whispercpp 1.3.7 → 1.3.8

data/ext/sources/ggml/src/ggml-hexagon/htp/hmx-ops.c

@@ -1,6 +0,0 @@
-// HMX operations compiled as a single translation unit.
-// This allows interprocedural optimizations within HMX ops without requiring global HTP LTO.
-
-#include "hmx-queue.c"
-#include "hmx-matmul-ops.c"
-#include "hmx-flash-attn-ops.c"

data/ext/sources/ggml/src/ggml-hexagon/htp/hmx-ops.h

@@ -1,88 +0,0 @@
-// HMX operation entry-point declarations.
-// Ported from htp-ops-lib/include/dsp/ops.h (renamed, benchmark kernels removed). (https://github.com/haozixu/htp-ops-lib)
-
-#ifndef HMX_OPS_H
-#define HMX_OPS_H
-
-#include <stddef.h>
-#include <stdint.h>
-
-#include "htp-ops.h"
-
-#ifdef __cplusplus
-extern "C" {
-#endif
-
-typedef struct {
-    float        *dst;
-    const float  *activation;
-    const __fp16 *permuted_weight;
-    int           m;
-    int           k;
-    int           n;
-    int           act_stride;
-    int           weight_stride;
-    int           dst_stride;
-    int           ne02;
-    int           ne03;
-    int           ne12;
-    int           ne13;
-    size_t        src0_nb2;
-    size_t        src0_nb3;
-    size_t        src1_nb2;
-    size_t        src1_nb3;
-    size_t        dst_nb2;
-    size_t        dst_nb3;
-} hmx_matmul_f16_f32_batched_params_t;
-
-// HMX matrix multiplication — tile-permuted FP16 weights, FP32 activation/output
-// act_stride: activation row stride in elements (= k for contiguous, or
-//             nb[1]/sizeof(float) for permuted tensors like attention Q).
-// weight_stride: weight row stride in elements (= k for compact weights, or
-//                nb[1]/sizeof(__fp16) for permuted KV-cache views used by QK).
-int hmx_matmul_f16_f32(struct htp_context *ctx,
-                                float *restrict dst,
-                                const float *activation,
-                                const __fp16 *permuted_weight,
-                                int m, int k, int n,
-                                int act_stride,
-                                int weight_stride);
-
-// Batched F16 wrapper over hmx_mat_mul_f16_f32.
-// Batch semantics match ggml_mul_mat(): src0 broadcasts to src1 in dims 2/3.
-int hmx_matmul_f16_f32_batched(struct htp_context *ctx, const hmx_matmul_f16_f32_batched_params_t *params);
-
-// HMX matrix multiplication — all supported weight types (F16/F32/Q4_0/Q4_1/Q8_0/IQ4_NL/MXFP4)
-int hmx_matmul_2d_f32(struct htp_context *ctx,
-                                      float *restrict dst,
-                                      const float *activation,
-                                      const uint8_t *permuted_weight,
-                                      int m, int k, int n,
-                                      int act_stride,
-                                      int weight_stride,
-                                      int weight_type);
-
-struct mmid_row_mapping;
-
-int hmx_matmul_id_2d_f32(struct htp_context *ctx,
-                                         float *restrict dst,
-                                         const float *activation,
-                                         const uint8_t *permuted_weight,
-                                         int m, int k, int n,
-                                         int ne11,
-                                         size_t act_nb1, size_t act_nb2,
-                                         size_t dst_nb1, size_t dst_nb2,
-                                         int weight_stride,
-                                         int weight_type,
-                                         const struct mmid_row_mapping *matrix_rows,
-                                         int cur_a,
-                                         int mapping_stride);
-
-// HMX flash attention
-int hmx_flash_attn_ext(struct htp_ops_context * octx);
-
-#ifdef __cplusplus
-}
-#endif
-
-#endif // HMX_OPS_H

data/ext/sources/ggml/src/ggml-hexagon/htp/hmx-profile.h

@@ -1,34 +0,0 @@
-// Conditional fine-grained profiling macros for HMX operations.
-//
-// Define ENABLE_PROFILE_TIMERS (via compiler flag or before including this
-// header) to instrument sub-operation latencies with HAP qtimer.  When the
-// macro is not defined the TIMER_* helpers expand to nothing so there is zero
-// overhead.
-//
-// Usage:
-//   TIMER_DEFINE(my_phase);          // declare accumulator variable
-//   TIMER_START(my_phase);           // snapshot start time
-//   ... work ...
-//   TIMER_STOP(my_phase);            // accumulate elapsed ticks
-//   FARF(ALWAYS, "my_phase: %lld us", TIMER_US(my_phase));
-
-#ifndef HMX_PROFILE_H
-#define HMX_PROFILE_H
-
-#include <HAP_perf.h>
-
-// #define ENABLE_PROFILE_TIMERS
-
-#if defined(ENABLE_PROFILE_TIMERS)
-#  define TIMER_DEFINE(name) int64_t name##_ticks = 0
-#  define TIMER_START(name)  int64_t name##_t0 = HAP_perf_get_qtimer_count()
-#  define TIMER_STOP(name)   name##_ticks += HAP_perf_get_qtimer_count() - name##_t0
-#  define TIMER_US(name)     HAP_perf_qtimer_count_to_us(name##_ticks)
-#else
-#  define TIMER_DEFINE(name)
-#  define TIMER_START(name)
-#  define TIMER_STOP(name)
-#  define TIMER_US(name)     0LL
-#endif
-
-#endif // HMX_PROFILE_H

data/ext/sources/ggml/src/ggml-hexagon/htp/vtcm-utils.h

@@ -1,16 +0,0 @@
-#ifndef VTCM_UTILS_H
-#define VTCM_UTILS_H
-
-#include "hex-utils.h"
-
-#include <assert.h>
-#include <stdint.h>
-#include <hexagon_types.h>
-
-static inline uint8_t *vtcm_seq_alloc(uint8_t **vtcm_ptr, size_t size) {
-    uint8_t *p = *vtcm_ptr;
-    *vtcm_ptr += size;
-    return p;
-}
-
-#endif // VTCM_UTILS_H

data/ext/sources/ggml/src/ggml-openvino/openvino/op/unary_gelu.cpp

@@ -1,25 +0,0 @@
-#include "../node_context.h"
-#include "../op_table.h"
-#include "../utils.h"
-
-#include <openvino/core/node_output.hpp>
-#include <openvino/op/gelu.hpp>
-
-namespace ov {
-namespace frontend {
-namespace ggml {
-namespace op {
-
-OutputVector translate_unary_gelu(const NodeContext & context) {
-    num_inputs_check(context, 1, 1);
-
-    auto input = context.get_input(0);
-    auto res = std::make_shared<ov::op::v7::Gelu>(input);
-
-    return rename_outputs_with_suffix({res}, context.get_name());
-}
-
-}  // namespace op
-}  // namespace ggml
-}  // namespace frontend
-}  // namespace ov

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/abs.comp

@@ -1,21 +0,0 @@
-#version 450
-
-#include "generic_head.glsl"
-#include "types.glsl"
-
-#extension GL_EXT_control_flow_attributes : enable
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
-layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
-
-void main() {
-    const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
-
-    if (i >= p.KX) {
-        return;
-    }
-
-    data_d[i] = D_TYPE(abs(float(data_a[i])));
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/ceil.comp

@@ -1,22 +0,0 @@
-#version 450
-
-#include "generic_head.glsl"
-#include "types.glsl"
-
-#extension GL_EXT_control_flow_attributes : enable
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
-layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
-
-void main() {
-    const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
-
-    if (i >= p.KX) {
-        return;
-    }
-
-    const float x = float(data_a[i]);
-    data_d[i] = D_TYPE(ceil(x));
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/clamp.comp

@@ -1,17 +0,0 @@
-#version 450
-
-#include "types.glsl"
-#include "generic_unary_head.glsl"
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-void main() {
-    const uint idx = get_idx();
-
-    if (idx >= p.ne) {
-        return;
-    }
-
-    const FLOAT_TYPE val = FLOAT_TYPE(data_a[get_aoffset() + src0_idx(idx)]);
-    data_d[get_doffset() + dst_idx(idx)] = D_TYPE(val < p.param1 ? p.param1 : (val > p.param2 ? p.param2 : val));
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/cos.comp

@@ -1,17 +0,0 @@
-#version 450
-
-#include "types.glsl"
-#include "generic_unary_head.glsl"
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-void main() {
-    const uint idx = get_idx();
-
-    if (idx >= p.ne) {
-        return;
-    }
-
-    const FLOAT_TYPE val = FLOAT_TYPE(data_a[get_aoffset() + src0_idx(idx)]);
-    data_d[get_doffset() + dst_idx(idx)] = D_TYPE(cos(val));
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/elu.comp

@@ -1,27 +0,0 @@
-#version 450
-
-#include "generic_head.glsl"
-#include "types.glsl"
-
-#extension GL_EXT_control_flow_attributes : enable
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
-layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
-
-void main() {
-    const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
-
-    if (i >= p.KX) {
-        return;
-    }
-
-    float x = float(data_a[i]);
-
-    if (x < 0.0f) {
-        x = exp(x) - 1;
-    }
-
-    data_d[i] = D_TYPE(x);
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/exp.comp

@@ -1,20 +0,0 @@
-#version 450
-
-#include "generic_head.glsl"
-#include "types.glsl"
-
-#extension GL_EXT_control_flow_attributes : enable
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
-layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
-
-void main() {
-    const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
-
-    if (i >= p.KX) {
-        return;
-    }
-    data_d[i] = D_TYPE(exp(float(data_a[i])));
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/floor.comp

@@ -1,22 +0,0 @@
-#version 450
-
-#include "generic_head.glsl"
-#include "types.glsl"
-
-#extension GL_EXT_control_flow_attributes : enable
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
-layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
-
-void main() {
-    const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
-
-    if (i >= p.KX) {
-        return;
-    }
-
-    const float x = float(data_a[i]);
-    data_d[i] = D_TYPE(floor(x));
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/gelu.comp

@@ -1,25 +0,0 @@
-#version 450
-
-#include "generic_head.glsl"
-#include "types.glsl"
-
-#extension GL_EXT_control_flow_attributes : enable
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
-layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
-
-void main() {
-    const float GELU_COEF_A    = 0.044715f;
-    const float SQRT_2_OVER_PI = 0.79788456080286535587989211986876f;
-    const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
-
-    if (i >= p.KX) {
-        return;
-    }
-
-    const float xi = float(data_a[i]);
-    const float val = SQRT_2_OVER_PI*xi*(1.0f + GELU_COEF_A*xi*xi);
-    data_d[i] = D_TYPE(0.5f*xi*(2.0f - 2.0f / (exp(2 * val) + 1)));
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/gelu_erf.comp

@@ -1,39 +0,0 @@
-#version 450
-
-#include "generic_head.glsl"
-#include "types.glsl"
-
-#extension GL_EXT_control_flow_attributes : enable
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
-layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
-
-void main() {
-    // based on Abramowitz and Stegun formula 7.1.26 or similar Hastings' approximation
-    // ref: https://www.johndcook.com/blog/python_erf/
-    const float p_erf  = 0.3275911f;
-    const float a1_erf = 0.254829592f;
-    const float a2_erf = -0.284496736f;
-    const float a3_erf = 1.421413741f;
-    const float a4_erf = -1.453152027f;
-    const float a5_erf = 1.061405429f;
-
-    const float SQRT_2_INV = 0.70710678118654752440084436210484f;
-    const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
-
-    if (i >= p.KX) {
-        return;
-    }
-
-    const float a = float(data_a[i]);
-    const float a_div_sqr2 = a * SQRT_2_INV;
-    const float sign_x = sign(a_div_sqr2);
-    const float x = abs(a_div_sqr2);
-    const float t = 1.0f / (1.0f + p_erf * x);
-    const float y = 1.0f - (((((a5_erf * t + a4_erf) * t) + a3_erf) * t + a2_erf) * t + a1_erf) * t * exp(-x * x);
-    const float erf_approx = sign_x * y;
-
-    data_d[i] = D_TYPE(0.5f * a * (1.0f + erf_approx));
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/gelu_quick.comp

@@ -1,23 +0,0 @@
-#version 450
-
-#include "generic_head.glsl"
-#include "types.glsl"
-
-#extension GL_EXT_control_flow_attributes : enable
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
-layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
-
-void main() {
-    const float GELU_QUICK_COEF = -1.702f;
-    const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
-
-    if (i >= p.KX) {
-        return;
-    }
-
-    const float x = float(data_a[i]);
-    data_d[i] = D_TYPE(x * (1.0f / (1.0f + exp(GELU_QUICK_COEF * x))));
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/hardsigmoid.comp

@@ -1,22 +0,0 @@
-#version 450
-
-#include "generic_head.glsl"
-#include "types.glsl"
-
-#extension GL_EXT_control_flow_attributes : enable
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
-layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
-
-void main() {
-    const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
-
-    if (i >= p.KX) {
-        return;
-    }
-
-    const float x = float(data_a[i]);
-    data_d[i] = D_TYPE(min(1.0f, max(0.0f, (x + 3.0f) / 6.0f)));
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/hardswish.comp

@@ -1,22 +0,0 @@
-#version 450
-
-#include "generic_head.glsl"
-#include "types.glsl"
-
-#extension GL_EXT_control_flow_attributes : enable
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
-layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
-
-void main() {
-    const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
-
-    if (i >= p.KX) {
-        return;
-    }
-
-    const float x = float(data_a[i]);
-    data_d[i] = D_TYPE(x * min(1.0f, max(0.0f, (x + 3.0f) / 6.0f)));
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/leaky_relu.comp

@@ -1,22 +0,0 @@
-#version 450
-
-#include "generic_head.glsl"
-#include "types.glsl"
-
-#extension GL_EXT_control_flow_attributes : enable
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
-layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
-
-void main() {
-    const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
-
-    if (i >= p.KX) {
-        return;
-    }
-
-    const float val = float(data_a[i]);
-    data_d[i] = D_TYPE(max(val, 0.0f) + min(val, 0.0f) * p.param1);
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/neg.comp

@@ -1,20 +0,0 @@
-#version 450
-
-#include "generic_head.glsl"
-#include "types.glsl"
-
-#extension GL_EXT_control_flow_attributes : enable
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
-layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
-
-void main() {
-    const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
-
-    if (i >= p.KX) {
-        return;
-    }
-    data_d[i] = D_TYPE(-float(data_a[i]));
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/relu.comp

@@ -1,21 +0,0 @@
-#version 450
-
-#include "generic_head.glsl"
-#include "types.glsl"
-
-#extension GL_EXT_control_flow_attributes : enable
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
-layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
-
-void main() {
-    const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
-
-    if (i >= p.KX) {
-        return;
-    }
-
-    data_d[i] = D_TYPE(max(float(data_a[i]), 0));
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/round.comp

@@ -1,29 +0,0 @@
-#version 450
-
-#include "generic_head.glsl"
-#include "types.glsl"
-
-#extension GL_EXT_control_flow_attributes : enable
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
-layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
-
-void main() {
-    const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
-
-    if (i >= p.KX) {
-        return;
-    }
-
-    const float x = float(data_a[i]);
-    float result;
-    // Round halfway cases away from zero as roundf does.
-    if (x >= 0.0) {
-        result = floor(x + 0.5);
-    } else {
-        result = ceil(x - 0.5);
-    }
-    data_d[i] = D_TYPE(result);
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/sgn.comp

@@ -1,21 +0,0 @@
-#version 450
-
-#include "generic_head.glsl"
-#include "types.glsl"
-
-#extension GL_EXT_control_flow_attributes : enable
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
-layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
-
-void main() {
-    const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
-
-    if (i >= p.KX) {
-        return;
-    }
-
-    data_d[i] = D_TYPE(sign(float(data_a[i])));
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/sigmoid.comp

@@ -1,20 +0,0 @@
-#version 450
-
-#include "generic_head.glsl"
-#include "types.glsl"
-
-#extension GL_EXT_control_flow_attributes : enable
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
-layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
-
-void main() {
-    const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
-
-    if (i >= p.KX) {
-        return;
-    }
-    data_d[i] = D_TYPE(1. / (1 + exp(-1. * float(data_a[i]))));
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/silu.comp

@@ -1,22 +0,0 @@
-#version 450
-
-#include "generic_head.glsl"
-#include "types.glsl"
-
-#extension GL_EXT_control_flow_attributes : enable
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
-layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
-
-void main() {
-    const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
-
-    if (i >= p.KX) {
-        return;
-    }
-
-    const float xi = float(data_a[i]);
-    data_d[i] = D_TYPE(xi / (1.0f + exp(-xi)));
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/sin.comp

@@ -1,17 +0,0 @@
-#version 450
-
-#include "types.glsl"
-#include "generic_unary_head.glsl"
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-void main() {
-    const uint idx = get_idx();
-
-    if (idx >= p.ne) {
-        return;
-    }
-
-    const FLOAT_TYPE val = FLOAT_TYPE(data_a[get_aoffset() + src0_idx(idx)]);
-    data_d[get_doffset() + dst_idx(idx)] = D_TYPE(sin(val));
-}

data/ext/sources/ggml/src/ggml-vulkan/vulkan-shaders/softplus.comp

@@ -1,23 +0,0 @@
-#version 450
-
-#include "generic_head.glsl"
-#include "types.glsl"
-
-#extension GL_EXT_control_flow_attributes : enable
-
-layout(local_size_x = 512, local_size_y = 1, local_size_z = 1) in;
-
-layout (binding = 0) readonly buffer X {A_TYPE data_a[];};
-layout (binding = 1) writeonly buffer D {D_TYPE data_d[];};
-
-void main() {
-    const uint i = gl_GlobalInvocationID.z * 262144 + gl_GlobalInvocationID.y * 512 + gl_GlobalInvocationID.x;
-
-    if (i >= p.KX) {
-        return;
-    }
-
-    const float x = float(data_a[i]);
-    const float result = (x > 20.0f) ? x : log(1.0f + exp(x));
-    data_d[i] = D_TYPE(result);
-}