perf/arm-cmn: Multi-DTC improvements
Status from this message: Applied to will (for-next/perf), thanks!
Get series with b4
b4 am [email protected]Versions newest first
Patches in v1 Applied
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/3 | perf/arm-cmn: Fix DTC domain detectiondrivers/ | +14 | 1 reply | |
| 2/3 | perf/arm-cmn: Rework DTC counters (again)drivers/ | +64 | 6 replies | |
| 3/3 | perf/arm-cmn: Enable per-DTC counter allocationdrivers/ | +10 | 1 reply |
Reviews trailers per patch
2 of 3 patches have a Reviewed-by or Acked-by.
| Patch | Ilkka Koskinen |
|---|---|
| 1/3 | R-b |
| 2/3 | R-b |
| 3/3 |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.
Cover letter [PATCH 0/3] perf/arm-cmn: Multi-DTC improvements
On larger CMN configurations with multiple Debug & Trace Controllers, we've so far ignored the notion of DTC domains, mostly since they were not software-discoverable in the original CMN-600 design. However this means that if the user wants to monitor lots of individual nodes across the whole mesh, we end up multiplexing events which could otherwise happily run in parallel if we allocated DTC counters per-domain. This mini-series finally bites the bullet to do that.
As usual I've only been able to personally test that it doesn't regress any behaviour on a single-DTC CMN-600, so it would be nice if anyone with a multi-domain CMN-650/CMN-700 setup could confirm that patches 1+2 alone do not visibly change any behaviour, and then patch #3 on top works as expected.
Thanks,
Robin.
Robin Murphy (3):
perf/arm-cmn: Fix DTC domain detection
perf/arm-cmn: Rework DTC counters (again)
perf/arm-cmn: Enable per-DTC counter allocation
drivers/perf/arm-cmn.c | 154 +++++++++++++++++++++++------------------ 1 file changed, 85 insertions(+), 69 deletions(-)
Signature · 1 line
2.39.2.101.g768bb238c484.dirty