build: add Pascal (sm_61) to CUDA architectures - #5
Merged
Conversation
The published CUDA whisper-server carried kernels for Turing (7.5) through Blackwell only. On Pascal cards (GTX 10 series, compute 6.1) the server starts, loads the model into VRAM, then aborts at the first kernel launch with 'no kernel image is available for execution on the device' — a two-day field debugging incident on a GTX 1080 Ti. Adding 61 ships Pascal SASS; its embedded PTX also covers Volta (70) via driver JIT. Release notes updated to match.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The published CUDA
whisper-servercarried kernels for Turing (7.5) through Blackwell only. On Pascal cards (GTX 10 series, compute 6.1) the server starts, detects the GPU, loads the full model into VRAM, then aborts at the first kernel launch withno kernel image is available for execution on the device— while the app's UI reports GPU acceleration active. Field-reported on a GTX 1080 Ti (two days lost to debugging).61toCUDA_ARCHITECTURES→ Pascal gets native SASS kernels; the embeddedcompute_61PTX also covers Volta (sm_70) via driver JIT as a bonus.Safe for live users: the OpenWhispr app pins release tag
0.0.8+ per-asset sha256 digests, so nothing ships until the app deliberately bumps its pin (together with lowering itsMIN_CUDA_COMPUTE_CAPgate). Next step after merge: dispatch a0.0.9build to verify compilation and measure the size delta.