Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Retry allocation with fallback flags #2451

Merged
merged 1 commit into from
Oct 6, 2024

Conversation

SRHMorris
Copy link
Contributor

This fixes issue #2450

I'm unsure what the exact cause is as I'm not overly familiar with Vulkan. So feel free to reject if you can think of a better solution.

@ggerganov
Copy link
Owner

@0cc4m What do you think about this patch? Should we merge it upstream?

@0cc4m
Copy link
Contributor

0cc4m commented Oct 6, 2024

@ggerganov Yeah, this should go upstream. I haven't tested it yet, but it looks good. It might even fix ggerganov/llama.cpp#9734.

Copy link
Owner

@ggerganov ggerganov left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I just tested it and it also fixes the issue reported in #2411 (comment)

@ggerganov ggerganov merged commit 9f346d0 into ggerganov:master Oct 6, 2024
44 checks passed
bygreencn added a commit to bygreencn/whisper.cpp that referenced this pull request Oct 12, 2024
# By Georgi Gerganov (18) and others
# Via Georgi Gerganov
* tag 'v1.7.1': (43 commits)
  release : v1.7.1
  vulkan : retry allocation with fallback flags (ggerganov#2451)
  release : v1.7.0
  scripts : bench v3-turbo
  whisper : remove mel leftover constants (396089f)
  whisper : zero-out the KV cache upon clear (ggerganov#2445)
  objc : fix build
  metal : zero-init buffer contexts (#0)
  whisper : revert mel-related changes (#0)
  whisper : adapt to latest ggml (skip) (#0)
  ggml : fix typo in example usage ggml_gallocr_new (ggml/984)
  ggml : fixes after sync (ggml/983)
  ggml-backend : add device and backend reg interfaces (llama/9707)
  Fixed dequant precision issues in Q4_1 and Q5_1 (llama/9711)
  ggml-backend : add device and backend reg interfaces (llama/9707)
  Initial cmake support of SYCL for AMD GPUs (llama/9658)
  vulkan : do not use tensor->extra (llama/9407)
  ggml/ex: calculate accuracy in graph, adapt MNIST (ggml/980)
  ggml: refactor cross entropy loss CPU impl. (ggml/976)
  scripts : sync ggml-backend.cpp
  ...

# Conflicts:
#	bindings/javascript/package.json
lyapple2008 pushed a commit to lyapple2008/whisper.cpp.mars that referenced this pull request Nov 2, 2024
Co-authored-by: Samuel Morris <samuel.morris@artlist.io>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants