Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

llava: Add ACC OP for GPU acceleration to the Vulkan backend in the LLAVA CLIP model. #8984

Merged
merged 8 commits into from
Aug 20, 2024
Merged

llava: Add ACC OP for GPU acceleration to the Vulkan backend in the LLAVA CLIP model. #8984

Show file tree
Hide file tree
Changes from 5 commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
4d56810
llava: Add ACC OP for GPU acceleration to the Vulkan backend in the L…
cyzero-kim Aug 11, 2024
b9fc678
fix-up coding style.
cyzero-kim Aug 11, 2024
e6e018d
Fix-up the missing initial parameter to resolve the compilation warning.
cyzero-kim Aug 12, 2024
12ab18b
[fix] Add missing parameters.
cyzero-kim Aug 15, 2024
f84cf24
Merge branch 'master' into vulkan
cyzero-kim Aug 15, 2024
065a9d8
[fix] Use nb1 and nb2 for dst.
cyzero-kim Aug 20, 2024
e85e232
Merge branch 'ggerganov:master' into vulkan
cyzero-kim Aug 20, 2024
a2d1f44
Fix check results ggml_acc call
0cc4m Aug 20, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 8 additions & 0 deletions examples/llava/clip.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,10 @@
#include "ggml-cann.h"
#endif

#ifdef GGML_USE_VULKAN
#include "ggml-vulkan.h"
#endif

#define STB_IMAGE_IMPLEMENTATION
#include "stb_image.h"

Expand Down Expand Up @@ -1125,6 +1129,10 @@ struct clip_ctx * clip_model_load(const char * fname, const int verbosity = 1) {
LOG_TEE("%s: CLIP using CANN backend\n", __func__);
#endif

#ifdef GGML_USE_VULKAN
new_clip->backend = ggml_backend_vk_init(0);
LOG_TEE("%s: CLIP using Vulkan backend\n", __func__);
#endif

if (!new_clip->backend) {
new_clip->backend = ggml_backend_cpu_init();
Expand Down
39 changes: 39 additions & 0 deletions ggml/src/ggml-vulkan.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -180,6 +180,7 @@ struct vk_device_struct {
vk_pipeline pipeline_mul_mat_vec_nc_f16_f32;
vk_pipeline pipeline_get_rows[GGML_TYPE_COUNT];
vk_pipeline pipeline_get_rows_f32[GGML_TYPE_COUNT];
vk_pipeline pipeline_acc_f32;
vk_pipeline pipeline_add_f32, pipeline_add_f16_f32_f16;
vk_pipeline pipeline_mul_f32;
vk_pipeline pipeline_div_f32;
Expand Down Expand Up @@ -1687,6 +1688,8 @@ static void ggml_vk_load_shaders(vk_device& device) {
ggml_vk_create_pipeline(device, device->pipeline_add_f32, "add_f32", add_f32_len, add_f32_data, "main", 3, sizeof(vk_op_binary_push_constants), {512, 1, 1}, {}, 1);
ggml_vk_create_pipeline(device, device->pipeline_add_f16_f32_f16, "add_f16_f32_f16", add_f16_f32_f16_len, add_f16_f32_f16_data, "main", 3, sizeof(vk_op_binary_push_constants), {512, 1, 1}, {}, 1);

ggml_vk_create_pipeline(device, device->pipeline_acc_f32, "acc_f32", acc_f32_len, acc_f32_data, "main", 3, sizeof(vk_op_binary_push_constants), {512, 1, 1}, {}, 1);

ggml_vk_create_pipeline(device, device->pipeline_mul_f32, "mul_f32", mul_f32_len, mul_f32_data, "main", 3, sizeof(vk_op_binary_push_constants), {512, 1, 1}, {}, 1);
ggml_vk_create_pipeline(device, device->pipeline_div_f32, "div_f32", div_f32_len, div_f32_data, "main", 3, sizeof(vk_op_binary_push_constants), {512, 1, 1}, {}, 1);

Expand Down Expand Up @@ -3971,6 +3974,11 @@ static vk_pipeline ggml_vk_op_get_pipeline(ggml_backend_vk_context * ctx, const
return ctx->device->pipeline_get_rows_f32[src0->type];
}
return nullptr;
case GGML_OP_ACC:
if (src0->type == GGML_TYPE_F32 && src1->type == GGML_TYPE_F32 && dst->type == GGML_TYPE_F32) {
return ctx->device->pipeline_acc_f32;
}
return nullptr;
case GGML_OP_ADD:
if (src0->type == GGML_TYPE_F32 && src1->type == GGML_TYPE_F32 && dst->type == GGML_TYPE_F32) {
return ctx->device->pipeline_add_f32;
Expand Down Expand Up @@ -4463,6 +4471,28 @@ static void ggml_vk_get_rows(ggml_backend_vk_context * ctx, vk_context& subctx,
}, dryrun);
}

static void ggml_vk_acc(ggml_backend_vk_context * ctx, vk_context& subctx, const ggml_tensor * src0, const ggml_tensor * src1, ggml_tensor * dst, bool dryrun = false) {
ggml_tensor_extra_gpu * extra = (ggml_tensor_extra_gpu *) dst->extra;
const uint32_t src0_type_size = ggml_type_size(src0->type);
const uint32_t src1_type_size = ggml_type_size(src1->type);
const uint32_t dst_type_size = ggml_type_size(dst->type);
const uint32_t d_offset = ((extra->offset + dst->view_offs) % ctx->device->properties.limits.minStorageBufferOffsetAlignment) / dst_type_size;
cyzero-kim marked this conversation as resolved.
Show resolved Hide resolved

int nb1 = dst->op_params[0] / 4; // 4 bytes of float32
int nb2 = dst->op_params[1] / 4; // 4 bytes of float32
// int nb3 = dst->op_params[2] / 4; // 4 bytes of float32 - unused
int offset = dst->op_params[3] / 4; // offset in bytes

ggml_vk_op_f32<vk_op_binary_push_constants>(ctx, subctx, src0, src1, nullptr, dst, GGML_OP_ACC, {
(uint32_t)ggml_nelements(src0),
(uint32_t)src0->ne[0], (uint32_t)src0->ne[1], (uint32_t)src0->ne[2],(uint32_t)src0->ne[3], (uint32_t)src0->nb[0] / src0_type_size, (uint32_t)nb1, (uint32_t)nb2, (uint32_t)src0->nb[3] / src0_type_size,
(uint32_t)src1->ne[0], (uint32_t)src1->ne[1], (uint32_t)src1->ne[2],(uint32_t)src1->ne[3], (uint32_t)src1->nb[0] / src1_type_size, (uint32_t)src1->nb[1] / src1_type_size, (uint32_t)src1->nb[2] / src1_type_size, (uint32_t)src1->nb[3] / src1_type_size,
(uint32_t) dst->ne[0], (uint32_t) dst->ne[1], (uint32_t) dst->ne[2],(uint32_t) dst->ne[3], (uint32_t) dst->nb[0] / dst_type_size, (uint32_t) dst->nb[1] / dst_type_size, (uint32_t) dst->nb[2] / dst_type_size, (uint32_t) dst->nb[3] / dst_type_size,
0cc4m marked this conversation as resolved.
Show resolved Hide resolved
d_offset,
0.0f, 0.0f, offset,
}, dryrun);
}

static void ggml_vk_add(ggml_backend_vk_context * ctx, vk_context& subctx, const ggml_tensor * src0, const ggml_tensor * src1, ggml_tensor * dst, bool dryrun = false) {
const uint32_t src0_type_size = ggml_type_size(src0->type);
const uint32_t src1_type_size = ggml_type_size(src1->type);
Expand Down Expand Up @@ -5621,6 +5651,7 @@ static void ggml_vk_build_graph(ggml_backend_vk_context * ctx, ggml_tensor * nod
case GGML_OP_REPEAT:
case GGML_OP_GET_ROWS:
case GGML_OP_ADD:
case GGML_OP_ACC:
case GGML_OP_MUL:
case GGML_OP_DIV:
case GGML_OP_CONCAT:
Expand Down Expand Up @@ -5668,6 +5699,10 @@ static void ggml_vk_build_graph(ggml_backend_vk_context * ctx, ggml_tensor * nod
case GGML_OP_REPEAT:
ggml_vk_repeat(ctx, compute_ctx, src0, node, dryrun);

break;
case GGML_OP_ACC:
ggml_vk_acc(ctx, compute_ctx, src0, src1, node, dryrun);

break;
case GGML_OP_GET_ROWS:
ggml_vk_get_rows(ctx, compute_ctx, src0, src1, node, dryrun);
Expand Down Expand Up @@ -5808,6 +5843,7 @@ static bool ggml_vk_compute_forward(ggml_backend_vk_context * ctx, ggml_tensor *

switch (tensor->op) {
case GGML_OP_ADD:
case GGML_OP_ACC:
case GGML_OP_GET_ROWS:
case GGML_OP_MUL:
case GGML_OP_DIV:
Expand Down Expand Up @@ -6539,6 +6575,7 @@ GGML_CALL static bool ggml_backend_vk_supports_op(ggml_backend_t backend, const
case GGML_OP_GROUP_NORM:
case GGML_OP_RMS_NORM:
case GGML_OP_ADD:
case GGML_OP_ACC:
case GGML_OP_MUL:
case GGML_OP_DIV:
case GGML_OP_CONCAT:
Expand Down Expand Up @@ -6995,6 +7032,8 @@ static void ggml_vk_check_results_0(ggml_tensor * tensor) {
tensor_clone = ggml_repeat(ggml_ctx, src0_clone, src1_clone);
} else if (tensor->op == GGML_OP_ADD) {
tensor_clone = ggml_add(ggml_ctx, src0_clone, src1_clone);
} else if (tensor->op == GGML_OP_ACC) {
tensor_clone = ggml_acc(ggml_ctx, src0_clone, src1_clone);
0cc4m marked this conversation as resolved.
Show resolved Hide resolved
} else if (tensor->op == GGML_OP_NORM) {
tensor_clone = ggml_norm(ggml_ctx, src0_clone, *(float *)tensor->op_params);
} else if (tensor->op == GGML_OP_GROUP_NORM) {
Expand Down
24 changes: 24 additions & 0 deletions ggml/src/vulkan-shaders/acc.comp
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,24 @@
#version 450

#include "types.comp"
#include "generic_binary_head.comp"

void main() {
const uint idx = gl_GlobalInvocationID.x;
if (idx >= p.ne) {
return;
}

const uint offset = p.param3;
const uint src1_i = idx - offset;
const uint oz = src1_i / p.nb02;
const uint oy = (src1_i - (oz * p.nb02)) / p.nb01;
const uint ox = src1_i % p.nb01;

if (ox < p.ne10 && oy < p.ne11 && oz < p.ne12) {
data_d[p.d_offset + dst_idx(idx)] = D_TYPE(FLOAT_TYPE(data_a[src0_idx(idx)]) + FLOAT_TYPE(data_b[ox + oy * p.ne10 + oz * p.ne10 * p.ne11]));
} else {
data_d[p.d_offset + dst_idx(idx)] = D_TYPE(FLOAT_TYPE(data_a[src0_idx(idx)]));
}
}

4 changes: 4 additions & 0 deletions ggml/src/vulkan-shaders/vulkan-shaders-gen.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -368,6 +368,10 @@ void process_shaders(std::vector<std::future<void>>& tasks) {
string_to_spv("add_f16_f32_f16", "add.comp", {{"A_TYPE", "float16_t"}, {"B_TYPE", "float"}, {"D_TYPE", "float16_t"}, {"FLOAT_TYPE", "float"}});
}));

tasks.push_back(std::async(std::launch::async, [] {
string_to_spv("acc_f32", "acc.comp", {{"A_TYPE", "float"}, {"B_TYPE", "float"}, {"D_TYPE", "float"}, {"FLOAT_TYPE", "float"}});
}));

tasks.push_back(std::async(std::launch::async, [] {
string_to_spv("split_k_reduce", "mul_mat_split_k_reduce.comp", {});
}));
Expand Down
Loading