Gpu timeout x86 macos #439

darenn1 · 2024-12-10T09:15:58Z

What happened?

A bug happened!

Steps to reproduce

step one...
step two...

What OS are you seeing the problem on?

MacOS

Relevant log output

options: {
  "path": "/Users/i/Documents/2024-12-10 10-14-54.wav",
  "lang": "en",
  "verbose": false,
  "n_threads": 4,
  "init_prompt": "",
  "temperature": 0.4,
  "translate": null,
  "max_text_ctx": null,
  "word_timestamps": false,
  "max_sentence_len": 1
}

Caused by:
   0: failed to transcribe
   1: Generic whisper error. Varies depending on the function. Error code: -6

Location:
    core/src/transcribe.rs:303:38
App Version: vibe 2.6.9
Commit Hash: 7ae85a2
Arch: x86_64
Platform: macos
Kernel Version: 14.6.1
OS: macos
OS Version: 14.6.1
Cuda Version: n/a
Models: ggml-medium-q8_0.bin
Default Model: ggml-medium-q8_0.bin"
Cargo features: 


CPU feature detection is not supported on this architecture.


<details>
<summary>logs</summary>


whisper_init_state: failed to load Core ML model from '/Users/i/Library/Application Support/github.com.thewh1teagle.vibe/ggml-medium-encoder.mlmodelc'
error: Caused GPU Timeout Error (00000002:kIOAccelCommandBufferCallbackErrorTimeout)
whisper_full_with_state: failed to encode
whisper_init_state: failed to load Core ML model from '/Users/i/Library/Application Support/github.com.thewh1teagle.vibe/ggml-medium-encoder.mlmodelc'
error: Caused GPU Timeout Error (00000002:kIOAccelCommandBufferCallbackErrorTimeout)
whisper_full_with_state: failed to encode
cmd: "/Applications/vibe.app/Contents/MacOS/../Resources/ffmpeg" "-i" "/var/folders/9v/9yn9yhf946s8b78ts288nvjw0000gn/T/vibe_temp_2024-12-10/gFFsUwc2i0.wav" "-ar" "16000" "-ac" "1" "-c:a" "pcm_s16le" "/var/folders/9v/9yn9yhf946s8b78ts288nvjw0000gn/T/vibe_temp_2024-12-10/2024-12-10 10-14-54.wav" "-hide_banner" "-y" "-loglevel" "error"
whisper_init_state: failed to load Core ML model from '/Users/i/Library/Application Support/github.com.thewh1teagle.vibe/ggml-medium-encoder.mlmodelc'
error: Caused GPU Timeout Error (00000002:kIOAccelCommandBufferCallbackErrorTimeout)
whisper_full_with_state: failed to encode

</details>

thewh1teagle · 2024-12-10T09:47:35Z

Hey
I don't have macos with x86 for test
Can you check if whisper.cpp works? See the debug.md in this repo
Or let me know if you need more instructions

thewh1teagle · 2024-12-11T12:00:51Z

I can try to fix it in Vibe, but I don't have macOS x86 to test it. If I create a fix, can you test it?

darenn1 · 2024-12-11T12:45:25Z

I can try to fix it in Vibe, but I don't have macOS x86 to test it. If I create a fix, can you test it?

yea for sure

dariusdarwish · 2024-12-11T12:57:58Z

this is my device: mac mini 2018 3.6 GHz Quad-Core Intel Core i3 Intel UHD Graphics 630 1536 MB

I'm here if you need help for testing.

dariusdarwish · 2024-12-11T12:59:00Z

What happened?
A bug happened!

Steps to reproduce
step one...
step two...
What OS are you seeing the problem on?
MacOS

Relevant log output

options: {
"path": "/Users/i/Documents/2024-12-10 10-14-54.wav",
"lang": "en",
"verbose": false,
"n_threads": 4,
"init_prompt": "",
"temperature": 0.4,
"translate": null,
"max_text_ctx": null,
"word_timestamps": false,
"max_sentence_len": 1
}

Caused by:
0: failed to transcribe
1: Generic whisper error. Varies depending on the function. Error code: -6

Location:
core/src/transcribe.rs:303:38
App Version: vibe 2.6.9
Commit Hash: 7ae85a2
Arch: x86_64
Platform: macos
Kernel Version: 14.6.1
OS: macos
OS Version: 14.6.1
Cuda Version: n/a
Models: ggml-medium-q8_0.bin
Default Model: ggml-medium-q8_0.bin"
Cargo features:

CPU feature detection is not supported on this architecture.

logs
whisper_init_state: failed to load Core ML model from '/Users/i/Library/Application Support/github.com.thewh1teagle.vibe/ggml-medium-encoder.mlmodelc'
error: Caused GPU Timeout Error (00000002:kIOAccelCommandBufferCallbackErrorTimeout)
whisper_full_with_state: failed to encode
whisper_init_state: failed to load Core ML model from '/Users/i/Library/Application Support/github.com.thewh1teagle.vibe/ggml-medium-encoder.mlmodelc'
error: Caused GPU Timeout Error (00000002:kIOAccelCommandBufferCallbackErrorTimeout)
whisper_full_with_state: failed to encode
cmd: "/Applications/vibe.app/Contents/MacOS/../Resources/ffmpeg" "-i" "/var/folders/9v/9yn9yhf946s8b78ts288nvjw0000gn/T/vibe_temp_2024-12-10/gFFsUwc2i0.wav" "-ar" "16000" "-ac" "1" "-c:a" "pcm_s16le" "/var/folders/9v/9yn9yhf946s8b78ts288nvjw0000gn/T/vibe_temp_2024-12-10/2024-12-10 10-14-54.wav" "-hide_banner" "-y" "-loglevel" "error"
whisper_init_state: failed to load Core ML model from '/Users/i/Library/Application Support/github.com.thewh1teagle.vibe/ggml-medium-encoder.mlmodelc'
error: Caused GPU Timeout Error (00000002:kIOAccelCommandBufferCallbackErrorTimeout)
whisper_full_with_state: failed to encode

dariusdarwish · 2024-12-11T13:09:12Z

@thewh1teagle Could you give me a beginner ready simple instruction to import Whisper.cpp?

thewh1teagle · 2024-12-11T13:26:36Z

Open the search (Command + Space) and search for terminal
Press Enter
Paste the following:

cd /tmp
wget https://github.com/thewh1teagle/vibe/raw/main/samples/single.wav -O single.wav
wget https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-tiny.bin -O ggml-tiny.bin
wget https://github.com/thewh1teagle/vibe/releases/download/v2.6.9/whisper-rs -O whisper-rs
chmod +x ./whisper-rs
reset
./whisper-rs ggml-tiny.bin single.wav

Wait for it few seconds to finish transcribe
Paste the logs here.

Note: you can safely do it I compiled myself the binary and it's hosted on this repository.

dariusdarwish · 2024-12-11T13:49:21Z

@thewh1teagle

dyld[9810]: Library not loaded: @rpath/libwhisper.1.dylib
Referenced from: /private/tmp/whisper-macos-x64 (built for macOS 14.0 which is newer than running OS)
Reason: tried: '/Volumes/Internal/audio/whisper.cpp/build/src/libwhisper.1.dylib' (no such file), '/System/Volumes/Preboot/Cryptexes/OS/Volumes/Internal/audio/whisper.cpp/build/src/libwhisper.1.dylib' (no such file), '/Volumes/Internal/audio/whisper.cpp/build/ggml/src/libwhisper.1.dylib' (no such file), '/System/Volumes/Preboot/Cryptexes/OS/Volumes/Internal/audio/whisper.cpp/build/ggml/src/libwhisper.1.dylib' (no such file), '/Volumes/Internal/audio/whisper.cpp/build/ggml/src/ggml-cpu/libwhisper.1.dylib' (no such file), '/System/Volumes/Preboot/Cryptexes/OS/Volumes/Internal/audio/whisper.cpp/build/ggml/src/ggml-cpu/libwhisper.1.dylib' (no such file), '/Volumes/Internal/audio/whisper.cpp/build/ggml/src/ggml-blas/libwhisper.1.dylib' (no such file), '/System/Volumes/Preboot/Cryptexes/OS/Volumes/Internal/audio/whisper.cpp/build/ggml/src/ggml-blas/libwhisper.1.dylib' (no such file), '/Volumes/Internal/audio/whisper.cpp/build/ggml/src/ggml-metal/libwhisper.1.dylib' (no such file), '/System/Volumes/Preboot/Cryptexes/OS/Volumes/Internal/audio/whisper.cpp/build/ggml/src/ggml-metal/libwhisper.1.dylib' (no such file), '/Volumes/Internal/audio/whisper.cpp/build/src/libwhisper.1.dylib' (no such file), '/System/Volumes/Preboot/Cryptexes/OS/Volumes/Internal/audio/whisper.cpp/build/src/libwhisper.1.dylib' (no such file), '/Volumes/Internal/audio/whisper.cpp/build/ggml/src/libwhisper.1.dylib' (no such file), '/System/Volumes/Preboot/Cryptexes/OS/Volumes/Internal/audio/whisper.cpp/build/ggml/src/libwhisper.1.dylib' (no such file), '/Volumes/Internal/audio/whisper.cpp/build/ggml/src/ggml-cpu/libwhisper.1.dylib' (no such file), '/System/Volumes/Preboot/Cryptexes/OS/Volumes/Internal/audio/whisper.cpp/build/ggml/src/ggml-cpu/libwhisper.1.dylib' (no such file), '/Volumes/Internal/audio/whisper.cpp/build/ggml/src/ggml-blas/libwhisper.1.dylib' (no such file), '/System/Volumes/Preboot/Cryptexes/OS/Volumes/Internal/audio/whisper.cpp/build/ggml/src/ggml-blas/libwhisper.1.dylib' (no such file), '/Volumes/Internal/audio/whisper.cpp/build/ggml/src/ggml-metal/libwhisper.1.dylib' (no such file), '/System/Volumes/Preboot/Cryptexes/OS/Volumes/Internal/audio/whisper.cpp/build/ggml/src/ggml-metal/libwhisper.1.dylib' (no such file)
zsh: abort ./whisper-macos-x64 -m ggml-tiny-q8_0.bin -f short.wav

thewh1teagle · 2024-12-11T14:08:38Z

@darenn1

I had mistake in the instructions. try to close the terminal and open again then paste the new command (edited)

dariusdarwish · 2024-12-11T14:15:36Z

@thewh1teagle whisper_init_from_file_with_params_no_state: loading model from 'ggml-tiny-q8_0.bin'
whisper_init_with_params_no_state: use gpu = 1
whisper_init_with_params_no_state: flash attn = 0
whisper_init_with_params_no_state: gpu_device = 0
whisper_init_with_params_no_state: dtw = 0
whisper_init_with_params_no_state: backends = 3
whisper_model_load: loading model
whisper_model_load: invalid model data (bad magic)
whisper_init_with_params_no_state: failed to load model
thread 'main' panicked at examples/basic_use.rs:26:10:
failed to load model: InitError
note: run with RUST_BACKTRACE=1 environment variable to display a backtrace

thewh1teagle · 2024-12-11T14:19:24Z

whisper_init_with_params_no_state: use gpu = 1
whisper_init_with_params_no_state: flash attn = 0
whisper_init_with_params_no_state: gpu_device = 0

The error still not relevant, I edited the commands again

dariusdarwish · 2024-12-11T14:23:03Z

@thewh1teagle
whisper_init_from_file_with_params_no_state: loading model from 'ggml-tiny.bin'
whisper_init_with_params_no_state: use gpu = 1
whisper_init_with_params_no_state: flash attn = 0
whisper_init_with_params_no_state: gpu_device = 0
whisper_init_with_params_no_state: dtw = 0
whisper_init_with_params_no_state: backends = 3
whisper_model_load: loading model
whisper_model_load: invalid model data (bad magic)
whisper_init_with_params_no_state: failed to load model
thread 'main' panicked at examples/basic_use.rs:26:10:
failed to load model: InitError
note: run with RUST_BACKTRACE=1 environment variable to display a backtrace

thewh1teagle · 2024-12-11T14:29:31Z

whisper_init_from_file_with_params_no_state: loading model from 'ggml-tiny.bin'

Very strange. you can try take the last line from the commands, omit ggml-tiny.ggml and drag and drop some model from models folder (open from vibe settings) drag into the terminal it will paste the full path. then hit Enter again

darenn1 added the bug Something isn't working label Dec 10, 2024

thewh1teagle mentioned this issue Dec 11, 2024

GPU timeout error on macOS x86 ggerganov/whisper.cpp#2621

Open

thewh1teagle changed the title ~~App reports bug~~ Gpu timeout x86 macos Dec 11, 2024

This was referenced Dec 11, 2024

Gpu timeout x86 macos #427

Closed

Gpu timeout x86 macos #441

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Gpu timeout x86 macos #439

Gpu timeout x86 macos #439

darenn1 commented Dec 10, 2024 •

edited by thewh1teagle

Loading

thewh1teagle commented Dec 10, 2024

thewh1teagle commented Dec 11, 2024

darenn1 commented Dec 11, 2024

dariusdarwish commented Dec 11, 2024

dariusdarwish commented Dec 11, 2024 •

edited by thewh1teagle

Loading

dariusdarwish commented Dec 11, 2024

thewh1teagle commented Dec 11, 2024 •

edited

Loading

dariusdarwish commented Dec 11, 2024 •

edited by thewh1teagle

Loading

thewh1teagle commented Dec 11, 2024 •

edited

Loading

dariusdarwish commented Dec 11, 2024

thewh1teagle commented Dec 11, 2024 •

edited

Loading

dariusdarwish commented Dec 11, 2024

thewh1teagle commented Dec 11, 2024 •

edited

Loading

Gpu timeout x86 macos #439

Gpu timeout x86 macos #439

Comments

darenn1 commented Dec 10, 2024 • edited by thewh1teagle Loading

What happened?

Steps to reproduce

What OS are you seeing the problem on?

Relevant log output

thewh1teagle commented Dec 10, 2024

thewh1teagle commented Dec 11, 2024

darenn1 commented Dec 11, 2024

dariusdarwish commented Dec 11, 2024

dariusdarwish commented Dec 11, 2024 • edited by thewh1teagle Loading

dariusdarwish commented Dec 11, 2024

thewh1teagle commented Dec 11, 2024 • edited Loading

dariusdarwish commented Dec 11, 2024 • edited by thewh1teagle Loading

thewh1teagle commented Dec 11, 2024 • edited Loading

dariusdarwish commented Dec 11, 2024

thewh1teagle commented Dec 11, 2024 • edited Loading

dariusdarwish commented Dec 11, 2024

thewh1teagle commented Dec 11, 2024 • edited Loading

darenn1 commented Dec 10, 2024 •

edited by thewh1teagle

Loading

dariusdarwish commented Dec 11, 2024 •

edited by thewh1teagle

Loading

thewh1teagle commented Dec 11, 2024 •

edited

Loading

dariusdarwish commented Dec 11, 2024 •

edited by thewh1teagle

Loading

thewh1teagle commented Dec 11, 2024 •

edited

Loading

thewh1teagle commented Dec 11, 2024 •

edited

Loading

thewh1teagle commented Dec 11, 2024 •

edited

Loading