Skip to content

RFC tracing: Teach FETCH_MTD_{symbol,deref} to handle per-cpu data

Oleg Nesterov lkml 1 version In review

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 RFC Latest3 patches 1 replyIn review

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/3tracing: Don't mangle sc->addr in update_symbol_cache()kernel/trace/trace_probe.c+3 -100 replies
2/3tracing: introduce {calc,parse}_probe_offset() for FETCH_MTD_{symbol,deref}kernel/trace/trace_probe.c+21 -60 replies
3/3tracing: Teach FETCH_MTD_{symbol,deref} to handle per-cpu datakernel/trace/trace_probe.c+32 -315 replies

Reviews trailers per patch

No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.

Read the v1 thread

Cover letter [RFC PATCH 0/3] tracing: Teach FETCH_MTD_{symbol,deref} to handle per-cpu data
Quoted 8 lines · Oleg Nesterov

After I looked at Namhyung's patches I learned that trace_probes has a rich set of ->fetch methods.

But perhaps it makes sense to add a bit more? The patch is simple and the idea looks natural. If this patch is accepted, I can try to add a bit more hacks to fetch the per-cpu data, but this needs a couple of simple preparations.

Before I redo/resend this patch, let me show the next changes which allow to dump the per-cpu data.

Not for inclusion! I am not sure if this is really useful, I am not sure about the syntax or implementation, etc. Just for discussion, please comment.

Oleg.