From: Ian Rogers <irogers@google.com>
To: "Peter Zijlstra" <peterz@infradead.org>,
"Ingo Molnar" <mingo@redhat.com>,
"Arnaldo Carvalho de Melo" <acme@kernel.org>,
"Mark Rutland" <mark.rutland@arm.com>,
"Alexander Shishkin" <alexander.shishkin@linux.intel.com>,
"Jiri Olsa" <jolsa@kernel.org>,
"Namhyung Kim" <namhyung@kernel.org>,
"Nathan Chancellor" <nathan@kernel.org>,
"Nick Desaulniers" <ndesaulniers@google.com>,
"Tom Rix" <trix@redhat.com>,
"Roberto Sassu" <roberto.sassu@huawei.com>,
"Quentin Monnet" <quentin@isovalent.com>,
"Andres Freund" <andres@anarazel.de>,
"Tiezhu Yang" <yangtiezhu@loongson.cn>,
"Pavithra Gurushankar" <gpavithrasha@gmail.com>,
"Yang Jihong" <yangjihong1@huawei.com>,
"Adrian Hunter" <adrian.hunter@intel.com>,
"Leo Yan" <leo.yan@linaro.org>, "Martin Liška" <mliska@suse.cz>,
linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org,
llvm@lists.linux.dev
Cc: Stephane Eranian <eranian@google.com>, Ian Rogers <irogers@google.com>
Subject: [PATCH v1 02/13] perf build: Make BUILD_BPF_SKEL default, rename to NO_BPF_SKEL
Date: Fri, 10 Mar 2023 22:57:42 -0800 [thread overview]
Message-ID: <20230311065753.3012826-3-irogers@google.com> (raw)
In-Reply-To: <20230311065753.3012826-1-irogers@google.com>
BPF skeleton support is now key to a number of perf features. Rather
than making it so that BPF support must be enabled for the build, make
this the default and error if the build lacks a clang and libbpf that
are sufficient. To avoid the error and build without BPF skeletons the
NO_BPF_SKEL=1 flag can be used. Add a build-options flag to 'perf
version' to enable detection of the BPF skeleton support and use this
in the offcpu shell test.
Signed-off-by: Ian Rogers <irogers@google.com>
---
tools/perf/Makefile.config | 19 ++++++++++++-------
tools/perf/Makefile.perf | 8 ++++----
tools/perf/builtin-lock.c | 2 +-
tools/perf/builtin-record.c | 2 +-
tools/perf/builtin-version.c | 1 +
tools/perf/tests/shell/record_offcpu.sh | 2 +-
6 files changed, 20 insertions(+), 14 deletions(-)
diff --git a/tools/perf/Makefile.config b/tools/perf/Makefile.config
index 5f11834f189e..33d62d542fd5 100644
--- a/tools/perf/Makefile.config
+++ b/tools/perf/Makefile.config
@@ -415,6 +415,7 @@ ifdef NO_LIBELF
NO_LIBUNWIND := 1
NO_LIBDW_DWARF_UNWIND := 1
NO_LIBBPF := 1
+ NO_BPF_SKEL := 1
NO_JVMTI := 1
else
ifeq ($(feature-libelf), 0)
@@ -662,18 +663,22 @@ ifndef NO_LIBBPF
msg := $(warning BPF API too old. Please install recent kernel headers. BPF support in 'perf record' is disabled.)
NO_LIBBPF := 1
endif
+else
+ NO_BPF_SKEL := 1
endif
-ifdef BUILD_BPF_SKEL
+ifndef NO_BPF_SKEL
$(call feature_check,clang-bpf-co-re)
ifeq ($(feature-clang-bpf-co-re), 0)
- dummy := $(error Error: clang too old/not installed. Please install recent clang to build with BUILD_BPF_SKEL)
- endif
- ifeq ($(filter -DHAVE_LIBBPF_SUPPORT, $(CFLAGS)),)
- dummy := $(error Error: BPF skeleton support requires libbpf)
+ dummy := $(error: ERROR: BPF skeletons unsupported. clang too old/not installed or build with NO_BPF_SKEL=1.)
+ else
+ ifeq ($(filter -DHAVE_LIBBPF_SUPPORT, $(CFLAGS)),)
+ dummy := $(error: ERROR: BPF skeletons unsupported. BPF skeleton support requires libbpf or build with NO_BPF_SKEL=1.)
+ else
+ $(call detected,CONFIG_PERF_BPF_SKEL)
+ CFLAGS += -DHAVE_BPF_SKEL
+ endif
endif
- $(call detected,CONFIG_PERF_BPF_SKEL)
- CFLAGS += -DHAVE_BPF_SKEL
endif
dwarf-post-unwind := 1
diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf
index 2fcee585b225..283ee4f56234 100644
--- a/tools/perf/Makefile.perf
+++ b/tools/perf/Makefile.perf
@@ -126,7 +126,7 @@ include ../scripts/utilities.mak
#
# Define NO_LIBDEBUGINFOD if you do not want support debuginfod
#
-# Define BUILD_BPF_SKEL to enable BPF skeletons
+# Define NO_BPF_SKEL to disable BPF skeletons
#
# As per kernel Makefile, avoid funny character set dependencies
@@ -1055,7 +1055,7 @@ SKELETONS += $(SKEL_OUT)/kwork_trace.skel.h
$(SKEL_TMP_OUT) $(LIBAPI_OUTPUT) $(LIBBPF_OUTPUT) $(LIBPERF_OUTPUT) $(LIBSUBCMD_OUTPUT) $(LIBSYMBOL_OUTPUT):
$(Q)$(MKDIR) -p $@
-ifdef BUILD_BPF_SKEL
+ifndef NO_BPF_SKEL
BPFTOOL := $(SKEL_TMP_OUT)/bootstrap/bpftool
BPF_INCLUDE := -I$(SKEL_TMP_OUT)/.. -I$(LIBBPF_INCLUDE)
@@ -1088,11 +1088,11 @@ bpf-skel: $(SKELETONS)
.PRECIOUS: $(SKEL_TMP_OUT)/%.bpf.o
-else # BUILD_BPF_SKEL
+else # NO_BPF_SKEL
bpf-skel:
-endif # BUILD_BPF_SKEL
+endif # NO_BPF_SKEL
bpf-skel-clean:
$(call QUIET_CLEAN, bpf-skel) $(RM) -r $(SKEL_TMP_OUT) $(SKELETONS)
diff --git a/tools/perf/builtin-lock.c b/tools/perf/builtin-lock.c
index 054997edd98b..240d49a85524 100644
--- a/tools/perf/builtin-lock.c
+++ b/tools/perf/builtin-lock.c
@@ -2371,7 +2371,7 @@ int cmd_lock(int argc, const char **argv)
#ifndef HAVE_BPF_SKEL
set_option_nobuild(contention_options, 'b', "use-bpf",
- "no BUILD_BPF_SKEL=1", false);
+ "NO_BPF_SKEL=1", false);
#endif
if (argc) {
argc = parse_options(argc, argv, contention_options,
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 8374117e66f6..495627437f97 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -3971,7 +3971,7 @@ int cmd_record(int argc, const char **argv)
#ifndef HAVE_BPF_SKEL
# define set_nobuild(s, l, m, c) set_option_nobuild(record_options, s, l, m, c)
- set_nobuild('\0', "off-cpu", "no BUILD_BPF_SKEL=1", true);
+ set_nobuild('\0', "off-cpu", "NO_BPF_SKEL=1", true);
# undef set_nobuild
#endif
diff --git a/tools/perf/builtin-version.c b/tools/perf/builtin-version.c
index a886929ec6e5..0d9cda238c07 100644
--- a/tools/perf/builtin-version.c
+++ b/tools/perf/builtin-version.c
@@ -83,6 +83,7 @@ static void library_status(void)
STATUS(HAVE_ZSTD_SUPPORT, zstd);
STATUS(HAVE_LIBPFM, libpfm4);
STATUS(HAVE_LIBTRACEEVENT, libtraceevent);
+ STATUS(HAVE_BPF_SKEL, BPF skeletons);
}
int cmd_version(int argc, const char **argv)
diff --git a/tools/perf/tests/shell/record_offcpu.sh b/tools/perf/tests/shell/record_offcpu.sh
index e01973d4e0fb..24f81ff85793 100755
--- a/tools/perf/tests/shell/record_offcpu.sh
+++ b/tools/perf/tests/shell/record_offcpu.sh
@@ -28,7 +28,7 @@ test_offcpu_priv() {
err=2
return
fi
- if perf record --off-cpu -o /dev/null --quiet true 2>&1 | grep BUILD_BPF_SKEL
+ if perf version --build-options 2>&1 | grep HAVE_BPF_SKEL | grep -q OFF
then
echo "off-cpu test [Skipped missing BPF support]"
err=2
--
2.40.0.rc1.284.g88254d51c5-goog
next prev parent reply other threads:[~2023-03-11 6:58 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-03-11 6:57 [PATCH v1 00/13] Perf tool build improvements Ian Rogers
2023-03-11 6:57 ` [PATCH v1 01/13] perf build: Support python/perf.so testing Ian Rogers
2023-03-11 6:57 ` Ian Rogers [this message]
2023-03-13 20:19 ` [PATCH v1 02/13] perf build: Make BUILD_BPF_SKEL default, rename to NO_BPF_SKEL Arnaldo Carvalho de Melo
2023-03-13 20:27 ` Ian Rogers
2023-03-13 20:34 ` Arnaldo Carvalho de Melo
2023-03-13 20:59 ` Arnaldo Carvalho de Melo
2023-03-13 21:05 ` Arnaldo Carvalho de Melo
2023-03-11 6:57 ` [PATCH v1 03/13] perf build: Remove unused HAVE_GLIBC_SUPPORT Ian Rogers
2023-03-11 6:57 ` [PATCH v1 04/13] perf build: Error if no libelf and NO_LIBELF isn't set Ian Rogers
2023-03-13 19:45 ` Arnaldo Carvalho de Melo
2023-03-13 20:18 ` Ian Rogers
2023-03-11 6:57 ` [PATCH v1 05/13] perf util: Remove weak sched_getcpu Ian Rogers
2023-03-11 6:57 ` [PATCH v1 06/13] perf build: Error if jevents won't work and NO_JEVENTS=1 isn't set Ian Rogers
2023-03-11 6:57 ` [PATCH v1 07/13] perf build: Make binutil libraries opt in Ian Rogers
2023-03-13 19:37 ` Arnaldo Carvalho de Melo
2023-03-11 6:57 ` [PATCH v1 08/13] tools build: Add feature test for abi::__cxa_demangle Ian Rogers
2023-03-11 6:57 ` [PATCH v1 09/13] perf symbol: Add abi::__cxa_demangle C++ demangling support Ian Rogers
2023-03-30 14:08 ` James Clark
2023-03-30 16:50 ` Ian Rogers
2023-03-30 19:03 ` Ian Rogers
2023-03-31 9:27 ` James Clark
2023-03-11 6:57 ` [PATCH v1 10/13] perf build: Switch libpfm4 to opt-out rather than opt-in Ian Rogers
2023-03-11 6:57 ` [PATCH v1 11/13] perf build: If libtraceevent isn't present error the build Ian Rogers
2023-03-11 6:57 ` [PATCH v1 12/13] tools headers: Make the difference output easier to read Ian Rogers
2023-03-13 19:35 ` Arnaldo Carvalho de Melo
2023-03-13 20:14 ` Ian Rogers
2023-03-11 6:57 ` [PATCH v1 13/13] perf build: Remove redundant NO_NEWT build option Ian Rogers
2023-03-14 12:11 ` [PATCH v1 00/13] Perf tool build improvements Arnaldo Carvalho de Melo
2023-03-14 12:12 ` Arnaldo Carvalho de Melo
2023-03-14 12:27 ` Arnaldo Carvalho de Melo
2023-03-14 12:29 ` Arnaldo Carvalho de Melo
2023-03-14 12:21 ` Adrian Hunter
2023-03-14 12:28 ` Arnaldo Carvalho de Melo
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230311065753.3012826-3-irogers@google.com \
--to=irogers@google.com \
--cc=acme@kernel.org \
--cc=adrian.hunter@intel.com \
--cc=alexander.shishkin@linux.intel.com \
--cc=andres@anarazel.de \
--cc=eranian@google.com \
--cc=gpavithrasha@gmail.com \
--cc=jolsa@kernel.org \
--cc=leo.yan@linaro.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=llvm@lists.linux.dev \
--cc=mark.rutland@arm.com \
--cc=mingo@redhat.com \
--cc=mliska@suse.cz \
--cc=namhyung@kernel.org \
--cc=nathan@kernel.org \
--cc=ndesaulniers@google.com \
--cc=peterz@infradead.org \
--cc=quentin@isovalent.com \
--cc=roberto.sassu@huawei.com \
--cc=trix@redhat.com \
--cc=yangjihong1@huawei.com \
--cc=yangtiezhu@loongson.cn \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox