Joel Sherrill commented: https://gitlab.rtems.org/rtems/tools/rtems-source-builder/-/issues/116#note_127530 This is in regards to what I am doing with the "next" tool chain. > 1. All tier 1 and 2 are build I am building all BSPs with all tests as part of making sure the tool chain is OK. > 2. No test regressions for ARM, AARCH64, i386, PowerPC and SPARC I haven't run tests for all but I plan to move from building all BSPs w/tests to doing a "cron sweep" before the tool configuration is trusted to move from "next" to "7". I also plan on setting up "next" on the OAR build sweep setup once it is ready > 3. LibBSD 14 and `main` (15). Tests for ARM (qemu) @opticron has been making sure libbsd builds and has fewer/no warnings on GCC 15. He should comment on testing. > 1. Deployment for `amd/amd-kra-k26`, `amd/amd-microzed`, > `gemini/gemini-powerpc-net-legacy-bsps`, `gemini/gemini-powerpc-libbsd`. Happy to build these but actual testing is up to someone else who has the hardware. > The selection provides a good indication on building vertical stacks in use. Yes. And the cron sweeper actually builds the vertical BSP stacks in the RSB. I suppose it needs to grow to include building deployment. But that will take time to get to. > Do we need to make sure EPICS builds? "We" needs to be the community or you. The side excursion with newlib breaking us for MIPS and AArch64 ilp32 has delayed putting out a community request for help testing. Also do not ignore other frameworks. cFS should be OK since I was told they use GCC 15 on some Linux targets already. But in a perfect world, users should build their stack and application while we are under development (e.g. before branching). -- View it on GitLab: https://gitlab.rtems.org/rtems/tools/rtems-source-builder/-/issues/116#note_127530 You're receiving this email because of your account on gitlab.rtems.org.
_______________________________________________ bugs mailing list [email protected] http://lists.rtems.org/mailman/listinfo/bugs
