leonarski_fandClaude Fable 5.1
546566cff4
Build Packages / Create release (push) Successful in 8s
Build Packages / build:portable:macos-arm64:nocuda (push) Successful in 5m12s
Build Packages / build:rugnux:linux-aarch64:cuda (push) Successful in 10m30s
Build Packages / build:portable:linux-x86_64:nocuda (push) Successful in 19m8s
Build Packages / build:portable:linux-x86_64:cuda (push) Successful in 23m7s
Build Packages / build:portable:windows-x86_64:nocuda (push) Successful in 23m26s
Build Packages / build:jfjoch:ubuntu2204:nocuda (push) Successful in 24m31s
Build Packages / build:portable:windows-x86_64:cuda (push) Successful in 27m40s
Build Packages / build:jfjoch:rocky8:nocuda (push) Successful in 28m31s
Build Packages / HDF5 consumer tests (DIALS, XDS) (push) Successful in 29m50s
Build Packages / build:jfjoch:rocky9:nocuda (push) Successful in 29m47s
Build Packages / build:jfjoch:ubuntu2404:nocuda (push) Successful in 23m42s
Build Packages / Generate python client (push) Successful in 32s
Build Packages / Build documentation (push) Successful in 1m0s
Build Packages / build:jfjoch:rocky8:cuda-sls9 (push) Successful in 24m45s
Build Packages / build:jfjoch:rocky9:cuda-sls9 (push) Successful in 29m59s
Build Packages / build:jfjoch:rocky8:cuda (push) Successful in 28m32s
Build Packages / build:jfjoch:ubuntu2204:cuda (push) Successful in 24m14s
Build Packages / build:jfjoch:ubuntu2404:cuda (push) Successful in 27m49s
Build Packages / build:jfjoch:rocky9:cuda (push) Successful in 30m1s
Build Packages / Unit tests (push) Successful in 1h33m58s
Build Packages / Validation Battery (push) Successful in 36m7s
ModelDensityGPU::Compute() no longer reads the pair count back to the host between its launches: the atoms
are counted per brick with integer atomics, the counts scanned into the bricks' segments, the segments
filled through atomics and then each brick put into model order by rank (one block per brick), which leaves
the pairs exactly as the stable radix sort of the atom-major fill left them - the gather sums in the same
order, bit for bit. The pair bound of the atoms set is checked on the host at SetAtoms() instead of on the
device at every evaluation, and the buffers shrink from four to three ints per pair.
With nothing coming back mid-evaluation, RefinementGPUEngine records the X-ray evaluation and the restraint
evaluation each as a CUDA graph on its stream (cudaStreamBeginCapture, thread-local mode) and replays it,
re-recording only where the shape changes (the atom count, a gradient asked or not, a restraint set
re-uploaded). Two graphs, not one, so that the host still fills the restraints' positions while the X-ray
half runs, as before. Where a stream cannot be captured the launches are queued as they were.
Outputs byte-identical to 037c32d9f on the in-house cytochrome C live run (MTZ arrays, HKL, models,
blobs, report). Workstation (RTX 5080): refinement 1.64 -> 1.50 s on that run.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_015CeEGgpKdfkCTJb3YNMJcu