leonarski_fandClaude Opus 5
6e805f53c0
Build Packages / build:viewer-tgz:cpu (push) Successful in 8m17s
Build Packages / build:viewer-tgz:cuda (push) Successful in 9m11s
Build Packages / build:rpm (ubuntu2404_nocuda) (push) Successful in 13m38s
Build Packages / build:rpm (rocky8_nocuda) (push) Successful in 13m57s
Build Packages / build:rpm (rocky8_sls9) (push) Successful in 13m57s
Build Packages / build:rpm (ubuntu2204_nocuda) (push) Successful in 14m13s
Build Packages / build:rpm (rocky9_nocuda) (push) Successful in 14m15s
Build Packages / build:rpm (rocky8) (push) Successful in 11m22s
Build Packages / build:rpm (rocky9_sls9) (push) Successful in 12m51s
Build Packages / XDS test (durin plugin) (push) Successful in 7m56s
Build Packages / Generate python client (push) Successful in 32s
Build Packages / Build documentation (push) Successful in 1m4s
Build Packages / Create release (push) Skipped
Build Packages / build:rpm (rocky9) (push) Successful in 13m23s
Build Packages / build:rpm (ubuntu2404) (push) Successful in 13m15s
Build Packages / build:rpm (ubuntu2204) (push) Successful in 13m53s
Build Packages / DIALS test (push) Successful in 14m21s
Build Packages / XDS test (neggia plugin) (push) Successful in 8m36s
Build Packages / XDS test (JFJoch plugin) (push) Successful in 9m16s
Build Packages / Unit tests (push) Successful in 1h15m16s
Build Packages / build:windows:nocuda (push) Failing after 2s
Build Packages / build:windows:cuda (push) Failing after 2s
Three independent costs, each measured, none changing a result. Across the 37-crystal regression set the run time halves (median per crystal 2.0x, total 2.3x) and every crystal's merge statistics are unchanged. The image copy back from the device moved the whole preprocessed frame - 72 MB on a large detector, every frame, per worker - to serve a single host consumer that reads only the strong pixels, at most a few hundred kilobytes of it. Give the buffer a Gather() so that consumer asks for the values it actually wants (a host loop on the CPU, a small kernel on the GPU), and copy the frame back only when a CPU spot finder will genuinely read it. The copy the other way was worse: it came from an unregistered vector, so the driver staged it through its own pinned pool with a host-side memcpy on the calling thread, which does not overlap and collapses under concurrency - 11.6 GB/s at one worker, 1.6 GB/s at eight. That, not any hardware limit, is why throughput stopped improving past four to eight workers. Pinning the decompression buffer once per worker fixes it: on a 18 Mpx dataset the image loop goes from 13.6 to 7.9 ms per image at 32 workers, and 32 workers now beat 8 instead of losing to them. Ceres was computing seventeen partial derivatives where five are free. The per-image rotation refinement frees the beam and the orientation and holds distance, detector angles, rotation axis and cell constant, but the cost function declared all seven blocks, so every residual evaluated in Jet<17> arithmetic. A residual exposing only the two free blocks - the same arithmetic, the constants baked in - halves refinement, and it is exact rather than merely close: dual coordinates evolve independently, so the residuals and the free Jacobian columns are unchanged bit for bit. The merge sorted an index array with a comparator that dereferenced a 1.6 GB array of 72-byte records, i.e. a random walk over memory, single-threaded, twice per two-pass run. Sorting a packed key instead is 2.4x. French-Wilson allocated its integration scratch per reflection and ran serially; it now takes caller-owned scratch and runs over chunks, 4.2x. The correction surfaces re-tested every observation for usability and parity on each of ~22 passes and re-allocated their accumulators each time; bucket the indices once and hoist the buffers. Also convert std::round to std::rint where the rounded value only ever enters a squared residual. The tie rules differ - away from zero against to even - so this is safe exactly where a tie flips the sign but not the magnitude, and unsafe wherever the value becomes a Miller index; those sites keep std::round. Verified over all 2^32 float bit patterns: 8388608 exact ties exist, and the squared residual is bitwise equal for every one of them. Worth little on its own here, because the rounding that dominates is in candidate refinement, where the value is an index and the substitution is not available. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>