On 1/27/25 1:06 PM, Vaibhav Jain wrote: > Hi Athira, > > Thanks for reviewing this patch series. My responses to your review > comment inline below: > > > Athira Rajeev <atrajeev@xxxxxxxxxxxxxxxxxx> writes: > >>> On 23 Jan 2025, at 5:37 PM, Vaibhav Jain <vaibhav@xxxxxxxxxxxxx> wrote: >>> >>> Introduce a new PMU named 'kvm-hv' to report Book3s kvm-hv specific > <snip> >> >> Hi Vaibhav >> >> All PMU specific code is under “arch/powerpc/perf in the kernel source. Here since we are introducing a kvm-hv specific PMU, can we please have it in arch/powerpc/perf ? > > Is it common convention to put PMU specific code in > arch/powerpc/perf across ppc achitecture variants ? If its there can you > please mention the reasons behind it. > My concern is about fragmentation. Would prefer to have the pmu code under perf folder. Secondly, we did handle module case for vpa-pmu. Maddy > Also the code for this PMU, will be part of kvm-hv kernel module as it > utilizes the functionality implemented there. Moving this PMU code to > arch/powerpc/perf will need this to be converted in yet another new > kernel module, adding a dependency to kvm-hv module and exporting a > bunch of functionality from kvm-hv. Which looks bit messy to me > > <snip> >
