summaryrefslogtreecommitdiff
path: root/tools/perf/builtin-trace.c
AgeCommit message (Collapse)Author
2025-03-24perf trace: Fix wrong size to bpf_map__update_elem callThomas Richter
In linux-next commit c760174401f6 ("perf cpumap: Reduce cpu size from int to int16_t") causes the perf tests 100 126 to fail on s390: Output before: # ./perf test 100 100: perf trace BTF general tests : FAILED! # The root cause is the change from int to int16_t for the cpu maps. The size of the CPU key value pair changes from four bytes to two bytes. However a two byte key size is not supported for bpf_map__update_elem(). Note: validate_map_op() in libbpf.c emits warning libbpf: map '__augmented_syscalls__': \ unexpected key size 2 provided, expected 4 when key size is set to int16_t. Therefore change to variable size back to 4 bytes for invocation of bpf_map__update_elem(). Output after: # ./perf test 100 100: perf trace BTF general tests : Ok # Fixes: c760174401f6 ("perf cpumap: Reduce cpu size from int to int16_t") Signed-off-by: Thomas Richter <tmricht@linux.ibm.com> Reviewed-by: Ian Rogers <irogers@google.com> Acked-by: Howard Chu <howardchu95@gmail.com> Cc: James Clark <james.clark@linaro.org> Link: https://lore.kernel.org/r/20250324152756.3879571-1-tmricht@linux.ibm.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2025-03-20perf trace: Fix evlist memory leakIan Rogers
Leak sanitizer was reporting a memory leak in the "perf record and replay" test. Add evlist__delete to trace__exit, also ensure trace__exit is called after trace__record. Signed-off-by: Ian Rogers <irogers@google.com> Reviewed-by: Namhyung Kim <namhyung@kernel.org> Acked-by: Arnaldo Carvalho de Melo <acme@kernel.org> Link: https://lore.kernel.org/r/20250319050741.269828-15-irogers@google.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2025-03-20perf trace: Fix BTF memory leakIan Rogers
Add missing btf__free in trace__exit. Signed-off-by: Ian Rogers <irogers@google.com> Reviewed-by: Namhyung Kim <namhyung@kernel.org> Acked-by: Arnaldo Carvalho de Melo <acme@kernel.org> Link: https://lore.kernel.org/r/20250319050741.269828-14-irogers@google.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2025-03-20perf trace: Make syscall table stableIan Rogers
Namhyung fixed the syscall table being reallocated and moving by reloading the system call pointer after a move: https://lore.kernel.org/lkml/Z9YHCzINiu4uBQ8B@google.com/ This could be brittle so this patch changes the syscall table to be an array of pointers of "struct syscall" that don't move. Remove unnecessary copies and searches with this change. Signed-off-by: Ian Rogers <irogers@google.com> Reviewed-by: Namhyung Kim <namhyung@kernel.org> Acked-by: Arnaldo Carvalho de Melo <acme@kernel.org> Link: https://lore.kernel.org/r/20250319050741.269828-13-irogers@google.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2025-03-20perf thread: Add support for reading the e_machine type for a threadIan Rogers
First try to read the e_machine from the dsos associated with the thread's maps. If live use the executable from /proc/pid/exe and read the e_machine from the ELF header. On failure use EM_HOST. Change builtin-trace syscall functions to pass e_machine from the thread rather than EM_HOST, so that in later patches when syscalltbl can use the e_machine the system calls are specific to the architecture. Signed-off-by: Ian Rogers <irogers@google.com> Reviewed-by: Namhyung Kim <namhyung@kernel.org> Acked-by: Arnaldo Carvalho de Melo <acme@kernel.org> Link: https://lore.kernel.org/r/20250319050741.269828-8-irogers@google.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2025-03-20perf syscalltbl: Remove struct syscalltblIan Rogers
The syscalltbl held entries of system call name and number pairs, generated from a native syscalltbl at start up. As there are gaps in the system call number there is a notion of index into the table. Going forward we want the system call table to be identifiable by a machine type, for example, i386 vs x86-64. Change the interface to the syscalltbl so (1) a (currently unused machine type of EM_HOST) is passed (2) the index to syscall number and system call name mapping is computed at build time. Two tables are used for this, an array of system call number to name, an array of system call numbers sorted by the system call name. The sorted array doesn't store strings in part to save memory and relocations. The index notion is carried forward and is an index into the sorted array of system call numbers, the data structures are opaque (held only in syscalltbl.c), and so the number of indices for a machine type is exposed as a new API. The arrays are computed in the syscalltbl.sh script and so no start-up time computation and storage is necessary. Signed-off-by: Ian Rogers <irogers@google.com> Reviewed-by: Howard Chu <howardchu95@gmail.com> Reviewed-by: Charlie Jenkins <charlie@rivosinc.com> Reviewed-by: Namhyung Kim <namhyung@kernel.org> Acked-by: Arnaldo Carvalho de Melo <acme@kernel.org> Link: https://lore.kernel.org/r/20250319050741.269828-6-irogers@google.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2025-03-20perf trace: Reorganize syscallsIan Rogers
Identify struct syscall information in the syscalls table by a machine type and syscall number, not just system call number. Having the machine type means that 32-bit system calls can be differentiated from 64-bit ones on a machine capable of both. Having a table for all machine types and all system call numbers would be too large, so maintain a sorted array of system calls as they are encountered. Signed-off-by: Ian Rogers <irogers@google.com> Reviewed-by: Howard Chu <howardchu95@gmail.com> Reviewed-by: Charlie Jenkins <charlie@rivosinc.com> Reviewed-by: Namhyung Kim <namhyung@kernel.org> Acked-by: Arnaldo Carvalho de Melo <acme@kernel.org> Link: https://lore.kernel.org/r/20250319050741.269828-5-irogers@google.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2025-02-27perf trace: Add missing perf_tool__init()Athira Rajeev
Perf trace on perf.data fails as below: ./perf trace record -- sleep 1 ./perf trace -i perf.data perf: Segmentation fault Segmentation fault (core dumped) Backtrace pointed to : ?? () perf_session.process_user_event () reader.read_event () perf_session.process_events () cmd_trace () run_builtin () handle_internal_command () main () Further debug pointed that, segmentation fault happens when trying to access id_index. Code snippet: case PERF_RECORD_ID_INDEX: err = tool->id_index(session, event); Since 'commit 15d4a6f41d72 ("perf tool: Remove perf_tool__fill_defaults()")', perf_tool__fill_defaults is removed. All tools are initialized using perf_tool__init() prior to use. But in builtin-trace, perf_tool__init is not used and hence the defaults are not initialized. Use perf_tool__init() in perf trace to handle the initialization. Reported-by: Tejas Manhas <Tejas.Manhas1@ibm.com> Signed-off-by: Athira Rajeev <atrajeev@linux.ibm.com> Link: https://lore.kernel.org/r/20250225113157.28836-1-atrajeev@linux.ibm.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2025-02-12perf sample: Make user_regs and intr_regs optionalIan Rogers
The struct dump_regs contains 512 bytes of cache_regs, meaning the two values in perf_sample contribute 1088 bytes of its total 1384 bytes size. Initializing this much memory has a cost reported by Tavian Barnes <tavianator@tavianator.com> as about 2.5% when running `perf script --itrace=i0`: https://lore.kernel.org/lkml/d841b97b3ad2ca8bcab07e4293375fb7c32dfce7.1736618095.git.tavianator@tavianator.com/ Adrian Hunter <adrian.hunter@intel.com> replied that the zero initialization was necessary and couldn't simply be removed. This patch aims to strike a middle ground of still zeroing the perf_sample, but removing 79% of its size by make user_regs and intr_regs optional pointers to zalloc-ed memory. To support the allocation accessors are created for user_regs and intr_regs. To support correct cleanup perf_sample__init and perf_sample__exit functions are created and added throughout the code base. Signed-off-by: Ian Rogers <irogers@google.com> Link: https://lore.kernel.org/r/20250113194345.1537821-1-irogers@google.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2025-02-12perf trace: Add --summary-mode optionNamhyung Kim
The --summary-mode option will select how to show the syscall summary at the end. By default, it'll show the summary for each thread and it's the same as if --summary-mode=thread is passed. The other option is to show total summary, which is --summary-mode=total. I'd like to have this instead of a separate option like --total-summary because we may want to add a new summary mode (by cgroup) later. $ sudo ./perf trace -as --summary-mode=total sleep 1 Summary of events: total, 21580 events syscall calls errors total min avg max stddev (msec) (msec) (msec) (msec) (%) --------------- -------- ------ -------- --------- --------- --------- ------ epoll_wait 1305 0 14716.712 0.000 11.277 551.529 8.87% futex 1256 89 13331.197 0.000 10.614 733.722 15.49% poll 669 0 6806.618 0.000 10.174 459.316 11.77% ppoll 220 0 3968.797 0.000 18.040 516.775 25.35% clock_nanosleep 1 0 1000.027 1000.027 1000.027 1000.027 0.00% epoll_pwait 21 0 592.783 0.000 28.228 522.293 88.29% nanosleep 16 0 60.515 0.000 3.782 10.123 33.33% ioctl 510 0 4.284 0.001 0.008 0.182 8.84% recvmsg 1434 775 3.497 0.001 0.002 0.174 6.37% write 1393 0 2.854 0.001 0.002 0.017 1.79% read 1063 100 2.236 0.000 0.002 0.083 5.11% ... Reviewed-by: Howard Chu <howardchu95@gmail.com> Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Link: https://lore.kernel.org/r/20250205205443.1986408-5-namhyung@kernel.org Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2025-02-12perf trace: Convert syscall_stats to hashmapNamhyung Kim
It was using a RBtree-based int-list as a hash and a custom resort logic for that. As we have hashmap, let's convert to it and add a custom sort function for the hashmap entries using an array. It should be faster and more light-weighted. It's also to prepare supporting system-wide syscall stats. No functional changes intended. Reviewed-by: Howard Chu <howardchu95@gmail.com> Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Link: https://lore.kernel.org/r/20250205205443.1986408-3-namhyung@kernel.org Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2025-02-12perf trace: Allocate syscall stats only if summary is onNamhyung Kim
The syscall stats are used only when summary is requested. Let's avoid unnecessary operations. While at it, let's pass 'trace' pointer directly instead of passing 'output' file pointer and 'summary' option in the 'trace' separately. Reviewed-by: Howard Chu <howardchu95@gmail.com> Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Link: https://lore.kernel.org/r/20250205205443.1986408-2-namhyung@kernel.org Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2025-02-05Merge tag 'v6.14-rc1' into perf-tools-nextNamhyung Kim
To get the various fixes in the current master. Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2025-01-28perf trace: Fix runtime error of index out of boundsHoward Chu
libtraceevent parses and returns an array of argument fields, sometimes larger than RAW_SYSCALL_ARGS_NUM (6) because it includes "__syscall_nr", idx will traverse to index 6 (7th element) whereas sc->fmt->arg holds 6 elements max, creating an out-of-bounds access. This runtime error is found by UBsan. The error message: $ sudo UBSAN_OPTIONS=print_stacktrace=1 ./perf trace -a --max-events=1 builtin-trace.c:1966:35: runtime error: index 6 out of bounds for type 'syscall_arg_fmt [6]' #0 0x5c04956be5fe in syscall__alloc_arg_fmts /home/howard/hw/linux-perf/tools/perf/builtin-trace.c:1966 #1 0x5c04956c0510 in trace__read_syscall_info /home/howard/hw/linux-perf/tools/perf/builtin-trace.c:2110 #2 0x5c04956c372b in trace__syscall_info /home/howard/hw/linux-perf/tools/perf/builtin-trace.c:2436 #3 0x5c04956d2f39 in trace__init_syscalls_bpf_prog_array_maps /home/howard/hw/linux-perf/tools/perf/builtin-trace.c:3897 #4 0x5c04956d6d25 in trace__run /home/howard/hw/linux-perf/tools/perf/builtin-trace.c:4335 #5 0x5c04956e112e in cmd_trace /home/howard/hw/linux-perf/tools/perf/builtin-trace.c:5502 #6 0x5c04956eda7d in run_builtin /home/howard/hw/linux-perf/tools/perf/perf.c:351 #7 0x5c04956ee0a8 in handle_internal_command /home/howard/hw/linux-perf/tools/perf/perf.c:404 #8 0x5c04956ee37f in run_argv /home/howard/hw/linux-perf/tools/perf/perf.c:448 #9 0x5c04956ee8e9 in main /home/howard/hw/linux-perf/tools/perf/perf.c:556 #10 0x79eb3622a3b7 in __libc_start_call_main ../sysdeps/nptl/libc_start_call_main.h:58 #11 0x79eb3622a47a in __libc_start_main_impl ../csu/libc-start.c:360 #12 0x5c04955422d4 in _start (/home/howard/hw/linux-perf/tools/perf/perf+0x4e02d4) (BuildId: 5b6cab2d59e96a4341741765ad6914a4d784dbc6) 0.000 ( 0.014 ms): Chrome_ChildIO/117244 write(fd: 238, buf: !, count: 1) = 1 Fixes: 5e58fcfaf4c6 ("perf trace: Allow allocating sc->arg_fmt even without the syscall tracepoint") Signed-off-by: Howard Chu <howardchu95@gmail.com> Link: https://lore.kernel.org/r/20250122025519.361873-1-howardchu95@gmail.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2025-01-24perf trace: Fix return value of trace__fprintf_tp_fieldsBenjamin Peterson
This function formerly returned twice the number of bytes printed. Signed-off-by: Benjamin Peterson <benjamin@engflow.com> Reviewed-by: Howard Chu <howardchu95@gmail.com> Link: https://lore.kernel.org/r/20250123-void-fprintf_tp_fields-v2-1-6038f8224987@engflow.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2025-01-10perf trace: Fix unaligned access for augmented argsNamhyung Kim
Some version of compilers reported unaligned accesses in perf trace when undefined-behavior sanitizer is on. I found that it uses raw data in the sample directly and assuming it's properly aligned. Unlike other sample fields, the raw data is not 8-byte aligned because there's a size field (u32) before the actual data. So I added a static buffer in syscall__augmented_args() and return it instead. This is not ideal but should work well as perf trace is single-threaded. A better approach would be aligning the raw data by adding a 4-byte data before the augmented args but I'm afraid it'd break the backward compatibility. Committer testing: To build with the undefined behaviour sanitizer: $ make CC=clang EXTRA_CFLAGS=-fsanitize=undefined -C tools/perf Checking if the resulting binary is instrumented: root@number:~# nm ~/bin/perf | grep ubsan | wc -l 113 root@number:~# nm ~/bin/perf | grep ubsan | tail -5 000000000043d5b0 t _ZN7__ubsanL19UBsanOnDeadlySignalEiPvS0_ 000000000043ce50 T _ZNK7__ubsan5Value12getSIntValueEv 000000000043cf40 T _ZNK7__ubsan5Value12getUIntValueEv 000000000043d140 T _ZNK7__ubsan5Value13getFloatValueEv 000000000043cfd0 T _ZNK7__ubsan5Value19getPositiveIntValueEv root@number:~# Now running something that will access timespec, as reported in the Closes URL: root@number:~# perf trace --max-events=1 -e *nano* sleep 1.1 trace/beauty/timespec.c:10:64: runtime error: member access within misaligned address 0x7fc583cfb2a4 for type 'struct augmented_arg', which requires 8 byte alignment 0x7fc583cfb2a4: note: pointer points here 99 99 11 00 10 00 00 00 00 00 00 00 01 00 00 00 00 00 00 00 01 e1 f5 05 00 00 00 00 00 00 00 00 ^ SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior trace/beauty/timespec.c:10:64 <SNIP> As Namhyung said we need to make the raw_data to be 64-bit aligned, probably we need to add a PERF_SAMPLE_ALIGNED_RAW with a 64-bit raw_size instead of the current u32 done at kernel/events/core.c, perf_output_sample(), that perf_output_put(handle, raw->size) where raw->size is an u32 and then the raw_data is always 64-bit unaligned... After the patch: root@number:~# perf trace -e *nano* sleep 1.1 0.000 (1100.064 ms): sleep/1984224 clock_nanosleep(rqtp: { .tv_sec: 1, .tv_nsec: 100000001 }, rmtp: 0x7fff5b3fe970) = 0 root@number:~# Closes: https://lore.kernel.org/r/Z2STgyD1p456Qqhg@google.com Reviewed-by: Howard Chu <howardchu95@gmail.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/r/20250102201248.790841-1-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2025-01-10perf tools: Remove dependency on libauditCharlie Jenkins
All architectures now support HAVE_SYSCALL_TABLE_SUPPORT, so the flag is no longer needed. With the removal of the flag, the related GENERIC_SYSCALL_TABLE can also be removed. libaudit was only used as a fallback for when HAVE_SYSCALL_TABLE_SUPPORT was not defined, so libaudit is also no longer needed for any architecture. Signed-off-by: Charlie Jenkins <charlie@rivosinc.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Christian Brauner <brauner@kernel.org> Cc: Guo Ren <guoren@kernel.org> Cc: Günther Noack <gnoack@google.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: James Clark <james.clark@linaro.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: John Garry <john.g.garry@oracle.com> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Leo Yan <leo.yan@linux.dev> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Mickaël Salaün <mic@digikod.net> Cc: Mike Leach <mike.leach@linaro.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Palmer Dabbelt <palmer@dabbelt.com> Cc: Paul Walmsley <paul.walmsley@sifive.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Will Deacon <will@kernel.org> Link: https://lore.kernel.org/r/20250108-perf_syscalltbl-v6-16-7543b5293098@rivosinc.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-12-18perf env: Move arch errno function to only use in envIan Rogers
Move arch_syscalls__strerrno_function out of builtin-trace.c to env.c so that there isn't a util to builtin function call. This allows the python.c stub to be removed. Also, remove declaration/prototype from env.h and make static to reduce scope. The include is moved inside ifdefs to avoid, "defined but unused warnings". Signed-off-by: Ian Rogers <irogers@google.com> Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Athira Rajeev <atrajeev@linux.vnet.ibm.com> Cc: Colin Ian King <colin.i.king@gmail.com> Cc: Dapeng Mi <dapeng1.mi@linux.intel.com> Cc: Howard Chu <howardchu95@gmail.com> Cc: Ilya Leoshkevich <iii@linux.ibm.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: James Clark <james.clark@linaro.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Michael Petlan <mpetlan@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Richter <tmricht@linux.ibm.com> Cc: Veronika Molnarova <vmolnaro@redhat.com> Cc: Weilin Wang <weilin.wang@intel.com> Link: https://lore.kernel.org/r/20241119011644.971342-15-irogers@google.com perf: perf python: Correctly throw IndexError Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-12-09perf evsel: Add/use accessor for tp_formatIan Rogers
Add an accessor function for tp_format. Rather than search+replace uses try to use a variable and reuse it. Add additional NULL checks when accessing/using the value. Make sure the PTR_ERR is nulled out on error path in evsel__newtp_idx. Reviewed-by: Namhyung Kim <namhyung@kernel.org> Signed-off-by: Ian Rogers <irogers@google.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Athira Rajeev <atrajeev@linux.vnet.ibm.com> Cc: Ben Gainey <ben.gainey@arm.com> Cc: Colin Ian King <colin.i.king@gmail.com> Cc: Dominique Martinet <asmadeus@codewreck.org> Cc: Ilkka Koskinen <ilkka@os.amperecomputing.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: James Clark <james.clark@linaro.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Oliver Upton <oliver.upton@linux.dev> Cc: Paran Lee <p4ranlee@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Steinar H. Gunderson <sesse@google.com> Cc: Steven Rostedt (VMware) <rostedt@goodmis.org> Cc: Thomas Falcon <thomas.falcon@intel.com> Cc: Weilin Wang <weilin.wang@intel.com> Cc: Yang Jihong <yangjihong@bytedance.com> Cc: Yang Li <yang.lee@linux.alibaba.com> Cc: Ze Gao <zegao2021@gmail.com> Cc: Zixian Cai <fzczx123@gmail.com> Cc: zhaimingbing <zhaimingbing@cmss.chinamobile.com> Link: https://lore.kernel.org/r/20241118225345.889810-6-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-11-14perf trace: Avoid garbage when not printing a syscall's argumentsBenjamin Peterson
syscall__scnprintf_args may not place anything in the output buffer (e.g., because the arguments are all zero). If that happened in trace__fprintf_sys_enter, its fprintf would receive an unitialized buffer leading to garbage output. Fix the problem by passing the (possibly zero) bounds of the argument buffer to the output fprintf. Fixes: a98392bb1e169a04 ("perf trace: Use beautifiers on syscalls:sys_enter_ handlers") Signed-off-by: Benjamin Peterson <benjamin@engflow.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Howard Chu <howardchu95@gmail.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/r/20241107232128.108981-2-benjamin@engflow.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-11-14perf trace: Do not lose last events in a raceBenjamin Peterson
If a perf trace event selector specifies a maximum number of events to output (i.e., "/nr=N/" syntax), the event printing handler, trace__event_handler, disables the event selector after the maximum number events are printed. Furthermore, trace__event_handler checked if the event selector was disabled before doing any work. This avoided exceeding the maximum number of events to print if more events were in the buffer before the selector was disabled. However, the event selector can be disabled for reasons other than exceeding the maximum number of events. In particular, when the traced subprocess exits, the main loop disables all event selectors. This meant the last events of a traced subprocess might be lost to the printing handler's short-circuiting logic. This nondeterministic problem could be seen by running the following many times: $ perf trace -e syscalls:sys_enter_exit_group true trace__event_handler should simply check for exceeding the maximum number of events to print rather than the state of the event selector. Fixes: a9c5e6c1e9bff42c ("perf trace: Introduce per-event maximum number of events property") Signed-off-by: Benjamin Peterson <benjamin@engflow.com> Tested-by: Howard Chu <howardchu95@gmail.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/r/20241107232128.108981-1-benjamin@engflow.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-11-14perf trace: Fix tracing itself, creating feedback loopsHoward Chu
There exists a pids_filtered map in augmented_raw_syscalls.bpf.c that ceases to provide functionality after the BPF skeleton migration done in: 5e6da6be3082f77b ("perf trace: Migrate BPF augmentation to use a skeleton") Before the migration, pid_filtered map works, courtesy of Arnaldo Carvalho de Melo <acme@kernel.org>: ⬢ [acme@toolbox perf-tools]$ git log --oneline -5 6f769c3458b6cf2d (HEAD) perf tests trace+probe_vfs_getname.sh: Accept quotes surrounding the filename 7777ac3dfe29f55d perf test trace+probe_vfs_getname.sh: Remove stray \ before / 33d9c5062113a4bd perf script python: Add stub for PMU symbol to the python binding e59fea47f83e8a9a perf symbols: Fix DSO kernel load and symbol process to correctly map DSO to its long_name, type and adjust_symbols 878460e8d0ff84a0 perf build: Remove -Wno-unused-but-set-variable from the flex flags when building with clang < 13.0.0 root@x1:/home/acme/git/perf-tools# perf trace -e /tmp/augmented_raw_syscalls.o -e write* --max-events=30 & [1] 180632 root@x1:/home/acme/git/perf-tools# 0.000 ( 0.051 ms): NetworkManager/1127 write(fd: 3, buf: 0x7ffeb508ef70, count: 8) = 8 0.115 ( 0.010 ms): NetworkManager/1127 write(fd: 3, buf: 0x7ffeb508ef70, count: 8) = 8 0.916 ( 0.068 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 246) = 246 1.699 ( 0.047 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 121) = 121 2.167 ( 0.041 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 121) = 121 2.739 ( 0.042 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 121) = 121 3.138 ( 0.027 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 121) = 121 3.477 ( 0.027 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 121) = 121 3.738 ( 0.023 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 121) = 121 3.946 ( 0.024 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 121) = 121 4.195 ( 0.024 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 121) = 121 4.212 ( 0.026 ms): NetworkManager/1127 write(fd: 3, buf: 0x7ffeb508ef70, count: 8) = 8 4.285 ( 0.006 ms): NetworkManager/1127 write(fd: 3, buf: 0x7ffeb508ef70, count: 8) = 8 4.445 ( 0.018 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 260) = 260 4.508 ( 0.009 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 124) = 124 4.592 ( 0.010 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 116) = 116 4.666 ( 0.009 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 130) = 130 4.715 ( 0.010 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 95) = 95 4.765 ( 0.007 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 102) = 102 4.815 ( 0.009 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 79) = 79 4.890 ( 0.008 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 57) = 57 4.937 ( 0.007 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 89) = 89 5.009 ( 0.010 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 112) = 112 5.059 ( 0.010 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 112) = 112 5.116 ( 0.007 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 79) = 79 5.152 ( 0.009 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 33) = 33 5.215 ( 0.008 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 37) = 37 5.293 ( 0.010 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 128) = 128 5.339 ( 0.009 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 89) = 89 5.384 ( 0.008 ms): sudo/156867 write(fd: 8, buf: 0x55cb4cd2f650, count: 100) = 100 [1]+ Done perf trace -e /tmp/augmented_raw_syscalls.o -e write* --max-events=30 root@x1:/home/acme/git/perf-tools# No events for the 'perf trace' (pid 180632), i.e. no feedback loop. If we leave it running: root@x1:/home/acme/git/perf-tools# perf trace -e /tmp/augmented_raw_syscalls.o -e landlock_add_rule & [1] 181068 root@x1:/home/acme/git/perf-tools# And then look at what maps it sets up: root@x1:/home/acme/git/perf-tools# bpftool map | grep pids_filtered -A3 1190: hash name pids_filtered flags 0x0 key 4B value 1B max_entries 64 memlock 7264B btf_id 1613 pids perf(181068) root@x1:/home/acme/git/perf-tools# And ask for dumping its contents: We see that we are _also_ setting it to filter those: root@x1:/home/acme/git/perf-tools# bpftool map dump id 1190 [{ "key": 181068, "value": 1 },{ "key": 156801, "value": 1 } ] Now testing the migration commit: perf $ git log commit 5e6da6be3082f77be06894a1a94d52a90b4007dc (HEAD) Author: Ian Rogers <irogers@google.com> Date: Thu Aug 10 11:48:51 2023 -0700 perf trace: Migrate BPF augmentation to use a skeleton perf $ ./perf trace -e write --max-events=10 & echo #! [1] 1808653 perf $ 0.000 ( 0.010 ms): :1808671/1808671 write(fd: 1, buf: 0x6003f5b26fc0, count: 11) = 11 0.162 ( ): perf/1808653 write(fd: 2, buf: 0x7fffc2174e50, count: 11) ... 0.174 ( ): perf/1808653 write(fd: 2, buf: 0x74ce21804563, count: 1) ... 0.184 ( ): perf/1808653 write(fd: 2, buf: 0x57b936589052, count: 5) The feedback loop is there. Keep it running, look into the bpf map: perf $ bpftool map | grep pids_filtered 10675: hash name pids_filtered flags 0x0 perf $ bpftool map dump id 10675 [] The map is empty. Now, this commit: 64917f4df048a064 ("perf trace: Use heuristic when deciding if a syscall tracepoint "const char *" field is really a string") Temporarily fixed the feedback loop for perf trace -e write, that's because before using the heuristic, write is hooked to sys_enter_openat: perf $ git log commit 83a0943b1870944612a8aa0049f910826ebfd4f7 (HEAD) Author: Arnaldo Carvalho de Melo <acme@redhat.com> Date: Thu Aug 17 12:11:51 2023 -0300 perf trace: Use the augmented_raw_syscall BPF skel only for tracing syscalls perf $ ./perf trace -e write --max-events=10 -v 2>&1 | grep Reusing Reusing "openat" BPF sys_enter augmenter for "write" And after the heuristic fix, it's unaugmented: perf $ git log commit 64917f4df048a0649ea7901c2321f020e71e6f24 (HEAD) Author: Arnaldo Carvalho de Melo <acme@redhat.com> Date: Thu Aug 17 15:14:21 2023 -0300 perf trace: Use heuristic when deciding if a syscall tracepoint "const char *" field is really a string perf $ ./perf trace -e write --max-events=10 -v 2>&1 | grep Reusing perf $ After using the heuristic, write is hooked to syscall_unaugmented, which returns 1. SEC("tp/raw_syscalls/sys_enter") int syscall_unaugmented(struct syscall_enter_args *args) { return 1; } If the BPF program returns 1, the tracepoint filter will filter it (since the tracepoint filter for perf is correctly set), but before the heuristic, when it was hooked to a sys_enter_openat(), which is a BPF program that calls bpf_perf_event_output() and writes to the buffer, it didn't get filtered, thus creating feedback loop. So switching write to unaugmented accidentally fixed the problem. But some syscalls are not so lucky, for example newfstatat: perf $ ./perf trace -e newfstatat --max-events=100 & echo #! [1] 2166948 457.718 ( ): perf/2166948 newfstatat(dfd: CWD, filename: "/proc/self/ns/mnt", statbuf: 0x7fff0132a9f0) ... 457.749 ( ): perf/2166948 newfstatat(dfd: CWD, filename: "/proc/2166950/ns/mnt", statbuf: 0x7fff0132aa80) ... 457.962 ( ): perf/2166948 newfstatat(dfd: CWD, filename: "/proc/self/ns/mnt", statbuf: 0x7fff0132a9f0) ... Currently, write is augmented by the new BTF general augmenter (which calls bpf_perf_event_output()). The problem, which luckily got fixed, resurfaced, and that’s how it was discovered. Fixes: 5e6da6be3082f77b ("perf trace: Migrate BPF augmentation to use a skeleton") Signed-off-by: Howard Chu <howardchu95@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: James Clark <james.clark@linaro.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/r/20241030052431.2220130-1-howardchu95@gmail.com [ Check if trace->skel is non-NULL, as it is only initialized if trace->trace_syscalls is set ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-11-08perf build: Include libtraceevent headers directly indicated by pkg-configYicong Yang
Currently the libtraceevent's found by pkg-config, which give the include path as: [root@localhost tmp]# pkg-config --cflags libtraceevent -I/usr/local/include/traceevent So we should include the libtraceevent headers directly without "traceevent/" prefix. Update all the users. Fixes: 0f0e1f445690 ("perf build: Use pkg-config for feature check for libtrace{event,fs}") Suggested-by: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/linux-perf-users/ZyF5_Hf1iL01kldE@google.com/ Signed-off-by: Yicong Yang <yangyicong@hisilicon.com> Cc: leo.yan@arm.com Cc: amadio@gentoo.org Cc: linuxarm@huawei.com Link: https://lore.kernel.org/r/20241105105649.45399-1-yangyicong@huawei.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2024-11-05perf trace: avoid garbage when not printing a trace event's argumentsBenjamin Peterson
trace__fprintf_tp_fields may not print any tracepoint arguments. E.g., if the argument values are all zero. Previously, this would result in a totally uninitialized buffer being passed to fprintf, which could lead to garbage on the console. Fix the problem by passing the number of initialized bytes fprintf. Fixes: f11b2803bb88 ("perf trace: Allow choosing how to augment the tracepoint arguments") Signed-off-by: Benjamin Peterson <benjamin@engflow.com> Tested-by: Howard Chu <howardchu95@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Link: https://lore.kernel.org/r/20241103204816.7834-1-benjamin@engflow.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2024-11-03Merge 'origin/master' into perf-tools-nextNamhyung Kim
To get the fixes in the perf-tools branch. Resolved a conflict due to RISC-V's syscall table change. Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2024-10-17perf color: Add printf format checking and resolve issuesIan Rogers
Add printf format checking to vararg printf routines in color.h. Resolve build errors/bugs that are found through this checking. Signed-off-by: Ian Rogers <irogers@google.com> Cc: Yicong Yang <yangyicong@hisilicon.com> Cc: Weilin Wang <weilin.wang@intel.com> Cc: Will Deacon <will@kernel.org> Cc: James Clark <james.clark@linaro.org> Cc: Mike Leach <mike.leach@linaro.org> Cc: Leo Yan <leo.yan@linux.dev> Cc: Sumanth Korikkar <sumanthk@linux.ibm.com> Cc: Thomas Richter <tmricht@linux.ibm.com> Cc: Tim Chen <tim.c.chen@linux.intel.com> Cc: John Garry <john.g.garry@oracle.com> Link: https://lore.kernel.org/r/20241017175356.783793-2-irogers@google.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2024-10-17perf trace: The return from 'write' isn't a pidArnaldo Carvalho de Melo
When adding a explicit beautifier for the 'write' syscall when the BPF based buffer collector was introduced there was a cut'n'paste error that carried the syscall_fmt->errpid setting from a nearby syscall (waitid) that returns a pid. So the write return was being suppressed by the return pretty printer, remove that field, reverting it back to the default return handler, that prints positive numbers as-is and interpret negative values as errnos. I actually introduced the problem while making Howard's original patch work just with the 'write' syscall, as we couldn't just look for any buffers, the ones that are filled in by the kernel couldn't use the same sys_enter BPF collector. Fixes: b257fac12f38d7f5 ("perf trace: Pretty print buffer data") Reported-by: James Clark <james.clark@linaro.org> Link: https://lore.kernel.org/lkml/bcf50648-3c7e-4513-8717-0d14492c53b9@linaro.org Link: https://lore.kernel.org/all/Zt8jTfzDYgBPvFCd@x1/#t Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alan Maguire <alan.maguire@oracle.com> Cc: Howard Chu <howardchu95@gmail.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-10-02perf trace: Keep exited threads for summaryMichael Petlan
Since 9ffa6c7512ca ("perf machine thread: Remove exited threads by default") perf cleans exited threads up, but as said, sometimes they are necessary to be kept. The mentioned commit does not cover all the cases, we also need the information to construct the summary table in perf-trace. Before: # perf trace -s true Summary of events: After: # perf trace -s -- true Summary of events: true (383382), 64 events, 91.4% syscall calls errors total min avg max stddev (msec) (msec) (msec) (msec) (%) --------------- -------- ------ -------- --------- --------- --------- ------ mmap 8 0 0.150 0.013 0.019 0.031 11.90% mprotect 3 0 0.045 0.014 0.015 0.017 6.47% openat 2 0 0.014 0.006 0.007 0.007 9.73% munmap 1 0 0.009 0.009 0.009 0.009 0.00% access 1 1 0.009 0.009 0.009 0.009 0.00% pread64 4 0 0.006 0.001 0.001 0.002 4.53% fstat 2 0 0.005 0.001 0.002 0.003 37.59% arch_prctl 2 1 0.003 0.001 0.002 0.002 25.91% read 1 0 0.003 0.003 0.003 0.003 0.00% close 2 0 0.003 0.001 0.001 0.001 3.86% brk 1 0 0.002 0.002 0.002 0.002 0.00% rseq 1 0 0.001 0.001 0.001 0.001 0.00% prlimit64 1 0 0.001 0.001 0.001 0.001 0.00% set_robust_list 1 0 0.001 0.001 0.001 0.001 0.00% set_tid_address 1 0 0.001 0.001 0.001 0.001 0.00% execve 1 0 0.000 0.000 0.000 0.000 0.00% [namhyung: simplified the condition] Fixes: 9ffa6c7512ca ("perf machine thread: Remove exited threads by default") Reported-by: Veronika Molnarova <vmolnaro@redhat.com> Signed-off-by: Michael Petlan <mpetlan@redhat.com> Link: https://lore.kernel.org/r/20240927151926.399474-1-mpetlan@redhat.com Signed-off-by: Namhyung Kim <namhyung@kernel.org>
2024-09-11perf trace: Mark the 'head' arg in the set_robust_list syscall as coming ↵Arnaldo Carvalho de Melo
from user space With that it uses the generic BTF based pretty printer: This one we need to think about, not being acquainted with this syscall, should we _traverse_ that list somehow? Would that be useful? root@number:~# perf trace -e set_robust_list sleep 1 0.000 ( 0.004 ms): sleep/1206493 set_robust_list(head: (struct robust_list_head){.list = (struct robust_list){.next = (struct robust_list *)0x7f48a9a02a20,},.futex_offset = (long int)-32,}, len: 24) = root@number:~# strace prints the default integer args: root@number:~# strace -e set_robust_list sleep 1 set_robust_list(0x7efd99559a20, 24) = 0 +++ exited with 0 +++ root@number:~# Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alan Maguire <alan.maguire@oracle.com> Cc: Howard Chu <howardchu95@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org Link: https://lore.kernel.org/lkml/ZuH6MquMraBvODRp@x1 Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-09-11perf trace: Mark the 'rseq' arg in the rseq syscall as coming from user spaceArnaldo Carvalho de Melo
With that it uses the generic BTF based pretty printer: root@number:~# grep -w rseq /sys/kernel/tracing/events/syscalls/sys_enter_rseq/format field:struct rseq * rseq; offset:16; size:8; signed:0; print fmt: "rseq: 0x%08lx, rseq_len: 0x%08lx, flags: 0x%08lx, sig: 0x%08lx", ((unsigned long)(REC->rseq)), ((unsigned long)(REC->rseq_len)), ((unsigned long)(REC->flags)), ((unsigned long)(REC->sig)) root@number:~# Before: root@number:~# perf trace -e rseq 0.000 ( 0.017 ms): Isolated Web C/1195452 rseq(rseq: 0x7ff0ecfe6fe0, rseq_len: 32, sig: 1392848979) = 0 74.018 ( 0.006 ms): :1195453/1195453 rseq(rseq: 0x7f2af20fffe0, rseq_len: 32, sig: 1392848979) = 0 1817.220 ( 0.009 ms): Isolated Web C/1195454 rseq(rseq: 0x7f5c9ec7dfe0, rseq_len: 32, sig: 1392848979) = 0 2515.526 ( 0.034 ms): :1195455/1195455 rseq(rseq: 0x7f61503fffe0, rseq_len: 32, sig: 1392848979) = 0 ^Croot@number:~# After: root@number:~# perf trace -e rseq 0.000 ( 0.019 ms): Isolated Web C/1197258 rseq(rseq: (struct rseq){.cpu_id_start = (__u32)4,.cpu_id = (__u32)4,.mm_cid = (__u32)5,}, rseq_len: 32, sig: 1392848979) = 0 1663.835 ( 0.019 ms): Isolated Web C/1197259 rseq(rseq: (struct rseq){.cpu_id_start = (__u32)24,.cpu_id = (__u32)24,.mm_cid = (__u32)2,}, rseq_len: 32, sig: 1392848979) = 0 4750.444 ( 0.018 ms): Isolated Web C/1197260 rseq(rseq: (struct rseq){.cpu_id_start = (__u32)8,.cpu_id = (__u32)8,.mm_cid = (__u32)4,}, rseq_len: 32, sig: 1392848979) = 0 4994.132 ( 0.018 ms): Isolated Web C/1197261 rseq(rseq: (struct rseq){.cpu_id_start = (__u32)10,.cpu_id = (__u32)10,.mm_cid = (__u32)1,}, rseq_len: 32, sig: 1392848979) = 0 4997.578 ( 0.011 ms): Isolated Web C/1197263 rseq(rseq: (struct rseq){.cpu_id_start = (__u32)16,.cpu_id = (__u32)16,.mm_cid = (__u32)4,}, rseq_len: 32, sig: 1392848979) = 0 4997.462 ( 0.014 ms): Isolated Web C/1197262 rseq(rseq: (struct rseq){.cpu_id_start = (__u32)17,.cpu_id = (__u32)17,.mm_cid = (__u32)3,}, rseq_len: 32, sig: 1392848979) = 0 ^Croot@number:~# We'll probably need to come up with some way for using the BTF info to synthesize a test that then gets used and captures the output of the 'perf trace' output to check if the arguments are the ones synthesized, randomically, for now, lets make do manually: root@number:~# cat ~acme/c/rseq.c #include <sys/syscall.h> /* Definition of SYS_* constants */ #include <linux/rseq.h> #include <errno.h> #include <string.h> #include <unistd.h> #include <stdint.h> #include <stdio.h> /* Provide own rseq stub because glibc doesn't */ __attribute__((weak)) int sys_rseq(struct rseq *rseq, __u32 rseq_len, int flags, __u32 sig) { return syscall(SYS_rseq, rseq, rseq_len, flags, sig); } int main(int argc, char *argv[]) { struct rseq rseq = { .cpu_id_start = 12, .cpu_id = 34, .rseq_cs = 56, .flags = 78, .node_id = 90, .mm_cid = 12, }; int err = sys_rseq(&rseq, sizeof(rseq), 98765, 0xdeadbeaf); printf("sys_rseq({ .cpu_id_start = 12, .cpu_id = 34, .rseq_cs = 56, .flags = 78, .node_id = 90, .mm_cid = 12, }, %d, 0) = %d (%s)\n", sizeof(rseq), err, strerror(errno)); return err; } root@number:~# perf trace -e rseq ~acme/c/rseq sys_rseq({ .cpu_id_start = 12, .cpu_id = 34, .rseq_cs = 56, .flags = 78, .node_id = 90, .mm_cid = 12, }, 32, 0) = -1 (Invalid argument) 0.000 ( 0.003 ms): rseq/1200640 rseq(rseq: (struct rseq){}, rseq_len: 32, sig: 1392848979) = 0.064 ( 0.001 ms): rseq/1200640 rseq(rseq: (struct rseq){.cpu_id_start = (__u32)12,.cpu_id = (__u32)34,.rseq_cs = (__u64)56,.flags = (__u32)78,.node_id = (__u32)90,.mm_cid = (__u32)12,}, rseq_len: 32, flags: 98765, sig: 3735928495) = -1 EINVAL (Invalid argument) root@number:~#root@number:~# cat ~acme/c/rseq.c #include <sys/syscall.h> /* Definition of SYS_* constants */ #include <linux/rseq.h> #include <errno.h> #include <string.h> #include <unistd.h> #include <stdint.h> #include <stdio.h> /* Provide own rseq stub because glibc doesn't */ __attribute__((weak)) int sys_rseq(struct rseq *rseq, __u32 rseq_len, int flags, __u32 sig) { return syscall(SYS_rseq, rseq, rseq_len, flags, sig); } int main(int argc, char *argv[]) { struct rseq rseq = { .cpu_id_start = 12, .cpu_id = 34, .rseq_cs = 56, .flags = 78, .node_id = 90, .mm_cid = 12, }; int err = sys_rseq(&rseq, sizeof(rseq), 98765, 0xdeadbeaf); printf("sys_rseq({ .cpu_id_start = 12, .cpu_id = 34, .rseq_cs = 56, .flags = 78, .node_id = 90, .mm_cid = 12, }, %d, 0) = %d (%s)\n", sizeof(rseq), err, strerror(errno)); return err; } root@number:~# perf trace -e rseq ~acme/c/rseq sys_rseq({ .cpu_id_start = 12, .cpu_id = 34, .rseq_cs = 56, .flags = 78, .node_id = 90, .mm_cid = 12, }, 32, 0) = -1 (Invalid argument) 0.000 ( 0.003 ms): rseq/1200640 rseq(rseq: (struct rseq){}, rseq_len: 32, sig: 1392848979) = 0.064 ( 0.001 ms): rseq/1200640 rseq(rseq: (struct rseq){.cpu_id_start = (__u32)12,.cpu_id = (__u32)34,.rseq_cs = (__u64)56,.flags = (__u32)78,.node_id = (__u32)90,.mm_cid = (__u32)12,}, rseq_len: 32, flags: 98765, sig: 3735928495) = -1 EINVAL (Invalid argument) root@number:~# Interesting, glibc seems to be using rseq here, as in addition to the totally fake one this test case uses, we have this one, around these other syscalls: 0.175 ( 0.001 ms): rseq/1201095 set_tid_address(tidptr: 0x7f6def759a10) = 1201095 (rseq) 0.177 ( 0.001 ms): rseq/1201095 set_robust_list(head: 0x7f6def759a20, len: 24) = 0 0.178 ( 0.001 ms): rseq/1201095 rseq(rseq: (struct rseq){}, rseq_len: 32, sig: 1392848979) = 0.231 ( 0.005 ms): rseq/1201095 mprotect(start: 0x7f6def93f000, len: 16384, prot: READ) = 0 0.238 ( 0.003 ms): rseq/1201095 mprotect(start: 0x403000, len: 4096, prot: READ) = 0 0.244 ( 0.004 ms): rseq/1201095 mprotect(start: 0x7f6def99c000, len: 8192, prot: READ) Matches strace (well, not really as the strace in fedora:40 doesn't know about rseq, printing just integer values in hex): set_robust_list(0x7fbc6acc7a20, 24) = 0 rseq(0x7fbc6acc8060, 0x20, 0, 0x53053053) = 0 mprotect(0x7fbc6aead000, 16384, PROT_READ) = 0 mprotect(0x403000, 4096, PROT_READ) = 0 mprotect(0x7fbc6af0a000, 8192, PROT_READ) = 0 prlimit64(0, RLIMIT_STACK, NULL, {rlim_cur=8192*1024, rlim_max=RLIM64_INFINITY}) = 0 munmap(0x7fbc6aebd000, 81563) = 0 rseq(0x7fff15bb9920, 0x20, 0x181cd, 0xdeadbeaf) = -1 EINVAL (Invalid argument) fstat(1, {st_mode=S_IFCHR|0620, st_rdev=makedev(0x88, 0x9), ...}) = 0 getrandom("\xd0\x34\x97\x17\x61\xc2\x2b\x10", 8, GRND_NONBLOCK) = 8 brk(NULL) = 0x18ff4000 brk(0x19015000) = 0x19015000 write(1, "sys_rseq({ .cpu_id_start = 12, ."..., 136sys_rseq({ .cpu_id_start = 12, .cpu_id = 34, .rseq_cs = 56, .flags = 78, .node_id = 90, .mm_cid = 12, }, 32, 0) = -1 (Invalid argument) ) = 136 exit_group(-1) = ? +++ exited with 255 +++ root@number:~# And also the focus for the v6.13 should be to have a better, strace like BTF pretty printer as one of the outputs we can get from the libbpf BTF dumper. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alan Maguire <alan.maguire@oracle.com> Cc: Howard Chu <howardchu95@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/lkml/ZuH2K1LLt1pIDkbd@x1 Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-09-11perf trace: If a syscall arg is marked as 'const', assume it is coming ↵Arnaldo Carvalho de Melo
_from_ userspace We need to decide where to copy syscall arg contents, if at the syscalls:sys_entry hook, meaning is something that is coming from user to kernel space, or if it is a response, i.e. if it is something the _kernel_ is filling in and thus going to userspace. Since we have 'const' used in those syscalls, and unsure about this being consistent, doing: root@number:~# echo $(grep const /sys/kernel/tracing/events/syscalls/sys_enter_*/format | grep struct | cut -c47- | cut -d'/' -f1) clock_nanosleep clock_settime epoll_pwait2 futex io_pgetevents landlock_create_ruleset listmount mq_getsetattr mq_notify mq_timedreceive mq_timedsend preadv2 preadv prlimit64 process_madvise process_vm_readv process_vm_readv process_vm_writev process_vm_writev pwritev2 pwritev readv rt_sigaction rt_sigtimedwait semtimedop statmount timerfd_settime timer_settime vmsplice writev root@number:~# Seems to indicate that we can use that for the ones that have the 'const' to mark it as coming from user space, do it. Most notable/frequent syscall that now gets BTF pretty printed in a system wide 'perf trace' session is: root@number:~# perf trace 21.160 ( ): MediaSu~isor #/1028597 futex(uaddr: 0x7f49e1dfe964, op: WAIT_BITSET|PRIVATE_FLAG, utime: (struct __kernel_timespec){.tv_sec = (__kernel_time64_t)50290,.tv_nsec = (long long int)810362837,}, val3: MATCH_ANY) ... 21.166 ( 0.000 ms): RemVidChild/6995 futex(uaddr: 0x7f49fcc7fa00, op: WAKE|PRIVATE_FLAG, val: 1) = 0 21.169 ( 0.001 ms): RemVidChild/6995 sendmsg(fd: 25<socket:[78915]>, msg: 0x7f49e9af9da0, flags: DONTWAIT) = 280 21.172 ( 0.289 ms): RemVidChild/6995 futex(uaddr: 0x7f49fcc7fa58, op: WAIT_BITSET|PRIVATE_FLAG|CLOCK_REALTIME, val3: MATCH_ANY) = 0 21.463 ( 0.000 ms): RemVidChild/6995 futex(uaddr: 0x7f49fcc7fa00, op: WAKE|PRIVATE_FLAG, val: 1) = 0 21.467 ( 0.001 ms): RemVidChild/6995 futex(uaddr: 0x7f49e28bb964, op: WAKE|PRIVATE_FLAG, val: 1) = 1 21.160 ( 0.314 ms): MediaSu~isor #/1028597 ... [continued]: futex()) = 0 21.469 ( ): RemVidChild/6995 futex(uaddr: 0x7f49fcc7fa5c, op: WAIT_BITSET|PRIVATE_FLAG|CLOCK_REALTIME, val3: MATCH_ANY) ... 21.475 ( 0.000 ms): MediaSu~isor #/1028597 futex(uaddr: 0x7f49d0223040, op: WAKE|PRIVATE_FLAG, val: 1) = 0 21.478 ( 0.001 ms): MediaSu~isor #/1028597 futex(uaddr: 0x7f49e26ac964, op: WAKE|PRIVATE_FLAG, val: 1) = 1 ^Croot@number:~# root@number:~# cat /sys/kernel/tracing/events/syscalls/sys_enter_futex/format name: sys_enter_futex ID: 454 format: field:unsigned short common_type; offset:0; size:2; signed:0; field:unsigned char common_flags; offset:2; size:1; signed:0; field:unsigned char common_preempt_count; offset:3; size:1; signed:0; field:int common_pid; offset:4; size:4; signed:1; field:int __syscall_nr; offset:8; size:4; signed:1; field:u32 * uaddr; offset:16; size:8; signed:0; field:int op; offset:24; size:8; signed:0; field:u32 val; offset:32; size:8; signed:0; field:const struct __kernel_timespec * utime; offset:40; size:8; signed:0; field:u32 * uaddr2; offset:48; size:8; signed:0; field:u32 val3; offset:56; size:8; signed:0; print fmt: "uaddr: 0x%08lx, op: 0x%08lx, val: 0x%08lx, utime: 0x%08lx, uaddr2: 0x%08lx, val3: 0x%08lx", ((unsigned long)(REC->uaddr)), ((unsigned long)(REC->op)), ((unsigned long)(REC->val)), ((unsigned long)(REC->utime)), ((unsigned long)(REC->uaddr2)), ((unsigned long)(REC->val3)) root@number:~# Suggested-by: Ian Rogers <irogers@google.com> Reviewed-by: Ian Rogers <irogers@google.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alan Maguire <alan.maguire@oracle.com> Cc: Howard Chu <howardchu95@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/lkml/CAP-5=fWnuQrrBoTn6Rrn6vM_xQ2fCoc9i-AitD7abTcNi-4o1Q@mail.gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-09-10perf trace: Mark the rlim arg in the prlimit64 and setrlimit syscalls as ↵Arnaldo Carvalho de Melo
coming from user space With that it uses the generic BTF based pretty printer: root@number:~# perf trace -e prlimit64 0.000 ( 0.004 ms): :3417020/3417020 prlimit64(resource: NOFILE, old_rlim: 0x7fb8842fe3b0) = 0 0.126 ( 0.003 ms): Chroot Helper/3417022 prlimit64(resource: NOFILE, old_rlim: 0x7fb8842fdfd0) = 0 12.557 ( 0.005 ms): firefox/3417020 prlimit64(resource: STACK, old_rlim: 0x7ffe9ade1b80) = 0 26.640 ( 0.006 ms): MainThread/3417020 prlimit64(resource: STACK, old_rlim: 0x7ffe9ade1780) = 0 27.553 ( 0.002 ms): Web Content/3417020 prlimit64(resource: AS, old_rlim: 0x7ffe9ade1660) = 0 29.405 ( 0.003 ms): Web Content/3417020 prlimit64(resource: NOFILE, old_rlim: 0x7ffe9ade0c80) = 0 30.471 ( 0.002 ms): Web Content/3417020 prlimit64(resource: RTTIME, old_rlim: 0x7ffe9ade1370) = 0 30.485 ( 0.001 ms): Web Content/3417020 prlimit64(resource: RTTIME, new_rlim: (struct rlimit64){.rlim_cur = (__u64)50000,.rlim_max = (__u64)200000,}) = 0 31.779 ( 0.001 ms): Web Content/3417020 prlimit64(resource: STACK, old_rlim: 0x7ffe9ade1670) = 0 ^Croot@number:~# Better than before, still needs improvements in the configurability of the libbpf BTF dumper to get it to the strace output standard. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alan Maguire <alan.maguire@oracle.com> Cc: Howard Chu <howardchu95@gmail.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/lkml/ZuBQI-f8CGpuhIdH@x1 Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-09-10perf trace: Support collecting 'union's with the BPF augmenterArnaldo Carvalho de Melo
And reuse the BTF based struct pretty printer, with that we can offer initial support for the 'bpf' syscall's second argument, a 'union bpf_attr' pointer. But this is not that satisfactory as the libbpf btf dumper will pretty print _all_ the union, we need to have a way to say that the first arg selects the type for the union member to be pretty printed, something like what pahole does translating the PERF_RECORD_ selector into a name, and using that name to find a matching struct. In the case of 'union bpf_attr' it would map PROG_LOAD to one of the union members, but unfortunately there is no such mapping: root@number:~# pahole bpf_attr union bpf_attr { struct { __u32 map_type; /* 0 4 */ __u32 key_size; /* 4 4 */ __u32 value_size; /* 8 4 */ __u32 max_entries; /* 12 4 */ __u32 map_flags; /* 16 4 */ __u32 inner_map_fd; /* 20 4 */ __u32 numa_node; /* 24 4 */ char map_name[16]; /* 28 16 */ __u32 map_ifindex; /* 44 4 */ __u32 btf_fd; /* 48 4 */ __u32 btf_key_type_id; /* 52 4 */ __u32 btf_value_type_id; /* 56 4 */ __u32 btf_vmlinux_value_type_id; /* 60 4 */ /* --- cacheline 1 boundary (64 bytes) --- */ __u64 map_extra; /* 64 8 */ __s32 value_type_btf_obj_fd; /* 72 4 */ __s32 map_token_fd; /* 76 4 */ }; /* 0 80 */ struct { __u32 map_fd; /* 0 4 */ /* XXX 4 bytes hole, try to pack */ __u64 key; /* 8 8 */ union { __u64 value; /* 16 8 */ __u64 next_key; /* 16 8 */ }; /* 16 8 */ __u64 flags; /* 24 8 */ }; /* 0 32 */ struct { __u64 in_batch; /* 0 8 */ __u64 out_batch; /* 8 8 */ __u64 keys; /* 16 8 */ __u64 values; /* 24 8 */ __u32 count; /* 32 4 */ __u32 map_fd; /* 36 4 */ __u64 elem_flags; /* 40 8 */ __u64 flags; /* 48 8 */ } batch; /* 0 56 */ struct { __u32 prog_type; /* 0 4 */ __u32 insn_cnt; /* 4 4 */ __u64 insns; /* 8 8 */ __u64 license; /* 16 8 */ __u32 log_level; /* 24 4 */ __u32 log_size; /* 28 4 */ __u64 log_buf; /* 32 8 */ __u32 kern_version; /* 40 4 */ __u32 prog_flags; /* 44 4 */ char prog_name[16]; /* 48 16 */ /* --- cacheline 1 boundary (64 bytes) --- */ __u32 prog_ifindex; /* 64 4 */ __u32 expected_attach_type; /* 68 4 */ __u32 prog_btf_fd; /* 72 4 */ __u32 func_info_rec_size; /* 76 4 */ __u64 func_info; /* 80 8 */ __u32 func_info_cnt; /* 88 4 */ __u32 line_info_rec_size; /* 92 4 */ __u64 line_info; /* 96 8 */ __u32 line_info_cnt; /* 104 4 */ __u32 attach_btf_id; /* 108 4 */ union { __u32 attach_prog_fd; /* 112 4 */ __u32 attach_btf_obj_fd; /* 112 4 */ }; /* 112 4 */ __u32 core_relo_cnt; /* 116 4 */ __u64 fd_array; /* 120 8 */ /* --- cacheline 2 boundary (128 bytes) --- */ __u64 core_relos; /* 128 8 */ __u32 core_relo_rec_size; /* 136 4 */ __u32 log_true_size; /* 140 4 */ __s32 prog_token_fd; /* 144 4 */ }; /* 0 152 */ struct { __u64 pathname; /* 0 8 */ __u32 bpf_fd; /* 8 4 */ __u32 file_flags; /* 12 4 */ __s32 path_fd; /* 16 4 */ }; /* 0 24 */ struct { union { __u32 target_fd; /* 0 4 */ __u32 target_ifindex; /* 0 4 */ }; /* 0 4 */ __u32 attach_bpf_fd; /* 4 4 */ __u32 attach_type; /* 8 4 */ __u32 attach_flags; /* 12 4 */ __u32 replace_bpf_fd; /* 16 4 */ union { __u32 relative_fd; /* 20 4 */ __u32 relative_id; /* 20 4 */ }; /* 20 4 */ __u64 expected_revision; /* 24 8 */ }; /* 0 32 */ struct { __u32 prog_fd; /* 0 4 */ __u32 retval; /* 4 4 */ __u32 data_size_in; /* 8 4 */ __u32 data_size_out; /* 12 4 */ __u64 data_in; /* 16 8 */ __u64 data_out; /* 24 8 */ __u32 repeat; /* 32 4 */ __u32 duration; /* 36 4 */ __u32 ctx_size_in; /* 40 4 */ __u32 ctx_size_out; /* 44 4 */ __u64 ctx_in; /* 48 8 */ __u64 ctx_out; /* 56 8 */ /* --- cacheline 1 boundary (64 bytes) --- */ __u32 flags; /* 64 4 */ __u32 cpu; /* 68 4 */ __u32 batch_size; /* 72 4 */ } test; /* 0 80 */ struct { union { __u32 start_id; /* 0 4 */ __u32 prog_id; /* 0 4 */ __u32 map_id; /* 0 4 */ __u32 btf_id; /* 0 4 */ __u32 link_id; /* 0 4 */ }; /* 0 4 */ __u32 next_id; /* 4 4 */ __u32 open_flags; /* 8 4 */ }; /* 0 12 */ struct { __u32 bpf_fd; /* 0 4 */ __u32 info_len; /* 4 4 */ __u64 info; /* 8 8 */ } info; /* 0 16 */ struct { union { __u32 target_fd; /* 0 4 */ __u32 target_ifindex; /* 0 4 */ }; /* 0 4 */ __u32 attach_type; /* 4 4 */ __u32 query_flags; /* 8 4 */ __u32 attach_flags; /* 12 4 */ __u64 prog_ids; /* 16 8 */ union { __u32 prog_cnt; /* 24 4 */ __u32 count; /* 24 4 */ }; /* 24 4 */ /* XXX 4 bytes hole, try to pack */ __u64 prog_attach_flags; /* 32 8 */ __u64 link_ids; /* 40 8 */ __u64 link_attach_flags; /* 48 8 */ __u64 revision; /* 56 8 */ } query; /* 0 64 */ struct { __u64 name; /* 0 8 */ __u32 prog_fd; /* 8 4 */ /* XXX 4 bytes hole, try to pack */ __u64 cookie; /* 16 8 */ } raw_tracepoint; /* 0 24 */ struct { __u64 btf; /* 0 8 */ __u64 btf_log_buf; /* 8 8 */ __u32 btf_size; /* 16 4 */ __u32 btf_log_size; /* 20 4 */ __u32 btf_log_level; /* 24 4 */ __u32 btf_log_true_size; /* 28 4 */ __u32 btf_flags; /* 32 4 */ __s32 btf_token_fd; /* 36 4 */ }; /* 0 40 */ struct { __u32 pid; /* 0 4 */ __u32 fd; /* 4 4 */ __u32 flags; /* 8 4 */ __u32 buf_len; /* 12 4 */ __u64 buf; /* 16 8 */ __u32 prog_id; /* 24 4 */ __u32 fd_type; /* 28 4 */ __u64 probe_offset; /* 32 8 */ __u64 probe_addr; /* 40 8 */ } task_fd_query; /* 0 48 */ struct { union { __u32 prog_fd; /* 0 4 */ __u32 map_fd; /* 0 4 */ }; /* 0 4 */ union { __u32 target_fd; /* 4 4 */ __u32 target_ifindex; /* 4 4 */ }; /* 4 4 */ __u32 attach_type; /* 8 4 */ __u32 flags; /* 12 4 */ union { __u32 target_btf_id; /* 16 4 */ struct { __u64 iter_info; /* 16 8 */ __u32 iter_info_len; /* 24 4 */ }; /* 16 16 */ struct { __u64 bpf_cookie; /* 16 8 */ } perf_event; /* 16 8 */ struct { __u32 flags; /* 16 4 */ __u32 cnt; /* 20 4 */ __u64 syms; /* 24 8 */ __u64 addrs; /* 32 8 */ __u64 cookies; /* 40 8 */ } kprobe_multi; /* 16 32 */ struct { __u32 target_btf_id; /* 16 4 */ /* XXX 4 bytes hole, try to pack */ __u64 cookie; /* 24 8 */ } tracing; /* 16 16 */ struct { __u32 pf; /* 16 4 */ __u32 hooknum; /* 20 4 */ __s32 priority; /* 24 4 */ __u32 flags; /* 28 4 */ } netfilter; /* 16 16 */ struct { union { __u32 relative_fd; /* 16 4 */ __u32 relative_id; /* 16 4 */ }; /* 16 4 */ /* XXX 4 bytes hole, try to pack */ __u64 expected_revision; /* 24 8 */ } tcx; /* 16 16 */ struct { __u64 path; /* 16 8 */ __u64 offsets; /* 24 8 */ __u64 ref_ctr_offsets; /* 32 8 */ __u64 cookies; /* 40 8 */ __u32 cnt; /* 48 4 */ __u32 flags; /* 52 4 */ __u32 pid; /* 56 4 */ } uprobe_multi; /* 16 48 */ struct { union { __u32 relative_fd; /* 16 4 */ __u32 relative_id; /* 16 4 */ }; /* 16 4 */ /* XXX 4 bytes hole, try to pack */ __u64 expected_revision; /* 24 8 */ } netkit; /* 16 16 */ }; /* 16 48 */ } link_create; /* 0 64 */ struct { __u32 link_fd; /* 0 4 */ union { __u32 new_prog_fd; /* 4 4 */ __u32 new_map_fd; /* 4 4 */ }; /* 4 4 */ __u32 flags; /* 8 4 */ union { __u32 old_prog_fd; /* 12 4 */ __u32 old_map_fd; /* 12 4 */ }; /* 12 4 */ } link_update; /* 0 16 */ struct { __u32 link_fd; /* 0 4 */ } link_detach; /* 0 4 */ struct { __u32 type; /* 0 4 */ } enable_stats; /* 0 4 */ struct { __u32 link_fd; /* 0 4 */ __u32 flags; /* 4 4 */ } iter_create; /* 0 8 */ struct { __u32 prog_fd; /* 0 4 */ __u32 map_fd; /* 4 4 */ __u32 flags; /* 8 4 */ } prog_bind_map; /* 0 12 */ struct { __u32 flags; /* 0 4 */ __u32 bpffs_fd; /* 4 4 */ } token_create; /* 0 8 */ }; root@number:~# So this is one case where BTF gets us only that far, not getting all the way to automate the pretty printing of unions designed like 'union bpf_attr', we will need a custom pretty printer for this union, as using the libbpf union BTF dumper is way too verbose: root@number:~# perf trace --max-events 1 -e bpf bpftool map 0.000 ( 0.054 ms): bpftool/3409073 bpf(cmd: PROG_LOAD, uattr: (union bpf_attr){(struct){.map_type = (__u32)1,.key_size = (__u32)2,.value_size = (__u32)2755142048,.max_entries = (__u32)32764,.map_flags = (__u32)150263906,.inner_map_fd = (__u32)21920,},(struct){.map_fd = (__u32)1,.key = (__u64)140723063628192,(union){.value = (__u64)94145833392226,.next_key = (__u64)94145833392226,},},.batch = (struct){.in_batch = (__u64)8589934593,.out_batch = (__u64)140723063628192,.keys = (__u64)94145833392226,},(struct){.prog_type = (__u32)1,.insn_cnt = (__u32)2,.insns = (__u64)140723063628192,.license = (__u64)94145833392226,},(struct){.pathname = (__u64)8589934593,.bpf_fd = (__u32)2755142048,.file_flags = (__u32)32764,.path_fd = (__s32)150263906,},(struct){(union){.target_fd = (__u32)1,.target_ifindex = (__u32)1,},.attach_bpf_fd = (__u32)2,.attach_type = (__u32)2755142048,.attach_flags = (__u32)32764,.replace_bpf_fd = (__u32)150263906,(union){.relative_fd = (__u32)21920,.relative_id = (__u32)21920,},},.test = (struct){.prog_fd = (__u32)1,.retval = (__u32)2,.data_size_in = (__u32)2755142048,.data_size_out = (__u32)32764,.data_in = (__u64)94145833392226,},(struct){(union){.start_id = (__u32)1,.prog_id = (__u32)1,.map_id = (__u32)1,.btf_id = (__u32)1,.link_id = (__u32)1,},.next_id = (__u32)2,.open_flags = (__u32)2755142048,},.info = (struct){.bpf_fd = (__u32)1,.info_len = (__u32)2,.info = (__u64)140723063628192,},.query = (struct){(union){.target_fd = (__u32)1,.target_ifindex = (__u32)1,},.attach_type = (__u32)2,.query_flags = (__u32)2755142048,.attach_flags = (__u32)32764,.prog_ids = (__u64)94145833392226,},.raw_tracepoint = (struct){.name = (__u64)8589934593,.prog_fd = (__u32)2755142048,.cookie = (__u64)94145833392226,},(struct){.btf = (__u64)8589934593,.btf_log_buf = (__u64)140723063628192,.btf_size = (__u32)150263906,.btf_log_size = (__u32)21920,},.task_fd_query = (struct){.pid = (__u32)1,.fd = (__u32)2,.flags = (__u32)2755142048,.buf_len = (__u32)32764,.buf = (__u64)94145833392226,},.link_create = (struct){(union){.prog_fd = (__u32)1,.map_fd = (__u32)1,},(u) = 3 root@number:~# 2: prog_array name hid_jmp_table flags 0x0 key 4B value 4B max_entries 1024 memlock 8440B owner_prog_type tracing owner jited 13: hash_of_maps name cgroup_hash flags 0x0 key 8B value 4B max_entries 2048 memlock 167584B pids systemd(1) 960: array name libbpf_global flags 0x0 key 4B value 32B max_entries 1 memlock 280B 961: array name pid_iter.rodata flags 0x480 key 4B value 4B max_entries 1 memlock 8192B btf_id 1846 frozen pids bpftool(3409073) 962: array name libbpf_det_bind flags 0x0 key 4B value 32B max_entries 1 memlock 280B root@number:~# For simpler unions this may be better than not seeing any payload, so keep it there. Acked-by: Howard Chu <howardchu95@gmail.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alan Maguire <alan.maguire@oracle.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/lkml/ZuBLat8cbadILNLA@x1 [ Removed needless parenteses in the if block leading to the trace__btf_scnprintf() call, as per Howard's review comments ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-09-10perf trace: Add --force-btf for debuggingHoward Chu
If --force-btf is enabled, prefer btf_dump general pretty printer to perf trace's customized pretty printers. Mostly for debug purposes. Committer testing: diff before/after shows we need several improvements to be able to compare the changes, first we need to cut off/disable mutable data such as pids and timestamps, then what is left are the buffer addresses passed from userspace, returned from kernel space, maybe we can ask 'perf trace' to go on making those reproducible. That would entail a Pointer Address Translation (PAT) like for networking, that would, for simple, reproducible if not for these details, workloads, that we would then use in our regression tests. Enough digression, this is one such diff: openat(dfd: CWD, filename: "/usr/share/locale/locale.alias", flags: RDONLY|CLOEXEC) = 3 -fstat(fd: 3, statbuf: 0x7fff01f212a0) = 0 -read(fd: 3, buf: 0x5596bab2d630, count: 4096) = 2998 -read(fd: 3, buf: 0x5596bab2d630, count: 4096) = 0 +fstat(fd: 3, statbuf: 0x7ffc163cf0e0) = 0 +read(fd: 3, buf: 0x55b4e0631630, count: 4096) = 2998 +read(fd: 3, buf: 0x55b4e0631630, count: 4096) = 0 close(fd: 3) = 0 openat(dfd: CWD, filename: "/usr/share/locale/en_US.UTF-8/LC_MESSAGES/coreutils.mo") = -1 ENOENT (No such file or directory) openat(dfd: CWD, filename: "/usr/share/locale/en_US.utf8/LC_MESSAGES/coreutils.mo") = -1 ENOENT (No such file or directory) @@ -45,7 +45,7 @@ openat(dfd: CWD, filename: "/usr/share/locale/en.UTF-8/LC_MESSAGES/coreutils.mo") = -1 ENOENT (No such file or directory) openat(dfd: CWD, filename: "/usr/share/locale/en.utf8/LC_MESSAGES/coreutils.mo") = -1 ENOENT (No such file or directory) openat(dfd: CWD, filename: "/usr/share/locale/en/LC_MESSAGES/coreutils.mo") = -1 ENOENT (No such file or directory) -{ .tv_sec: 1, .tv_nsec: 0 }, rmtp: 0x7fff01f21990) = 0 +(struct __kernel_timespec){.tv_sec = (__kernel_time64_t)1,}, rmtp: 0x7ffc163cf7d0) = The problem more close to our hands is to make the libbpf BTF pretty printer to have a mode that closely resembles what we're trying to resemble: strace output. Being able to run something with 'perf trace' and with 'strace' and get the exact same output should be of interest of anybody wanting to have strace and 'perf trace' regression tested against each other. That last part is 'perf trace' shot at being something so useful as strace... ;-) Suggested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Howard Chu <howardchu95@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/r/20240824163322.60796-8-howardchu95@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-09-10perf trace: Pretty print buffer dataHoward Chu
Define TRACE_AUG_MAX_BUF in trace_augment.h data, which is the maximum buffer size we can augment. BPF will include this header too. Print buffer in a way that's different than just printing a string, we print all the control characters in \digits (such as \0 for null, and \10 for newline, LF). For character that has a bigger value than 127, we print the digits instead of the character itself as well. Committer notes: Simplified the buffer scnprintf to avoid using multiple buffers as discussed in the patch review thread. We can't really all 'buf' args to SCA_BUF as we're collecting so far just on the sys_enter path, so we would be printing the previous 'read' arg buffer contents, not what the kernel puts there. So instead of: static int syscall_fmt__cmp(const void *name, const void *fmtp) @@ -1987,8 +1989,6 @@ syscall_arg_fmt__init_array(struct syscall_arg_fmt *arg, struct tep_format_field - else if (strstr(field->type, "char *") && strstr(field->name, "buf")) - arg->scnprintf = SCA_BUF; Do: static const struct syscall_fmt syscall_fmts[] = { + { .name = "write", .errpid = true, + .arg = { [1] = { .scnprintf = SCA_BUF /* buf */, from_user = true, }, }, }, Signed-off-by: Howard Chu <howardchu95@gmail.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/r/20240815013626.935097-8-howardchu95@gmail.com Link: https://lore.kernel.org/r/20240824163322.60796-6-howardchu95@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-09-10perf trace: Pretty print struct dataHoward Chu
Change the arg->augmented.args to arg->augmented.args->value to skip the header for customized pretty printers, since we collect data in BPF using the general augment_sys_enter(), which always adds the header. Use btf_dump API to pretty print augmented struct pointer. Prefer existed pretty-printer than btf general pretty-printer. set compact = true and skip_names = true, so that no newline character and argument name are printed. Committer notes: Simplified the btf_dump_snprintf callback to avoid using multiple buffers, as discussed in the thread accessible via the Link tag below. Also made it do: dump_data_opts.skip_names = !arg->trace->show_arg_names; I.e. show the type and struct field names according to that tunable, we probably need another tunable just for this, but for now if the user wants to see syscall names in addition to its value, it makes sense to see the struct field names according to that tunable. Committer testing: The following have explicitely set beautifiers (SCA_FILENAME, SCA_SOCKADDR and SCA_PERF_ATTR), SCA_FILENAME is here just because we have been wiring up the "renameat2" ("renameat" until recently), so it doesn't use the introduced generic fallback (btf_struct_scnprintf(), see the definition of SCA_PERF_ATTR, SCA_SOCKADDR to see the more feature rich beautifiers, that are not using BTF): root@number:~# rm -f 987654 ; touch 123456 ; perf trace -e rename* mv 123456 987654 0.000 ( 0.039 ms): mv/258478 renameat2(olddfd: CWD, oldname: "123456", newdfd: CWD, newname: "987654", flags: NOREPLACE) = 0 root@number:~# perf trace -e connect,sendto ping -c 1 www.google.com 0.000 ( 0.014 ms): ping/258481 connect(fd: 5, uservaddr: { .family: LOCAL, path: /run/systemd/resolve/io.systemd.Resolve }, addrlen: 42) = 0 0.040 ( 0.003 ms): ping/258481 sendto(fd: 5, buff: 0x55bc317a6980, len: 97, flags: DONTWAIT|NOSIGNAL) = 97 18.742 ( 0.020 ms): ping/258481 sendto(fd: 5, buff: 0x7ffc04768df0, len: 20, addr: { .family: NETLINK }, addr_len: 0xc) = 20 PING www.google.com (142.251.129.68) 56(84) bytes of data. 18.783 ( 0.012 ms): ping/258481 connect(fd: 5, uservaddr: { .family: INET6, port: 0, addr: 2800:3f0:4004:810::2004 }, addrlen: 28) = 0 18.797 ( 0.001 ms): ping/258481 connect(fd: 5, uservaddr: { .family: UNSPEC }, addrlen: 16) = 0 18.800 ( 0.004 ms): ping/258481 connect(fd: 5, uservaddr: { .family: INET, port: 0, addr: 142.251.129.68 }, addrlen: 16) = 0 18.815 ( 0.002 ms): ping/258481 connect(fd: 5, uservaddr: { .family: INET, port: 1025, addr: 142.251.129.68 }, addrlen: 16) = 0 18.862 ( 0.023 ms): ping/258481 sendto(fd: 3, buff: 0x55bc317a0ac0, len: 64, addr: { .family: INET, port: 0, addr: 142.251.129.68 }, addr_len: 0x10) = 64 63.330 ( 0.038 ms): ping/258481 connect(fd: 5, uservaddr: { .family: LOCAL, path: /run/systemd/resolve/io.systemd.Resolve }, addrlen: 42) = 0 63.435 ( 0.010 ms): ping/258481 sendto(fd: 5, buff: 0x55bc317a8340, len: 110, flags: DONTWAIT|NOSIGNAL) = 110 64 bytes from rio07s07-in-f4.1e100.net (142.251.129.68): icmp_seq=1 ttl=49 time=44.2 ms --- www.google.com ping statistics --- 1 packets transmitted, 1 received, 0% packet loss, time 0ms rtt min/avg/max/mdev = 44.158/44.158/44.158/0.000 ms root@number:~# perf trace -e perf_event_open perf stat -e instructions,cache-misses,syscalls:sys_enter*sleep* sleep 1.23456789 0.000 ( 0.010 ms): :258487/258487 perf_event_open(attr_uptr: { type: 0 (PERF_TYPE_HARDWARE), config: 0xa00000000, disabled: 1, { bp_len, config2 }: 0x900000000, branch_sample_type: USER|COUNTERS, sample_regs_user: 0x3f1b7ffffffff, sample_stack_user: 258487, clockid: -599052088, sample_regs_intr: 0x60a000003eb, sample_max_stack: 14, sig_data: 120259084288 }, cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 3 0.016 ( 0.002 ms): :258487/258487 perf_event_open(attr_uptr: { type: 0 (PERF_TYPE_HARDWARE), config: 0x400000000, disabled: 1, { bp_len, config2 }: 0x900000000, branch_sample_type: USER|COUNTERS, sample_regs_user: 0x3f1b7ffffffff, sample_stack_user: 258487, clockid: -599044082, sample_regs_intr: 0x60a000003eb, sample_max_stack: 14, sig_data: 120259084288 }, cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 4 1.838 ( 0.006 ms): perf/258487 perf_event_open(attr_uptr: { type: 0 (PERF_TYPE_HARDWARE), size: 136, config: 0xa00000001, sample_type: IDENTIFIER, read_format: TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING, disabled: 1, inherit: 1, enable_on_exec: 1, exclude_guest: 1 }, pid: 258488 (perf), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 5 1.846 ( 0.002 ms): perf/258487 perf_event_open(attr_uptr: { type: 0 (PERF_TYPE_HARDWARE), size: 136, config: 0x400000001, sample_type: IDENTIFIER, read_format: TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING, disabled: 1, inherit: 1, enable_on_exec: 1, exclude_guest: 1 }, pid: 258488 (perf), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 6 1.849 ( 0.002 ms): perf/258487 perf_event_open(attr_uptr: { type: 0 (PERF_TYPE_HARDWARE), size: 136, config: 0xa00000003, sample_type: IDENTIFIER, read_format: TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING, disabled: 1, inherit: 1, enable_on_exec: 1, exclude_guest: 1 }, pid: 258488 (perf), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 7 1.851 ( 0.002 ms): perf/258487 perf_event_open(attr_uptr: { type: 0 (PERF_TYPE_HARDWARE), size: 136, config: 0x400000003, sample_type: IDENTIFIER, read_format: TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING, disabled: 1, inherit: 1, enable_on_exec: 1, exclude_guest: 1 }, pid: 258488 (perf), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 9 1.853 ( 0.600 ms): perf/258487 perf_event_open(attr_uptr: { type: 2 (tracepoint), size: 136, config: 0x190 (syscalls:sys_enter_nanosleep), sample_type: IDENTIFIER, read_format: TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING, disabled: 1, inherit: 1, enable_on_exec: 1, exclude_guest: 1 }, pid: 258488 (perf), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 10 2.456 ( 0.016 ms): perf/258487 perf_event_open(attr_uptr: { type: 2 (tracepoint), size: 136, config: 0x196 (syscalls:sys_enter_clock_nanosleep), sample_type: IDENTIFIER, read_format: TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING, disabled: 1, inherit: 1, enable_on_exec: 1, exclude_guest: 1 }, pid: 258488 (perf), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 11 Performance counter stats for 'sleep 1.23456789': 1,402,839 cpu_atom/instructions/ <not counted> cpu_core/instructions/ (0.00%) 11,066 cpu_atom/cache-misses/ <not counted> cpu_core/cache-misses/ (0.00%) 0 syscalls:sys_enter_nanosleep 1 syscalls:sys_enter_clock_nanosleep 1.236246714 seconds time elapsed 0.000000000 seconds user 0.001308000 seconds sys root@number:~# Now if we use it even for the ones we have a specific beautifier in tools/perf/trace/beauty, i.e. use btf_struct_scnprintf() for all structs, by adding the following patch: @@ -2316,7 +2316,7 @@ static size_t syscall__scnprintf_args(struct syscall *sc, char *bf, size_t size, default_scnprintf = sc->arg_fmt[arg.idx].scnprintf; - if (default_scnprintf == NULL || default_scnprintf == SCA_PTR) { + if (1 || (default_scnprintf == NULL || default_scnprintf == SCA_PTR)) { btf_printed = trace__btf_scnprintf(trace, &arg, bf + printed, size - printed, val, field->type); if (btf_printed) { We get: root@number:~# perf trace -e connect,sendto ping -c 1 www.google.com PING www.google.com (142.251.129.68) 56(84) bytes of data. 0.000 ( 0.015 ms): ping/283259 connect(fd: 5, uservaddr: (struct sockaddr){.sa_family = (sa_family_t)1,(union){.sa_data_min = (char[14])['/','r','u','n','/','s','y','s','t','e','m','d','/','r',],},}, addrlen: 42) = 0 0.046 ( 0.004 ms): ping/283259 sendto(fd: 5, buff: 0x559b008ae980, len: 97, flags: DONTWAIT|NOSIGNAL) = 97 0.353 ( 0.012 ms): ping/283259 sendto(fd: 5, buff: 0x7ffc01294960, len: 20, addr: (struct sockaddr){.sa_family = (sa_family_t)16,}, addr_len: 0xc) = 20 0.377 ( 0.006 ms): ping/283259 connect(fd: 5, uservaddr: (struct sockaddr){.sa_family = (sa_family_t)2,}, addrlen: 16) = 0 0.388 ( 0.010 ms): ping/283259 connect(fd: 5, uservaddr: (struct sockaddr){.sa_family = (sa_family_t)10,}, addrlen: 28) = 0 0.402 ( 0.001 ms): ping/283259 connect(fd: 5, uservaddr: (struct sockaddr){.sa_family = (sa_family_t)2,(union){.sa_data_min = (char[14])[4,1,142,251,129,'D',],},}, addrlen: 16) = 0 0.425 ( 0.045 ms): ping/283259 sendto(fd: 3, buff: 0x559b008a8ac0, len: 64, addr: (struct sockaddr){.sa_family = (sa_family_t)2,}, addr_len: 0x10) = 64 64 bytes from rio07s07-in-f4.1e100.net (142.251.129.68): icmp_seq=1 ttl=49 time=44.1 ms --- www.google.com ping statistics --- 1 packets transmitted, 1 received, 0% packet loss, time 0ms rtt min/avg/max/mdev = 44.113/44.113/44.113/0.000 ms 44.849 ( 0.038 ms): ping/283259 connect(fd: 5, uservaddr: (struct sockaddr){.sa_family = (sa_family_t)1,(union){.sa_data_min = (char[14])['/','r','u','n','/','s','y','s','t','e','m','d','/','r',],},}, addrlen: 42) = 0 44.927 ( 0.006 ms): ping/283259 sendto(fd: 5, buff: 0x559b008b03d0, len: 110, flags: DONTWAIT|NOSIGNAL) = 110 root@number:~# Which looks sane, i.e.: 18.800 ( 0.004 ms): ping/258481 connect(fd: 5, uservaddr: { .family: INET, port: 0, addr: 142.251.129.68 }, addrlen: 16) = 0 Becomes: 0.402 ( 0.001 ms): ping/283259 connect(fd: 5, uservaddr: (struct sockaddr){.sa_family = (sa_family_t)2,(union){.sa_data_min = (char[14])[4,1,142,251,129,'D',],},}, addrlen: 16) = 0 And. #define AF_UNIX 1 /* Unix domain sockets */ #define AF_LOCAL 1 /* POSIX name for AF_UNIX */ #define AF_INET 2 /* Internet IP Protocol */ <SNIP> #define AF_INET6 10 /* IP version 6 */ And 'D' == 68, so the preexisting sockaddr BPF collector is working with the new generic BTF pretty printer (btf_struct_scnprintf()), its just that it doesn't know about 'struct sockaddr' besides what is in BTF, i.e. its an array of bytes, not an IPv4 address that needs extra massaging. Ditto for the 'struct perf_event_attr' case: 1.851 ( 0.002 ms): perf/258487 perf_event_open(attr_uptr: { type: 0 (PERF_TYPE_HARDWARE), size: 136, config: 0x400000003, sample_type: IDENTIFIER, read_format: TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING, disabled: 1, inherit: 1, enable_on_exec: 1, exclude_guest: 1 }, pid: 258488 (perf), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 9 Becomes: 2.081 ( 0.002 ms): :283304/283304 perf_event_open(attr_uptr: (struct perf_event_attr){.size = (__u32)136,.config = (__u64)17179869187,.sample_type = (__u64)65536,.read_format = (__u64)3,.disabled = (__u64)0x1,.inherit = (__u64)0x1,.enable_on_exec = (__u64)0x1,.exclude_guest = (__u64)0x1,}, pid: 283305 (sleep), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 9 hex(17179869187) = 0x400000003, etc. read_format: TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING is enum perf_event_read_format { PERF_FORMAT_TOTAL_TIME_ENABLED = 1U << 0, PERF_FORMAT_TOTAL_TIME_RUNNING = 1U << 1, and so on. We need to work with the libbpf btf dump api to get one output that matches the 'perf trace'/strace expectations/format, but having this in this current form is already an improvement to 'perf trace', so lets improve from what we have. Signed-off-by: Howard Chu <howardchu95@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/r/20240815013626.935097-7-howardchu95@gmail.com Link: https://lore.kernel.org/r/20240824163322.60796-5-howardchu95@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-09-10perf trace: Add trace__bpf_sys_enter_beauty_map() to prepare for fetching ↵Howard Chu
data in BPF Set up beauty_map, load it to BPF, in such format: if argument No.3 is a struct of size 32 bytes (of syscall number 114) beauty_map[114][2] = 32; if argument No.3 is a string (of syscall number 114) beauty_map[114][2] = 1; if argument No.3 is a buffer, its size is indicated by argument No.4 (of syscall number 114) beauty_map[114][2] = -4; /* -1 ~ -6, we'll read this buffer size in BPF */ Committer notes: Moved syscall_arg_fmt__cache_btf_struct() from a ifdef HAVE_LIBBPF_SUPPORT to closer to where it is used, that is ifdef'ed on HAVE_BPF_SKEL and thus breaks the build when building with BUILD_BPF_SKEL=0, as detected using 'make -C tools/perf build-test'. Also add 'struct beauty_map_enter' to tools/perf/util/bpf_skel/augmented_raw_syscalls.bpf.c as we're using it in this patch, otherwise we get this while trying to build at this point in the original patch series: builtin-trace.c: In function ‘trace__init_syscalls_bpf_prog_array_maps’: builtin-trace.c:3725:58: error: ‘struct <anonymous>’ has no member named ‘beauty_map_enter’ 3725 | int beauty_map_fd = bpf_map__fd(trace->skel->maps.beauty_map_enter); | We also have to take into account syscall_arg_fmt.from_user when telling the kernel what to copy in the sys_enter generic collector, we don't want to collect bogus data in buffers that will only be available to us at sys_exit time, i.e. after the kernel has filled it, so leave this for when we have such a sys_exit based collector. Committer testing: Not wired up yet, so all continues to work, using the existing BPF collector and userspace beautifiers that are augmentation aware: root@number:~# rm -f 987654 ; touch 123456 ; perf trace -e rename* mv 123456 987654 0.000 ( 0.031 ms): mv/20888 renameat2(olddfd: CWD, oldname: "123456", newdfd: CWD, newname: "987654", flags: NOREPLACE) = 0 root@number:~# perf trace -e connect,sendto ping -c 1 www.google.com 0.000 ( 0.014 ms): ping/20892 connect(fd: 5, uservaddr: { .family: LOCAL, path: /run/systemd/resolve/io.systemd.Resolve }, addrlen: 42) = 0 0.040 ( 0.003 ms): ping/20892 sendto(fd: 5, buff: 0x560b4ff17980, len: 97, flags: DONTWAIT|NOSIGNAL) = 97 0.480 ( 0.017 ms): ping/20892 sendto(fd: 5, buff: 0x7ffd82d07150, len: 20, addr: { .family: NETLINK }, addr_len: 0xc) = 20 0.526 ( 0.014 ms): ping/20892 connect(fd: 5, uservaddr: { .family: INET6, port: 0, addr: 2800:3f0:4004:810::2004 }, addrlen: 28) = 0 0.542 ( 0.002 ms): ping/20892 connect(fd: 5, uservaddr: { .family: UNSPEC }, addrlen: 16) = 0 0.544 ( 0.004 ms): ping/20892 connect(fd: 5, uservaddr: { .family: INET, port: 0, addr: 142.251.135.100 }, addrlen: 16) = 0 0.559 ( 0.002 ms): ping/20892 connect(fd: 5, uservaddr: { .family: INET, port: 1025, addr: 142.251.135.100 }, addrlen: 16PING www.google.com (142.251.135.100) 56(84) bytes of data. ) = 0 0.589 ( 0.058 ms): ping/20892 sendto(fd: 3, buff: 0x560b4ff11ac0, len: 64, addr: { .family: INET, port: 0, addr: 142.251.135.100 }, addr_len: 0x10) = 64 45.250 ( 0.029 ms): ping/20892 connect(fd: 5, uservaddr: { .family: LOCAL, path: /run/systemd/resolve/io.systemd.Resolve }, addrlen: 42) = 0 45.344 ( 0.012 ms): ping/20892 sendto(fd: 5, buff: 0x560b4ff19340, len: 111, flags: DONTWAIT|NOSIGNAL) = 111 64 bytes from rio09s08-in-f4.1e100.net (142.251.135.100): icmp_seq=1 ttl=49 time=44.4 ms --- www.google.com ping statistics --- 1 packets transmitted, 1 received, 0% packet loss, time 0ms rtt min/avg/max/mdev = 44.361/44.361/44.361/0.000 ms root@number:~# Signed-off-by: Howard Chu <howardchu95@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/r/20240815013626.935097-4-howardchu95@gmail.com Link: https://lore.kernel.org/r/20240824163322.60796-3-howardchu95@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-09-10perf trace: Mark bpf's attr as from_userArnaldo Carvalho de Melo
This one has no specific pretty printer right now, so will be handled by the generic BTF based one later in this patch series. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-09-09perf trace: Introduce SCA_TIMESPEC_FROM_USER() to set .from_user = trueArnaldo Carvalho de Melo
Paving the way for the generic BPF BTF based syscall arg augmenter. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Howard Chu <howardchu95@gmail.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-09-09perf trace: Introduce SCA_SOCKADDR_FROM_USER() to set .from_user = trueArnaldo Carvalho de Melo
Paving the way for the generic BPF BTF based syscall arg augmenter. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Howard Chu <howardchu95@gmail.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-09-09perf trace: Introduce SCA_PERF_ATTR_FROM_USER() to set .from_user = trueArnaldo Carvalho de Melo
Paving the way for the generic BPF BTF based syscall arg augmenter. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Howard Chu <howardchu95@gmail.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-09-09perf trace: Mark which syscall arguments go from user space to kernel spaceArnaldo Carvalho de Melo
We need to know where to collect it in the BPF augmenters, if in the sys_enter hook or in the sys_exit hook. Start with the SCA_FILENAME one, that is just from user to kernel space. The alternative, better, but takes a bit more time than I have now, is to use the __user information that is already in the syscall args and encoded in BTF via a tag, do it later. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Howard Chu <howardchu95@gmail.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-08-28perf trace: Pass the richer 'struct syscall_arg' pointer to ↵Arnaldo Carvalho de Melo
trace__btf_scnprintf() Since we'll need it later in the current patch series and we can get the syscall_arg_fmt from syscall_arg->fmt. Based-on-a-patch-by: Howard Chu <howardchu95@gmail.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/lkml/Zsd8vqCrTh5h69rp@x1 Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-08-12perf tool: Constify tool pointersIan Rogers
The tool pointer (to a struct largely of function pointers) is passed around but is unchanged except at initialization. Change parameter and variable types to be const to lower the possibilities of what could happen with a tool. Reviewed-by: Adrian Hunter <adrian.hunter@intel.com> Signed-off-by: Ian Rogers <irogers@google.com> Tested-by: Adrian Hunter <adrian.hunter@intel.com> Tested-by: Leo Yan <leo.yan@arm.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Anshuman Khandual <anshuman.khandual@arm.com> Cc: Athira Rajeev <atrajeev@linux.vnet.ibm.com> Cc: Huacai Chen <chenhuacai@kernel.org> Cc: Ilkka Koskinen <ilkka@os.amperecomputing.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: James Clark <james.clark@arm.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: John Garry <john.g.garry@oracle.com> Cc: Jonathan Cameron <jonathan.cameron@huawei.com> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Leo Yan <leo.yan@linux.dev> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Mike Leach <mike.leach@linaro.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Nick Desaulniers <ndesaulniers@google.com> Cc: Nick Terrell <terrelln@fb.com> Cc: Oliver Upton <oliver.upton@linux.dev> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Song Liu <song@kernel.org> Cc: Sun Haiyong <sunhaiyong@loongson.cn> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: Will Deacon <will@kernel.org> Cc: Yanteng Si <siyanteng@loongson.cn> Cc: Yicong Yang <yangyicong@hisilicon.com> Cc: linux-arm-kernel@lists.infradead.org Link: https://lore.kernel.org/r/20240812204720.631678-4-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-08-01perf bpf-filter: Pass 'target' to perf_bpf_filter__prepare()Namhyung Kim
This is needed to prepare target-specific actions in the later patch. We want to reuse the pinned BPF program and map for regular users to profile their own processes. Signed-off-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: KP Singh <kpsingh@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Song Liu <song@kernel.org> Cc: Stephane Eranian <eranian@google.com> Cc: bpf@vger.kernel.org Link: https://lore.kernel.org/r/20240703223035.2024586-3-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-07-31perf trace: Remove arg_fmt->is_enum, we can get that from the BTF typeArnaldo Carvalho de Melo
This is to pave the way for other BTF types, i.e. we try to find BTF type then use things like btf_is_enum(btf_type) that we cached to find the right strtoul and scnprintf routines. For now only enum is supported, all the other types simple return zero for scnprintf which makes it have the same behaviour as when BTF isn't available, i.e. fallback to no pretty printing. Ditto for strtoul. root@x1:~# perf test -v enum 124: perf trace enum augmentation tests : Ok root@x1:~# perf test -v enum 124: perf trace enum augmentation tests : Ok root@x1:~# perf test -v enum 124: perf trace enum augmentation tests : Ok root@x1:~# perf test -v enum 124: perf trace enum augmentation tests : Ok root@x1:~# perf test -v enum 124: perf trace enum augmentation tests : Ok root@x1:~# Signed-off-by: Howard Chu <howardchu95@gmail.com> Tested-by: Howard Chu <howardchu95@gmail.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Howard Chu <howardchu95@gmail.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/r/20240624181345.124764-9-howardchu95@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-07-31perf trace: Introduce trace__btf_scnprintf()Arnaldo Carvalho de Melo
To have a central place that will look at the BTF type and call the right scnprintf routine or return zero, meaning BTF pretty printing isn't available or not implemented for a specific type. Signed-off-by: Howard Chu <howardchu95@gmail.com> Tested-by: Howard Chu <howardchu95@gmail.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/r/20240624181345.124764-8-howardchu95@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-07-31perf trace: Filter enum arguments with enum namesHoward Chu
Before: perf $ ./perf trace -e timer:hrtimer_start --filter='mode!=HRTIMER_MODE_ABS_PINNED_HARD' --max-events=1 No resolver (strtoul) for "mode" in "timer:hrtimer_start", can't set filter "(mode!=HRTIMER_MODE_ABS_PINNED_HARD) && (common_pid != 281988)" After: perf $ ./perf trace -e timer:hrtimer_start --filter='mode!=HRTIMER_MODE_ABS_PINNED_HARD' --max-events=1 0.000 :0/0 timer:hrtimer_start(hrtimer: 0xffff9498a6ca5f18, function: 0xffffffffa77a5be0, expires: 12351248764875, softexpires: 12351248764875, mode: HRTIMER_MODE_ABS) && and ||: perf $ ./perf trace -e timer:hrtimer_start --filter='mode != HRTIMER_MODE_ABS_PINNED_HARD && mode != HRTIMER_MODE_ABS' --max-events=1 0.000 Hyprland/534 timer:hrtimer_start(hrtimer: 0xffff9497801a84d0, function: 0xffffffffc04cdbe0, expires: 12639434638458, softexpires: 12639433638458, mode: HRTIMER_MODE_REL) perf $ ./perf trace -e timer:hrtimer_start --filter='mode == HRTIMER_MODE_REL || mode == HRTIMER_MODE_PINNED' --max-events=1 0.000 ldlck-test/60639 timer:hrtimer_start(hrtimer: 0xffffb16404ee7bf8, function: 0xffffffffa7790420, expires: 12772614418016, softexpires: 12772614368016, mode: HRTIMER_MODE_REL) Switching it up, using both enum name and integer value(--filter='mode == HRTIMER_MODE_ABS_PINNED_HARD || mode == 0'): perf $ ./perf trace -e timer:hrtimer_start --filter='mode == HRTIMER_MODE_ABS_PINNED_HARD || mode == 0' --max-events=3 0.000 :0/0 timer:hrtimer_start(hrtimer: 0xffff9498a6ca5f18, function: 0xffffffffa77a5be0, expires: 12601748739825, softexpires: 12601748739825, mode: HRTIMER_MODE_ABS_PINNED_HARD) 0.036 :0/0 timer:hrtimer_start(hrtimer: 0xffff9498a6ca5f18, function: 0xffffffffa77a5be0, expires: 12518758748124, softexpires: 12518758748124, mode: HRTIMER_MODE_ABS_PINNED_HARD) 0.172 tmux: server/41881 timer:hrtimer_start(hrtimer: 0xffffb164081e7838, function: 0xffffffffa7790420, expires: 12518768255836, softexpires: 12518768205836, mode: HRTIMER_MODE_ABS) P.S. perf $ pahole hrtimer_mode enum hrtimer_mode { HRTIMER_MODE_ABS = 0, HRTIMER_MODE_REL = 1, HRTIMER_MODE_PINNED = 2, HRTIMER_MODE_SOFT = 4, HRTIMER_MODE_HARD = 8, HRTIMER_MODE_ABS_PINNED = 2, HRTIMER_MODE_REL_PINNED = 3, HRTIMER_MODE_ABS_SOFT = 4, HRTIMER_MODE_REL_SOFT = 5, HRTIMER_MODE_ABS_PINNED_SOFT = 6, HRTIMER_MODE_REL_PINNED_SOFT = 7, HRTIMER_MODE_ABS_HARD = 8, HRTIMER_MODE_REL_HARD = 9, HRTIMER_MODE_ABS_PINNED_HARD = 10, HRTIMER_MODE_REL_PINNED_HARD = 11, }; Committer testing: root@x1:~# perf trace -e timer:hrtimer_start --filter='mode != HRTIMER_MODE_ABS' --max-events=2 0.000 :0/0 timer:hrtimer_start(hrtimer: 0xffff8d4eff2a5050, function: 0xffffffff9e22ddd0, expires: 241502326000000, softexpires: 241502326000000, mode: HRTIMER_MODE_ABS_PINNED_HARD) 18446744073709.488 :0/0 timer:hrtimer_start(hrtimer: 0xffff8d4eff425050, function: 0xffffffff9e22ddd0, expires: 241501814000000, softexpires: 241501814000000, mode: HRTIMER_MODE_ABS_PINNED_HARD) root@x1:~# perf trace -e timer:hrtimer_start --filter='mode != HRTIMER_MODE_ABS && mode != HRTIMER_MODE_ABS_PINNED_HARD' --max-events=2 0.000 podman/510644 timer:hrtimer_start(hrtimer: 0xffffa2024f5f7dd0, function: 0xffffffff9e2170c0, expires: 241530497418194, softexpires: 241530497368194, mode: HRTIMER_MODE_REL) 40.251 gnome-shell/2484 timer:hrtimer_start(hrtimer: 0xffff8d48bda17650, function: 0xffffffffc0661550, expires: 241550528619247, softexpires: 241550527619247, mode: HRTIMER_MODE_REL) root@x1:~# perf trace -v -e timer:hrtimer_start --filter='mode != HRTIMER_MODE_ABS && mode != HRTIMER_MODE_ABS_PINNED_HARD && mode != HRTIMER_MODE_REL' --max-events=2 Using CPUID GenuineIntel-6-BA-3 vmlinux BTF loaded <SNIP> 0 0xa 0x1 New filter for timer:hrtimer_start: (mode != 0 && mode != 0xa && mode != 0x1) && (common_pid != 524049 && common_pid != 4041) mmap size 528384B ^Croot@x1:~# Suggested-by: Arnaldo Carvalho de Melo <acme@kernel.org> Signed-off-by: Howard Chu <howardchu95@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/lkml/ZnCcliuecJABD5FN@x1 Link: https://lore.kernel.org/r/20240624181345.124764-5-howardchu95@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-07-31perf trace: Augment non-syscall tracepoints with enum arguments with BTFHoward Chu
Before: perf $ ./perf trace -e timer:hrtimer_start --max-events=1 0.000 :0/0 timer:hrtimer_start(hrtimer: 0xffff974466c25f18, function: 0xffffffff89da5be0, expires: 377432432256753, softexpires: 377432432256753, mode: 10) After: perf $ ./perf trace -e timer:hrtimer_start --max-events=1 0.000 :0/0 timer:hrtimer_start(hrtimer: 0xffff9498a6ca5f18, function: 0xffffffffa77a5be0, expires: 4382442895089, softexpires: 4382442895089, mode: HRTIMER_MODE_ABS_PINNED_HARD) in which HRTIMER_MODE_ABS_PINNED_HARD is: perf $ pahole hrtimer_mode enum hrtimer_mode { HRTIMER_MODE_ABS = 0, HRTIMER_MODE_REL = 1, HRTIMER_MODE_PINNED = 2, HRTIMER_MODE_SOFT = 4, HRTIMER_MODE_HARD = 8, HRTIMER_MODE_ABS_PINNED = 2, HRTIMER_MODE_REL_PINNED = 3, HRTIMER_MODE_ABS_SOFT = 4, HRTIMER_MODE_REL_SOFT = 5, HRTIMER_MODE_ABS_PINNED_SOFT = 6, HRTIMER_MODE_REL_PINNED_SOFT = 7, HRTIMER_MODE_ABS_HARD = 8, HRTIMER_MODE_REL_HARD = 9, HRTIMER_MODE_ABS_PINNED_HARD = 10, HRTIMER_MODE_REL_PINNED_HARD = 11, }; Can also be tested by ./perf trace -e pagemap:mm_lru_insertion,timer:hrtimer_start,timer:hrtimer_init,skb:kfree_skb --max-events=10 (Chose these 4 events because they happen quite frequently.) However some enum arguments may not be contained in vmlinux BTF. To see what enum arguments are supported, use: vmlinux_dir $ bpftool btf dump file /sys/kernel/btf/vmlinux > vmlinux vmlinux_dir $ while read l; do grep "ENUM '$l'" vmlinux; done < <(grep field:enum /sys/kernel/tracing/events/*/*/format | awk '{print $3}' | sort | uniq) | awk '{print $3}' | sed "s/'\(.*\)'/\1/g" dev_pm_qos_req_type error_detector hrtimer_mode i2c_slave_event ieee80211_bss_type lru_list migrate_mode nl80211_auth_type nl80211_band nl80211_iftype numa_vmaskip_reason pm_qos_req_action pwm_polarity skb_drop_reason thermal_trip_type xen_lazy_mode xen_mc_extend_args xen_mc_flush_reason zone_type And what tracepoints have these enum types as their arguments: vmlinux_dir $ while read l; do grep "ENUM '$l'" vmlinux; done < <(grep field:enum /sys/kernel/tracing/events/*/*/format | awk '{print $3}' | sort | uniq) | awk '{print $3}' | sed "s/'\(.*\)'/\1/g" > good_enums vmlinux_dir $ cat good_enums dev_pm_qos_req_type error_detector hrtimer_mode i2c_slave_event ieee80211_bss_type lru_list migrate_mode nl80211_auth_type nl80211_band nl80211_iftype numa_vmaskip_reason pm_qos_req_action pwm_polarity skb_drop_reason thermal_trip_type xen_lazy_mode xen_mc_extend_args xen_mc_flush_reason zone_type vmlinux_dir $ grep -f good_enums -l /sys/kernel/tracing/events/*/*/format /sys/kernel/tracing/events/cfg80211/cfg80211_chandef_dfs_required/format /sys/kernel/tracing/events/cfg80211/cfg80211_ch_switch_notify/format /sys/kernel/tracing/events/cfg80211/cfg80211_ch_switch_started_notify/format /sys/kernel/tracing/events/cfg80211/cfg80211_get_bss/format /sys/kernel/tracing/events/cfg80211/cfg80211_ibss_joined/format /sys/kernel/tracing/events/cfg80211/cfg80211_inform_bss_frame/format /sys/kernel/tracing/events/cfg80211/cfg80211_radar_event/format /sys/kernel/tracing/events/cfg80211/cfg80211_ready_on_channel_expired/format /sys/kernel/tracing/events/cfg80211/cfg80211_ready_on_channel/format /sys/kernel/tracing/events/cfg80211/cfg80211_reg_can_beacon/format /sys/kernel/tracing/events/cfg80211/cfg80211_return_bss/format /sys/kernel/tracing/events/cfg80211/cfg80211_tx_mgmt_expired/format /sys/kernel/tracing/events/cfg80211/rdev_add_virtual_intf/format /sys/kernel/tracing/events/cfg80211/rdev_auth/format /sys/kernel/tracing/events/cfg80211/rdev_change_virtual_intf/format /sys/kernel/tracing/events/cfg80211/rdev_channel_switch/format /sys/kernel/tracing/events/cfg80211/rdev_connect/format /sys/kernel/tracing/events/cfg80211/rdev_inform_bss/format /sys/kernel/tracing/events/cfg80211/rdev_libertas_set_mesh_channel/format /sys/kernel/tracing/events/cfg80211/rdev_mgmt_tx/format /sys/kernel/tracing/events/cfg80211/rdev_remain_on_channel/format /sys/kernel/tracing/events/cfg80211/rdev_return_chandef/format /sys/kernel/tracing/events/cfg80211/rdev_return_int_survey_info/format /sys/kernel/tracing/events/cfg80211/rdev_set_ap_chanwidth/format /sys/kernel/tracing/events/cfg80211/rdev_set_monitor_channel/format /sys/kernel/tracing/events/cfg80211/rdev_set_radar_background/format /sys/kernel/tracing/events/cfg80211/rdev_start_ap/format /sys/kernel/tracing/events/cfg80211/rdev_start_radar_detection/format /sys/kernel/tracing/events/cfg80211/rdev_tdls_channel_switch/format /sys/kernel/tracing/events/compaction/mm_compaction_defer_compaction/format /sys/kernel/tracing/events/compaction/mm_compaction_deferred/format /sys/kernel/tracing/events/compaction/mm_compaction_defer_reset/format /sys/kernel/tracing/events/compaction/mm_compaction_finished/format /sys/kernel/tracing/events/compaction/mm_compaction_kcompactd_wake/format /sys/kernel/tracing/events/compaction/mm_compaction_suitable/format /sys/kernel/tracing/events/compaction/mm_compaction_wakeup_kcompactd/format /sys/kernel/tracing/events/error_report/error_report_end/format /sys/kernel/tracing/events/i2c_slave/i2c_slave/format /sys/kernel/tracing/events/migrate/mm_migrate_pages/format /sys/kernel/tracing/events/migrate/mm_migrate_pages_start/format /sys/kernel/tracing/events/pagemap/mm_lru_insertion/format /sys/kernel/tracing/events/power/dev_pm_qos_add_request/format /sys/kernel/tracing/events/power/dev_pm_qos_remove_request/format /sys/kernel/tracing/events/power/dev_pm_qos_update_request/format /sys/kernel/tracing/events/power/pm_qos_update_flags/format /sys/kernel/tracing/events/power/pm_qos_update_target/format /sys/kernel/tracing/events/pwm/pwm_apply/format /sys/kernel/tracing/events/pwm/pwm_get/format /sys/kernel/tracing/events/sched/sched_skip_vma_numa/format /sys/kernel/tracing/events/skb/kfree_skb/format /sys/kernel/tracing/events/thermal/thermal_zone_trip/format /sys/kernel/tracing/events/timer/hrtimer_init/format /sys/kernel/tracing/events/timer/hrtimer_start/format /sys/kernel/tracing/events/xen/xen_mc_batch/format /sys/kernel/tracing/events/xen/xen_mc_extend_args/format /sys/kernel/tracing/events/xen/xen_mc_flush_reason/format /sys/kernel/tracing/events/xen/xen_mc_issue/format Committer testing: root@x1:~# perf trace -e timer:hrtimer_start --max-events=2 0.000 :0/0 timer:hrtimer_start(hrtimer: 0xffff8d4eff225050, function: 0xffffffff9e22ddd0, expires: 241152380000000, softexpires: 241152380000000, mode: HRTIMER_MODE_ABS) 0.028 :0/0 timer:hrtimer_start(hrtimer: 0xffff8d4eff225050, function: 0xffffffff9e22ddd0, expires: 241153654000000, softexpires: 241153654000000, mode: HRTIMER_MODE_ABS_PINNED_HARD) root@x1:~# Suggested-by: Arnaldo Carvalho de Melo <acme@kernel.org> Reviewed-by: Arnaldo Carvalho de Melo <acme@kernel.org> Signed-off-by: Howard Chu <howardchu95@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/lkml/20240615032743.112750-1-howardchu95@gmail.com Link: https://lore.kernel.org/r/20240624181345.124764-4-howardchu95@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2024-07-31perf trace: BTF-based enum pretty printing for syscall argsHoward Chu
In this patch, BTF is used to turn enum value to the corresponding name. There is only one system call that uses enum value as its argument, that is `landlock_add_rule()`. The vmlinux btf is loaded lazily, when user decided to trace the `landlock_add_rule` syscall. But if one decide to run `perf trace` without any arguments, the behaviour is to trace `landlock_add_rule`, so vmlinux btf will be loaded by default. The laziest behaviour is to load vmlinux btf when a `landlock_add_rule` syscall hits. But I think you could lose some samples when loading vmlinux btf at run time, for it can delay the handling of other samples. I might need your precious opinions on this... before: ``` perf $ ./perf trace -e landlock_add_rule 0.000 ( 0.008 ms): ldlck-test/438194 landlock_add_rule(rule_type: 2) = -1 EBADFD (File descriptor in bad state) 0.010 ( 0.001 ms): ldlck-test/438194 landlock_add_rule(rule_type: 1) = -1 EBADFD (File descriptor in bad state) ``` after: ``` perf $ ./perf trace -e landlock_add_rule 0.000 ( 0.029 ms): ldlck-test/438194 landlock_add_rule(rule_type: LANDLOCK_RULE_NET_PORT) = -1 EBADFD (File descriptor in bad state) 0.036 ( 0.004 ms): ldlck-test/438194 landlock_add_rule(rule_type: LANDLOCK_RULE_PATH_BENEATH) = -1 EBADFD (File descriptor in bad state) ``` Committer notes: Made it build with NO_LIBBPF=1, simplified btf_enum_fprintf(), see [1] for the discussion. Signed-off-by: Howard Chu <howardchu95@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Günther Noack <gnoack@google.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Mickaël Salaün <mic@digikod.net> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/lkml/20240613022757.3589783-1-howardchu95@gmail.com Link: https://lore.kernel.org/lkml/ZnXAhFflUl_LV1QY@x1 # [1] Link: https://lore.kernel.org/r/20240624181345.124764-3-howardchu95@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
generated by cgit (git 2.34.1) at 2025-05-12 05:06:09 +0000