[LITMUS^RT] New release: 2012.3
Björn Brandenburg
bbb at mpi-sws.org
Wed Jan 16 18:07:00 CET 2013
On Jan 16, 2013, at 5:31 PM, Glenn Elliott <gelliott at cs.unc.edu> wrote:
> Also, and probably the biggest issue going forward, testing on ARM is difficult since ARM kernels are so fragmented. We have to port Litmus to each unique kernel devised for each of our ARM platforms.
One more reason to rebase to a newer Linux version; they consolidated the ARM support in recent versions.
> At UNC, we have three. From the sounds of it, MPI has a Freescale ARM. So that's four different ARM kernels to test. Do we need some sort of policy, or is the decision "do better next time" good enough?
Do better next time is a good start. At the very least, everyone should make an effort to compile staging from time to time for their "usual" target platform, which would reveal such basic issues. At some point, however, we have to just accept that we are a small team and can't possibly test everything all the time. So some oversights will happen, and we'll correct them in the next release. Not ideal, but manageable.
I'm thinking that we should do another 3.0-based release shortly after ECRTS. I have some other bug fixes RTAS-releated work that I need to rebase, which I'd like to push out some time in the near future, too.
Thanks,
Björn
More information about the litmus-dev
mailing list