Pádraig Brady <[email protected]> writes: > Cool. Some quick benchmarks shows about a 10% efficiency win, > and that should go up for larger memory usage in the original process.
Nice! I'll have a patch for sort soon since I think fixing the 'sort --compress-program ...' issue you mentioned previously would be a major improvement [1]. I have a rough patch that just calls abort on errors, so the hard work should be done. Collin [1] https://lists.gnu.org/archive/html/coreutils/2025-10/msg00089.html
