llama.cpp
Arm AArch64: optimized GEMV and GEMM kernels for q4_0_q8_0, and q8_0_q8_0 quantization
#5780
Merged

Arm AArch64: optimized GEMV and GEMM kernels for q4_0_q8_0, and q8_0_q8_0 quantization #5780

Dibakar
Jeximo
cebtenzzre
cebtenzzre commented on 2024-02-29
Dibakar
Jeximo
slaren
Dibakar
Jeximo
USBhost
ggerganov
Dibakar
Dibakar
slaren
Dibakar
slaren
ggerganov
snadampal
akingoverlook
akingoverlook
akingoverlook
akingoverlook
Dibakar
Confetti-lxy
Dibakar Dibakar force pushed 2 years ago
Dibakar
Dibakar
Jeximo
akingoverlook
akingoverlook
akingoverlook commented on 2024-05-07
akingoverlook
akingoverlook requested changes on 2024-05-07
akingoverlook
akingoverlook
AGSaidi
mofosyne mofosyne added enhancement
mofosyne mofosyne added Review Complexity : Medium
github-actions
Dibakar
Dibakar
ggerganov
ggerganov commented on 2024-05-17
github-actions github-actions added build
github-actions github-actions added examples
github-actions github-actions added ggml
Dibakar
Dibakar Dibakar force pushed 2 years ago
Dibakar
ggerganov
ggerganov commented on 2024-06-08
Dibakar
ggerganov
Dibakar
Dibakar
Dibakar
ggerganov
ggerganov approved these changes on 2024-07-07
Dibakar Arm AArch64: optimized GEMV and GEMM kernels for q4_0_q8_0, and q8_0_…
002e36ea
Dibakar Arm AArch64: add optimized GEMV and GEMM asm kernels for q4_0_q8_0 qu…
340ef07f
Dibakar Arm AArch64: add optimized GEMV and GEMM asm kernels for q4_0_q8_0 qu…
81215ff4
Dibakar Arm AArch64: add optimized GEMV and GEMM asm kernels for q4_0_q8_0 qu…
6c8d8266
Dibakar Arm AArch64: add optimized GEMV and GEMM asm kernels for q4_0_q8_0 qu…
43e12974
Dibakar Arm AArch64: add copyright claim only to ggml-aarch64.cpp and ggml-aa…
441ab649
Dibakar Arm AArch64: minor code refactoring for rebase
8ee67791
Dibakar Arm AArch64: minor code refactoring for resolving a build issue with …
a657246d
Dibakar Arm AArch64: minor code refactoring to split the Q4_0_AARC64 type int…
746b57f4
Dibakar Arm AArch64: minor code change for resolving a build issue with serve…
5d10c218
Dibakar retrigger checks
7ac03e5f
Dibakar Arm AArch64: minor code changes for rebase
e2c1c47f
Dibakar Arm AArch64: minor changes to skip the pr#7433 vec_dot code for arm c…
79b6cdfe
Dibakar Arm AArch64: remove stale LLAMA_QKK_64 from CMakeLists.txt and delete…
3c1ad5fe
Dibakar Arm AArch64: add reference scalar gemm and gemv, and avoid dynamic me…
a7055b7b
Dibakar Arm AArch64: add multithreaded quantization support for the new types…
cce236bc
Dibakar Arm AArch64: minor code refactoring
7a706067
Dibakar Arm AArch64: simplify logic for calling gemm and gemv functions in gg…
ffbfabb5
Dibakar Arm AArch64: minimize changes in ggml_compute_forward_mul_mat
cbbfd69f
Dibakar Arm AArch64: minor code refactoring, and add reference scalar code to…
35646445
Dibakar Arm AArch64: minor code refactoring
110d143e
Dibakar Arm AArch64: minor code refactoring
4ff0b223
Dibakar Arm AArch64: minor code refactoring
42724b4d
Dibakar rebase on the latest master commit 3fd62a6 and adapt to the new direc…
e5f4713d
Dibakar Dibakar force pushed to e5f4713d 2 years ago
Dibakar
ggerganov
ggerganov commented on 2024-07-09
Dibakar Arm AArch64: remove a redundant comment
c2595d0b
Dibakar Arm AArch64: add pragma in ggml-aarch64.c to turn -Woverlength-string…
a7abb785
ggerganov
Dibakar Arm AArch64: use __aarch64__ check to guard 64-bit neon kernels
0e84ef1a
Dibakar
Dibakar Arm AArch64: update docs/build.md README to include compile time flag…
c653eb1f
github-actions github-actions added documentation
ggerganov ggerganov merged 0f1a39f3 into master 2 years ago
ggerganov
Dibakar
AndreasKunar
AndreasKunar
RhinoDevel

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone