Skip to content

GEMM driver and kernel#29

Merged
asroy merged 6 commits into
developfrom
gemm_driver
Sep 5, 2021
Merged

GEMM driver and kernel#29
asroy merged 6 commits into
developfrom
gemm_driver

Conversation

@asroy
Copy link
Copy Markdown
Contributor

@asroy asroy commented Sep 4, 2021

Add GEMM driver and GEMM kernel for
TT: A[m, k] * B [k, n] = C[m, n]
TN: A[m, k] * B [n, k] = C[m, n]
NT: A[k, m] * B [k, n] = C[m, n]
NN: A[k, m] * B [n, k] = C[m, n]

MI-100 fp16, unlocked frequency (initialized with random integer value)
image

@asroy asroy merged commit 1961390 into develop Sep 5, 2021
asroy added a commit that referenced this pull request Oct 6, 2021
…sing from int32_t (#29)

* overhaul vector_type, make int8x4_t real vector instead of aliasing from int32_t
@illsilin illsilin deleted the gemm_driver branch December 8, 2023 16:03
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant