Hello Oliver, On Saturday 14 of October 2023 12:57:53 Oliver Hartkopp wrote: > Hello Pavel, > > is there any news on this latency issue? > > I've not seen any can-gw related changes between 6.2 and 6.6. > > The only change for linux/net/can/gw.c is this patch: > > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/? >id=2a30b2bd01c23a7eeace3a3f82c2817227099805 I am keeping eye on the results daily and it seems that run run-231002-045216-hist+6.6.0-rc3-rt5-ge31516c1e553+flood-kern-prio-fd-load https://canbus.pages.fel.cvut.cz/can-latester/inspect.html?kernel=rt&load=1&flood=1&fd=1&prio=1&kern=1 is really outlayer. At the end it is single one, I have interpretted the graph incorrectly because so much outstanding value at the end looked as flat increase covering two consecutive runs. So the kernel GW average and maximas follow previous trend after this single peak. So the peak could be related to some transitional state in RT development causing some problem with priorities etc., or can be result of some other problems in the whole setup. I am analyzing some problems with lost messages in some cases of RT runs which seems to be related more to some problem in testing system, setup before run, FPGA reload etc. which cause bus error or something similar with initial suspicion on monitoring side problem. But I do not have conclusion yet. Published runs are complete with no message lost and statistic/trends seems to be without significant change from start of the measurement in May. The change/increase of trends before May has well understood reason, we have updated stress testing, include more sources and tuned priorities for user GW etc... In alonger term perspective, initial setup testing data from April should/will be removed/masked from public data to not provide false assumptions. We probably start new series when 2024 year starts. We will see how data capacity and viewing will work work and slower as data set is extended. If I notice some significant change in more consecutive runs, I try to check it and send information. In the fact, we have caught one real problem in RT already. Best wishes, Pavel -- Pavel Pisa phone: +420 603531357 e-mail: pisa@xxxxxxxxxxxxxxxx Department of Control Engineering FEE CVUT Karlovo namesti 13, 121 35, Prague 2 university: http://control.fel.cvut.cz/ personal: http://cmp.felk.cvut.cz/~pisa social: https://social.kernel.org/ppisa projects: https://www.openhub.net/accounts/ppisa CAN related:http://canbus.pages.fel.cvut.cz/ RISC-V education: https://comparch.edu.cvut.cz/ Open Technologies Research Education and Exchange Services https://gitlab.fel.cvut.cz/otrees/org/-/wikis/home
