tools: bpftool: use PERF_SAMPLE_TIME instead of reading the clock
authorJakub Kicinski <jakub.kicinski@netronome.com>
Thu, 10 May 2018 17:24:37 +0000 (10:24 -0700)
committerDaniel Borkmann <daniel@iogearbox.net>
Thu, 10 May 2018 23:40:51 +0000 (01:40 +0200)
commite3687510fce2a5b039133be8979b2c8e8cbffa67
tree2bccacf26947c659d816faafcc6eeaabf980cbca
parentcb9c28ef579debbc51e48b37f2534390f5da8467
tools: bpftool: use PERF_SAMPLE_TIME instead of reading the clock

Ask the kernel to include sample time in each even instead of
reading the clock.  This is also more accurate because our
clock reading was done when user space would dump the buffer,
not when sample was produced.

Suggested-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
tools/bpf/bpftool/map_perf_ring.c