| Following are my steps to build the ARM64 version of xnu-4570.1.46, hope this is helpfull for saving time. | |
| 1. Use Xcode 9.0 | |
| 2. Preparation is same as macOS, and there is a guide: https://0xcc.re/building-xnu-kernel-macosx-sierrra-10-12-x/ | |
| 3. There is an ARM64 version libfirehose: https://github.com/Proteas/install_firehose_lib | |
| 4. Copy and edit the ARM64 config(CFLAGS, LDFLAGS) from darwin-on-arm/xnu to your target project | |
| 5. Example CFLAGS: -Darm64 -DARM64 -D__arm64__ -D__ARM64__ -DLP64 -DCONFIG_EMBEDDED -mkernel -DARM64_BOARD_CONFIG_T8011=1 | |
| 6. Fix compiling stage errors by directly importing the missing headers or editing the code | |
| 7. Fix linking stage errors by implementing place holder funcitons for: chudxnu_cpu_alloc, etc | |
| 8. If missing symbol __divti3 in linking stage, get the runtime from llvm. |
| import tiktoken | |
| import langdetect | |
| T = tiktoken.get_encoding("o200k_base") | |
| length_dict = {} | |
| for i in range(T.n_vocab): | |
| try: | |
| length_dict[i] = len(T.decode([i])) | |
| except: |
Apple restricted the access to Xcode LLM (Predictive code completion) feature on China models of Mac. This guide provides a way to bypass that restriction. It's verified on macOS 15.0 Beta (24A5264n), but there is no guarentee that it will always work on later macOS versions.
csrutil enable --without debug command in recovery mode).
| # download FluxCFGPipline | |
| !wget https://raw.githubusercontent.com/linoytsaban/diffusers/refs/heads/dreambooth-lora-flux-exploration/examples/community/pipeline_flux_with_cfg.py | |
| # load pipeline | |
| import diffusers | |
| import torch | |
| from pipeline_flux_with_cfg import FluxCFGPipeline | |
| pipe = FluxCFGPipeline.from_pretrained("black-forest-labs/FLUX.1-dev", | |
| torch_dtype=torch.bfloat16) |