root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-29 04:44:34 +00:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

vulkan : replace ggml_diag_mask_inf with ggml_add (custom -inf mask)

Browse Source
This commit is contained in:
Jared Van Bortel 2023-11-23 17:18:27 -05:00
parent af00cca08e
commit 71565eb0c3
2 changed files with 82 additions and 21 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

59
ggml-vulkan.cpp
View File

@ -579,29 +579,48 @@ uint32_t safe_divide(uint32_t a, uint32_t b) {
return a / b;
}
void ggml_vk_add(kp::Sequence& seq,
const std::shared_ptr<kp::Tensor>& inA,
const std::shared_ptr<kp::Tensor>& inB,
const std::shared_ptr<kp::Tensor>& out,
uint32_t inAOff, uint32_t inBOff, uint32_t outOff,
uint32_t size) {
void ggml_vk_add(
kp::Sequence& seq,
const std::shared_ptr<kp::Tensor>& inA,
const std::shared_ptr<kp::Tensor>& inB,
const std::shared_ptr<kp::Tensor>& out,
uint32_t inAOff, uint32_t inBOff, uint32_t outOff,
int32_t ne00, int32_t ne01, int32_t ne02, int32_t ne03,
int32_t nb00, int32_t nb01, int32_t nb02, int32_t nb03,
int32_t ne10, int32_t ne11, int32_t ne12, int32_t ne13,
int32_t nb10, int32_t nb11, int32_t nb12, int32_t nb13,
int32_t ne0,
int32_t nb0, int32_t nb1, int32_t nb2, int32_t nb3
) {
const static auto spirv = getSpirvShader(kp::shader_data::op_add_comp_spv,
kp::shader_data::op_add_comp_spv_len);
struct PushConstants {
uint32_t inAOff, inBOff, outOff;
int32_t ne00;
int32_t nb00, nb01, nb02, nb03;
int32_t ne10, ne11, ne12, ne13;
int32_t nb10, nb11, nb12, nb13;
int32_t ne0;
int32_t nb0, nb1, nb2, nb3;
} const pushConsts {
safe_divide(inAOff, 4), safe_divide(inBOff, 4), safe_divide(outOff, 4)
safe_divide(inAOff, 4), safe_divide(inBOff, 4), safe_divide(outOff, 4),
ne00,
nb00, nb01, nb02, nb03,
ne10, ne11, ne12, ne13,
nb10, nb11, nb12, nb13,
ne0,
nb0, nb1, nb2, nb3
};
std::shared_ptr<kp::Algorithm> s_algo = nullptr;
if (!komputeManager()->hasAlgorithm(__func__))
s_algo = komputeManager()->algorithm<float, PushConstants>(__func__, s_kompute_context->pool.get(), {inA, inB, out}, spirv, {size}, {}, {pushConsts});
else {
if (!komputeManager()->hasAlgorithm(__func__)) {
s_algo = komputeManager()->algorithm<float, PushConstants>(__func__, s_kompute_context->pool.get(), {inA, inB, out}, spirv, {unsigned(ne01), unsigned(ne02), unsigned(ne03)}, {}, {pushConsts});
} else {
s_algo = komputeManager()->getAlgorithm(__func__);
s_algo->setTensors({inA, inB, out});
s_algo->setWorkgroup({size});
s_algo->setWorkgroup({unsigned(ne01), unsigned(ne02), unsigned(ne03)});
s_algo->setPushConstants<PushConstants>({pushConsts});
s_algo->updateDescriptors(s_kompute_context->pool.get());
}
@ -1315,12 +1334,12 @@ void ggml_vk_graph_compute(struct ggml_kompute_context * ctx, struct ggml_cgraph
const int32_t ne10 = src1 ? src1->ne[0] : 0;
const int32_t ne11 = src1 ? src1->ne[1] : 0;
const int32_t ne12 = src1 ? src1->ne[2] : 0;
// const int32_t ne13 = src1 ? src1->ne[3] : 0;
const int32_t ne13 = src1 ? src1->ne[3] : 0;
// const uint32_t nb10 = src1 ? src1->nb[0] : 0;
const uint32_t nb10 = src1 ? src1->nb[0] : 0;
const uint32_t nb11 = src1 ? src1->nb[1] : 0;
const uint32_t nb12 = src1 ? src1->nb[2] : 0;
// const uint32_t nb13 = src1 ? src1->nb[3] : 0;
const uint32_t nb13 = src1 ? src1->nb[3] : 0;
const int32_t ne0 = dst ? dst->ne[0] : 0;
const int32_t ne1 = dst ? dst->ne[1] : 0;
@ -1354,11 +1373,19 @@ void ggml_vk_graph_compute(struct ggml_kompute_context * ctx, struct ggml_cgraph
} break;
case GGML_OP_ADD:
{
if (ggml_nelements(src1) == ne10) {
if (ggml_nelements(src1) == ne10 && ne00 % 4 == 0) {
// src1 is a row
ggml_vk_addrow(seq, id_src0, id_src1, id_dst, off_src0, off_src1, off_dst, ggml_nelements(dst)/4, ne00);
} else {
ggml_vk_add(seq, id_src0, id_src1, id_dst, off_src0, off_src1, off_dst, ggml_nelements(dst)/4);
ggml_vk_add(
seq, id_src0, id_src1, id_dst, off_src0, off_src1, off_dst,
ne00, ne01, ne02, ne03,
nb00, nb01, nb02, nb03,
ne10, ne11, ne12, ne13,
nb10, nb11, nb12, nb13,
ne0,
nb0, nb1, nb2, nb3
);
}
} break;
case GGML_OP_MUL:

44
kompute/op_add.comp
View File

@ -10,7 +10,7 @@
#include "common.comp"
layout(local_size_x = 1) in;
layout(local_size_x = 1024) in;
layout(binding = 0) buffer restrict readonly tensorInA { float inA[]; };
layout(binding = 1) buffer restrict readonly tensorInB { float inB[]; };
@ -20,13 +20,47 @@ layout(push_constant) uniform PushConstants {
uint inAOff;
uint inBOff;
uint outOff;
int ne00;
int nb00;
int nb01;
int nb02;
int nb03;
int ne10;
int ne11;
int ne12;
int ne13;
int nb10;
int nb11;
int nb12;
int nb13;
int ne0;
int nb0;
int nb1;
int nb2;
int nb3;
} pcs;
// general-purpose kernel for addition of two tensors
// pros: works for non-contiguous tensors, supports broadcast across dims 1, 2 and 3
// cons: not very efficient
void main() {
const uint baseIndex = gl_WorkGroupID.x * 4;
const uint i03 = gl_WorkGroupID.z;
const uint i02 = gl_WorkGroupID.y;
const uint i01 = gl_WorkGroupID.x;
for (uint x = 0; x < 4; x++) {
const uint i = baseIndex + x;
out_[i + pcs.outOff] = inA[i + pcs.inAOff] + inB[i + pcs.inBOff];
const uint i13 = i03 % pcs.ne13;
const uint i12 = i02 % pcs.ne12;
const uint i11 = i01 % pcs.ne11;
uint src0_off = uint((i03*pcs.nb03 + i02*pcs.nb02 + i01*pcs.nb01 + gl_SubgroupInvocationID.x*pcs.nb00) / 4);
uint src1_off = uint((i13*pcs.nb13 + i12*pcs.nb12 + i11*pcs.nb11 + gl_SubgroupInvocationID.x*pcs.nb10) / 4);
uint dst_off = uint((i03*pcs.nb3 + i02*pcs.nb2 + i01*pcs.nb1 + gl_SubgroupInvocationID.x*pcs.nb0 ) / 4);
for (uint i0 = gl_LocalInvocationID.x; i0 < pcs.ne0; i0 += gl_WorkGroupSize.x) {
out_[pcs.outOff + dst_off] = inA[pcs.inAOff + src0_off] + inB[pcs.inBOff + src1_off];
src0_off += gl_WorkGroupSize.x*pcs.ne00;
src1_off += gl_WorkGroupSize.x*pcs.ne10;
dst_off += gl_WorkGroupSize.x*pcs.ne0;
}
}