kernel/tools/perf/util/bpf-loader.c, branch linux-rolling-stable

perf parse-events: Remove BPF event support

2023-08-15T19:41:48Z

New features like the BPF --filter support in perf record have made the BPF event functionality somewhat redundant. As shown by commit fcb027c1a4f6 ("perf tools: Revert enable indices setting syntax for BPF map") and commit 14e4b9f4289a ("perf trace: Raw augmented syscalls fix libbpf 1.0+ compatibility") the BPF event support hasn't been well maintained and it adds considerable complexity in areas like event parsing, not least as '/' is a separator for event modifiers as well as in paths. This patch removes support in the event parser for BPF events and then the associated functions are removed. This leads to the removal of whole source files like bpf-loader.c. Removing support means that augmented syscalls in perf trace is broken, this will be fixed in a later commit adding support using BPF skeletons. The removal of BPF events causes an unused label warning from flex generated code, so update build to ignore it: ``` util/parse-events-flex.c:2704:1: error: label ‘find_rule’ defined but not used [-Werror=unused-label] 2704 | find_rule: /* we branch to this label when backing up */ ``` Committer notes: Extracted from a larger patch that was also removing the support for linking with libllvm and libclang, that were an alternative to using an external clang execution to compile the .c event source code into BPF bytecode. Testing it: # perf trace -e /home/acme/git/perf/tools/perf/examples/bpf/augmented_raw_syscalls.c event syntax error: '/home/acme/git/perf/tools/perf/examples/bpf/augmented_raw_syscalls.c' \___ Bad event or PMU Unabled to find PMU or event on a PMU of 'home' Initial error: event syntax error: '/home/acme/git/perf/tools/perf/examples/bpf/augmented_raw_syscalls.c' \___ Cannot find PMU `home'. Missing kernel support? Run 'perf list' for a list of valid events Usage: perf trace [] [] or: perf trace [] -- [] or: perf trace record [] [] or: perf trace record [] -- [] -e, --event event/syscall selector. use 'perf list' to list available events # Signed-off-by: Ian Rogers Acked-by: Jiri Olsa Tested-by: Arnaldo Carvalho de Melo Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Andi Kleen Cc: Andrii Nakryiko Cc: Anshuman Khandual Cc: Athira Rajeev Cc: Brendan Gregg Cc: Carsten Haitzler Cc: Eduard Zingerman Cc: Fangrui Song Cc: He Kuang Cc: Ingo Molnar Cc: James Clark Cc: Kan Liang Cc: Leo Yan Cc: Madhavan Srinivasan Cc: Mark Rutland Cc: Namhyung Kim Cc: Nathan Chancellor Cc: Naveen N. Rao Cc: Nick Desaulniers Cc: Peter Zijlstra Cc: Ravi Bangoria Cc: Rob Herring Cc: Tiezhu Yang Cc: Tom Rix Cc: Wang Nan Cc: Wang ShaoBo Cc: Yang Jihong Cc: Yonghong Song Cc: YueHaibing Cc: bpf@vger.kernel.org Cc: llvm@lists.linux.dev Link: https://lore.kernel.org/r/20230810184853.2860737-2-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo Signed-off-by: Arnaldo Carvalho de Melo

perf bpf: Remove support for embedding clang for compiling BPF events (-e foo.c)

2023-08-15T19:41:48Z

This never was in the default build for perf, is difficult to maintain as it uses clang/llvm internals so ditch it, keeping, for now, the external compilation of .c BPF into .o bytecode and its subsequent loading, that is also going to be removed, do it separately to help bisection and to properly document what is being removed and why. Committer notes: Extracted from a larger patch and removed some leftovers, namely deleting these now unused feature tests: tools/build/feature/test-clang.cpp tools/build/feature/test-cxx.cpp tools/build/feature/test-llvm-version.cpp tools/build/feature/test-llvm.cpp Testing the use of BPF events after applying this patch: To use the external clang/llvm toolchain to compile a .c event and then use libbpf to load it, to get the syscalls:sys_enter_open* tracepoints and read the filename pointer, putting it into the ring buffer right after the usual tracepoint payload for 'perf trace' to then print it: [root@quaco ~]# perf trace -e /home/acme/git/perf-tools-next/tools/perf/examples/bpf/augmented_raw_syscalls.c,open* --max-events=10 0.000 systemd-oomd/959 openat(dfd: CWD, filename: "/proc/meminfo", flags: RDONLY|CLOEXEC) = 12 0.083 abrt-dump-jour/1453 openat(dfd: CWD, filename: "/var/log/journal/d6a97235307247e09f13f326fb607e3c/system.journal", flags: RDONLY|CLOEXEC|NONBLOCK) = 4 0.063 abrt-dump-jour/1454 openat(dfd: CWD, filename: "/var/log/journal/d6a97235307247e09f13f326fb607e3c/system.journal", flags: RDONLY|CLOEXEC|NONBLOCK) = 4 0.082 abrt-dump-jour/1455 openat(dfd: CWD, filename: "/var/log/journal/d6a97235307247e09f13f326fb607e3c/system.journal", flags: RDONLY|CLOEXEC|NONBLOCK) = 4 250.124 systemd-oomd/959 openat(dfd: CWD, filename: "/proc/meminfo", flags: RDONLY|CLOEXEC) = 12 250.521 systemd-oomd/959 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1000.slice/user@1000.service/app.slice/memory.pressure", flags: RDONLY|CLOEXEC) = 12 251.047 systemd-oomd/959 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1000.slice/user@1000.service/app.slice/memory.current", flags: RDONLY|CLOEXEC) = 12 251.162 systemd-oomd/959 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1000.slice/user@1000.service/app.slice/memory.min", flags: RDONLY|CLOEXEC) = 12 251.242 systemd-oomd/959 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1000.slice/user@1000.service/app.slice/memory.low", flags: RDONLY|CLOEXEC) = 12 251.353 systemd-oomd/959 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1000.slice/user@1000.service/app.slice/memory.swap.current", flags: RDONLY|CLOEXEC) = 12 [root@quaco ~]# Same thing, but with a prebuilt .o BPF bytecode: [root@quaco ~]# perf trace -e /home/acme/git/perf-tools-next/tools/perf/examples/bpf/augmented_raw_syscalls.o,open* --max-events=10 0.000 systemd-oomd/959 openat(dfd: CWD, filename: "/proc/meminfo", flags: RDONLY|CLOEXEC) = 12 0.083 abrt-dump-jour/1453 openat(dfd: CWD, filename: "/var/log/journal/d6a97235307247e09f13f326fb607e3c/system.journal", flags: RDONLY|CLOEXEC|NONBLOCK) = 4 0.083 abrt-dump-jour/1455 openat(dfd: CWD, filename: "/var/log/journal/d6a97235307247e09f13f326fb607e3c/system.journal", flags: RDONLY|CLOEXEC|NONBLOCK) = 4 0.062 abrt-dump-jour/1454 openat(dfd: CWD, filename: "/var/log/journal/d6a97235307247e09f13f326fb607e3c/system.journal", flags: RDONLY|CLOEXEC|NONBLOCK) = 4 249.985 systemd-oomd/959 openat(dfd: CWD, filename: "/proc/meminfo", flags: RDONLY|CLOEXEC) = 12 466.763 thermald/1234 openat(dfd: CWD, filename: "/sys/class/powercap/intel-rapl/intel-rapl:0/intel-rapl:0:2/energy_uj") = 13 467.145 thermald/1234 openat(dfd: CWD, filename: "/sys/class/powercap/intel-rapl/intel-rapl:0/energy_uj") = 13 467.311 thermald/1234 openat(dfd: CWD, filename: "/sys/class/thermal/thermal_zone2/temp") = 13 500.040 cgroupify/24006 openat(dfd: 4, filename: ".", flags: RDONLY|CLOEXEC|DIRECTORY|NONBLOCK) = 5 500.295 cgroupify/24006 openat(dfd: 4, filename: "24616/cgroup.procs") = 5 [root@quaco ~]# Signed-off-by: Ian Rogers Tested-by: Arnaldo Carvalho de Melo Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Andi Kleen Cc: Andrii Nakryiko Cc: Anshuman Khandual Cc: Athira Rajeev Cc: Brendan Gregg Cc: Carsten Haitzler Cc: Eduard Zingerman Cc: Fangrui Song Cc: He Kuang Cc: Ingo Molnar Cc: James Clark Cc: Jiri Olsa Cc: Kan Liang Cc: Leo Yan Cc: Madhavan Srinivasan Cc: Mark Rutland Cc: Namhyung Kim Cc: Nathan Chancellor Cc: "Naveen N. Rao" Cc: Nick Desaulniers Cc: Peter Zijlstra Cc: Ravi Bangoria Cc: Rob Herring Cc: Tiezhu Yang Cc: Tom Rix Cc: Wang Nan Cc: Wang ShaoBo Cc: Yang Jihong Cc: Yonghong Song Cc: YueHaibing Link: https://lore.kernel.org/lkml/ZNZWsAXg2px1sm2h@kernel.org Signed-off-by: Arnaldo Carvalho de Melo

perf parse-events: Remove array remnants

2023-08-03T20:01:25Z

parse_events_array was set up by event term parsing, which no longer exists. Remove this struct and references to it. Signed-off-by: Ian Rogers Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Eduard Zingerman Cc: He Kuang Cc: Ingo Molnar Cc: Jiri Olsa Cc: Kan Liang Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Rob Herring Cc: Wang Nan Cc: Wang ShaoBo Cc: YueHaibing Cc: bpf@vger.kernel.org Link: https://lore.kernel.org/r/20230728001212.457900-4-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo

perf bpf-loader: Remove unneeded diagnostic pragma

2023-07-28T22:01:17Z

Added during the progress to libbpf 1.0 the deprecated functions are no longer used and so the pragma can be removed. Signed-off-by: Ian Rogers Acked-by: James Clark Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Andrii Nakryiko Cc: Eduard Zingerman Cc: Gaosheng Cui Cc: Ingo Molnar Cc: Jiri Olsa Cc: Kan Liang Cc: Mark Rutland Cc: Namhyung Kim Cc: Nathan Chancellor Cc: Nick Desaulniers Cc: Peter Zijlstra Cc: Rob Herring Cc: Tom Rix Cc: bpf@vger.kernel.org Cc: llvm@lists.linux.dev Link: https://lore.kernel.org/r/20230728064917.767761-2-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo

perf bpf: Remove pre libbpf 1.0 conditional logic

2023-03-14T11:29:46Z

Tests are no longer applicable as libbpf 1.0 can be assumed. Signed-off-by: Ian Rogers Acked/Tested-by: Jiri Olsa Acked/Tested-by: Jiri Olsa Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Andres Freund Cc: Andrii Nakryiko Cc: Christy Lee Cc: Ingo Molnar Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Quentin Monnet Cc: Roberto Sassu Cc: bpf@vger.kernel.org Link: https://lore.kernel.org/r/20230116010115.490713-2-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo

Merge remote-tracking branch 'torvalds/master' into perf/core

2022-12-16T12:53:53Z

To resolve a trivial merge conflict with c302378bc157f6a7 ("libbpf: Hashmap interface update to allow both long and void* keys/values"), where a function present upstream was removed in the perf tools development tree. Signed-off-by: Arnaldo Carvalho de Melo

perf expr: Tidy hashmap dependency

2022-11-16T15:17:15Z

hashmap.h comes from libbpf but isn't installed with its headers. Always use the header file of the code in util. Change the hashmap.h dependency in expr.h to a forward declaration, add the necessary header file includes in the C files. Signed-off-by: Ian Rogers Acked-by: Namhyung Kim Cc: Alexander Shishkin Cc: Andrii Nakryiko Cc: Jiri Olsa Cc: Mark Rutland Cc: Masahiro Yamada Cc: Nick Desaulniers Cc: Nicolas Schier Cc: Peter Zijlstra Cc: Stephane Eranian Cc: bpf@vger.kernel.org Link: http://lore.kernel.org/lkml/20221109184914.1357295-12-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo

libbpf: Hashmap interface update to allow both long and void* keys/values

2022-11-10T04:45:14Z

An update for libbpf's hashmap interface from void* -> void* to a polymorphic one, allowing both long and void* keys and values. This simplifies many use cases in libbpf as hashmaps there are mostly integer to integer. Perf copies hashmap implementation from libbpf and has to be updated as well. Changes to libbpf, selftests/bpf and perf are packed as a single commit to avoid compilation issues with any future bisect. Polymorphic interface is acheived by hiding hashmap interface functions behind auxiliary macros that take care of necessary type casts, for example: #define hashmap_cast_ptr(p) \ ({ \ _Static_assert((p) == NULL || sizeof(*(p)) == sizeof(long),\ #p " pointee should be a long-sized integer or a pointer"); \ (long *)(p); \ }) bool hashmap_find(const struct hashmap *map, long key, long *value); #define hashmap__find(map, key, value) \ hashmap_find((map), (long)(key), hashmap_cast_ptr(value)) - hashmap__find macro casts key and value parameters to long and long* respectively - hashmap_cast_ptr ensures that value pointer points to a memory of appropriate size. This hack was suggested by Andrii Nakryiko in [1]. This is a follow up for [2]. [1] https://lore.kernel.org/bpf/CAEf4BzZ8KFneEJxFAaNCCFPGqp20hSpS2aCj76uRk3-qZUH5xg@mail.gmail.com/ [2] https://lore.kernel.org/bpf/af1facf9-7bc8-8a3d-0db4-7b3f333589a2@meta.com/T/#m65b28f1d6d969fcd318b556db6a3ad499a42607d Signed-off-by: Eduard Zingerman Signed-off-by: Andrii Nakryiko Link: https://lore.kernel.org/bpf/20221109142611.879983-2-eddyz87@gmail.com

perf bpf: Fix build with libbpf 0.7.0 by checking if bpf_program__set_insns() is available

2022-10-25T20:40:48Z

During the transition to libbpf 1.0 some functions that perf used were deprecated and finally removed from libbpf, so bpf_program__set_insns() was introduced for perf to continue to use its bpf loader. But when build with LIBBPF_DYNAMIC=1 we now need to check if that function is available so that perf can build with older libbpf versions, even if the end result is emitting a warning to the user that the use of the perf BPF loader requires a newer libbpf, since bpf_program__set_insns() touches libbpf objects internal state. This affects only 'perf trace' when using bpf C code or pre-compiled bytecode as an event. Noticed on RHEL9, that has libbpf 0.7.0, where bpf_program__set_insns() isn't available. Signed-off-by: Arnaldo Carvalho de Melo

perf hashmap: Tidy hashmap dependency

2022-10-04T11:55:19Z

When libbpf is present the build uses definitions in libbpf hashmap.c, however, libbpf's hashmap.h wasn't being used. Switch to using the correct hashmap.h dependent on the define HAVE_LIBBPF_SUPPORT. This was the original intent in: https://lore.kernel.org/lkml/20200515221732.44078-8-irogers@google.com/ Signed-off-by: Ian Rogers Cc: Alexander Shishkin Cc: Jiri Olsa Cc: Mark Rutland Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lore.kernel.org/lkml/20220824050604.352156-1-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo