Skip to content

perf_counter patches

Peter Zijlstra lkml 1 version In review

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 Latest9 patches 14 repliesIn review

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/9perf_counter: unify and fix delayed counter wakeuparch/powerpc/include/asm/hw_irq.h arch/powerpc/include/asm/paca.h arch/powerpc/kernel/asm-offsets.c +10 more+133 -1414 replies
2/9perf_counter: fix update_userpage()include/linux/perf_counter.h kernel/perf_counter.c+58 -159 replies
3/9perf_counter: kerneltop: simplify data_head readDocumentation/perf_counter/kerneltop.c+1 -130 replies
4/9perf_counter: executable mmap() informationinclude/linux/perf_counter.h kernel/perf_counter.c mm/mmap.c+177 -20 replies
5/9perf_counter: kerneltop: parse the mmap data streamDocumentation/perf_counter/kerneltop.c+44 -60 replies
6/9perf_counter: powerpc: only reserve PMU hardware when we need itarch/powerpc/kernel/perf_counter.c+42 -50 replies
7/9perf_counter: make it possible for hw_perf_counter_init to return error codesarch/powerpc/kernel/perf_counter.c kernel/perf_counter.c+28 -191 reply
8/9perf_counter tools: optionally scale counter values in perfstat modeDocumentation/perf_counter/kerneltop.c+45 -110 replies
9/9RFC perf_counter: event overlow handlinginclude/linux/perf_counter.h kernel/perf_counter.c+101 -270 replies

Reviews trailers per patch

1 of 9 patches has a Reviewed-by or Acked-by.

PatchPeter ZijlstraPaul Mackerras
1/9
2/9A-b
3/9
4/9
5/9
6/9
7/9
8/9
9/9Maybe-Signed-off-by

Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.

Read the v1 thread

Cover letter [PATCH 0/9] perf_counter patches

This set fixes 2 problems I found while going over the perf_counter code.
- the delayed wakeups were incomplete (and broken on x86)
- update_userpage() wasn't used properly

It also adds tracking of the mmap() data so that we can profile userspace.

I also folded in Paulus' pending patches (I had to fix up a few conflict, please double check).

Finally, the writable control page patch... still not convinced we need that.