSERT(false);
+    } else {
+        fprintf(stderr, "Unsupported KV type combination for head_size 256.\n");
+        fprintf(stderr, "Only f16 is supported.\n");
+        GGML_ASSERT(false);
+    }
+}
+
 template <int D, int parallel_blocks>
 void launch_fattn(ggml_backend_cuda_context & ctx, ggml_tensor * dst, fattn_kernel_t fattn_kernel, int nwarps, int cols_per_block) {
     const ggml_tensor * Q = dst->src[0];
@@ -94,8 +630,6 @@ void launch_fattn(ggml_backend_cuda_context & ctx, ggml_tensor * dst, fattn_kern
     ggml_tensor * KQV = dst;
 
     GGML_ASSERT(Q->type == GGML_TYPE_F32);
-    GGML_ASSERT(K->type == GGML_TYPE_F16);
-    GGML_ASSERT(V->type == GGML_TYPE_F16);
     GGML_ASSERT(KQV->type == GGML_TYPE_F32);
 
     GGML_ASSERT(!mask || mask->type == GGML_TYPE_F16);
@@ -143,6 +677,7 @@ void launch_fattn(ggml_backend_cuda_context & ctx, ggml_tensor * dst, fattn_kern
         mask ? mask->ne[1] : 0, mask ?  mask->nb[1] : 0,
         Q->nb[1], Q->nb[2], Q->nb[3],
         K->nb[1], K->nb[2], K->nb[3],
+        V->nb[1], V->nb[2], V->nb[3],
         KQV->ne[0], KQV->ne[1], KQV->ne[2], KQV->ne[3]
     );
     CUDA_CHECK(cudaGetLastError());

ggml-cuda/fattn-tile-f16.cu

@@ -36,6 +36,9 @@ static __global__ void flash_attn_tile_ext_f16(
         const int nb11,
         const int nb12,
         const int nb13,
+        const int nb21,
+        const int nb22,
+        const int nb23,
         const int ne0,
         const int ne1,
         const int ne2,

ggml-cuda/fattn-tile-f32.cu

@@ -36,6 +36,9 @@ static __global__ void flash_attn_tile_ext_f32(
         const int nb11,
         const int nb12,
         const int nb13,
+        const int nb21,
+        const int nb22,
+        const int nb23,
         const int ne0,
         const int ne1,
         const int ne2,

ggml-cuda/fattn-vec-f16.cuh

@@ -1,5 +1,395 @@
 #include "common.cuh"
+#include "fattn-common.cuh"
 
-void ggml_cuda_flash_attn_ext_vec_f16(ggml_backend_cuda_context & ctx, ggml_tensor * dst);
+template<int D, int ncols, int parallel_blocks, ggml_type type_K, ggml_type type_V> // D == head size
+#if !(defined(GGML_USE_HIPBLAS) && defined(__HIP_PLATFORM_AMD__))
+__launch_bounds__(D, 1)
+#endif // !(defined(GGML_USE_HIPBLAS) && defined(__HIP_PLATFORM_AMD__))
+static __global__ void flash_attn_vec_ext_f16(
+        const char * __restrict__ Q,
+        const char * __restrict__ K,
+        const char * __restrict__ V,
+        const char * __restrict__ mask,
+        float      * __restrict__ dst,
+        float2     * __restrict__ dst_meta,
+        const float scale,
+        const float max_bias,
+        const float m0,
+        const float m1,
+        const uint32_t n_head_log2,
+        const int ne00,
+        const int ne01,
+        const int ne02,
+        const int ne03,
+        const int ne10,
+        const int ne11,
+        const int ne12,
+        const int ne13,
+        const int ne31,
+        const int nb31,
+        const int nb01,
+        const int nb02,
+        const int nb03,
+        const int nb11,
+        const int nb12,
+        const int nb13,
+        const int nb21,
+        const int nb22,
+        const int nb23,
+        const int ne0,
+        const int ne1,
+        const int ne2,
+        const int ne3) {
+#if FP16_AVAILABLE
+    //In this kernel Q, K, V are matrices while i, j, k are matrix indices.
 
-void ggml_cuda_flash_attn_ext_vec_f16_no_mma(ggml_backend_cuda_context & ctx, ggml_tensor * dst);
+    constexpr vec_dot_KQ_f16_t vec_dot_KQ = get_vec_dot_KQ_f16<D>(type_K);
+    constexpr bool Q_q8_1 = type_K != GGML_TYPE_F16;
+    constexpr dequantize_1_f16_t dequantize_1_v = get_dequantize_1_f16(type_V);
+
+    const int ic0 = (blockIdx.x / parallel_blocks) * ncols; // Index of the Q/QKV column to work on.
+    const int ip  =  blockIdx.x % parallel_blocks; // Index in group of blocks running for the same column in parallel.
+
+    const int gqa_ratio = ne02 / ne12; // With grouped query attention there are > 1 Q matrices per K, V matrix.
+    Q += nb02* blockIdx.y              + nb01*ic0;
+    K += nb12*(blockIdx.y / gqa_ratio);
+    V += nb22*(blockIdx.y / gqa_ratio);
+
+    const half * maskh = (const half   *)  mask + ne11*ic0;
+
+    const float slopef = get_alibi_slope(max_bias, blockIdx.y, n_head_log2, m0, m1);
+    const half  slopeh = __float2half(slopef);
+
+    static_assert(D % (2*WARP_SIZE) == 0, "D not divisible by 2*WARP_SIZE == 64.");
+    constexpr int nwarps = D / WARP_SIZE;
+    const int tid = WARP_SIZE*threadIdx.y + threadIdx.x;
+    __builtin_assume(tid < D);
+
+    __shared__ half KQ[ncols*D];
+    half2 * KQ2 = (half2 *) KQ;
+
+    half kqmax[ncols];
+#pragma unroll
+    for (int j = 0; j < ncols; ++j) {
+        kqmax[j] = -HALF_MAX_HALF;
+    }
+    half kqsum[ncols] = {0.0f};
+
+    __shared__ half kqmax_shared[ncols][WARP_SIZE];
+    __shared__ half kqsum_shared[ncols][WARP_SIZE];
+#pragma unroll
+    for (int j = 0; j < ncols; ++j) {
+        if (threadIdx.y == 0) {
+            kqmax_shared[j][threadIdx.x] = -HALF_MAX_HALF;
+            kqsum_shared[j][threadIdx.x] = 0.0f;
+        }
+    }
+    __syncthreads();
+
+    // Convert Q to half2 (f16 K) or q8_1 (quantized K) and store in registers:
+    half2  Q_h2[ncols][D/(2*WARP_SIZE)];
+    int   Q_i32[ncols][D/(sizeof(int)*QK8_1) == 0 ? 1 : D/(sizeof(int)*QK8_1)];
+    half2  Q_ds[ncols][D/QK8_1 == 0 ? 1 : D/QK8_1];
+    if (Q_q8_1) {
+#pragma unroll
+        for (int j0 = 0; j0 < ncols; j0 += nwarps) {
+            const int j = j0 + threadIdx.y;
+
+            if (j0 + nwarps > ncols && j >= ncols) {
+                break;
+            }
+
+            // Reuse KQ as temporary storage for converting Q to q8_1:
+            int   * tmp_q_i32 = (int   *) &KQ[j*D];
+            half2 * tmp_q_ds  = (half2 *) (tmp_q_i32 + D/sizeof(int));
+
+            // Set memory to zero if out of bounds:
+            if (ncols > 2 && ic0 + j >= ne01) {
+#pragma unroll
+                for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) {
+                    const int i = i0 + threadIdx.x;
+
+                    tmp_q_i32[i] = 0;
+                }
+                if (threadIdx.x < D/QK8_1) {
+                    tmp_q_ds[threadIdx.x] = make_half2(0.0f, 0.0f);
+                }
+                continue;
+            }
+
+            const float * Q_f = (const float *) (Q + j*nb01);
+#pragma unroll
+            for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) {
+                quantize_q8_1_to_shared<half2>(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds);
+            }
+        }
+
+        __syncthreads();
+
+#pragma unroll
+        for (int j = 0; j < ncols; ++j) {
+            int   * tmp_q_i32 = (int   *) &KQ[j*D];
+            half2 * tmp_q_ds  = (half2 *) (tmp_q_i32 + D/sizeof(int));
+
+#pragma unroll
+            for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) {
+                const int i = i0 + threadIdx.x;
+
+                Q_i32[j][i0/WARP_SIZE] = tmp_q_i32[i];
+                Q_ds[j][i0/WARP_SIZE]  = tmp_q_ds[i/QI8_1];
+            }
+        }
+
+        __syncthreads();
+    } else {
+#pragma unroll
+        for (int j = 0; j < ncols; ++j) {
+            const float2 * Q_f2_j = (const float2 *) (Q + j*nb01);
+
+#pragma unroll
+            for (int i0 = 0; i0 < D/2; i0 += WARP_SIZE) {
+                const int i = i0 + threadIdx.x;
+
+                const float2 tmp = ncols <= 2 || ic0 + j < ne01 ? Q_f2_j[i] : make_float2(0.0f, 0.0f);
+                Q_h2[j][i0/WARP_SIZE] = make_half2(scale, scale) * make_half2(tmp.x, tmp.y);
+            }
+        }
+    }
+
+
+#pragma unroll
+    for (int j = 0; j < ncols; ++j) {
+        KQ[j*D + tid] = -HALF_MAX_HALF;
+    }
+
+    half2 VKQ[ncols] = {{0.0f, 0.0f}};
+
+    const int k_start = parallel_blocks == 1 ? 0 : ip*D;
+    for (int k_VKQ_0 = k_start; k_VKQ_0 < ne11; k_VKQ_0 += parallel_blocks*D) {
+        // Calculate KQ tile and keep track of new maximum KQ values:
+
+        // For unknown reasons using a half array of size 1 for kqmax_new causes a performance regression,
+        // see https://github.com/ggerganov/llama.cpp/pull/7061 .
+        // Therefore this variable is defined twice but only used once (so that the compiler can optimize out the unused variable).
+        half kqmax_new = kqmax[0];
+        half kqmax_new_arr[ncols];
+#pragma unroll
+        for (int j = 0; j < ncols; ++j) {
+            kqmax_new_arr[j] = kqmax[j];
+        }
+
+#pragma unroll
+        for (int i_KQ_0 = 0; i_KQ_0 < D; i_KQ_0 += nwarps) {
+            const int i_KQ = i_KQ_0 + threadIdx.y;
+
+            if ((i_KQ_0 + nwarps > D && i_KQ >= D) || (FATTN_KQ_STRIDE % D != 0 && k_VKQ_0 + i_KQ >= ne11)) {
+                break;
+            }
+
+#pragma unroll
+            for (int j = 0; j < ncols; ++j) {
+                half sum = vec_dot_KQ(K + (k_VKQ_0 + i_KQ)*nb11, Q_h2[j], Q_i32[j], Q_ds[j]);
+                sum = warp_reduce_sum(sum);
+                sum += mask ? slopeh*maskh[j*ne11 + k_VKQ_0 + i_KQ] : __float2half(0.0f);
+
+                if (ncols == 1) {
+                    kqmax_new        = ggml_cuda_hmax(kqmax_new,        sum);
+                } else {
+                    kqmax_new_arr[j] = ggml_cuda_hmax(kqmax_new_arr[j], sum);
+                }
+
+                if (threadIdx.x == 0) {
+                    KQ[j*D + i_KQ] = sum;
+                }
+            }
+        }
+
+#pragma unroll
+        for (int j = 0; j < ncols; ++j) {
+            half kqmax_new_j = ncols == 1 ? kqmax_new : kqmax_new_arr[j];
+
+            kqmax_new_j = warp_reduce_max(kqmax_new_j);
+            if (threadIdx.x == 0) {
+                kqmax_shared[j][threadIdx.y] = kqmax_new_j;
+            }
+        }
+
+        __syncthreads();
+
+#pragma unroll
+        for (int j = 0; j < ncols; ++j) {
+            half kqmax_new_j = kqmax_shared[j][threadIdx.x];
+            kqmax_new_j = warp_reduce_max(kqmax_new_j);
+
+            const half KQ_max_scale = hexp(kqmax[j] - kqmax_new_j);
+            kqmax[j] = kqmax_new_j;
+
+            const half val = hexp(KQ[j*D + tid] - kqmax[j]);
+            kqsum[j] = kqsum[j]*KQ_max_scale + val;
+            KQ[j*D + tid] = val;
+
+            VKQ[j] *= __half2half2(KQ_max_scale);
+        }
+
+        __syncthreads();
+
+#pragma unroll
+        for (int k0 = 0; k0 < D; k0 += 2) {
+            if (FATTN_KQ_STRIDE % D != 0 && k_VKQ_0 + k0 >= ne11) {
+                break;
+            }
+
+            half2 V_k;
+            reinterpret_cast<half&>(V_k.x) = dequantize_1_v(V + (k_VKQ_0 + k0 + 0)*nb21, tid);
+            reinterpret_cast<half&>(V_k.y) = dequantize_1_v(V + (k_VKQ_0 + k0 + 1)*nb21, tid);
+#pragma unroll
+            for (int j = 0; j < ncols; ++j) {
+                VKQ[j] += V_k*KQ2[j*(D/2) + k0/2];
+            }
+        }
+
+        __syncthreads();
+    }
+
+#pragma unroll
+    for (int j = 0; j < ncols; ++j) {
+        kqsum[j] = warp_reduce_sum(kqsum[j]);
+        if (threadIdx.x == 0) {
+            kqsum_shared[j][threadIdx.y] = kqsum[j];
+        }
+    }
+
+    __syncthreads();
+
+#pragma unroll
+    for (int j_VKQ = 0; j_VKQ < ncols; ++j_VKQ) {
+        if (ncols > 2 && ic0 + j_VKQ >= ne01) {
+            break;
+        }
+
+        kqsum[j_VKQ] = kqsum_shared[j_VKQ][threadIdx.x];
+        kqsum[j_VKQ] = warp_reduce_sum(kqsum[j_VKQ]);
+
+        half dst_val = (__low2half(VKQ[j_VKQ]) + __high2half(VKQ[j_VKQ]));
+        if (parallel_blocks == 1) {
+            dst_val /= kqsum[j_VKQ];
+        }
+        const int j_dst = (ic0 + j_VKQ)*parallel_blocks + ip;
+        dst[j_dst*D*gridDim.y + D*blockIdx.y + tid] = dst_val;
+    }
+
+    if (parallel_blocks != 1 && tid < ncols && (ncols <= 2 || ic0 + tid < ne01)) {
+        dst_meta[(ic0 + tid)*gridDim.y*parallel_blocks + blockIdx.y*parallel_blocks + ip] = make_float2(kqmax[tid], kqsum[tid]);
+    }
+#else
+   NO_DEVICE_CODE;
+#endif // FP16_AVAILABLE
+}
+
+template <int D, int cols_per_block, int parallel_blocks, ggml_type type_K, ggml_type type_V>
+void ggml_cuda_flash_attn_ext_vec_f16_case_impl(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
+    constexpr int nwarps = D/WARP_SIZE;
+    fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f16<D, cols_per_block, parallel_blocks, type_K, type_V>;
+    launch_fattn<D, parallel_blocks>(ctx, dst, fattn_kernel, nwarps, cols_per_block);
+}
+
+template <int D, ggml_type type_K, ggml_type type_V>
+void ggml_cuda_flash_attn_ext_vec_f16_case(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
+    ggml_tensor * KQV = dst;
+    ggml_tensor * Q   = dst->src[0];
+    ggml_tensor * K   = dst->src[1];
+    ggml_tensor * V   = dst->src[2];
+
+    const int32_t precision = KQV->op_params[2];
+    GGML_ASSERT(precision == GGML_PREC_DEFAULT);
+
+    GGML_ASSERT(K->type == type_K);
+    GGML_ASSERT(V->type == type_V);
+
+    if (Q->ne[1] == 1) {
+        constexpr int cols_per_block  = 1;
+        constexpr int parallel_blocks = 4;
+        ggml_cuda_flash_attn_ext_vec_f16_case_impl<D, cols_per_block, parallel_blocks, type_K, type_V>(ctx, dst);
+        return;
+    }
+
+    if (Q->ne[1] == 2) {
+        constexpr int cols_per_block  = 2;
+        constexpr int parallel_blocks = 4;
+        ggml_cuda_flash_attn_ext_vec_f16_case_impl<D, cols_per_block, parallel_blocks, type_K, type_V>(ctx, dst);
+        return;
+    }
+
+    if (Q->ne[1] <= 4) {
+        constexpr int cols_per_block  = 4;
+        constexpr int parallel_blocks = 4;
+        ggml_cuda_flash_attn_ext_vec_f16_case_impl<D, cols_per_block, parallel_blocks, type_K, type_V>(ctx, dst);
+        return;
+    }
+
+    if (Q->ne[1] <= 8) {
+        constexpr int cols_per_block  = 8;
+        constexpr int parallel_blocks = 4;
+        ggml_cuda_flash_attn_ext_vec_f16_case_impl<D, cols_per_block, parallel_blocks, type_K, type_V>(ctx, dst);
+        return;
+    }
+
+    constexpr int cols_per_block  = 8;
+    constexpr int parallel_blocks = 1;
+    ggml_cuda_flash_attn_ext_vec_f16_case_impl<D, cols_per_block, parallel_blocks, type_K, type_V>(ctx, dst);
+}
+
+#define DECL_FATTN_VEC_F16_CASE(D, type_K, type_V)                          \
+    template void ggml_cuda_flash_attn_ext_vec_f16_case                     \
+    <D, type_K, type_V>(ggml_backend_cuda_context & ctx, ggml_tensor * dst) \
+
+extern DECL_FATTN_VEC_F16_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q4_0);
+extern DECL_FATTN_VEC_F16_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q4_1);
+extern DECL_FATTN_VEC_F16_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q5_0);
+extern DECL_FATTN_VEC_F16_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q5_1);
+extern DECL_FATTN_VEC_F16_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q8_0);
+extern DECL_FATTN_VEC_F16_CASE( 64, GGML_TYPE_F16, GGML_TYPE_F16);
+
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q4_0);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q4_0);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q4_0);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q4_0);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q4_0);
+
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_1);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q4_1);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q4_1);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q4_1);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q4_1);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q4_1);
+
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q5_0);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q5_0);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q5_0);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q5_0);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q5_0);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q5_0);
+
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q5_1);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q5_1);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q5_1);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q5_1);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q5_1);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q5_1);
+
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q8_0);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q8_0);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q8_0);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q8_0);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q8_0);
+
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_F16);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_F16);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_F16);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_F16);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_F16);
+extern DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_F16,  GGML_TYPE_F16);
+
+extern DECL_FATTN_VEC_F16_CASE(256, GGML_TYPE_F16, GGML_TYPE_F16);

ggml-cuda/fattn-vec-f32.cuh

@@ -1,3 +1,376 @@
 #include "common.cuh"
+#include "fattn-common.cuh"
 
-void ggml_cuda_flash_attn_ext_vec_f32(ggml_backend_cuda_context & ctx, ggml_tensor * dst);
+template<int D, int ncols, int parallel_blocks, ggml_type type_K, ggml_type type_V> // D == head size
+#if !(defined(GGML_USE_HIPBLAS) && defined(__HIP_PLATFORM_AMD__))
+__launch_bounds__(D, 1)
+#endif // !(defined(GGML_USE_HIPBLAS) && defined(__HIP_PLATFORM_AMD__))
+static __global__ void flash_attn_vec_ext_f32(
+        const char * __restrict__ Q,
+        const char * __restrict__ K,
+        const char * __restrict__ V,
+        const char * __restrict__ mask,
+        float      * __restrict__ dst,
+        float2     * __restrict__ dst_meta,
+        const float scale,
+        const float max_bias,
+        const float m0,
+        const float m1,
+        const uint32_t n_head_log2,
+        const int ne00,
+        const int ne01,
+        const int ne02,
+        const int ne03,
+        const int ne10,
+        const int ne11,
+        const int ne12,
+        const int ne13,
+        const int ne31,
+        const int nb31,
+        const int nb01,
+        const int nb02,
+        const int nb03,
+        const int nb11,
+        const int nb12,
+        const int nb13,
+        const int nb21,
+        const int nb22,
+        const int nb23,
+        const int ne0,
+        const int ne1,
+        const int ne2,
+        const int ne3) {
+    //In this kernel Q, K, V are matrices while i, j, k are matrix indices.
+
+    constexpr vec_dot_KQ_f32_t vec_dot_KQ = get_vec_dot_KQ_f32<D>(type_K);
+    constexpr bool Q_q8_1 = type_K != GGML_TYPE_F16;
+    constexpr dequantize_1_f32_t dequantize_1_v = get_dequantize_1_f32(type_V);
+
+    const int ic0 = (blockIdx.x / parallel_blocks) * ncols; // Index of the Q/QKV column to work on.
+    const int ip  =  blockIdx.x % parallel_blocks; // Index in group of blocks running for the same column in parallel.
+
+    const int gqa_ratio = ne02 / ne12; // With grouped query attention there are > 1 Q matrices per K, V matrix.
+    Q += nb02* blockIdx.y              + nb01*ic0;
+    K += nb12*(blockIdx.y / gqa_ratio);
+    V += nb22*(blockIdx.y / gqa_ratio); // K and V have same shape
+    const half * maskh = (const half   *)  mask + ne11*ic0;
+
+    const float slope = get_alibi_slope(max_bias, blockIdx.y, n_head_log2, m0, m1);
+
+    static_assert(D % (2*WARP_SIZE) == 0, "D not divisible by 2*WARP_SIZE == 64.");
+    constexpr int nwarps = D / WARP_SIZE;
+    const int tid = WARP_SIZE*threadIdx.y + threadIdx.x;
+    __builtin_assume(tid < D);
+
+    __shared__ float KQ[ncols*D];
+#pragma unroll
+    for (int j = 0; j < ncols; ++j) {
+        KQ[j*D + tid] = -FLT_MAX/2.0f;
+    }
+
+    float kqmax[ncols];
+#pragma unroll
+    for (int j = 0; j < ncols; ++j) {
+        kqmax[j] = -FLT_MAX/2.0f;
+    }
+    float kqsum[ncols] = {0.0f};
+
+    __shared__ float kqmax_shared[ncols][WARP_SIZE];
+    __shared__ float kqsum_shared[ncols][WARP_SIZE];
+#pragma unroll
+    for (int j = 0; j < ncols; ++j) {
+        if (threadIdx.y == 0) {
+            kqmax_shared[j][threadIdx.x] = -FLT_MAX/2.0f;
+            kqsum_shared[j][threadIdx.x] = 0.0f;
+        }
+    }
+    __syncthreads();
+
+    // Convert Q to float2 (f16 K) or q8_1 (quantized K) and store in registers:
+    float2  Q_f2[ncols][D/(2*WARP_SIZE)];
+    int    Q_i32[ncols][D/(sizeof(int)*QK8_1) == 0 ? 1 : D >= D/(sizeof(int)*QK8_1)];
+    float2  Q_ds[ncols][D/QK8_1 == 0 ? 1 : D/QK8_1];
+    if (Q_q8_1) {
+#pragma unroll
+        for (int j0 = 0; j0 < ncols; j0 += nwarps) {
+            const int j = j0 + threadIdx.y;
+
+            if (j0 + nwarps > ncols && j >= ncols) {
+                break;
+            }
+
+            // Reuse KQ as temporary storage for converting Q to q8_1:
+            int    * tmp_q_i32 = (int    *) &KQ[j*D];
+            float2 * tmp_q_ds  = (float2 *) (tmp_q_i32 + D/sizeof(int));
+
+            // Set memory to zero if out of bounds:
+            if (ncols > 2 && ic0 + j >= ne01) {
+#pragma unroll
+                for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) {
+                    const int i = i0 + threadIdx.x;
+
+                    tmp_q_i32[i] = 0;
+                }
+                if (threadIdx.x < D/QK8_1) {
+                    tmp_q_ds[threadIdx.x] = make_float2(0.0f, 0.0f);
+                }
+                continue;
+            }
+
+            const float * Q_f = (const float *) (Q + j*nb01);
+#pragma unroll
+            for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) {
+                quantize_q8_1_to_shared<float2>(Q_f + 4*i0, scale, tmp_q_i32, tmp_q_ds);
+            }
+        }
+
+        __syncthreads();
+
+#pragma unroll
+        for (int j = 0; j < ncols; ++j) {
+            int    * tmp_q_i32 = (int    *) &KQ[j*D];
+            float2 * tmp_q_ds  = (float2 *) (tmp_q_i32 + D/sizeof(int));
+
+#pragma unroll
+            for (int i0 = 0; i0 < D/sizeof(int); i0 += WARP_SIZE) {
+                const int i = i0 + threadIdx.x;
+
+                Q_i32[j][i0/WARP_SIZE] = tmp_q_i32[i];
+                Q_ds[j][i0/WARP_SIZE]  = tmp_q_ds[i/QI8_1];
+            }
+        }
+
+        __syncthreads();
+    } else {
+#pragma unroll
+        for (int j = 0; j < ncols; ++j) {
+            const float2 * Q_f2_j = (const float2 *) (Q + j*nb01);
+#pragma unroll
+            for (int i0 = 0; i0 < D/2; i0 += WARP_SIZE) {
+                const int i = i0 + threadIdx.x;
+
+                Q_f2[j][i0/WARP_SIZE]    = ncols <= 2 || ic0 + j ? Q_f2_j[i] : make_float2(0.0f, 0.0f);
+                Q_f2[j][i0/WARP_SIZE].x *= scale;
+                Q_f2[j][i0/WARP_SIZE].y *= scale;
+            }
+        }
+    }
+
+    float VKQ[ncols] = {0.0f};
+
+    const int k_start = parallel_blocks == 1 ? 0 : ip*D;
+    for (int k_VKQ_0 = k_start; k_VKQ_0 < ne11; k_VKQ_0 += parallel_blocks*D) {
+        // Calculate KQ tile and keep track of new maximum KQ values:
+
+        float kqmax_new_arr[ncols];
+#pragma unroll
+        for (int j = 0; j < ncols; ++j) {
+            kqmax_new_arr[j] = kqmax[j];
+        }
+
+#pragma unroll
+        for (int i_KQ_0 = 0; i_KQ_0 < D; i_KQ_0 += nwarps) {
+            const int i_KQ = i_KQ_0 + threadIdx.y;
+
+            if ((i_KQ_0 + nwarps > D && i_KQ >= D) || (FATTN_KQ_STRIDE % D != 0 && k_VKQ_0 + i_KQ >= ne11)) {
+                break;
+            }
+
+#pragma unroll
+            for (int j = 0; j < ncols; ++j) {
+                float sum = vec_dot_KQ(K + (k_VKQ_0 + i_KQ)*nb11, Q_f2[j], Q_i32[j], Q_ds[j]);
+                sum = warp_reduce_sum(sum);
+                sum += mask ? slope*__half2float(maskh[j*ne11 + k_VKQ_0 + i_KQ]) : 0.0f;
+
+                kqmax_new_arr[j] = fmaxf(kqmax_new_arr[j], sum);
+
+                if (threadIdx.x == 0) {
+                    KQ[j*D + i_KQ] = sum;
+                }
+            }
+        }
+
+#pragma unroll
+        for (int j = 0; j < ncols; ++j) {
+            float kqmax_new_j = kqmax_new_arr[j];
+
+            kqmax_new_j = warp_reduce_max(kqmax_new_j);
+            if (threadIdx.x == 0) {
+                kqmax_shared[j][threadIdx.y] = kqmax_new_j;
+            }
+        }
+
+        __syncthreads();
+
+#pragma unroll
+        for (int j = 0; j < ncols; ++j) {
+            float kqmax_new_j = kqmax_shared[j][threadIdx.x];
+            kqmax_new_j = warp_reduce_max(kqmax_new_j);
+
+            const float KQ_max_scale = expf(kqmax[j] - kqmax_new_j);
+            kqmax[j] = kqmax_new_j;
+
+            const float val = expf(KQ[j*D + tid] - kqmax[j]);
+            kqsum[j] = kqsum[j]*KQ_max_scale + val;
+            KQ[j*D + tid] = val;
+
+            VKQ[j] *= KQ_max_scale;
+        }
+
+        __syncthreads();
+
+#pragma unroll
+        for (int k = 0; k < D; ++k) {
+            if (FATTN_KQ_STRIDE % D != 0 && k_VKQ_0 + k >= ne11) {
+                break;
+            }
+
+            const float V_ki = dequantize_1_v(V + (k_VKQ_0 + k)*nb21, tid);
+#pragma unroll
+            for (int j = 0; j < ncols; ++j) {
+                VKQ[j] += V_ki*KQ[j*D + k];
+            }
+        }
+
+        __syncthreads();
+    }
+
+#pragma unroll
+    for (int j = 0; j < ncols; ++j) {
+        kqsum[j] = warp_reduce_sum(kqsum[j]);
+        if (threadIdx.x == 0) {
+            kqsum_shared[j][threadIdx.y] = kqsum[j];
+        }
+    }
+
+    __syncthreads();
+
+#pragma unroll
+    for (int j_VKQ = 0; j_VKQ < ncols; ++j_VKQ) {
+        if (ncols > 2 && ic0 + j_VKQ >= ne01) {
+            break;
+        }
+
+        kqsum[j_VKQ] = kqsum_shared[j_VKQ][threadIdx.x];
+        kqsum[j_VKQ] = warp_reduce_sum(kqsum[j_VKQ]);
+
+        float dst_val = VKQ[j_VKQ];
+        if (parallel_blocks == 1) {
+            dst_val /= kqsum[j_VKQ];
+        }
+        const int j_dst = (ic0 + j_VKQ)*parallel_blocks + ip;
+        dst[j_dst*D*gridDim.y + D*blockIdx.y + tid] = dst_val;
+    }
+
+    if (parallel_blocks != 1 && tid < ncols && (ncols <= 2 || ic0 + tid < ne01)) {
+        dst_meta[(ic0 + tid)*gridDim.y*parallel_blocks + blockIdx.y*parallel_blocks + ip] = make_float2(kqmax[tid], kqsum[tid]);
+    }
+}
+
+template <int D, int cols_per_block, int parallel_blocks, ggml_type type_K, ggml_type type_V>
+void ggml_cuda_flash_attn_ext_vec_f32_case_impl(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
+    constexpr int nwarps = D/WARP_SIZE;
+    fattn_kernel_t fattn_kernel = flash_attn_vec_ext_f32<D, cols_per_block, parallel_blocks, type_K, type_V>;
+    launch_fattn<D, parallel_blocks>(ctx, dst, fattn_kernel, nwarps, cols_per_block);
+}
+
+template <int D, ggml_type type_K, ggml_type type_V>
+void ggml_cuda_flash_attn_ext_vec_f32_case(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
+    ggml_tensor * KQV = dst;
+    ggml_tensor * Q   = dst->src[0];
+    ggml_tensor * K   = dst->src[1];
+    ggml_tensor * V   = dst->src[2];
+
+    const int32_t precision = KQV->op_params[2];
+    GGML_ASSERT(precision == GGML_PREC_DEFAULT);
+
+    GGML_ASSERT(K->type == type_K);
+    GGML_ASSERT(V->type == type_V);
+
+    if (Q->ne[1] == 1) {
+        constexpr int cols_per_block  = 1;
+        constexpr int parallel_blocks = 4;
+        ggml_cuda_flash_attn_ext_vec_f32_case_impl<D, cols_per_block, parallel_blocks, type_K, type_V>(ctx, dst);
+        return;
+    }
+
+    if (Q->ne[1] == 2) {
+        constexpr int cols_per_block  = 2;
+        constexpr int parallel_blocks = 4;
+        ggml_cuda_flash_attn_ext_vec_f32_case_impl<D, cols_per_block, parallel_blocks, type_K, type_V>(ctx, dst);
+        return;
+    }
+
+    if (Q->ne[1] <= 4) {
+        constexpr int cols_per_block  = 4;
+        constexpr int parallel_blocks = 4;
+        ggml_cuda_flash_attn_ext_vec_f32_case_impl<D, cols_per_block, parallel_blocks, type_K, type_V>(ctx, dst);
+        return;
+    }
+
+    if (Q->ne[1] <= 8) {
+        constexpr int cols_per_block  = 8;
+        constexpr int parallel_blocks = 4;
+        ggml_cuda_flash_attn_ext_vec_f32_case_impl<D, cols_per_block, parallel_blocks, type_K, type_V>(ctx, dst);
+        return;
+    }
+
+    constexpr int cols_per_block  = 8;
+    constexpr int parallel_blocks = 1;
+    ggml_cuda_flash_attn_ext_vec_f32_case_impl<D, cols_per_block, parallel_blocks, type_K, type_V>(ctx, dst);
+}
+
+#define DECL_FATTN_VEC_F32_CASE(D, type_K, type_V)                          \
+    template void ggml_cuda_flash_attn_ext_vec_f32_case                     \
+    <D, type_K, type_V>(ggml_backend_cuda_context & ctx, ggml_tensor * dst) \
+
+extern DECL_FATTN_VEC_F32_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q4_0);
+extern DECL_FATTN_VEC_F32_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q4_1);
+extern DECL_FATTN_VEC_F32_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q5_0);
+extern DECL_FATTN_VEC_F32_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q5_1);
+extern DECL_FATTN_VEC_F32_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q8_0);
+extern DECL_FATTN_VEC_F32_CASE( 64, GGML_TYPE_F16, GGML_TYPE_F16);
+
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q4_0);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q4_0);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q4_0);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q4_0);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q4_0);
+
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_1);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q4_1);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q4_1);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q4_1);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q4_1);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q4_1);
+
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q5_0);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q5_0);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q5_0);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q5_0);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q5_0);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q5_0);
+
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q5_1);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q5_1);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q5_1);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q5_1);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q5_1);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q5_1);
+
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q8_0);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q8_0);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q8_0);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q8_0);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q8_0);
+
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_F16);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_F16);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_F16);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_F16);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_F16);
+extern DECL_FATTN_VEC_F32_CASE(128, GGML_TYPE_F16,  GGML_TYPE_F16);
+
+extern DECL_FATTN_VEC_F32_CASE(256, GGML_TYPE_F16, GGML_TYPE_F16);

ggml-cuda/fattn-wmma-f16.cuh

@@ -0,0 +1,490 @@
+#include "common.cuh"
+#include "fattn-common.cuh"
+
+#if FP16_MMA_AVAILABLE
+#include <mma.h>
+#endif
+
+// D == head size, VKQ_stride == num VKQ rows calculated in parallel:
+template<int D, int ncols, int nwarps, int VKQ_stride, int parallel_blocks, typename KQ_acc_t>
+#if !(defined(GGML_USE_HIPBLAS) && defined(__HIP_PLATFORM_AMD__))
+__launch_bounds__(nwarps*WARP_SIZE, 1)
+#endif // !(defined(GGML_USE_HIPBLAS) && defined(__HIP_PLATFORM_AMD__))
+static __global__ void flash_attn_ext_f16(
+        const char * __restrict__ Q,
+        const char * __restrict__ K,
+        const char * __restrict__ V,
+        const char * __restrict__ mask,
+        float      * __restrict__ dst,
+        float2     * __restrict__ dst_meta,
+        const float scale,
+        const float max_bias,
+        const float m0,
+        const float m1,
+        const uint32_t n_head_log2,
+        const int ne00,
+        const int ne01,
+        const int ne02,
+        const int ne03,
+        const int ne10,
+        const int ne11,
+        const int ne12,
+        const int ne13,
+        const int ne31,
+        const int nb31,
+        const int nb01,
+        const int nb02,
+        const int nb03,
+        const int nb11,
+        const int nb12,
+        const int nb13,
+        const int nb21,
+        const int nb22,
+        const int nb23,
+        const int ne0,
+        const int ne1,
+        const int ne2,
+        const int ne3) {
+#if FP16_MMA_AVAILABLE
+    //In this kernel Q, K, V are matrices while i, j, k are matrix indices.
+
+    const int ic0 = ncols*(blockIdx.x / parallel_blocks); // Index of the first Q/QKV column to work on.
+    const int ip  =        blockIdx.x % parallel_blocks;  // Index in group of blocks running for the same column in parallel.
+
+    static_assert(D <= FATTN_KQ_STRIDE, "D must be <= FATTN_KQ_STRIDE.");
+    static_assert(ncols == 8 || ncols % 16 == 0, "ncols must be 8 or a multiple of 16.");
+    constexpr int frag_m = ncols == 8 ? 32 : 16;
+    constexpr int frag_n = ncols == 8 ?  8 : 16;
+    static_assert(D % frag_m == 0, "If ncols == 8 then D % frag_m must be 0.");
+    typedef nvcuda::wmma::fragment<nvcuda::wmma::matrix_a,    frag_m, frag_n, 16, half, nvcuda::wmma::row_major> frag_a_K;
+    typedef nvcuda::wmma::fragment<nvcuda::wmma::matrix_a,    frag_m, frag_n, 16, half, nvcuda::wmma::col_major> frag_a_V;
+    typedef nvcuda::wmma::fragment<nvcuda::wmma::matrix_b,    frag_m, frag_n, 16, half, nvcuda::wmma::col_major> frag_b;
+    typedef nvcuda::wmma::fragment<nvcuda::wmma::accumulator, frag_m, frag_n, 16, KQ_acc_t>                      frag_c_KQ;
+    typedef nvcuda::wmma::fragment<nvcuda::wmma::accumulator, frag_m, frag_n, 16, half>                          frag_c_VKQ;
+
+    constexpr int KQ_stride_tc  = nwarps*frag_m; // Number of KQ rows calculated in parallel.
+    constexpr int VKQ_ratio = KQ_stride_tc/VKQ_stride; // Number of parallel VKQ accumulators needed to keep all warps busy.
+    static_assert(VKQ_ratio <= nwarps, "VKQ_ratio must be <= nwarps.");
+
+    // Pad internal representation of KQ, KQV to reduce shared memory bank conflicts:
+    constexpr int D_padded = D + 8;
+    constexpr int kqs_padded = FATTN_KQ_STRIDE + 8;
+    constexpr int kqar = sizeof(KQ_acc_t)/sizeof(half);
+
+    const int gqa_ratio = ne02 / ne12; // With grouped query attention there are > 1 Q matrices per K, V matrix.
+    const float * Q_f   = (const float *) (Q + nb02* blockIdx.y              + nb01*ic0);
+    const half  * K_h   = (const half  *) (K + nb12*(blockIdx.y / gqa_ratio));
+    const half  * V_h   = (const half  *) (V + nb12*(blockIdx.y / gqa_ratio)); // K and V have same shape
+    const half  * maskh = (const half  *)  mask + (nb31/sizeof(half))* ic0;
+    const half2 * mask2 = (const half2 *)  mask + (nb31/sizeof(half))*(ic0/2);
+
+    const int stride_Q  = nb01 / sizeof(float);
+    const int stride_KV = nb11 / sizeof(half);
+
+    const float slopef = get_alibi_slope(max_bias, blockIdx.y, n_head_log2, m0, m1);
+    const half  slopeh = __float2half(slopef);
+    const half2 slope2 = make_half2(slopef, slopef);
+
+    frag_b Q_b[D/16][ncols/frag_n];
+
+    // A single buffer for temporarily holding tiles of KQ and VKQ parts:
+    constexpr int mem_KQ = ncols*kqs_padded*kqar;
+    constexpr int mem_VKQ_parts = VKQ_ratio*ncols*D_padded;
+    __shared__ half KQ[mem_KQ >= mem_VKQ_parts ? mem_KQ : mem_VKQ_parts];
+    float * KQ_f = (float *) KQ;
+    half2 * KQ2 = (half2 *) KQ;
+
+    float    KQ_rowsum_f[ncols/nwarps] = {0.0f};
+    float       KQ_max_f[ncols/nwarps];
+    float KQ_max_scale_f[ncols/nwarps] = {0.0f};
+
+#pragma unroll
+    for (int j = 0; j < ncols/nwarps; ++j) {
+        KQ_max_f[j] = -FLT_MAX/2.0f;
+    }
+
+    half2    KQ_rowsum_h2[ncols/nwarps] = {{0.0f, 0.0f}};
+    half2       KQ_max_h2[ncols/nwarps];
+    half2 KQ_max_scale_h2[ncols/nwarps] = {{0.0f, 0.0f}};
+
+#pragma unroll
+    for (int j = 0; j < ncols/nwarps; ++j) {
+        KQ_max_h2[j] = make_half2(-HALF_MAX_HALF, -HALF_MAX_HALF);
+    }
+
+    __shared__ half VKQ[ncols*D_padded]; // Accumulator for final VKQ slice.
+    half2 * VKQ2 = (half2 *) VKQ;
+#pragma unroll
+    for (int j0 = 0; j0 < ncols; j0 += nwarps) {
+        const int j = j0 + threadIdx.y;
+#pragma unroll
+        for (int i0 = 0; i0 < D/2; i0 += WARP_SIZE) {
+            const int i = i0 + threadIdx.x;
+            if (i0 + WARP_SIZE > D/2 && i >= D/2) {
+                break;
+            }
+            VKQ2[j*(D_padded/2) + i] = make_half2(0.0f, 0.0f);
+        }
+    }
+
+    // Convert Q to half and apply scale, temporarily store in KQ:
+#pragma unroll
+    for (int j0 = 0; j0 < ncols; j0 += nwarps) {
+        const int j = j0 + threadIdx.y;
+#pragma unroll
+        for (int i0 = 0; i0 < D; i0 += WARP_SIZE) {
+            const int i = i0 + threadIdx.x;
+            if (i0 + WARP_SIZE > D && i >= D) {
+                break;
+            }
+            KQ[j*D_padded + i] = ic0 + j < ne01 ? Q_f[j*stride_Q + i] * scale : 0.0f;
+        }
+    }
+
+    __syncthreads();
+
+    // Load Q into tensor core fragments/registers since it will be used frequently:
+#pragma unroll
+    for (int i0 = 0; i0 < D; i0 += 16) {
+#pragma unroll
+        for (int j0 = 0; j0 < ncols; j0 += frag_n) {
+            nvcuda::wmma::load_matrix_sync(Q_b[i0/16][j0/frag_n], KQ + j0*D_padded + i0, D_padded);
+        }
+    }
+
+    __syncthreads();
+
+    // Iterate over ne11 == previous tokens:
+    for (int k_VKQ_0 = ip*FATTN_KQ_STRIDE; k_VKQ_0 < ne11; k_VKQ_0 += parallel_blocks*FATTN_KQ_STRIDE) {
+        // Calculate tile of KQ:
+#pragma unroll
+        for (int i_KQ_0 = 0; i_KQ_0 < FATTN_KQ_STRIDE; i_KQ_0 += KQ_stride_tc) {
+            frag_c_KQ KQ_c[ncols/frag_n];
+#pragma unroll
+            for (int j = 0; j < ncols/frag_n; ++j) {
+                nvcuda::wmma::fill_fragment(KQ_c[j], 0.0f);
+            }
+#pragma unroll
+            for (int k_KQ_0 = 0; k_KQ_0 < D; k_KQ_0 += 16) {
+                frag_a_K K_a;
+                nvcuda::wmma::load_matrix_sync(K_a, K_h + (k_VKQ_0 + i_KQ_0 + frag_m*threadIdx.y)*stride_KV + k_KQ_0, stride_KV);
+#pragma unroll
+                for (int j = 0; j < ncols/frag_n; ++j) {
+                    nvcuda::wmma::mma_sync(KQ_c[j], K_a, Q_b[k_KQ_0/16][j], KQ_c[j]);
+                }
+            }
+#pragma unroll
+            for (int j0 = 0; j0 < ncols; j0 += frag_n) {
+                nvcuda::wmma::store_matrix_sync((KQ_acc_t *) KQ + j0*kqs_padded + i_KQ_0 + frag_m*threadIdx.y, KQ_c[j0/frag_n], kqs_padded, nvcuda::wmma::mem_col_major);
+            }
+        }
+
+        __syncthreads();
+
+        // Calculate softmax for each KQ column using the current max. value.
+        // The divisor is stored in KQ_rowsum and will be applied at the end.
+#pragma unroll
+        for (int j0 = 0; j0 < ncols; j0 += nwarps) {
+            const int j = j0 + threadIdx.y;
+
+            if (std::is_same<KQ_acc_t, float>::value) {
+                float KQ_f_tmp[FATTN_KQ_STRIDE / WARP_SIZE];
+#pragma unroll
+                for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += WARP_SIZE) {
+                    const int k = k0 + threadIdx.x;
+
+                    KQ_f_tmp[k0/WARP_SIZE] = KQ_f[j*kqs_padded + k];
+                }
+
+                float KQ_max_new = KQ_max_f[j0/nwarps];
+#pragma unroll
+                for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += WARP_SIZE) {
+                    const int k = k0 + threadIdx.x;
+
+                    KQ_f_tmp[k0/WARP_SIZE] += mask ? __half2float(slopeh*maskh[j*(nb31/sizeof(half)) + k_VKQ_0 + k]) : 0.0f;
+                    KQ_max_new = max(KQ_max_new, KQ_f_tmp[k0/WARP_SIZE]);
+                }
+                KQ_max_new = warp_reduce_max(KQ_max_new);
+
+                const float diff = KQ_max_f[j0/nwarps] - KQ_max_new;
+                KQ_max_scale_f[j0/nwarps] = expf(diff);
+                if (diff <= SOFTMAX_FTZ_THRESHOLD) {
+                    KQ_max_scale_f[j0/nwarps] = 0.0f;
+                }
+                KQ_max_f[j0/nwarps] = KQ_max_new;
+
+                float KQ_rowsum_add = 0.0f;
+#pragma unroll
+                for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += WARP_SIZE) {
+                    const int k = k0 + threadIdx.x;
+
+                    const float diff = KQ_f_tmp[k0/WARP_SIZE] - KQ_max_f[j0/nwarps];
+                    KQ_f_tmp[k0/WARP_SIZE] = expf(diff);
+                    if (diff <= SOFTMAX_FTZ_THRESHOLD) {
+                        KQ_f_tmp[k0/WARP_SIZE] = 0.0f;
+                    }
+                    KQ_rowsum_add += KQ_f_tmp[k0/WARP_SIZE];
+                    KQ[j*(kqar*kqs_padded) + k] = KQ_f_tmp[k0/WARP_SIZE];
+                }
+                KQ_rowsum_add = warp_reduce_sum(KQ_rowsum_add);
+
+                // Scale previous KQ_rowsum to account for a potential increase in KQ_max:
+                KQ_rowsum_f[j0/nwarps] = KQ_max_scale_f[j0/nwarps]*KQ_rowsum_f[j0/nwarps] + KQ_rowsum_add;
+            } else {
+                half2 KQ2_tmp[FATTN_KQ_STRIDE/(2*WARP_SIZE)];
+#pragma unroll
+                for (int k0 = 0; k0 < FATTN_KQ_STRIDE/2; k0 += WARP_SIZE) {
+                    const int k = k0 + threadIdx.x;
+
+                    KQ2_tmp[k0/WARP_SIZE] = KQ2[j*(kqs_padded/2) + k];
+                }
+
+                half2 KQ_max_new = KQ_max_h2[j0/nwarps];
+#pragma unroll
+                for (int k0 = 0; k0 < FATTN_KQ_STRIDE/2; k0 += WARP_SIZE) {
+                    const int k = k0 + threadIdx.x;
+
+                    KQ2_tmp[k0/WARP_SIZE] += mask ? slope2*mask2[(j*ne11 + k_VKQ_0)/2 + k] : make_half2(0.0f, 0.0f);
+                    KQ_max_new = ggml_cuda_hmax2(KQ_max_new, KQ2_tmp[k0/WARP_SIZE]);
+                }
+                KQ_max_new = __half2half2(warp_reduce_max(ggml_cuda_hmax(__low2half(KQ_max_new), __high2half(KQ_max_new))));
+                const half2 diff = KQ_max_h2[j0/nwarps] - KQ_max_new;
+                KQ_max_scale_h2[j0/nwarps] = h2exp(diff);
+                const uint32_t ftz_mask = __hgt2_mask(diff, make_half2(SOFTMAX_FTZ_THRESHOLD, SOFTMAX_FTZ_THRESHOLD));
+                *((uint32_t *) &KQ_max_scale_h2[j0/nwarps]) &= ftz_mask;
+                KQ_max_h2[j0/nwarps] = KQ_max_new;
+
+                half2 KQ_rowsum_add = make_half2(0.0f, 0.0f);
+#pragma unroll
+                for (int k0 = 0; k0 < FATTN_KQ_STRIDE/2; k0 += WARP_SIZE) {
+                    const int k = k0 + threadIdx.x;
+
+                    const half2 diff = KQ2_tmp[k0/WARP_SIZE] - KQ_max_h2[j0/nwarps];
+                    KQ2_tmp[k0/WARP_SIZE] = h2exp(diff);
+                    const uint32_t ftz_mask = __hgt2_mask(diff, make_half2(SOFTMAX_FTZ_THRESHOLD, SOFTMAX_FTZ_THRESHOLD));
+                    *((uint32_t *) &KQ2_tmp[k0/WARP_SIZE]) &= ftz_mask;
+                    KQ_rowsum_add += KQ2_tmp[k0/WARP_SIZE];
+                    KQ2[j*(kqs_padded/2) + k] = KQ2_tmp[k0/WARP_SIZE];
+                }
+                KQ_rowsum_add = warp_reduce_sum(KQ_rowsum_add);
+
+                // Scale previous KQ_rowsum to account for a potential increase in KQ_max:
+                KQ_rowsum_h2[j0/nwarps] = KQ_max_scale_h2[j0/nwarps]*KQ_rowsum_h2[j0/nwarps] + KQ_rowsum_add;
+            }
+        }
+
+        __syncthreads();
+
+        frag_b KQ_b[FATTN_KQ_STRIDE/(VKQ_ratio*16)][ncols/frag_n];
+#pragma unroll
+        for (int j0 = 0; j0 < ncols; j0 += frag_n) {
+#pragma unroll
+            for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += VKQ_ratio*16) {
+                const int k = k0 + (threadIdx.y % VKQ_ratio)*16;
+                nvcuda::wmma::load_matrix_sync(
+                    KQ_b[k0/(VKQ_ratio*16)][j0/frag_n],
+                    KQ + j0*(kqar*kqs_padded) + k,
+                    kqar*kqs_padded);
+            }
+        }
+
+        frag_c_VKQ VKQ_c[D/VKQ_stride][ncols/frag_n];
+#pragma unroll
+        for (int i_VKQ_0 = 0; i_VKQ_0 < D; i_VKQ_0 += VKQ_stride) {
+#pragma unroll
+            for (int j = 0; j < ncols/frag_n; ++j) {
+                nvcuda::wmma::fill_fragment(VKQ_c[i_VKQ_0/VKQ_stride][j], 0.0f);
+            }
+
+#pragma unroll
+            for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += VKQ_ratio*16) {
+                const int k = k0 + (threadIdx.y % VKQ_ratio)*16;
+
+                frag_a_V v_a;
+                nvcuda::wmma::load_matrix_sync(v_a, V_h + (k_VKQ_0 + k)*stride_KV + i_VKQ_0 + frag_m*(threadIdx.y/VKQ_ratio), stride_KV);
+#pragma unroll
+                for (int j = 0; j < ncols/frag_n; ++j) {
+                    nvcuda::wmma::mma_sync(VKQ_c[i_VKQ_0/VKQ_stride][j], v_a, KQ_b[k0/(VKQ_ratio*16)][j], VKQ_c[i_VKQ_0/VKQ_stride][j]);
+                }
+            }
+        }
+
+        __syncthreads();
+
+        const int offset_k = (threadIdx.y % VKQ_ratio) * (ncols*D_padded);
+#pragma unroll
+        for (int i_KQ_0 = 0; i_KQ_0 < D; i_KQ_0 += VKQ_stride) {
+#pragma unroll
+            for (int j0 = 0; j0 < ncols; j0 += frag_n) {
+                nvcuda::wmma::store_matrix_sync(
+                    KQ + offset_k + j0*D_padded + i_KQ_0 + frag_m*(threadIdx.y/VKQ_ratio),
+                    VKQ_c[i_KQ_0/VKQ_stride][j0/frag_n],
+                    D_padded, nvcuda::wmma::mem_col_major);
+            }
+        }
+
+        __syncthreads();
+
+#pragma unroll
+        for (int j0 = 0; j0 < ncols; j0 += nwarps) {
+            const int j = j0 + threadIdx.y;
+
+            half2 VKQ_scale;
+            if (std::is_same<KQ_acc_t, float>::value) {
+                VKQ_scale = make_half2(KQ_max_scale_f[j0/nwarps], KQ_max_scale_f[j0/nwarps]);
+            } else {
+                VKQ_scale = KQ_max_scale_h2[j0/nwarps];
+            }
+
+#pragma unroll
+            for (int i0 = 0; i0 < D/2; i0 += WARP_SIZE) {
+                const int i = i0 + threadIdx.x;
+                if (i0 + WARP_SIZE > D/2 && i >= D/2) {
+                    break;
+                }
+
+                half2 VKQ_add = make_half2(0.0f, 0.0f);
+#pragma unroll
+                for (int l = 0; l < VKQ_ratio; ++l) {
+                    VKQ_add += KQ2[l*(ncols*D_padded/2) + j*(D_padded/2) + i];
+                }
+                VKQ2[j*(D_padded/2) + i] = VKQ_scale*VKQ2[j*(D_padded/2) + i] + VKQ_add;
+            }
+        }
+
+        __syncthreads();
+    }
+
+#pragma unroll
+    for (int j0 = 0; j0 < ncols; j0 += nwarps) {
+        const int j_VKQ = j0 + threadIdx.y;
+        if (ic0 + j_VKQ >= ne01) {
+            return;
+        }
+        const int j_dst = (ic0 + j_VKQ)*parallel_blocks + ip;
+
+        float KQ_rowsum_j;
+        if (std::is_same<KQ_acc_t, float>::value) {
+            KQ_rowsum_j = KQ_rowsum_f[j0/nwarps];
+        } else {
+            KQ_rowsum_j = __low2float(KQ_rowsum_h2[j0/nwarps]) + __high2float(KQ_rowsum_h2[j0/nwarps]);
+        }
+
+#pragma unroll
+        for (int i0 = 0; i0 < D; i0 += WARP_SIZE) {
+            const int i = i0 + threadIdx.x;
+            if (i0 + WARP_SIZE > D && i >= D) {
+                break;
+            }
+            float dst_val = VKQ[j_VKQ*D_padded + i];
+            if (parallel_blocks == 1) {
+                dst_val /= KQ_rowsum_j;
+            }
+            dst[j_dst*gridDim.y*D + blockIdx.y*D + i] = dst_val;
+        }
+
+        if (parallel_blocks == 1 || threadIdx.x != 0) {
+            continue;
+        }
+
+        float2 dst_meta_val;
+        if (std::is_same<KQ_acc_t, float>::value) {
+            dst_meta_val.x = KQ_max_f[j0/nwarps];
+        } else {
+            dst_meta_val.x = __low2float(KQ_max_h2[j0/nwarps]);
+        }
+        dst_meta_val.y = KQ_rowsum_j;
+        dst_meta[(ic0 + j_VKQ)*gridDim.y*parallel_blocks + blockIdx.y*parallel_blocks + ip] = dst_meta_val;
+    }
+#else
+   NO_DEVICE_CODE;
+#endif // FP16_MMA_AVAILABLE
+}
+
+constexpr int get_max_power_of_2(int x) {
+    return x % 2 == 0 ? 2*get_max_power_of_2(x/2) : 1;
+}
+
+static_assert(get_max_power_of_2(1) == 1, "Test failed.");
+static_assert(get_max_power_of_2(2) == 2, "Test failed.");
+static_assert(get_max_power_of_2(4) == 4, "Test failed.");
+static_assert(get_max_power_of_2(6) == 2, "Test failed.");
+
+// Number of VKQ rows calculated in parallel:
+constexpr int get_VKQ_stride(int D, int nwarps, int frag_m) {
+    return (get_max_power_of_2(D/frag_m) < nwarps ? get_max_power_of_2(D/frag_m) : nwarps)*frag_m;
+}
+
+static_assert(get_VKQ_stride(128, 1, 32) ==  32, "Test failed.");
+static_assert(get_VKQ_stride(128, 2, 32) ==  64, "Test failed.");
+static_assert(get_VKQ_stride(128, 4, 32) == 128, "Test failed.");
+static_assert(get_VKQ_stride( 64, 1, 32) ==  32, "Test failed.");
+static_assert(get_VKQ_stride( 64, 2, 32) ==  64, "Test failed.");
+static_assert(get_VKQ_stride( 64, 4, 32) ==  64, "Test failed.");
+static_assert(get_VKQ_stride( 80, 1, 16) ==  16, "Test failed.");
+static_assert(get_VKQ_stride( 80, 2, 16) ==  16, "Test failed.");
+static_assert(get_VKQ_stride( 80, 4, 16) ==  16, "Test failed.");
+
+template <int D, int cols_per_block, typename KQ_acc_t>
+void ggml_cuda_flash_attn_ext_wmma_f16_case(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
+    const ggml_tensor * Q = dst->src[0];
+
+    constexpr int nwarps = 4;
+
+    constexpr int frag_m = cols_per_block == 8 && D % 32 == 0 ? 32 : 16;
+    const int blocks_num_pb1 = ((Q->ne[1] + cols_per_block - 1) / cols_per_block)*Q->ne[2]*Q->ne[3];
+    const int nsm = ggml_cuda_info().devices[ggml_cuda_get_device()].nsm;
+
+    if (4*blocks_num_pb1 < 2*nsm) {
+        constexpr int parallel_blocks = 4;
+        fattn_kernel_t fattn_kernel = flash_attn_ext_f16<D, cols_per_block, nwarps, get_VKQ_stride(D, nwarps, frag_m), parallel_blocks, KQ_acc_t>;
+        launch_fattn<D, parallel_blocks>(ctx, dst, fattn_kernel, nwarps, cols_per_block);
+        return;
+    }
+    if (2*blocks_num_pb1 < 2*nsm) {
+        constexpr int parallel_blocks = 2;
+        fattn_kernel_t fattn_kernel = flash_attn_ext_f16<D, cols_per_block, nwarps, get_VKQ_stride(D, nwarps, frag_m), parallel_blocks, KQ_acc_t>;
+        launch_fattn<D, parallel_blocks>(ctx, dst, fattn_kernel, nwarps, cols_per_block);
+        return;
+    }
+    constexpr int parallel_blocks = 1;
+    fattn_kernel_t fattn_kernel = flash_attn_ext_f16<D, cols_per_block, nwarps, get_VKQ_stride(D, nwarps, frag_m), parallel_blocks, KQ_acc_t>;
+    launch_fattn<D, parallel_blocks>(ctx, dst, fattn_kernel, nwarps, cols_per_block);
+}
+
+#define DECL_FATTN_WMMA_F16_CASE(D, cols_per_block, KQ_acc_t)                         \
+    template void ggml_cuda_flash_attn_ext_wmma_f16_case                              \
+    <D, cols_per_block, KQ_acc_t>(ggml_backend_cuda_context & ctx, ggml_tensor * dst) \
+
+extern DECL_FATTN_WMMA_F16_CASE( 64, 16, float);
+extern DECL_FATTN_WMMA_F16_CASE( 80, 16, float);
+extern DECL_FATTN_WMMA_F16_CASE( 96, 16, float);
+extern DECL_FATTN_WMMA_F16_CASE(112, 16, float);
+extern DECL_FATTN_WMMA_F16_CASE(128, 16, float);
+extern DECL_FATTN_WMMA_F16_CASE(256, 16, float);
+
+extern DECL_FATTN_WMMA_F16_CASE( 64, 32, float);
+extern DECL_FATTN_WMMA_F16_CASE( 80, 32, float);
+extern DECL_FATTN_WMMA_F16_CASE( 96, 32, float);
+extern DECL_FATTN_WMMA_F16_CASE(112, 32, float);
+extern DECL_FATTN_WMMA_F16_CASE(128, 32, float);
+// extern DECL_FATTN_WMMA_F16_CASE(256, 16, float);
+
+extern DECL_FATTN_WMMA_F16_CASE( 64,  8, half);
+extern DECL_FATTN_WMMA_F16_CASE( 96,  8, half);
+extern DECL_FATTN_WMMA_F16_CASE(128,  8, half);
+extern DECL_FATTN_WMMA_F16_CASE(256,  8, half);
+
+extern DECL_FATTN_WMMA_F16_CASE( 64, 16, half);
+extern DECL_FATTN_WMMA_F16_CASE( 80, 16, half);
+extern DECL_FATTN_WMMA_F16_CASE( 96, 16, half);
+extern DECL_FATTN_WMMA_F16_CASE(112, 16, half);
+extern DECL_FATTN_WMMA_F16_CASE(128, 16, half);
+extern DECL_FATTN_WMMA_F16_CASE(256, 16, half);
+
+extern DECL_FATTN_WMMA_F16_CASE( 64, 32, half);
+extern DECL_FATTN_WMMA_F16_CASE( 80, 32, half);
+extern DECL_FATTN_WMMA_F16_CASE( 96, 32, half);
+extern DECL_FATTN_WMMA_F16_CASE(112, 32, half);
+extern DECL_FATTN_WMMA_F16_CASE(128, 32, half);
+extern DECL_FATTN_WMMA_F16_CASE(256, 16, half);

ggml-cuda/fattn.cu

@@ -4,519 +4,38 @@
 #include "fattn-tile-f32.cuh"
 #include "fattn-vec-f16.cuh"
 #include "fattn-vec-f32.cuh"
+#include "fattn-wmma-f16.cuh"
 #include "fattn.cuh"
 
 #include <cstdint>
 
-#if FP16_MMA_AVAILABLE
-#include <mma.h>
-#endif
-
-// D == head size, VKQ_stride == num VKQ rows calculated in parallel:
-template<int D, int ncols, int nwarps, int VKQ_stride, int parallel_blocks, typename KQ_acc_t>
-#if !(defined(GGML_USE_HIPBLAS) && defined(__HIP_PLATFORM_AMD__))
-__launch_bounds__(nwarps*WARP_SIZE, 1)
-#endif // !(defined(GGML_USE_HIPBLAS) && defined(__HIP_PLATFORM_AMD__))
-static __global__ void flash_attn_ext_f16(
-        const char * __restrict__ Q,
-        const char * __restrict__ K,
-        const char * __restrict__ V,
-        const char * __restrict__ mask,
-        float      * __restrict__ dst,
-        float2     * __restrict__ dst_meta,
-        const float scale,
-        const float max_bias,
-        const float m0,
-        const float m1,
-        const uint32_t n_head_log2,
-        const int ne00,
-        const int ne01,
-        const int ne02,
-        const int ne03,
-        const int ne10,
-        const int ne11,
-        const int ne12,
-        const int ne13,
-        const int ne31,
-        const int nb31,
-        const int nb01,
-        const int nb02,
-        const int nb03,
-        const int nb11,
-        const int nb12,
-        const int nb13,
-        const int ne0,
-        const int ne1,
-        const int ne2,
-        const int ne3) {
-#if FP16_MMA_AVAILABLE
-    //In this kernel Q, K, V are matrices while i, j, k are matrix indices.
-
-    const int ic0 = ncols*(blockIdx.x / parallel_blocks); // Index of the first Q/QKV column to work on.
-    const int ip  =        blockIdx.x % parallel_blocks;  // Index in group of blocks running for the same column in parallel.
-
-    static_assert(D <= FATTN_KQ_STRIDE, "D must be <= FATTN_KQ_STRIDE.");
-    static_assert(ncols == 8 || ncols % 16 == 0, "ncols must be 8 or a multiple of 16.");
-    constexpr int frag_m = ncols == 8 ? 32 : 16;
-    constexpr int frag_n = ncols == 8 ?  8 : 16;
-    static_assert(D % frag_m == 0, "If ncols == 8 then D % frag_m must be 0.");
-    typedef nvcuda::wmma::fragment<nvcuda::wmma::matrix_a,    frag_m, frag_n, 16, half, nvcuda::wmma::row_major> frag_a_K;
-    typedef nvcuda::wmma::fragment<nvcuda::wmma::matrix_a,    frag_m, frag_n, 16, half, nvcuda::wmma::col_major> frag_a_V;
-    typedef nvcuda::wmma::fragment<nvcuda::wmma::matrix_b,    frag_m, frag_n, 16, half, nvcuda::wmma::col_major> frag_b;
-    typedef nvcuda::wmma::fragment<nvcuda::wmma::accumulator, frag_m, frag_n, 16, KQ_acc_t>                      frag_c_KQ;
-    typedef nvcuda::wmma::fragment<nvcuda::wmma::accumulator, frag_m, frag_n, 16, half>                          frag_c_VKQ;
-
-    constexpr int KQ_stride_tc  = nwarps*frag_m; // Number of KQ rows calculated in parallel.
-    constexpr int VKQ_ratio = KQ_stride_tc/VKQ_stride; // Number of parallel VKQ accumulators needed to keep all warps busy.
-    static_assert(VKQ_ratio <= nwarps, "VKQ_ratio must be <= nwarps.");
-
-    // Pad internal representation of KQ, KQV to reduce shared memory bank conflicts:
-    constexpr int D_padded = D + 8;
-    constexpr int kqs_padded = FATTN_KQ_STRIDE + 8;
-    constexpr int kqar = sizeof(KQ_acc_t)/sizeof(half);
-
-    const int gqa_ratio = ne02 / ne12; // With grouped query attention there are > 1 Q matrices per K, V matrix.
-    const float * Q_f   = (const float *) (Q + nb02* blockIdx.y              + nb01*ic0);
-    const half  * K_h   = (const half  *) (K + nb12*(blockIdx.y / gqa_ratio));
-    const half  * V_h   = (const half  *) (V + nb12*(blockIdx.y / gqa_ratio)); // K and V have same shape
-    const half  * maskh = (const half  *)  mask + (nb31/sizeof(half))* ic0;
-    const half2 * mask2 = (const half2 *)  mask + (nb31/sizeof(half))*(ic0/2);
-
-    const int stride_Q  = nb01 / sizeof(float);
-    const int stride_KV = nb11 / sizeof(half);
-
-    const float slopef = get_alibi_slope(max_bias, blockIdx.y, n_head_log2, m0, m1);
-    const half  slopeh = __float2half(slopef);
-    const half2 slope2 = make_half2(slopef, slopef);
-
-    frag_b Q_b[D/16][ncols/frag_n];
-
-    // A single buffer for temporarily holding tiles of KQ and VKQ parts:
-    constexpr int mem_KQ = ncols*kqs_padded*kqar;
-    constexpr int mem_VKQ_parts = VKQ_ratio*ncols*D_padded;
-    __shared__ half KQ[mem_KQ >= mem_VKQ_parts ? mem_KQ : mem_VKQ_parts];
-    float * KQ_f = (float *) KQ;
-    half2 * KQ2 = (half2 *) KQ;
-
-    float    KQ_rowsum_f[ncols/nwarps] = {0.0f};
-    float       KQ_max_f[ncols/nwarps];
-    float KQ_max_scale_f[ncols/nwarps] = {0.0f};
-
-#pragma unroll
-    for (int j = 0; j < ncols/nwarps; ++j) {
-        KQ_max_f[j] = -FLT_MAX/2.0f;
-    }
-
-    half2    KQ_rowsum_h2[ncols/nwarps] = {{0.0f, 0.0f}};
-    half2       KQ_max_h2[ncols/nwarps];
-    half2 KQ_max_scale_h2[ncols/nwarps] = {{0.0f, 0.0f}};
-
-#pragma unroll
-    for (int j = 0; j < ncols/nwarps; ++j) {
-        KQ_max_h2[j] = make_half2(-HALF_MAX_HALF, -HALF_MAX_HALF);
-    }
-
-    __shared__ half VKQ[ncols*D_padded]; // Accumulator for final VKQ slice.
-    half2 * VKQ2 = (half2 *) VKQ;
-#pragma unroll
-    for (int j0 = 0; j0 < ncols; j0 += nwarps) {
-        const int j = j0 + threadIdx.y;
-#pragma unroll
-        for (int i0 = 0; i0 < D/2; i0 += WARP_SIZE) {
-            const int i = i0 + threadIdx.x;
-            if (i0 + WARP_SIZE > D/2 && i >= D/2) {
-                break;
-            }
-            VKQ2[j*(D_padded/2) + i] = make_half2(0.0f, 0.0f);
-        }
-    }
-
-    // Convert Q to half and apply scale, temporarily store in KQ:
-#pragma unroll
-    for (int j0 = 0; j0 < ncols; j0 += nwarps) {
-        const int j = j0 + threadIdx.y;
-#pragma unroll
-        for (int i0 = 0; i0 < D; i0 += WARP_SIZE) {
-            const int i = i0 + threadIdx.x;
-            if (i0 + WARP_SIZE > D && i >= D) {
-                break;
-            }
-            KQ[j*D_padded + i] = ic0 + j < ne01 ? Q_f[j*stride_Q + i] * scale : 0.0f;
-        }
-    }
-
-    __syncthreads();
-
-    // Load Q into tensor core fragments/registers since it will be used frequently:
-#pragma unroll
-    for (int i0 = 0; i0 < D; i0 += 16) {
-#pragma unroll
-        for (int j0 = 0; j0 < ncols; j0 += frag_n) {
-            nvcuda::wmma::load_matrix_sync(Q_b[i0/16][j0/frag_n], KQ + j0*D_padded + i0, D_padded);
-        }
-    }
-
-    __syncthreads();
-
-    // Iterate over ne11 == previous tokens:
-    for (int k_VKQ_0 = ip*FATTN_KQ_STRIDE; k_VKQ_0 < ne11; k_VKQ_0 += parallel_blocks*FATTN_KQ_STRIDE) {
-        // Calculate tile of KQ:
-#pragma unroll
-        for (int i_KQ_0 = 0; i_KQ_0 < FATTN_KQ_STRIDE; i_KQ_0 += KQ_stride_tc) {
-            frag_c_KQ KQ_c[ncols/frag_n];
-#pragma unroll
-            for (int j = 0; j < ncols/frag_n; ++j) {
-                nvcuda::wmma::fill_fragment(KQ_c[j], 0.0f);
-            }
-#pragma unroll
-            for (int k_KQ_0 = 0; k_KQ_0 < D; k_KQ_0 += 16) {
-                frag_a_K K_a;
-                nvcuda::wmma::load_matrix_sync(K_a, K_h + (k_VKQ_0 + i_KQ_0 + frag_m*threadIdx.y)*stride_KV + k_KQ_0, stride_KV);
-#pragma unroll
-                for (int j = 0; j < ncols/frag_n; ++j) {
-                    nvcuda::wmma::mma_sync(KQ_c[j], K_a, Q_b[k_KQ_0/16][j], KQ_c[j]);
-                }
-            }
-#pragma unroll
-            for (int j0 = 0; j0 < ncols; j0 += frag_n) {
-                nvcuda::wmma::store_matrix_sync((KQ_acc_t *) KQ + j0*kqs_padded + i_KQ_0 + frag_m*threadIdx.y, KQ_c[j0/frag_n], kqs_padded, nvcuda::wmma::mem_col_major);
-            }
-        }
-
-        __syncthreads();
-
-        // Calculate softmax for each KQ column using the current max. value.
-        // The divisor is stored in KQ_rowsum and will be applied at the end.
-#pragma unroll
-        for (int j0 = 0; j0 < ncols; j0 += nwarps) {
-            const int j = j0 + threadIdx.y;
-
-            if (std::is_same<KQ_acc_t, float>::value) {
-                float KQ_f_tmp[FATTN_KQ_STRIDE / WARP_SIZE];
-#pragma unroll
-                for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += WARP_SIZE) {
-                    const int k = k0 + threadIdx.x;
-
-                    KQ_f_tmp[k0/WARP_SIZE] = KQ_f[j*kqs_padded + k];
-                }
-
-                float KQ_max_new = KQ_max_f[j0/nwarps];
-#pragma unroll
-                for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += WARP_SIZE) {
-                    const int k = k0 + threadIdx.x;
-
-                    KQ_f_tmp[k0/WARP_SIZE] += mask ? __half2float(slopeh*maskh[j*(nb31/sizeof(half)) + k_VKQ_0 + k]) : 0.0f;
-                    KQ_max_new = max(KQ_max_new, KQ_f_tmp[k0/WARP_SIZE]);
-                }
-                KQ_max_new = warp_reduce_max(KQ_max_new);
-
-                const float diff = KQ_max_f[j0/nwarps] - KQ_max_new;
-                KQ_max_scale_f[j0/nwarps] = expf(diff);
-                if (diff <= SOFTMAX_FTZ_THRESHOLD) {
-                    KQ_max_scale_f[j0/nwarps] = 0.0f;
-                }
-                KQ_max_f[j0/nwarps] = KQ_max_new;
-
-                float KQ_rowsum_add = 0.0f;
-#pragma unroll
-                for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += WARP_SIZE) {
-                    const int k = k0 + threadIdx.x;
-
-                    const float diff = KQ_f_tmp[k0/WARP_SIZE] - KQ_max_f[j0/nwarps];
-                    KQ_f_tmp[k0/WARP_SIZE] = expf(diff);
-                    if (diff <= SOFTMAX_FTZ_THRESHOLD) {
-                        KQ_f_tmp[k0/WARP_SIZE] = 0.0f;
-                    }
-                    KQ_rowsum_add += KQ_f_tmp[k0/WARP_SIZE];
-                    KQ[j*(kqar*kqs_padded) + k] = KQ_f_tmp[k0/WARP_SIZE];
-                }
-                KQ_rowsum_add = warp_reduce_sum(KQ_rowsum_add);
-
-                // Scale previous KQ_rowsum to account for a potential increase in KQ_max:
-                KQ_rowsum_f[j0/nwarps] = KQ_max_scale_f[j0/nwarps]*KQ_rowsum_f[j0/nwarps] + KQ_rowsum_add;
-            } else {
-                half2 KQ2_tmp[FATTN_KQ_STRIDE/(2*WARP_SIZE)];
-#pragma unroll
-                for (int k0 = 0; k0 < FATTN_KQ_STRIDE/2; k0 += WARP_SIZE) {
-                    const int k = k0 + threadIdx.x;
-
-                    KQ2_tmp[k0/WARP_SIZE] = KQ2[j*(kqs_padded/2) + k];
-                }
-
-                half2 KQ_max_new = KQ_max_h2[j0/nwarps];
-#pragma unroll
-                for (int k0 = 0; k0 < FATTN_KQ_STRIDE/2; k0 += WARP_SIZE) {
-                    const int k = k0 + threadIdx.x;
-
-                    KQ2_tmp[k0/WARP_SIZE] += mask ? slope2*mask2[(j*ne11 + k_VKQ_0)/2 + k] : make_half2(0.0f, 0.0f);
-                    KQ_max_new = ggml_cuda_hmax2(KQ_max_new, KQ2_tmp[k0/WARP_SIZE]);
-                }
-                KQ_max_new = __half2half2(warp_reduce_max(ggml_cuda_hmax(__low2half(KQ_max_new), __high2half(KQ_max_new))));
-                const half2 diff = KQ_max_h2[j0/nwarps] - KQ_max_new;
-                KQ_max_scale_h2[j0/nwarps] = h2exp(diff);
-                const uint32_t ftz_mask = __hgt2_mask(diff, make_half2(SOFTMAX_FTZ_THRESHOLD, SOFTMAX_FTZ_THRESHOLD));
-                *((uint32_t *) &KQ_max_scale_h2[j0/nwarps]) &= ftz_mask;
-                KQ_max_h2[j0/nwarps] = KQ_max_new;
-
-                half2 KQ_rowsum_add = make_half2(0.0f, 0.0f);
-#pragma unroll
-                for (int k0 = 0; k0 < FATTN_KQ_STRIDE/2; k0 += WARP_SIZE) {
-                    const int k = k0 + threadIdx.x;
-
-                    const half2 diff = KQ2_tmp[k0/WARP_SIZE] - KQ_max_h2[j0/nwarps];
-                    KQ2_tmp[k0/WARP_SIZE] = h2exp(diff);
-                    const uint32_t ftz_mask = __hgt2_mask(diff, make_half2(SOFTMAX_FTZ_THRESHOLD, SOFTMAX_FTZ_THRESHOLD));
-                    *((uint32_t *) &KQ2_tmp[k0/WARP_SIZE]) &= ftz_mask;
-                    KQ_rowsum_add += KQ2_tmp[k0/WARP_SIZE];
-                    KQ2[j*(kqs_padded/2) + k] = KQ2_tmp[k0/WARP_SIZE];
-                }
-                KQ_rowsum_add = warp_reduce_sum(KQ_rowsum_add);
-
-                // Scale previous KQ_rowsum to account for a potential increase in KQ_max:
-                KQ_rowsum_h2[j0/nwarps] = KQ_max_scale_h2[j0/nwarps]*KQ_rowsum_h2[j0/nwarps] + KQ_rowsum_add;
-            }
-        }
-
-        __syncthreads();
-
-        frag_b KQ_b[FATTN_KQ_STRIDE/(VKQ_ratio*16)][ncols/frag_n];
-#pragma unroll
-        for (int j0 = 0; j0 < ncols; j0 += frag_n) {
-#pragma unroll
-            for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += VKQ_ratio*16) {
-                const int k = k0 + (threadIdx.y % VKQ_ratio)*16;
-                nvcuda::wmma::load_matrix_sync(
-                    KQ_b[k0/(VKQ_ratio*16)][j0/frag_n],
-                    KQ + j0*(kqar*kqs_padded) + k,
-                    kqar*kqs_padded);
-            }
-        }
-
-        frag_c_VKQ VKQ_c[D/VKQ_stride][ncols/frag_n];
-#pragma unroll
-        for (int i_VKQ_0 = 0; i_VKQ_0 < D; i_VKQ_0 += VKQ_stride) {
-#pragma unroll
-            for (int j = 0; j < ncols/frag_n; ++j) {
-                nvcuda::wmma::fill_fragment(VKQ_c[i_VKQ_0/VKQ_stride][j], 0.0f);
-            }
-
-#pragma unroll
-            for (int k0 = 0; k0 < FATTN_KQ_STRIDE; k0 += VKQ_ratio*16) {
-                const int k = k0 + (threadIdx.y % VKQ_ratio)*16;
-
-                frag_a_V v_a;
-                nvcuda::wmma::load_matrix_sync(v_a, V_h + (k_VKQ_0 + k)*stride_KV + i_VKQ_0 + frag_m*(threadIdx.y/VKQ_ratio), stride_KV);
-#pragma unroll
-                for (int j = 0; j < ncols/frag_n; ++j) {
-                    nvcuda::wmma::mma_sync(VKQ_c[i_VKQ_0/VKQ_stride][j], v_a, KQ_b[k0/(VKQ_ratio*16)][j], VKQ_c[i_VKQ_0/VKQ_stride][j]);
-                }
-            }
-        }
-
-        __syncthreads();
-
-        const int offset_k = (threadIdx.y % VKQ_ratio) * (ncols*D_padded);
-#pragma unroll
-        for (int i_KQ_0 = 0; i_KQ_0 < D; i_KQ_0 += VKQ_stride) {
-#pragma unroll
-            for (int j0 = 0; j0 < ncols; j0 += frag_n) {
-                nvcuda::wmma::store_matrix_sync(
-                    KQ + offset_k + j0*D_padded + i_KQ_0 + frag_m*(threadIdx.y/VKQ_ratio),
-                    VKQ_c[i_KQ_0/VKQ_stride][j0/frag_n],
-                    D_padded, nvcuda::wmma::mem_col_major);
-            }
-        }
-
-        __syncthreads();
-
-#pragma unroll
-        for (int j0 = 0; j0 < ncols; j0 += nwarps) {
-            const int j = j0 + threadIdx.y;
-
-            half2 VKQ_scale;
-            if (std::is_same<KQ_acc_t, float>::value) {
-                VKQ_scale = make_half2(KQ_max_scale_f[j0/nwarps], KQ_max_scale_f[j0/nwarps]);
-            } else {
-                VKQ_scale = KQ_max_scale_h2[j0/nwarps];
-            }
-
-#pragma unroll
-            for (int i0 = 0; i0 < D/2; i0 += WARP_SIZE) {
-                const int i = i0 + threadIdx.x;
-                if (i0 + WARP_SIZE > D/2 && i >= D/2) {
-                    break;
-                }
-
-                half2 VKQ_add = make_half2(0.0f, 0.0f);
-#pragma unroll
-                for (int l = 0; l < VKQ_ratio; ++l) {
-                    VKQ_add += KQ2[l*(ncols*D_padded/2) + j*(D_padded/2) + i];
-                }
-                VKQ2[j*(D_padded/2) + i] = VKQ_scale*VKQ2[j*(D_padded/2) + i] + VKQ_add;
-            }
-        }
-
-        __syncthreads();
-    }
-
-#pragma unroll
-    for (int j0 = 0; j0 < ncols; j0 += nwarps) {
-        const int j_VKQ = j0 + threadIdx.y;
-        if (ic0 + j_VKQ >= ne01) {
-            return;
-        }
-        const int j_dst = (ic0 + j_VKQ)*parallel_blocks + ip;
-
-        float KQ_rowsum_j;
-        if (std::is_same<KQ_acc_t, float>::value) {
-            KQ_rowsum_j = KQ_rowsum_f[j0/nwarps];
-        } else {
-            KQ_rowsum_j = __low2float(KQ_rowsum_h2[j0/nwarps]) + __high2float(KQ_rowsum_h2[j0/nwarps]);
-        }
-
-#pragma unroll
-        for (int i0 = 0; i0 < D; i0 += WARP_SIZE) {
-            const int i = i0 + threadIdx.x;
-            if (i0 + WARP_SIZE > D && i >= D) {
-                break;
-            }
-            float dst_val = VKQ[j_VKQ*D_padded + i];
-            if (parallel_blocks == 1) {
-                dst_val /= KQ_rowsum_j;
-            }
-            dst[j_dst*gridDim.y*D + blockIdx.y*D + i] = dst_val;
-        }
-
-        if (parallel_blocks == 1 || threadIdx.x != 0) {
-            continue;
-        }
-
-        float2 dst_meta_val;
-        if (std::is_same<KQ_acc_t, float>::value) {
-            dst_meta_val.x = KQ_max_f[j0/nwarps];
-        } else {
-            dst_meta_val.x = __low2float(KQ_max_h2[j0/nwarps]);
-        }
-        dst_meta_val.y = KQ_rowsum_j;
-        dst_meta[(ic0 + j_VKQ)*gridDim.y*parallel_blocks + blockIdx.y*parallel_blocks + ip] = dst_meta_val;
-    }
-#else
-   NO_DEVICE_CODE;
-#endif // FP16_MMA_AVAILABLE
-}
-
-constexpr int get_max_power_of_2(int x) {
-    return x % 2 == 0 ? 2*get_max_power_of_2(x/2) : 1;
-}
-
-static_assert(get_max_power_of_2(1) == 1, "Test failed.");
-static_assert(get_max_power_of_2(2) == 2, "Test failed.");
-static_assert(get_max_power_of_2(4) == 4, "Test failed.");
-static_assert(get_max_power_of_2(6) == 2, "Test failed.");
-
-// Number of VKQ rows calculated in parallel:
-constexpr int get_VKQ_stride(int D, int nwarps, int frag_m) {
-    return (get_max_power_of_2(D/frag_m) < nwarps ? get_max_power_of_2(D/frag_m) : nwarps)*frag_m;
-}
-
-static_assert(get_VKQ_stride(128, 1, 32) ==  32, "Test failed.");
-static_assert(get_VKQ_stride(128, 2, 32) ==  64, "Test failed.");
-static_assert(get_VKQ_stride(128, 4, 32) == 128, "Test failed.");
-static_assert(get_VKQ_stride( 64, 1, 32) ==  32, "Test failed.");
-static_assert(get_VKQ_stride( 64, 2, 32) ==  64, "Test failed.");
-static_assert(get_VKQ_stride( 64, 4, 32) ==  64, "Test failed.");
-static_assert(get_VKQ_stride( 80, 1, 16) ==  16, "Test failed.");
-static_assert(get_VKQ_stride( 80, 2, 16) ==  16, "Test failed.");
-static_assert(get_VKQ_stride( 80, 4, 16) ==  16, "Test failed.");
-
-template <int D, int cols_per_block, int nwarps, typename KQ_acc_t>
-void launch_fattn_f16(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
-    const ggml_tensor * Q = dst->src[0];
-
-    constexpr int frag_m = cols_per_block == 8 && D % 32 == 0 ? 32 : 16;
-    const int blocks_num_pb1 = ((Q->ne[1] + cols_per_block - 1) / cols_per_block)*Q->ne[2]*Q->ne[3];
-    const int nsm = ggml_cuda_info().devices[ggml_cuda_get_device()].nsm;
-
-    if (4*blocks_num_pb1 < 2*nsm) {
-        constexpr int parallel_blocks = 4;
-        fattn_kernel_t fattn_kernel = flash_attn_ext_f16<D, cols_per_block, nwarps, get_VKQ_stride(D, nwarps, frag_m), parallel_blocks, KQ_acc_t>;
-        launch_fattn<D, parallel_blocks>(ctx, dst, fattn_kernel, nwarps, cols_per_block);
-        return;
-    }
-    if (2*blocks_num_pb1 < 2*nsm) {
-        constexpr int parallel_blocks = 2;
-        fattn_kernel_t fattn_kernel = flash_attn_ext_f16<D, cols_per_block, nwarps, get_VKQ_stride(D, nwarps, frag_m), parallel_blocks, KQ_acc_t>;
-        launch_fattn<D, parallel_blocks>(ctx, dst, fattn_kernel, nwarps, cols_per_block);
-        return;
-    }
-    constexpr int parallel_blocks = 1;
-    fattn_kernel_t fattn_kernel = flash_attn_ext_f16<D, cols_per_block, nwarps, get_VKQ_stride(D, nwarps, frag_m), parallel_blocks, KQ_acc_t>;
-    launch_fattn<D, parallel_blocks>(ctx, dst, fattn_kernel, nwarps, cols_per_block);
-}
-
-void ggml_cuda_flash_attn_ext(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
+static void ggml_cuda_flash_attn_ext_wmma_f16(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
     const ggml_tensor * KQV = dst;
     const ggml_tensor * Q   = dst->src[0];
 
-    ggml_cuda_set_device(ctx.device);
-    const int cc = ggml_cuda_info().devices[ggml_cuda_get_device()].cc;
     const int32_t precision = KQV->op_params[2];
 
-    // On AMD the tile kernels perform poorly, use the vec kernel instead:
-    if (cc >= CC_OFFSET_AMD) {
-        if (precision == GGML_PREC_DEFAULT) {
-            ggml_cuda_flash_attn_ext_vec_f16_no_mma(ctx, dst);
-        } else {
-            ggml_cuda_flash_attn_ext_vec_f32(ctx, dst);
-        }
-        return;
-    }
-
-    if (!fast_fp16_available(cc)) {
-        if (Q->ne[1] <= 8) {
-            ggml_cuda_flash_attn_ext_vec_f32(ctx, dst);
-        } else {
-            ggml_cuda_flash_attn_ext_tile_f32(ctx, dst);
-        }
-        return;
-    }
-
-    if (!fp16_mma_available(cc)) {
-        if (Q->ne[1] <= 8) {
-            ggml_cuda_flash_attn_ext_vec_f16_no_mma(ctx, dst);
-        } else {
-            ggml_cuda_flash_attn_ext_tile_f16(ctx, dst);
-        }
-        return;
-    }
-
     if (precision != GGML_PREC_DEFAULT) {
-        if (Q->ne[1] == 1 && (Q->ne[0] == 64 || Q->ne[0] == 128)) {
-            ggml_cuda_flash_attn_ext_vec_f32(ctx, dst);
-            return;
-        }
-
         if (Q->ne[1] <= 32 || Q->ne[0] > 128) {
             constexpr int cols_per_block = 16;
-            constexpr int nwarps         =  4;
             switch (Q->ne[0]) {
                 case 64:
-                    launch_fattn_f16< 64, cols_per_block, nwarps, float>(ctx, dst);
+                    ggml_cuda_flash_attn_ext_wmma_f16_case< 64, cols_per_block, float>(ctx, dst);
                     break;
                 case 80:
-                    launch_fattn_f16< 80, cols_per_block, nwarps, float>(ctx, dst);
+                    ggml_cuda_flash_attn_ext_wmma_f16_case< 80, cols_per_block, float>(ctx, dst);
                     break;
                 case 96:
-                    launch_fattn_f16< 96, cols_per_block, nwarps, float>(ctx, dst);
+                    ggml_cuda_flash_attn_ext_wmma_f16_case< 96, cols_per_block, float>(ctx, dst);
                     break;
                 case 112:
-                    launch_fattn_f16<112, cols_per_block, nwarps, float>(ctx, dst);
+                    ggml_cuda_flash_attn_ext_wmma_f16_case<112, cols_per_block, float>(ctx, dst);
                     break;
                 case 128:
-                    launch_fattn_f16<128, cols_per_block, nwarps, float>(ctx, dst);
+                    ggml_cuda_flash_attn_ext_wmma_f16_case<128, cols_per_block, float>(ctx, dst);
                     break;
                 case 256:
-                    launch_fattn_f16<256, cols_per_block, nwarps, float>(ctx, dst);
+                    ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, float>(ctx, dst);
                     break;
                 default:
                     GGML_ASSERT(false);
@@ -524,25 +43,24 @@ void ggml_cuda_flash_attn_ext(ggml_backend_cuda_context & ctx, ggml_tensor * dst
             }
         } else {
             constexpr int cols_per_block = 32;
-            constexpr int nwarps         =  4;
             switch (Q->ne[0]) {
                 case 64:
-                    launch_fattn_f16< 64, cols_per_block, nwarps, float>(ctx, dst);
+                    ggml_cuda_flash_attn_ext_wmma_f16_case< 64, cols_per_block, float>(ctx, dst);
                     break;
                 case 80:
-                    launch_fattn_f16< 80, cols_per_block, nwarps, float>(ctx, dst);
+                    ggml_cuda_flash_attn_ext_wmma_f16_case< 80, cols_per_block, float>(ctx, dst);
                     break;
                 case 96:
-                    launch_fattn_f16< 96, cols_per_block, nwarps, float>(ctx, dst);
+                    ggml_cuda_flash_attn_ext_wmma_f16_case< 96, cols_per_block, float>(ctx, dst);
                     break;
                 case 112:
-                    launch_fattn_f16<112, cols_per_block, nwarps, float>(ctx, dst);
+                    ggml_cuda_flash_attn_ext_wmma_f16_case<112, cols_per_block, float>(ctx, dst);
                     break;
                 case 128:
-                    launch_fattn_f16<128, cols_per_block, nwarps, float>(ctx, dst);
+                    ggml_cuda_flash_attn_ext_wmma_f16_case<128, cols_per_block, float>(ctx, dst);
                     break;
                 // case 256:
-                //     launch_fattn_f16<256, cols_per_block, nwarps, float>(ctx, dst);
+                //     ggml_cuda_flash_attn_ext_wmma_f16_case<128, cols_per_block, float>(ctx, dst);
                 //     break;
                 default:
                     GGML_ASSERT(false);
@@ -552,26 +70,20 @@ void ggml_cuda_flash_attn_ext(ggml_backend_cuda_context & ctx, ggml_tensor * dst
         return;
     }
 
-    if (Q->ne[1] == 1 && Q->ne[0] % (2*WARP_SIZE) == 0) {
-        ggml_cuda_flash_attn_ext_vec_f16(ctx, dst);
-        return;
-    }
-
     if (Q->ne[1] <= 8 && Q->ne[0] % WARP_SIZE == 0) {
         constexpr int cols_per_block = 8;
-        constexpr int nwarps         = 4;
         switch (Q->ne[0]) {
             case 64:
-                launch_fattn_f16< 64, cols_per_block, nwarps, half>(ctx, dst);
+                ggml_cuda_flash_attn_ext_wmma_f16_case< 64, cols_per_block, half>(ctx, dst);
                 break;
             case 96:
-                launch_fattn_f16< 96, cols_per_block, nwarps, half>(ctx, dst);
+                ggml_cuda_flash_attn_ext_wmma_f16_case< 96, cols_per_block, half>(ctx, dst);
                 break;
             case 128:
-                launch_fattn_f16<128, cols_per_block, nwarps, half>(ctx, dst);
+                ggml_cuda_flash_attn_ext_wmma_f16_case<128, cols_per_block, half>(ctx, dst);
                 break;
             case 256:
-                launch_fattn_f16<256, cols_per_block, nwarps, half>(ctx, dst);
+                ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst);
                 break;
             default:
                 GGML_ASSERT(false);
@@ -582,25 +94,24 @@ void ggml_cuda_flash_attn_ext(ggml_backend_cuda_context & ctx, ggml_tensor * dst
 
     if (Q->ne[1] <= 32) {
         constexpr int cols_per_block = 16;
-        constexpr int nwarps         =  4;
         switch (Q->ne[0]) {
             case 64:
-                launch_fattn_f16< 64, cols_per_block, nwarps, half>(ctx, dst);
+                ggml_cuda_flash_attn_ext_wmma_f16_case< 64, cols_per_block, half>(ctx, dst);
                 break;
             case 80:
-                launch_fattn_f16< 80, cols_per_block, nwarps, half>(ctx, dst);
+                ggml_cuda_flash_attn_ext_wmma_f16_case< 80, cols_per_block, half>(ctx, dst);
                 break;
             case 96:
-                launch_fattn_f16< 96, cols_per_block, nwarps, half>(ctx, dst);
+                ggml_cuda_flash_attn_ext_wmma_f16_case< 96, cols_per_block, half>(ctx, dst);
                 break;
             case 112:
-                launch_fattn_f16<112, cols_per_block, nwarps, half>(ctx, dst);
+                ggml_cuda_flash_attn_ext_wmma_f16_case<112, cols_per_block, half>(ctx, dst);
                 break;
             case 128:
-                launch_fattn_f16<128, cols_per_block, nwarps, half>(ctx, dst);
+                ggml_cuda_flash_attn_ext_wmma_f16_case<128, cols_per_block, half>(ctx, dst);
                 break;
             case 256:
-                launch_fattn_f16<256, cols_per_block, nwarps, half>(ctx, dst);
+                ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst);
                 break;
             default:
                 GGML_ASSERT(false);
@@ -610,29 +121,229 @@ void ggml_cuda_flash_attn_ext(ggml_backend_cuda_context & ctx, ggml_tensor * dst
     }
 
     constexpr int cols_per_block = 32;
-    constexpr int nwarps         =  4;
     switch (Q->ne[0]) {
         case 64:
-            launch_fattn_f16< 64, cols_per_block, nwarps, half>(ctx, dst);
+            ggml_cuda_flash_attn_ext_wmma_f16_case< 64, cols_per_block, half>(ctx, dst);
             break;
         case 80:
-            launch_fattn_f16< 80, cols_per_block, nwarps, half>(ctx, dst);
+            ggml_cuda_flash_attn_ext_wmma_f16_case< 80, cols_per_block, half>(ctx, dst);
             break;
         case 96:
-            launch_fattn_f16< 96, cols_per_block, nwarps, half>(ctx, dst);
+            ggml_cuda_flash_attn_ext_wmma_f16_case< 96, cols_per_block, half>(ctx, dst);
             break;
         case 112:
-            launch_fattn_f16<112, cols_per_block, nwarps, half>(ctx, dst);
+            ggml_cuda_flash_attn_ext_wmma_f16_case<112, cols_per_block, half>(ctx, dst);
             break;
         case 128:
-            launch_fattn_f16<128, cols_per_block, nwarps, half>(ctx, dst);
+            ggml_cuda_flash_attn_ext_wmma_f16_case<128, cols_per_block, half>(ctx, dst);
             break;
         case 256:
-            launch_fattn_f16<256, cols_per_block, nwarps, half>(ctx, dst);
+            ggml_cuda_flash_attn_ext_wmma_f16_case<256, cols_per_block, half>(ctx, dst);
             break;
         default:
             GGML_ASSERT(false);
             break;
     }
-    return;
+}
+#define FATTN_VEC_F16_CASE(D, type_K, type_V)                               \
+    if (Q->ne[0] == (D) && K->type == (type_K) && V->type == (type_V)) {    \
+        ggml_cuda_flash_attn_ext_vec_f16_case<D, type_K, type_V>(ctx, dst); \
+        return;                                                             \
+    }                                                                       \
+
+static void ggml_cuda_flash_attn_ext_vec_f16(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
+    ggml_tensor * Q = dst->src[1];
+    ggml_tensor * K = dst->src[1];
+    ggml_tensor * V = dst->src[2];
+
+#ifdef GGML_CUDA_FA_ALL_QUANTS
+    FATTN_VEC_F16_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q4_0)
+    FATTN_VEC_F16_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q4_1)
+    FATTN_VEC_F16_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q5_0)
+    FATTN_VEC_F16_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q5_1)
+    FATTN_VEC_F16_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q8_0)
+    FATTN_VEC_F16_CASE( 64, GGML_TYPE_F16, GGML_TYPE_F16 )
+
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q4_0)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q4_0)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q4_0)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q4_0)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q4_0)
+
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_1)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q4_1)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q4_1)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q4_1)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q4_1)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q4_1)
+
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q5_0)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q5_0)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q5_0)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q5_0)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q5_0)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q5_0)
+
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q5_1)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q5_1)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q5_1)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q5_1)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q5_1)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q5_1)
+
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q8_0)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q8_0)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q8_0)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q8_0)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q8_0)
+
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_F16)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_F16)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_F16)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_F16)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_F16)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_F16,  GGML_TYPE_F16)
+
+    FATTN_VEC_F16_CASE(256, GGML_TYPE_F16, GGML_TYPE_F16)
+#else
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0)
+
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0)
+
+    FATTN_VEC_F16_CASE( 64, GGML_TYPE_F16, GGML_TYPE_F16)
+    FATTN_VEC_F16_CASE(128, GGML_TYPE_F16, GGML_TYPE_F16)
+    FATTN_VEC_F16_CASE(256, GGML_TYPE_F16, GGML_TYPE_F16)
+#endif // GGML_CUDA_FA_ALL_QUANTS
+
+    on_no_fattn_vec_case(Q->ne[0]);
+}
+
+#define FATTN_VEC_F32_CASE(D, type_K, type_V)                               \
+    if (Q->ne[0] == (D) && K->type == (type_K) && V->type == (type_V)) {    \
+        ggml_cuda_flash_attn_ext_vec_f32_case<D, type_K, type_V>(ctx, dst); \
+        return;                                                             \
+    }                                                                       \
+
+static void ggml_cuda_flash_attn_ext_vec_f32(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
+    ggml_tensor * Q = dst->src[1];
+    ggml_tensor * K = dst->src[1];
+    ggml_tensor * V = dst->src[2];
+
+#ifdef GGML_CUDA_FA_ALL_QUANTS
+    FATTN_VEC_F32_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q4_0)
+    FATTN_VEC_F32_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q4_1)
+    FATTN_VEC_F32_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q5_0)
+    FATTN_VEC_F32_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q5_1)
+    FATTN_VEC_F32_CASE( 64, GGML_TYPE_F16, GGML_TYPE_Q8_0)
+    FATTN_VEC_F32_CASE( 64, GGML_TYPE_F16, GGML_TYPE_F16)
+
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q4_0)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q4_0)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q4_0)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q4_0)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q4_0)
+
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_1)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q4_1)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q4_1)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q4_1)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q4_1)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q4_1)
+
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q5_0)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q5_0)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q5_0)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q5_0)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q5_0)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q5_0)
+
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q5_1)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q5_1)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q5_1)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q5_1)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q5_1)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q5_1)
+
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q8_0)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q8_0)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q8_0)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q8_0)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_F16,  GGML_TYPE_Q8_0)
+
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_F16)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_F16)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_F16)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_F16)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_F16)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_F16,  GGML_TYPE_F16)
+
+    FATTN_VEC_F32_CASE(256, GGML_TYPE_F16, GGML_TYPE_F16)
+#else
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0)
+
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0)
+
+    FATTN_VEC_F32_CASE( 64, GGML_TYPE_F16, GGML_TYPE_F16)
+    FATTN_VEC_F32_CASE(128, GGML_TYPE_F16, GGML_TYPE_F16)
+    FATTN_VEC_F32_CASE(256, GGML_TYPE_F16, GGML_TYPE_F16)
+#endif // GGML_CUDA_FA_ALL_QUANTS
+
+    on_no_fattn_vec_case(Q->ne[0]);
+}
+
+void ggml_cuda_flash_attn_ext(ggml_backend_cuda_context & ctx, ggml_tensor * dst) {
+    const ggml_tensor * KQV = dst;
+    const ggml_tensor * Q   = dst->src[0];
+    const ggml_tensor * K   = dst->src[1];
+    const ggml_tensor * V   = dst->src[2];
+
+    ggml_cuda_set_device(ctx.device);
+    const int cc = ggml_cuda_info().devices[ggml_cuda_get_device()].cc;
+    const int32_t precision = KQV->op_params[2];
+
+    const bool quantized_KV = ggml_is_quantized(K->type) || ggml_is_quantized(V->type);
+
+    // On AMD the tile kernels perform poorly, use the vec kernel instead:
+    if (cc >= CC_OFFSET_AMD || quantized_KV) {
+        if (precision == GGML_PREC_DEFAULT && fast_fp16_available(cc)) {
+            ggml_cuda_flash_attn_ext_vec_f16(ctx, dst);
+        } else {
+            ggml_cuda_flash_attn_ext_vec_f32(ctx, dst);
+        }
+        return;
+    }
+
+    if (!fast_fp16_available(cc)) {
+        if (Q->ne[1] <= 8) {
+            ggml_cuda_flash_attn_ext_vec_f32(ctx, dst);
+        } else {
+            ggml_cuda_flash_attn_ext_tile_f32(ctx, dst);
+        }
+        return;
+    }
+
+    if (!fp16_mma_available(cc)) {
+        if (Q->ne[1] <= 8) {
+            ggml_cuda_flash_attn_ext_vec_f16(ctx, dst);
+        } else {
+            ggml_cuda_flash_attn_ext_tile_f16(ctx, dst);
+        }
+        return;
+    }
+
+    if (Q->ne[1] == 1 && Q->ne[0] % (2*WARP_SIZE) == 0) {
+        if (precision == GGML_PREC_DEFAULT) {
+            ggml_cuda_flash_attn_ext_vec_f16(ctx, dst);
+            return;
+        } else if(Q->ne[0] <= 128) {
+            ggml_cuda_flash_attn_ext_vec_f32(ctx, dst);
+            return;
+        }
+    }
+
+    ggml_cuda_flash_attn_ext_wmma_f16(ctx, dst);
 }

ggml-cuda/mmq.cu

@@ -386,7 +386,7 @@ static __device__ __forceinline__ float vec_dot_q5_0_q8_1_mul_mat(
         u[2*l+1] = y_qs[j * WARP_SIZE + (kyqs + l + QI5_0) % WARP_SIZE];
     }
 
-    return vec_dot_q8_0_q8_1_impl<QR5_0*VDR_Q5_0_Q8_1_MMQ>
+    return vec_dot_q8_0_q8_1_impl<float, QR5_0*VDR_Q5_0_Q8_1_MMQ>
         (&x_ql[i * (2*WARP_SIZE + 1) + 2 * k], u, x_dmf[index_bx], y_df[j * (WARP_SIZE/QI8_1) + (2*k/QI8_1) % (WARP_SIZE/QI8_1)]);
 }
 
@@ -547,7 +547,7 @@ static __device__ __forceinline__ float vec_dot_q8_0_q8_1_mul_mat(
     const float * x_dmf = (const float *) x_dm;
     const float * y_df  = (const float *) y_ds;
 
-    return vec_dot_q8_0_q8_1_impl<VDR_Q8_0_Q8_1_MMQ>
+    return vec_dot_q8_0_q8_1_impl<float, VDR_Q8_0_Q8_1_MMQ>
         (&x_ql[i * (WARP_SIZE + 1) + k], &y_qs[j * WARP_SIZE + k], x_dmf[i * (WARP_SIZE/QI8_0) + i/QI8_0 + k/QI8_0],
          y_df[j * (WARP_SIZE/QI8_1) + k/QI8_1]);
 }

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-f16.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_F16, GGML_TYPE_F16);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q4_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q4_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q4_1.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q4_1);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q5_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q5_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q5_1.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q5_1);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-f16-q8_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_F16, GGML_TYPE_Q8_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-f16.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_F16);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q4_1.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q4_1);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q5_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q5_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q5_1.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q5_1);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_0-q8_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_0, GGML_TYPE_Q8_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-f16.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_F16);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q4_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q4_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q4_1.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q4_1);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q5_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q5_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q5_1.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q5_1);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q4_1-q8_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q4_1, GGML_TYPE_Q8_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-f16.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_F16);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q4_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q4_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q4_1.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q4_1);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q5_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q5_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q5_1.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q5_1);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_0-q8_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_0, GGML_TYPE_Q8_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-f16.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_F16);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q4_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q4_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q4_1.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q4_1);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q5_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q5_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q5_1.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q5_1);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q5_1-q8_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q5_1, GGML_TYPE_Q8_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-f16.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_F16);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q4_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q4_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q4_1.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q4_1);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q5_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q5_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q5_1.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q5_1);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs128-q8_0-q8_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(128, GGML_TYPE_Q8_0, GGML_TYPE_Q8_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs256-f16-f16.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(256, GGML_TYPE_F16, GGML_TYPE_F16);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-f16.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(64, GGML_TYPE_F16, GGML_TYPE_F16);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q4_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(64, GGML_TYPE_F16, GGML_TYPE_Q4_0);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q4_1.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(64, GGML_TYPE_F16, GGML_TYPE_Q4_1);

ggml-cuda/template-instances/fattn-vec-f16-instance-hs64-f16-q5_0.cu

@@ -0,0 +1,5 @@
+// This file has been autogenerated by generate-variants.py, do not edit manually.
+
+#include "../fattn-vec-f16.cuh"
+
+DECL_FATTN_VEC_F16_CASE(64, GGML_TYPE_F16, GGML_TYPE_Q5_0);