On Thu, 5 Dec 2013 16:29:37 -0800 Aaron Fabbri <ajfabbri@xxxxxxxxx> wrote: > [This patch is rebased on latest rt-tests.git, and is available in the > "delivery" branch at https://github.com/ajfabbri/af-rt-tests.git] > > Change -a (affinity) option to take a set of CPUs, instead of just one. > > e.g. > cyclictest -a4,6-8 -t5 > > will use 5 threads, assigned round-robin to the set of CPUs {4,6,7,8}. > CPU 4 will get threads 1 and 5, CPU 6 gets thread 2, CPU 7 gets thread 3, and > CPU 8 gets thread 4. > > Although this builds without libnuma, libnuma >= v2 is required for these > arbitrary CPU sets. With libnuma v1, the -a option behaves as before. As > before, compiling without libnuma is supported. The command usage help is fixed > up at compile time to always show the correct usage of the -a option. > > Also note that, since numa_parse_cpustring_all() wasn't available in early > libnuma v2 versions, we use numa_parse_cpustring(). This means you'll have to > use taskset in some cases (isolcpus kernel parameter) to add the desired CPUs to > the set of allowed cores, e.g.: > > taskset -c4-6 cyclictest -a4-6 > > Tested without libnuma (numactl), and with versions 1.0.2 and 2.0.9-rc3. > > Signed-off-by: Aaron Fabbri <ajfabbri@xxxxxxxxx> Thanks, I cherry-picked the commit from your git tree (the included patch was too badly line-wrapped for me to use it). The only issue I've seen so far was that in display_help() your #endif picked up the -A option which is not dependent on libnuma behavior, so I fixed that up. I'll do some testing and push a new rt-tests out soon. Clark
Attachment:
signature.asc
Description: PGP signature
