[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250530062408.1438861-1-howardchu95@gmail.com>
Date: Thu, 29 May 2025 23:24:06 -0700
From: Howard Chu <howardchu95@...il.com>
To: acme@...nel.org
Cc: mingo@...hat.com,
namhyung@...nel.org,
mark.rutland@....com,
alexander.shishkin@...ux.intel.com,
jolsa@...nel.org,
irogers@...gle.com,
adrian.hunter@...el.com,
peterz@...radead.org,
kan.liang@...ux.intel.com,
linux-perf-users@...r.kernel.org,
linux-kernel@...r.kernel.org,
Howard Chu <howardchu95@...il.com>
Subject: [PATCH v1 0/2] perf trace: Enhance task filtering and remove an unused function
Task filtering is flawed on perf trace. Spawned threads can't get
augmented, and the BPF program collects more data than it needs. This
patch series adds a pids_targeted BPF map to better filter the tasks,
and makes the bpf-output event '__augmented_syscalls__' system-wide. It
also removes evlist__has_bpf_output() that's no longer useful.
Howard Chu (2):
perf trace: Enhance task filtering
perf evlist: Remove evlist__has_bpf_output()
tools/perf/builtin-trace.c | 52 ++++++++++++++++---
.../bpf_skel/augmented_raw_syscalls.bpf.c | 35 ++++++++++---
tools/perf/util/evlist.c | 14 +----
tools/perf/util/evlist.h | 1 -
4 files changed, 73 insertions(+), 29 deletions(-)
--
2.45.2
Powered by blists - more mailing lists