[Kernel] Add GPU kernels. #372
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
pujiang2018
reviewed
May 15, 2024
| if (nbytes == 0) { return nullptr; } | ||
|
|
||
| void *data; | ||
|
|
||
| #ifdef GPU | ||
| if (device != nullptr) { | ||
| data = sycl::malloc_device<char>(nbytes, *static_cast<sycl::queue *>(device)); |
There was a problem hiding this comment.
The allocation may fail, need to deal with the fail case?
src/layers/attention.h
Outdated
| #ifdef GPU | ||
| sycl::queue *q = static_cast<sycl::queue *>(ctx->device); | ||
| int64_t size = ctx->batchSize * ctx->inputSeqLen * qkvCols * sizeof(float); | ||
| q->memcpy(qkvMatMul.Data(), query.Data(), size).wait(); |
There was a problem hiding this comment.
why do we need a copy here?
src/models/model_factory.h
Outdated
There was a problem hiding this comment.
Do not need to change this file.
src/layers/mlp_llama.h
Outdated
| @@ -275,8 +275,7 @@ class LlamaMLP : public SingletonBase<LlamaMLP<WeiT>> { | |||
| } | |||
| } | |||
|
|
|||
| template <typename T1, typename T2> | |||
| void catGateUpProj(DecoderContext *ctx, hpj::Matrix<T1> &input, hpj::Matrix<T2> &output, hpj::Matrix<T2> &siluBuf) { | |||
| void catGateUpProj(DecoderContext *ctx, hpj::Matrix<InT> &input, hpj::Matrix<ImT> &output, hpj::Matrix<ImT> &siluBuf) { | |||
There was a problem hiding this comment.
changed because of compiler error/warning? suggest using OutT for output.
| @@ -349,8 +349,11 @@ class MMHelper { | |||
| // W8A8 | |||
| else if constexpr (std::is_same_v<WeiT, w8a8_t>) { | |||
| using dt = dnnl::memory::data_type; | |||
| dnnl::engine eng(dnnl::engine::kind::cpu, 0); | |||
| dnnl::stream stm(eng); | |||
There was a problem hiding this comment.
why now need to create an engine and stream every time calling into the function? it may impact the performance.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.