Hi, I pushed an update (hopefully ok because bots were saturating the logs and such so we ran out of disk space). Apart from fixes for issues mentioned / encountered the last weeks we also added a goodie file for lm so that Jim can go forward. I don't know how many users process multiple documents at the same time, but as a side effect of improving performance of processing the test suite (I run 8 parallel processes for it). Here is some fun: context --parallel test-1.tex test-2.tex or with output: context --parallel test-*.tex --terminal a bit more distinct: context --parallel test-*.tex --terminal=ansi if you have more cores: context --parallel test-*.tex --terminal=ansi --runners=16 A tex process needs one core so a rule of thumb is the nofcores multiplied by 2 if hyperthreading is available. On linux one can get 100% saturation, on windows a bit less (the test suite went down from 800 sec for 2200 files to 425 and on linux (wsl) we get to some 260 sec. (A 2018 laptop with 4 xeon cores that hyperthread; I can't test on a faster machine.) Profiled binaries perform some 10% faster. Hans ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | www.pragma-ade.nl | www.pragma-pod.nl -----------------------------------------------------------------