thek0tyara/llama-cpp-turboquant
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 4

Updated tests to run synchronous

Browse source
This commit is contained in:
Alejandro Saucedo 2021-03-05 08:33:53 +00:00
parent 71129392c2
commit 33a0ef933b
3 changed files with 33 additions and 36 deletions
Download patch file Download diff file Expand all files Collapse all files

7
test/TestMultipleAlgoExecutions.cpp
View file

@ -16,8 +16,6 @@ TEST(TestMultipleAlgoExecutions, TestEndToEndFunctionality)
std::string shader = (R"(
#version 450
#extension GL_EXT_shader_atomic_float: enable
layout (local_size_x = 1) in;
// The input tensors bind index is relative to index in parameter passed
@ -36,8 +34,8 @@ TEST(TestMultipleAlgoExecutions, TestEndToEndFunctionality)
void main() {
uint index = gl_GlobalInvocationID.x;
atomicAdd(out_a[index], in_a[index] * in_b[index]);
atomicAdd(out_b[index], const_one * push_const.val);
out_a[index] += in_a[index] * in_b[index];
out_b[index] += const_one * push_const.val;
}
)");
@ -57,6 +55,7 @@ TEST(TestMultipleAlgoExecutions, TestEndToEndFunctionality)
mgr.sequence()
->record<kp::OpTensorSyncDevice>(params)
->record<kp::OpAlgoDispatch>(algorithm)
->eval()
->record<kp::OpAlgoDispatch>(algorithm, pushConstsB)
->eval();