4 Commits

Include renames

Author	SHA1	Message	Date
leonarski_fandClaude Opus 5	6e4c0ce202	image_preprocessing: decode bitshuffle+LZ4 on the GPU ... The pipeline decompressed each image on the host and uploaded the result. On an 18 Mpx rotation dataset that made the host-to-device copy the bottleneck of the whole per-image loop: nsys puts the copies at 78% of the loop against 39% for every kernel combined - 3600 transfers of 72.4 MB - and they ran at only 12.5 GB/s of an available 27-28 because the host-side decompression was itself saturating host memory bandwidth. The GPU was mostly waiting. So the compressed chunk goes across instead, about 4 MB rather than 72 MB, and is decoded on the device. That removes the transfer and the host decompression that was throttling it, in one change. Measured on an idle machine, a run goes from 45.11 s to 24.97 s - 1.81x - with the merged output unchanged. THE APPROACH IS JON WRIGHT'S (ESRF): "Experiences with GPU decompression for bitshuffle + LZ4 data", HDF5 User Group 2021, and github.com/jonwright/ bslz4decoders. The kernels here are ours, but the idea and the demonstration that it is worth doing are his. Cited in docs/ACKNOWLEDGEMENT.md and in the new section 0 of docs/CPU_DATA_ANALYSIS.md. Two kernels mirror the CPU decoder. LZ4 runs one WARP per bitshuffle block: every lane parses the same sequence stream (a broadcast read, no divergence) and the literal and match copies are split across the 32 lanes so the stores coalesce; an overlapping match is treated as a pattern of period offset sourced from bytes that already precede the write position, which keeps it parallel rather than a serial byte loop. One thread per block instead measured 13x slower. The bitshuffle inverse then un-transposes each byte-plane through shared memory and interleaves the planes back into elements. Only BSHUF_LZ4 is decoded on the device. The zstd variants have no device decoder, and neither has an uncompressed or float image; Supports() returns false for those and the caller decompresses on the host exactly as before. The fallback is explicit, so a format we cannot decode on the device is a slower path and never a wrong answer. Tests hold the device decoder against the CPU one byte for byte, on data from the production compressor, for every element size the detectors emit - including the 8-bit DECTRIS modes, which take bitshuf_decode_block's separate elem_size == 1 branch - plus a many-block frame, the formats it must decline, and malformed containers, which must throw rather than run off a buffer. Battery: 37 crystals, no failures, identical to the host-decode run. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>	2026-08-03 00:16:36 +02:00
leonarski_f	1c4dfd03e2	v1.0.0-rc.123 (#30) ... This is an UNSTABLE release. * jfjoch_broker: Use newer version of Google Ceres for (potential) CUDA 13 compatibility * jfjoch_broker: Improve performance of generating preview images, especially for large detectors (9M-16M) * jfjoch_viewer: Improve performance of displaying images, especially for large detectors (9M-16M) * jfjoch_viewer: Add more color schemes for better image readability * HDF5: Common mutex for reading and writing HDF5 if both operations were to happen in the same executable * HDF5: suppress warning if path (upstream group) doesn't exists when checking if leaf exists Reviewed-on: #30 Co-authored-by: Filip Leonarski <filip.leonarski@psi.ch> Co-committed-by: Filip Leonarski <filip.leonarski@psi.ch>	2026-01-30 13:43:09 +01:00
leonarski_f	06c5b9cf7f	1.0.0-rc.65	2025-07-16 20:19:48 +02:00
leonarski_f	28d224afab	version 1.0.0-rc.25	2024-11-22 21:25:20 +01:00