llama.cpp
Perf audit and tweaking
#22864
Closed

Perf audit and tweaking #22864

sanmai wants to merge 18 commits into ggml-org:master from sanmai:feat/perf-bulk
sanmai
sanmai Implement ggml_sycl_pool_vmm
c288a6d2
sanmai Update ggml_sycl_device_info
27789e34
sanmai Remove qptr
5a792059
sanmai update logging, fix padding
966a8a42
sanmai simplify
92d6cdfb
sanmai condense
48809a17
sanmai wording
782b58b2
sanmai clarify
79db6400
sanmai use SYCL_EXT_ONEAPI_VIRTUAL_MEM
a12d2d4f
sanmai refactor
4ba009e3
sanmai simplify
aca782e5
sanmai clean
71599ced
sanmai try wrap physical_mem constructor
b0512ad2
sanmai Optimization outline
017c7c3f
sanmai add another audit
5388dd41
sanmai Merge branch 'ggml_sycl_pool_vmm' into feat/perf-bulk
d8df3952
sanmai sycl : multi-stream MoE expert loop in mul_mat_id
7277d77b
sanmai Merge branch 'master' into feat/perf-bulk
9c5e467b
sanmai sanmai closed this 5 days ago
sanmai sanmai deleted the feat/perf-bulk branch 5 days ago
github-actions github-actions added ggml
github-actions github-actions added SYCL

Login to write a write a comment.

Login via GitHub

Reviewers
No reviews
Assignees
No one assigned
Labels
Milestone