On Thursday, 1 September 2022 06:28:55 -03 Volker Hilsheimer wrote: > * stress tests for data races: if your test doesn’t expose any race > conditions if you run with QThread::idealThreadCount threads, then it’s > unlikely that it will expose races if you run with more threads. But with > time sharing, the threads might run a lot longer than you expect. See e.g. > https://codereview.qt-project.org/c/qt/qtbase/+/421391
Also, be careful with scaling. If your code has quadratic scaling, 8 cores isn't twice as bad as 4 cores, compared to 2; it's actually 16x worse. And if I run on 48 cores on a 2-socket NUMA system, it's 576x worse. Or worse. -- Thiago Macieira - thiago.macieira (AT) intel.com Cloud Software Architect - Intel DCAI Cloud Engineering _______________________________________________ Development mailing list [email protected] https://lists.qt-project.org/listinfo/development
