Skip to content

Bug: llama-bench csv output truncated #9312

Closed
@jeberger

Description

@jeberger

What happened?

Since rev. 506122d, llama-bench -o csv output gets truncated just after the CPU info and before the closing double quote. It is in particular missing the timing results:

build_commit,build_number,cuda,vulkan,kompute,metal,sycl,rpc,gpu_blas,blas,cpu_info,gpu_info,model_filename,model_type,model_size,model_n_params,n_batch,n_ubatch,n_threads,cpu_mask,cpu_strict,poll,type_k,type_v,n_gpu_layers,split_mode,main_gpu,no_kv_offload,flash_attn,tensor_split,use_mmap,embeddings,n_prompt,n_gen,test_time,avg_ns,stddev_ns,avg_ts,stddev_ts
"506122d","3538","0","0","0","0","0","0","0","0","13th Gen Intel(R) Core(TM) i7-13850HX

Name and Version

> ./bin/Release/llama-cli.exe --version
version: 3538 (506122d)
built with MSVC 19.41.34120.0 for x64

and

> ./bin/llama-cli.exe --version
version: 3538 (506122d)
built with  for x86_64-w64-mingw32

What operating system are you seeing the problem on?

Windows

Relevant log output

No response

Metadata

Metadata

Assignees

No one assigned

    Labels

    bug-unconfirmedmedium severityUsed to report medium severity bugs in llama.cpp (e.g. Malfunctioning Features but still useable)

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions