git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit

]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit

overview / pkg / ggml / sources / llama.cpp / commit

author	leo-pony <redacted>
	Thu, 28 Nov 2024 07:25:24 +0000 (15:25 +0800)
committer	GitHub <redacted>
	Thu, 28 Nov 2024 07:25:24 +0000 (15:25 +0800)
commit	605fa66c509f9f117bd654cf0b9b3ea08bb86e80
tree	342a1b8ac45dcf3dd453eeb12898dc5fe36500f7	tree
parent	b7420131bf8ab3e067bc660439ab1ab18be7edbd	commit \| diff

CANN: Fix SOC_TYPE compile bug (#10519)

* CANN: Fix the bug build fail on Ascend310P under two cases:
1) Manual specify SOC_TYPE
2) Under some unusual compile environment

* Update the cann backend News content: Support F16 and F32 data type model for Ascend 310P NPU.

* fix CANN compile fail bug: the assert in ascend kernel function doesn't supportted on some CANN version

Packaging of ggml-org/llama.cpp

RSS Atom

docs/backend/CANN.md		diff \| blob \| history
ggml/src/ggml-cann/CMakeLists.txt		diff \| blob \| history
ggml/src/ggml-cann/kernels/CMakeLists.txt		diff \| blob \| history
ggml/src/ggml-cann/kernels/dup.cpp		diff \| blob \| history
ggml/src/ggml-cann/kernels/get_row_q4_0.cpp		diff \| blob \| history
ggml/src/ggml-cann/kernels/quantize_f16_q8_0.cpp		diff \| blob \| history
ggml/src/ggml-cann/kernels/quantize_f32_q8_0.cpp		diff \| blob \| history
ggml/src/ggml-cann/kernels/quantize_float_to_q4_0.cpp		diff \| blob \| history