- 18 Aug, 2023 1 commit
-
-
Arnaldo Carvalho de Melo authored
It is possible to use 'perf trace' with tracepoints and in that case we can't initialize/use the augmented_raw_syscalls BPF skel. For instance, this usecase: # perf trace -e sched:*exec --max-events=5 ? ( ): NetworkManager/1183 ... [continued]: poll()) = 1 0.043 ( 0.007 ms): NetworkManager/1183 epoll_wait(epfd: 17<anon_inode:[eventpoll]>, events: 0x55555f90e920, maxevents: 6) = 0 0.060 ( 0.007 ms): NetworkManager/1183 write(fd: 3<anon_inode:[eventfd]>, buf: 0x7ffc5a27cd30, count: 8) = 8 0.073 ( 0.005 ms): NetworkManager/1183 epoll_wait(epfd: 24<anon_inode:[eventpoll]>, events: 0x7ffc5a27cd20, maxevents: 2) = 1 0.082 ( 0.010 ms): NetworkManager/1183 recvmmsg(fd: 26<socket:[30298]>, mmsg: 0x7ffc5a27caa0, vlen: 8) = 1 # Where we want to trace just some sched tracepoints ending in 'exec' ends up tracing all syscalls. Fix it by checking existing trace->trace_syscalls boolean to see if we need the augmenter. A followup patch will move those sections of code used only with the augmenter to separate functions, to get it cleaner and remove the goto, done just for reviewing purposes. With this patch in place the previous behaviour is restored: no syscalls when we have other events and no syscall names: [root@quaco ~]# perf probe do_filp_open "filename=pathname->name:string" Added new event: probe:do_filp_open (on do_filp_open with filename=pathname->name:string) You can now use it in all perf tools, such as: perf record -e probe:do_filp_open -aR sleep 1 [root@quaco ~]# perf trace --max-events=10 -e probe:do_filp_open sleep 1 0.000 sleep/455122 probe:do_filp_open(__probe_ip: -1186560412, filename: "/etc/ld.so.cache") 0.056 sleep/455122 probe:do_filp_open(__probe_ip: -1186560412, filename: "/lib64/libc.so.6") 0.481 sleep/455122 probe:do_filp_open(__probe_ip: -1186560412, filename: "/usr/lib/locale/locale-archive") 0.501 sleep/455122 probe:do_filp_open(__probe_ip: -1186560412, filename: "/usr/share/locale/locale.alias") 0.572 sleep/455122 probe:do_filp_open(__probe_ip: -1186560412, filename: "/usr/lib/locale/en_US.UTF-8/LC_IDENTIFICATION") 0.581 sleep/455122 probe:do_filp_open(__probe_ip: -1186560412, filename: "/usr/lib/locale/en_US.utf8/LC_IDENTIFICATION") 0.616 sleep/455122 probe:do_filp_open(__probe_ip: -1186560412, filename: "/usr/lib64/gconv/gconv-modules.cache") 0.656 sleep/455122 probe:do_filp_open(__probe_ip: -1186560412, filename: "/usr/lib/locale/en_US.UTF-8/LC_MEASUREMENT") 0.664 sleep/455122 probe:do_filp_open(__probe_ip: -1186560412, filename: "/usr/lib/locale/en_US.utf8/LC_MEASUREMENT") 0.696 sleep/455122 probe:do_filp_open(__probe_ip: -1186560412, filename: "/usr/lib/locale/en_US.UTF-8/LC_TELEPHONE") [root@quaco ~]# As well as mixing syscalls with tracepoints, getting the syscall tracepoints used augmented using the BPF skel: [root@quaco ~]# perf trace --max-events=10 -e open*,probe:do_filp_open sleep 1 0.000 ( ): sleep/455124 openat(dfd: CWD, filename: "/etc/ld.so.cache", flags: RDONLY|CLOEXEC) ... 0.005 ( ): sleep/455124 probe:do_filp_open(__probe_ip: -1186560412, filename: "/etc/ld.so.cache") 0.000 ( 0.011 ms): sleep/455124 ... [continued]: openat()) = 3 0.031 ( ): sleep/455124 openat(dfd: CWD, filename: "/lib64/libc.so.6", flags: RDONLY|CLOEXEC) ... 0.033 ( ): sleep/455124 probe:do_filp_open(__probe_ip: -1186560412, filename: "/lib64/libc.so.6") 0.031 ( 0.006 ms): sleep/455124 ... [continued]: openat()) = 3 0.258 ( ): sleep/455124 openat(dfd: CWD, filename: "/usr/lib/locale/locale-archive", flags: RDONLY|CLOEXEC) ... 0.261 ( ): sleep/455124 probe:do_filp_open(__probe_ip: -1186560412, filename: "/usr/lib/locale/locale-archive") 0.258 ( 0.006 ms): sleep/455124 ... [continued]: openat()) = -1 ENOENT (No such file or directory) 0.272 ( ): sleep/455124 openat(dfd: CWD, filename: "/usr/share/locale/locale.alias", flags: RDONLY|CLOEXEC) ... 0.273 ( ): sleep/455124 probe:do_filp_open(__probe_ip: -1186560412, filename: "/usr/share/locale/locale.alias") A final note: the probe:do_filp_open uses a kprobe (probably optimized as its in the start of a function) that uses the kprobe_tracer mechanism in the kernel to collect the pathname->name string and stash it into the tracepoint created by 'perf probe' for that: [root@quaco ~]# cat /sys/kernel/debug/tracing/kprobe_events p:probe/do_filp_open _text+4621920 filename=+0(+0(%si)):string [root@quaco ~]# While the syscalls:sys_enter_openat tracepoint gets its string from a BPF program attached to raw_syscalls:sys_enter that tail calls into another BPF program that knows the types for the openat syscall args and thus can bpf_probe_read it right after the normal sys_enter/sys_enter_openat tracepoint payload that comes prefixed with whatever perf_event_open asked for (CPU, timestamp, etc): [root@quaco ~]# bpftool prog | grep -E "sys_enter |sys_enter_opena" -A3 3176: tracepoint name sys_enter tag 0bc3fc9d11754ba1 gpl loaded_at 2023-08-17T12:32:20-0300 uid 0 xlated 272B jited 257B memlock 4096B map_ids 2462,2466,2463 btf_id 2976 -- 3180: tracepoint name sys_enter_opena tag 19dd077f00ec2f58 gpl loaded_at 2023-08-17T12:32:20-0300 uid 0 xlated 328B jited 206B memlock 4096B map_ids 2466,2465 btf_id 2976 [root@quaco ~]# Fixes: 5e6da6be ("perf trace: Migrate BPF augmentation to use a skeleton") Reviewed-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: Andrii Nakryiko <andrii@kernel.org> Cc: Anshuman Khandual <anshuman.khandual@arm.com> Cc: Athira Jajeev <atrajeev@linux.vnet.ibm.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Carsten Haitzler <carsten.haitzler@arm.com> Cc: Eduard Zingerman <eddyz87@gmail.com> Cc: Fangrui Song <maskray@google.com> Cc: He Kuang <hekuang@huawei.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: James Clark <james.clark@arm.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Leo Yan <leo.yan@linaro.org> Cc: Madhavan Srinivasan <maddy@linux.ibm.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Nathan Chancellor <nathan@kernel.org> Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com> Cc: Nick Desaulniers <ndesaulniers@google.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ravi Bangoria <ravi.bangoria@amd.com> Cc: Rob Herring <robh@kernel.org> Cc: Tiezhu Yang <yangtiezhu@loongson.cn> Cc: Tom Rix <trix@redhat.com> Cc: Wang Nan <wangnan0@huawei.com> Cc: Wang ShaoBo <bobo.shaobowang@huawei.com> Cc: Yang Jihong <yangjihong1@huawei.com> Cc: Yonghong Song <yhs@fb.com> Cc: YueHaibing <yuehaibing@huawei.com> Link: https://lore.kernel.org/lkml/ZN4+s2Wl+zYmXTDj@kernel.orgSigned-off-by:
Arnaldo Carvalho de Melo <acme@redhat.com>
-
- 17 Aug, 2023 7 commits
-
-
Arnaldo Carvalho de Melo authored
While debugging a segfault on 'perf lock contention' without an available perf.data file I noticed that it was basically calling: perf_session__delete(ERR_PTR(-1)) Resulting in: (gdb) run lock contention Starting program: /root/bin/perf lock contention [Thread debugging using libthread_db enabled] Using host libthread_db library "/lib64/libthread_db.so.1". failed to open perf.data: No such file or directory (try 'perf record' first) Initializing perf session failed Program received signal SIGSEGV, Segmentation fault. 0x00000000005e7515 in auxtrace__free (session=0xffffffffffffffff) at util/auxtrace.c:2858 2858 if (!session->auxtrace) (gdb) p session $1 = (struct perf_session *) 0xffffffffffffffff (gdb) bt #0 0x00000000005e7515 in auxtrace__free (session=0xffffffffffffffff) at util/auxtrace.c:2858 #1 0x000000000057bb4d in perf_session__delete (session=0xffffffffffffffff) at util/session.c:300 #2 0x000000000047c421 in __cmd_contention (argc=0, argv=0x7fffffffe200) at builtin-lock.c:2161 #3 0x000000000047dc95 in cmd_lock (argc=0, argv=0x7fffffffe200) at builtin-lock.c:2604 #4 0x0000000000501466 in run_builtin (p=0xe597a8 <commands+552>, argc=2, argv=0x7fffffffe200) at perf.c:322 #5 0x00000000005016d5 in handle_internal_command (argc=2, argv=0x7fffffffe200) at perf.c:375 #6 0x0000000000501824 in run_argv (argcp=0x7fffffffe02c, argv=0x7fffffffe020) at perf.c:419 #7 0x0000000000501b11 in main (argc=2, argv=0x7fffffffe200) at perf.c:535 (gdb) So just set it to NULL after using PTR_ERR(session) to decode the error as perf_session__delete(NULL) is supported. Fixes: eef4fee5 ("perf lock: Dynamically allocate lockhash_table") Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: K Prateek Nayak <kprateek.nayak@amd.com> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Leo Yan <leo.yan@linaro.org> Cc: Mamatha Inamdar <mamatha4@linux.vnet.ibm.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ravi Bangoria <ravi.bangoria@amd.com> Cc: Ross Zwisler <zwisler@chromium.org> Cc: Sean Christopherson <seanjc@google.com> Cc: Steven Rostedt (VMware) <rostedt@goodmis.org> Cc: Tiezhu Yang <yangtiezhu@loongson.cn> Cc: Yang Jihong <yangjihong1@huawei.com> Link: https://lore.kernel.org/lkml/ZN4R1AYfsD2J8lRs@kernel.orgSigned-off-by:
-
Arnaldo Carvalho de Melo authored
While debugging a segfault on 'perf lock contention' without an available perf.data file I noticed that it was basically calling: perf_session__delete(ERR_PTR(-1)) Resulting in: (gdb) run lock contention Starting program: /root/bin/perf lock contention [Thread debugging using libthread_db enabled] Using host libthread_db library "/lib64/libthread_db.so.1". failed to open perf.data: No such file or directory (try 'perf record' first) Initializing perf session failed Program received signal SIGSEGV, Segmentation fault. 0x00000000005e7515 in auxtrace__free (session=0xffffffffffffffff) at util/auxtrace.c:2858 2858 if (!session->auxtrace) (gdb) p session $1 = (struct perf_session *) 0xffffffffffffffff (gdb) bt #0 0x00000000005e7515 in auxtrace__free (session=0xffffffffffffffff) at util/auxtrace.c:2858 #1 0x000000000057bb4d in perf_session__delete (session=0xffffffffffffffff) at util/session.c:300 #2 0x000000000047c421 in __cmd_contention (argc=0, argv=0x7fffffffe200) at builtin-lock.c:2161 #3 0x000000000047dc95 in cmd_lock (argc=0, argv=0x7fffffffe200) at builtin-lock.c:2604 #4 0x0000000000501466 in run_builtin (p=0xe597a8 <commands+552>, argc=2, argv=0x7fffffffe200) at perf.c:322 #5 0x00000000005016d5 in handle_internal_command (argc=2, argv=0x7fffffffe200) at perf.c:375 #6 0x0000000000501824 in run_argv (argcp=0x7fffffffe02c, argv=0x7fffffffe020) at perf.c:419 #7 0x0000000000501b11 in main (argc=2, argv=0x7fffffffe200) at perf.c:535 (gdb) So just set it to NULL after using PTR_ERR(session) to decode the error as perf_session__delete(NULL) is supported. The same problem was found in 'perf top' after an audit of all perf_session__new() failure handling. Fixes: 6ef81c55 ("perf session: Return error code for perf_session__new() function on failure") Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Alexey Budankov <alexey.budankov@linux.intel.com> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Jeremie Galarneau <jeremie.galarneau@efficios.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kate Stewart <kstewart@linuxfoundation.org> Cc: Mamatha Inamdar <mamatha4@linux.vnet.ibm.com> Cc: Mukesh Ojha <mojha@codeaurora.org> Cc: Nageswara R Sastry <rnsastry@linux.vnet.ibm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ravi Bangoria <ravi.bangoria@linux.ibm.com> Cc: Shawn Landden <shawn@git.icu> Cc: Song Liu <songliubraving@fb.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Tzvetomir Stoyanov <tstoyanov@vmware.com> Link: https://lore.kernel.org/lkml/ZN4Q2rxxsL08A8rd@kernel.orgSigned-off-by:
-
James Clark authored
Apart from some slight naming and grouping differences, the new metrics are functionally the same as the existing ones. Any missing metrics were manually appended to the end of the auto generated file. For the events, the new data includes descriptions that may have product specific details and new groupings that will be consistent with other products. After generating the metrics from the telemetry repo [1], the following manual steps were performed: * Change the topdown expressions to compare on CPUID and use #slots so that the same data can be shared between N2 and V2. Apart from these modifications, the expressions now match more closely with the Arm telemetry data which will hopefully make future updates easier. * Append some metrics from the old N2/V2 data that aren't present in the telemetry data. These will possibly be added to the telemetry-solution repo at a later time: l3d_cache_mpki, l3d_cache_miss_rate, branch_pki, ipc_rate, spec_ipc, retired_rate, wasted_rate, branch_immed_spec_rate, branch_return_spec_rate, branch_indirect_spec_rate [1]: https://gitlab.arm.com/telemetry-solution/telemetry-solution/-/blob/main/data/pmu/cpu/neoverse/neoverse-n2.jsonSigned-off-by:James Clark <james.clark@arm.com> Reviewed-by:
John Garry <john.g.garry@oracle.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andrii Nakryiko <andrii@kernel.org> Cc: Eduard Zingerman <eddyz87@gmail.com> Cc: Haixin Yu <yuhaixin.yhx@linux.alibaba.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jing Zhang <renyu.zj@linux.alibaba.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kajol Jain <kjain@linux.ibm.com> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Leo Yan <leo.yan@linaro.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Mike Leach <mike.leach@linaro.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Nick Forrington <nick.forrington@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Rob Herring <robh@kernel.org> Cc: Sohom Datta <sohomdatta1@gmail.com> Cc: Will Deacon <will@kernel.org> Cc: linux-arm-kernel@lists.infradead.org Link: https://lore.kernel.org/r/20230816114841.1679234-7-james.clark@arm.comSigned-off-by:
-
James Clark authored
N2 r0p3 doesn't require the workaround [1], so gating on (#slots - 5) no longer works because all N2s have 5 slots. Use the new expression builtin that allows calling strcmp_cpuid_str() and comparing CPUIDs in metric formulas. In this case, the commented formula looks like this: strcmp_cpuid_str(0x410fd493) # greater than or equal to N2 r0p3 | strcmp_cpuid_str(0x410fd490) ^ 1 # OR NOT any version of N2 [1]: https://gitlab.arm.com/telemetry-solution/telemetry-solution/-/blob/main/data/pmu/cpu/neoverse/neoverse-n2-r0p3.jsonSigned-off-by:
-
James Clark authored
This will allow writing formulas that are conditional on a specific CPU type or CPU version. It calls through to the existing strcmp_cpuid_str() function in Perf which has a default weak version, and an arch specific version for x86 and arm64. The function takes an 'ID' type value, which is a string. But in this case Arm CPU IDs are hex numbers prefixed with '0x'. metric.py assumes strings are only used by event names, and that they can't start with a number ('0'), so an additional change has to be made to the regex to convert hex numbers back to 'ID' types. Signed-off-by: -
James Clark authored
Now that variant and revision fields are taken into account the behavior is slightly more complicated so add a test to ensure that this behaves as expected. Reviewed-by:
-
James Clark authored
Currently variant and revision fields are masked out of the MIDR so it's not possible to compare different versions of the same CPU. In a later commit a workaround will be removed just for N2 r0p3, so enable comparisons on version. This has the side effect of changing the MIDR stored in the header of the perf.data file to no longer have masked version fields. It also affects the lookups in mapfile.csv, but as that currently only has zeroed version fields, it has no actual effect. The mapfile.csv documentation also states to zero the version fields, so unless this isn't done it will continue to have no effect. There is an existing weak default strcmp_cpuid_str() function, and an x86 version. This adds another version for arm64. Signed-off-by:
-
- 16 Aug, 2023 17 commits
-
-
Arnaldo Carvalho de Melo authored
This works with: $ clang -v clang version 14.0.5 (Fedora 14.0.5-2.fc36) $ But not with: $ clang -v clang version 16.0.6 (Fedora 16.0.6-2.fc38) $ [root@quaco ~]# perf trace -e connect*,sendto* ping -c 10 localhost libbpf: prog 'sys_enter_sendto': BPF program load failed: Permission denied libbpf: prog 'sys_enter_sendto': -- BEGIN PROG LOAD LOG -- reg type unsupported for arg#0 function sys_enter_sendto#59 0: R1=ctx(off=0,imm=0) R10=fp0 ; int sys_enter_sendto(struct syscall_enter_args *args) 0: (bf) r6 = r1 ; R1=ctx(off=0,imm=0) R6_w=ctx(off=0,imm=0) 1: (b7) r1 = 0 ; R1_w=0 ; int key = 0; 2: (63) *(u32 *)(r10 -4) = r1 ; R1_w=0 R10=fp0 fp-8=0000???? 3: (bf) r2 = r10 ; R2_w=fp0 R10=fp0 ; 4: (07) r2 += -4 ; R2_w=fp-4 ; return bpf_map_lookup_elem(&augmented_args_tmp, &key); 5: (18) r1 = 0xffff8de5a5b8bc00 ; R1_w=map_ptr(off=0,ks=4,vs=8272,imm=0) 7: (85) call bpf_map_lookup_elem#1 ; R0_w=map_value_or_null(id=1,off=0,ks=4,vs=8272,imm=0) 8: (bf) r7 = r0 ; R0_w=map_value_or_null(id=1,off=0,ks=4,vs=8272,imm=0) R7_w=map_value_or_null(id=1,off=0,ks=4,vs=8272,imm=0) 9: (b7) r0 = 1 ; R0_w=1 ; if (augmented_args == NULL) 10: (15) if r7 == 0x0 goto pc+25 ; R7_w=map_value(off=0,ks=4,vs=8272,imm=0) ; unsigned int socklen = args->args[5]; 11: (79) r1 = *(u64 *)(r6 +56) ; R1_w=scalar() R6_w=ctx(off=0,imm=0) ; 12: (bf) r2 = r1 ; R1_w=scalar(id=2) R2_w=scalar(id=2) 13: (67) r2 <<= 32 ; R2_w=scalar(smax=9223372032559808512,umax=18446744069414584320,var_off=(0x0; 0xffffffff00000000),s32_min=0,s32_max=0,u32_max=0) 14: (77) r2 >>= 32 ; R2_w=scalar(umax=4294967295,var_off=(0x0; 0xffffffff)) 15: (b7) r8 = 128 ; R8=128 ; if (socklen > sizeof(augmented_args->saddr)) 16: (25) if r2 > 0x80 goto pc+1 ; R2=scalar(umax=128,var_off=(0x0; 0xff)) 17: (bf) r8 = r1 ; R1=scalar(id=2) R8_w=scalar(id=2) ; const void *sockaddr_arg = (const void *)args->args[4]; 18: (79) r3 = *(u64 *)(r6 +48) ; R3_w=scalar() R6=ctx(off=0,imm=0) ; bpf_probe_read(&augmented_args->saddr, socklen, sockaddr_arg); 19: (bf) r1 = r7 ; R1_w=map_value(off=0,ks=4,vs=8272,imm=0) R7=map_value(off=0,ks=4,vs=8272,imm=0) 20: (07) r1 += 64 ; R1_w=map_value(off=64,ks=4,vs=8272,imm=0) ; bpf_probe_read(&augmented_args->saddr, socklen, sockaddr_arg); 21: (bf) r2 = r8 ; R2_w=scalar(id=2) R8_w=scalar(id=2) 22: (85) call bpf_probe_read#4 R2 min value is negative, either use unsigned or 'var &= const' processed 22 insns (limit 1000000) max_states_per_insn 0 total_states 1 peak_states 1 mark_read 1 -- END PROG LOAD LOG -- libbpf: prog 'sys_enter_sendto': failed to load: -13 libbpf: failed to load object 'augmented_raw_syscalls_bpf' libbpf: failed to load BPF skeleton 'augmented_raw_syscalls_bpf': -13 So use the suggested &= variant since sizeof(saddr) == 128 bytes. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Signed-off-by: -
Kajol Jain authored
Update JSON/events for power10 platform with additional metrics. Signed-off-by:
Kajol Jain <kjain@linux.ibm.com> Cc: Athira Rajeev <atrajeev@linux.vnet.ibm.com> Cc: Disha Goel <disgoel@linux.ibm.com> Cc: Ian Rogers <irogers@google.com> Cc: Madhavan Srinivasan <maddy@linux.ibm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: linuxppc-dev@lists.ozlabs.org Link: https://lore.kernel.org/r/20230814112803.1508296-7-kjain@linux.ibm.comSigned-off-by:
-
Kajol Jain authored
Update metric event name for some of the JSON/metric events for power10 platform. Fixes: 3ca3af7d ("perf vendor events power10: Add metric events JSON file for power10 platform") Signed-off-by:
-
Kajol Jain authored
Update JSON/events for power10 platform with additional events. Signed-off-by:
-
Kajol Jain authored
Move some of the power10 JSON/events to appropriate files. Fixes: 32daa5d7 ("perf vendor events: Initial JSON/events list for power10 platform") Signed-off-by:
-
Kajol Jain authored
Drop STORES_PER_INST metric event for the power10 platform, as the metric expression of STORES_PER_INST metric event using dropped event PM_ST_FIN. Fixes: 3ca3af7d ("perf vendor events power10: Add metric events JSON file for power10 platform") Signed-off-by:
-
Kajol Jain authored
Drop some of the JSON/events for power10 platform due to counter data mismatch. Fixes: 32daa5d7 ("perf vendor events: Initial JSON/events list for power10 platform") Signed-off-by:
-
Kajol Jain authored
Update the description for some of the JSON/events for power10 platform. Fixes: 32daa5d7 ("perf vendor events: Initial JSON/events list for power10 platform") Signed-off-by:
-
Alexandre Ghiti authored
riscv now supports mmaping hardware counters to userspace so adapt the test to run on this architecture. Reviewed-by:
Andrew Jones <ajones@ventanamicro.com> Reviewed-by:
Atish Patra <atishp@rivosinc.com> Reviewed-by:
Alexandre Ghiti <alexghiti@rivosinc.com> Cc: Albert Ou <aou@eecs.berkeley.edu> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Anup Patel <anup@brainfault.org> Cc: Atish Patra <atishp@atishpatra.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Mark Rutland <mark.rutland@arm.com> 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: Rob Herring <robh@kernel.org> Cc: Rémi Denis-Courmont <remi@remlab.net> Cc: Will Deacon <will@kernel.org> Cc: linux-arm-kernel@lists.infradead.org Cc: linux-doc@vger.kernel.org Cc: linux-riscv@lists.infradead.org Link: https://lore.kernel.org/r/20230802080328.1213905-11-alexghiti@rivosinc.comSigned-off-by:
-
Alexandre Ghiti authored
riscv now supports mmaping hardware counters so add what's needed to take advantage of that in libperf. Reviewed-by:
-
Leo Yan authored
util/perf_regs.h includes another perf_regs.h: #include <perf_regs.h> Here it includes architecture specific header, for example, if we build arm64 target, the header tools/perf/arch/arm64/include/perf_regs.h is included. We use this implicit way to include architecture specific header, which is not directive; furthermore, util/perf_regs.c is coupled with the architecture specific definitions. This patch moves out arch specific header from util/perf_regs.h for generalizing the 'util' folder, as a result, the source files in 'arch' folder explicitly include architecture's perf_regs.h. Signed-off-by:
Leo Yan <leo.yan@linaro.org> Acked-by:
-
Leo Yan authored
The macros PERF_REGS_MAX and PERF_REGS_MASK are architecture specific, let's remove them from the common file util/perf_regs.c. As a side effect, the weak functions arch__intr_reg_mask() and arch__user_reg_mask() just return zeros, every arch defines its own functions in the 'arch' folder for returning right values. Note, we don't need to return intr/user register masks dynamically, this is because these two functions are invoked during recording phase but not decoding phase, they are always invoked on the native environment, thus we don't need to parse them dynamically. Signed-off-by:
-
Leo Yan authored
The macros PERF_REG_{IP|SP} have been replaced by using functions perf_arch_reg_{ip|sp}(), remove them! Signed-off-by: -
Leo Yan authored
We use perf_arch_reg_ip() and perf_arch_reg_sp() to substitute macros for obtaining the register numbers of SP and IP. This modification enables cross analysis in the unwinding, therefore, the unwinding is not restricted to the predefined values by the macros. Consequently, the macros LIBUNWIND__ARCH_REG_{IP|SP} are removed since they are no longer used. Committer notes: Add missing "util/env.h" header to make sure we have the definition for perf_env__arch(), that when built with NO_LIBUNWIND=1 isn't available, i.e. it was being included by sheer luck. Signed-off-by: -
Leo Yan authored
The current code uses macros PERF_REG_IP and PERF_REG_SP for parsing registers and we build perf with these macros statically, which means it only can correctly analyze CPU registers for the native architecture and fails to support cross analysis (e.g. we build perf on x86 and cannot analyze Arm64's registers). We need to generalize util/perf_regs.c for support multi architectures, as a first step, this commit introduces new functions perf_arch_reg_ip() and perf_arch_reg_sp(), these two functions dynamically return IP and SP register index respectively according to the parameter "arch". Every architecture has its own functions (like __perf_reg_ip_arm64 and __perf_reg_sp_arm64), these architecture specific functions are defined in each arch source file under folder util/perf-regs-arch; at the end all of them are built into the tool for cross analysis. Committer notes: Make DWARF_MINIMAL_REGS() an inline function, so that we can use the __maybe_unused attribute for the 'arch' parameter, as this will avoid a build failure when that variable is unused in the callers. That happens when building on unsupported architectures, the ones without HAVE_PERF_REGS_SUPPORT defined. Signed-off-by:
-
Leo Yan authored
Every architecture has a specific register parsing function for returning register name based on register index, to support cross analysis (e.g. we use perf x86 binary to parse Arm64's perf data), we build all these register parsing functions into the tool, this is why we place all related functions into util/perf_regs.c. Unfortunately, since util/perf_regs.c needs to include every arch's perf_regs.h, this easily introduces duplicated definitions coming from multiple headers, finally it's fragile for building and difficult for maintenance. We cannot simply move these register parsing functions into the corresponding 'arch' folder, the folder is only conditionally built based on the target architecture. Therefore, this commit creates a new folder util/perf-regs-arch/ and uses a dedicated source file to keep every architecture's register parsing function to avoid definition conflicts. This is only a refactoring, no functionality change is expected. Committer notes: Had to add util/perf-regs-arch/*.c to tools/perf/util/python-ext-sources to keep 'perf test python' passing. Signed-off-by:
-
Changbin Du authored
Fix the format of unordered lists so the can wrap properly. Signed-off-by:
Changbin Du <changbin.du@huawei.com> Acked-by:
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: Mark Rutland <mark.rutland@arm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/r/20230718085242.3090797-1-changbin.du@huawei.comSigned-off-by:
-
- 15 Aug, 2023 15 commits
-
-
James Clark authored
Metrics will be published here [1] going forwards, but they have slightly different scale units. To allow autogenerated metrics to be added more easily, update the scale units to match. The more detailed descriptions have also been taken and added to the common file. [1]: https://gitlab.arm.com/telemetry-solution/telemetry-solution/-/tree/main/data/pmu/cpu/Reviewed-by:
-
James Clark authored
linux/bitfield.h can be included as long as linux/kernel.h is included first, so change the order of the includes and drop the duplicate macro. Reviewed-by:
-
Adrian Hunter authored
Add perf_dlfilter_fns.al_cleanup() to do addr_location__exit() on data passed via perf_dlfilter_fns.resolve_address(). Add dlfilter-test-api-v2 to the "dlfilter C API" test to test it. Update documentation, clarifying that data returned by APIs should not be dereferenced after filter_event() and filter_event_early() return. Fixes: 0dd5041c ("perf addr_location: Add init/exit/copy functions") Reviewed-by:
-
Arnaldo Carvalho de Melo authored
As thread__find_symbol_fb() will end up calling thread__find_map() and it in turn will call these on uninitialized memory: maps__zput(al->maps); map__zput(al->map); thread__zput(al->thread); Fixes: 0dd5041c ("perf addr_location: Add init/exit/copy functions") Reviewed-by: -
Adrian Hunter authored
Extend the "dlfilter C API" test to test perf_dlfilter_fns.resolve_address(). The test currently fails, but passes after a subsequent patch. Reviewed-by:
-
Wei Li authored
'audit-libs-python' is the package for python2, update it for python3. On Ubuntu and Fedora, the new package is 'python3-audit'. Signed-off-by:
Wei Li <liwei391@huawei.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: Li Bin <huawei.libin@huawei.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/20230815131805.1237491-1-liwei391@huawei.comSigned-off-by:
-
Wei Li authored
In the result of "perf script syscall-counts" on arm64, the syscall events are not resolved currently. Add "aarch64" to audit uname list to support name parsing. * After the patch: [root@localhost ~]# perf script syscall-counts sleep 1 Press control+C to stop and show the summary syscall events: event count ---------------------------------------- ----------- mmap 6 close 5 mprotect 4 brk 3 newfstatat 3 openat 3 getrandom 1 prlimit64 1 munmap 1 clock_nanosleep 1 set_robust_list 1 set_tid_address 1 exit_group 1 read 1 faccessat 1 Signed-off-by:
-
Yang Jihong authored
The `file` parameter in evsel__intval() is checked repeatedly, fix it. No functional change. Signed-off-by:
Yang Jihong <yangjihong1@huawei.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> Cc: Ravi Bangoria <ravi.bangoria@amd.com> Cc: Sandipan Das <sandipan.das@amd.com> Link: https://lore.kernel.org/r/20230815221009.3641751-1-yangjihong1@huawei.comSigned-off-by:
-
Namhyung Kim authored
$ sudo ./perf test 'sample filter' -v 94: perf record sample filtering (by BPF) tests : --- start --- test child forked, pid 3817527 Checking BPF-filter privilege Basic bpf-filter test Basic bpf-filter test [Success] Failing bpf-filter test Error: task-clock event does not have PERF_SAMPLE_CPU Failing bpf-filter test [Success] Group bpf-filter test Error: task-clock event does not have PERF_SAMPLE_CPU Error: task-clock event does not have PERF_SAMPLE_CODE_PAGE_SIZE Group bpf-filter test [Success] test child finished with 0 ---- end ---- perf record sample filtering (by BPF) tests: Ok Reviewed-by:
Namhyung Kim <namhyung@kernel.org> Tested-by:
-
Namhyung Kim authored
For logical OR operator, the actual sample_flags are in the 'groups' list so it needs to check entries in the list instead. Otherwise it would show the following error message. $ sudo perf record -a -e cycles:p --filter 'period > 100 || weight > 0' sleep 1 Error: cycles:p event does not have sample flags 0 failed to set filter "BPF" on event cycles:p with 2 (No such file or directory) Actually it should warn on 'weight' is used without WEIGHT flag. Error: cycles:p event does not have PERF_SAMPLE_WEIGHT Hint: please add -W option to perf record failed to set filter "BPF" on event cycles:p with 2 (No such file or directory) Fixes: 4310551b ("perf bpf filter: Show warning for missing sample flags") Reviewed-by:
-
Ian Rogers authored
Now tools/perf/examples/bpf/augmented_syscalls.c is tools/perf/util/bpf_skel/augmented_syscalls.bpf.c and not enabled as a BPF event, tidy the comments to reflect this. Signed-off-by:
Jiri Olsa <jolsa@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Andrii Nakryiko <andrii@kernel.org> Cc: Anshuman Khandual <anshuman.khandual@arm.com> Cc: Athira Rajeev <atrajeev@linux.vnet.ibm.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Carsten Haitzler <carsten.haitzler@arm.com> Cc: Eduard Zingerman <eddyz87@gmail.com> Cc: Fangrui Song <maskray@google.com> Cc: He Kuang <hekuang@huawei.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: James Clark <james.clark@arm.com> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Leo Yan <leo.yan@linaro.org> Cc: Madhavan Srinivasan <maddy@linux.ibm.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Nathan Chancellor <nathan@kernel.org> Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com> Cc: Nick Desaulniers <ndesaulniers@google.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ravi Bangoria <ravi.bangoria@amd.com> Cc: Rob Herring <robh@kernel.org> Cc: Tiezhu Yang <yangtiezhu@loongson.cn> Cc: Tom Rix <trix@redhat.com> Cc: Wang Nan <wangnan0@huawei.com> Cc: Wang ShaoBo <bobo.shaobowang@huawei.com> Cc: Yang Jihong <yangjihong1@huawei.com> Cc: Yonghong Song <yhs@fb.com> Cc: YueHaibing <yuehaibing@huawei.com> Cc: bpf@vger.kernel.org Cc: llvm@lists.linux.dev Link: https://lore.kernel.org/r/20230810184853.2860737-5-irogers@google.comSigned-off-by:
-
Ian Rogers authored
The examples were used to give demonstrations of BPF events but such functionality is now subsumed by using --filter with 'perf record' or the direct use of BPF skeletons. Signed-off-by:
-
Ian Rogers authored
Previously a BPF event of augmented_raw_syscalls.c could be used to enable augmentation of syscalls by perf trace. As BPF events are no longer supported, switch to using a BPF skeleton which when attached explicitly opens the sysenter and sysexit tracepoints. The dump map is removed as debugging wasn't supported by the augmentation and bpf_printk can be used when necessary. Remove tools/perf/examples/bpf/augmented_raw_syscalls.c so that the rename/migration to a BPF skeleton captures that this was the source. Committer notes: Some minor stylistic changes to help visualizing the diff. Use libbpf_strerror when failing to load the augmented raw syscalls BPF. Use bpf_object__for_each_program(prog, trace.skel->obj) to disable auto attachment for all but the sys_enter, sys_exit tracepoints, to avoid having to add extra lines as we go adding support for more pointer receiving syscalls. Committer testing: # perf trace -e open* --max-events=10 0.000 ( 0.022 ms): systemd-oomd/1151 openat(dfd: CWD, filename: "/proc/meminfo", flags: RDONLY|CLOEXEC) = 11 208.833 ( ): gnome-terminal/3223 openat(dfd: CWD, filename: "/proc/51250/cmdline") ... 249.993 ( 0.024 ms): systemd-oomd/1151 openat(dfd: CWD, filename: "/proc/meminfo", flags: RDONLY|CLOEXEC) = 11 250.118 ( 0.030 ms): systemd-oomd/1151 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1000.slice/user@1000.service/memory.pressure", flags: RDONLY|CLOEXEC) = 11 250.205 ( 0.016 ms): systemd-oomd/1151 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1000.slice/user@1000.service/memory.current", flags: RDONLY|CLOEXEC) = 11 250.244 ( 0.014 ms): systemd-oomd/1151 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1000.slice/user@1000.service/memory.min", flags: RDONLY|CLOEXEC) = 11 250.282 ( 0.014 ms): systemd-oomd/1151 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1000.slice/user@1000.service/memory.low", flags: RDONLY|CLOEXEC) = 11 250.320 ( 0.014 ms): systemd-oomd/1151 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1000.slice/user@1000.service/memory.swap.current", flags: RDONLY|CLOEXEC) = 11 250.355 ( 0.014 ms): systemd-oomd/1151 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1000.slice/user@1000.service/memory.stat", flags: RDONLY|CLOEXEC) = 11 250.717 ( 0.016 ms): systemd-oomd/1151 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1001.slice/user@1001.service/memory.pressure", flags: RDONLY|CLOEXEC) = 11 # # perf trace -e *nanosleep* --max-events=10 ? ( ): SCTP timer/28304 ... [continued]: clock_nanosleep()) = 0 0.007 (10.058 ms): SCTP timer/28304 clock_nanosleep(rqtp: { .tv_sec: 0, .tv_nsec: 10000000 }, rmtp: 0x7f0466b78de0) = 0 10.069 ( ): SCTP timer/28304 clock_nanosleep(rqtp: { .tv_sec: 0, .tv_nsec: 10000000 }, rmtp: 0x7f0466b78de0) ... 10.069 (10.056 ms): SCTP timer/28304 ... [continued]: clock_nanosleep()) = 0 17.059 ( ): podman/3572 nanosleep(rqtp: 0x7fc4f4d75be0) ... 17.059 (10.061 ms): podman/3572 ... [continued]: nanosleep()) = 0 20.131 (10.059 ms): SCTP timer/28304 clock_nanosleep(rqtp: { .tv_sec: 0, .tv_nsec: 10000000 }, rmtp: 0x7f0466b78de0) = 0 30.195 (10.038 ms): SCTP timer/28304 clock_nanosleep(rqtp: { .tv_sec: 0, .tv_nsec: 10000000 }, rmtp: 0x7f0466b78de0) = 0 40.238 (10.057 ms): SCTP timer/28304 clock_nanosleep(rqtp: { .tv_sec: 0, .tv_nsec: 10000000 }, rmtp: 0x7f0466b78de0) = 0 50.301 ( ): SCTP timer/28304 clock_nanosleep(rqtp: { .tv_sec: 0, .tv_nsec: 10000000 }, rmtp: 0x7f0466b78de0) ... # # perf trace -e perf_event* -- perf stat -e instructions,cycles,cache-misses sleep 0.1 0.000 ( 0.011 ms): perf/51331 perf_event_open(attr_uptr: { type: 0 (PERF_TYPE_HARDWARE), size: 136, config: 0x1 (PERF_COUNT_HW_INSTRUCTIONS), sample_type: IDENTIFIER, read_format: TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING, disabled: 1, inherit: 1, enable_on_exec: 1, exclude_guest: 1 }, pid: 51332 (perf), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 3 0.013 ( 0.003 ms): perf/51331 perf_event_open(attr_uptr: { type: 0 (PERF_TYPE_HARDWARE), size: 136, config: 0 (PERF_COUNT_HW_CPU_CYCLES), sample_type: IDENTIFIER, read_format: TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING, disabled: 1, inherit: 1, enable_on_exec: 1, exclude_guest: 1 }, pid: 51332 (perf), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 4 0.017 ( 0.002 ms): perf/51331 perf_event_open(attr_uptr: { type: 0 (PERF_TYPE_HARDWARE), size: 136, config: 0x3 (PERF_COUNT_HW_CACHE_MISSES), sample_type: IDENTIFIER, read_format: TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING, disabled: 1, inherit: 1, enable_on_exec: 1, exclude_guest: 1 }, pid: 51332 (perf), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 5 Performance counter stats for 'sleep 0.1': 1,495,051 instructions # 1.11 insn per cycle 1,347,641 cycles 35,424 cache-misses 0.100935279 seconds time elapsed 0.000924000 seconds user 0.000000000 seconds sys # # perf trace -e connect* ssh localhost 0.000 ( 0.012 ms): ssh/51346 connect(fd: 4, uservaddr: { .family: LOCAL, path: /var/lib/sss/pipes/nss }, addrlen: 110) = -1 ECONNREFUSED (Connection refused) 0.118 ( 0.004 ms): ssh/51346 connect(fd: 6, uservaddr: { .family: LOCAL, path: /var/lib/sss/pipes/nss }, addrlen: 110) = -1 ECONNREFUSED (Connection refused) 0.399 ( 0.007 ms): ssh/51346 connect(fd: 4, uservaddr: { .family: LOCAL, path: /var/lib/sss/pipes/nss }, addrlen: 110) = -1 ECONNREFUSED (Connection refused) 0.426 ( 0.003 ms): ssh/51346 connect(fd: 4, uservaddr: { .family: LOCAL, path: /var/lib/sss/pipes/nss }, addrlen: 110) = -1 ECONNREFUSED (Connection refused) 0.754 ( 0.009 ms): ssh/51346 connect(fd: 4, uservaddr: { .family: INET, port: 22, addr: 127.0.0.1 }, addrlen: 16) = 0 0.771 ( 0.010 ms): ssh/51346 connect(fd: 4, uservaddr: { .family: INET6, port: 22, addr: ::1 }, addrlen: 28) = 0 0.798 ( 0.053 ms): ssh/51346 connect(fd: 4, uservaddr: { .family: INET6, port: 22, addr: ::1 }, addrlen: 28) = 0 0.870 ( 0.004 ms): ssh/51346 connect(fd: 5, uservaddr: { .family: LOCAL, path: /var/lib/sss/pipes/nss }, addrlen: 110) = -1 ECONNREFUSED (Connection refused) 0.904 ( 0.003 ms): ssh/51346 connect(fd: 5, uservaddr: { .family: LOCAL, path: /var/lib/sss/pipes/nss }, addrlen: 110) = -1 ECONNREFUSED (Connection refused) 0.930 ( 0.003 ms): ssh/51346 connect(fd: 5, uservaddr: { .family: LOCAL, path: /var/lib/sss/pipes/nss }, addrlen: 110) = -1 ECONNREFUSED (Connection refused) 0.957 ( 0.003 ms): ssh/51346 connect(fd: 5, uservaddr: { .family: LOCAL, path: /var/lib/sss/pipes/nss }, addrlen: 110) = -1 ECONNREFUSED (Connection refused) 0.981 ( 0.003 ms): ssh/51346 connect(fd: 5, uservaddr: { .family: LOCAL, path: /var/lib/sss/pipes/nss }, addrlen: 110) = -1 ECONNREFUSED (Connection refused) 1.006 ( 0.004 ms): ssh/51346 connect(fd: 5, uservaddr: { .family: LOCAL, path: /var/lib/sss/pipes/nss }, addrlen: 110) = -1 ECONNREFUSED (Connection refused) 1.036 ( 0.005 ms): ssh/51346 connect(fd: 5, uservaddr: { .family: LOCAL, path: /var/lib/sss/pipes/nss }, addrlen: 110) = -1 ECONNREFUSED (Connection refused) 65.077 ( 0.022 ms): ssh/51346 connect(fd: 5, uservaddr: { .family: LOCAL, path: /var/run/.heim_org.h5l.kcm-socket }, addrlen: 110) = 0 66.608 ( 0.014 ms): ssh/51346 connect(fd: 5, uservaddr: { .family: LOCAL, path: /var/run/.heim_org.h5l.kcm-socket }, addrlen: 110) = 0 root@localhost's password: # # perf trace -e sendto* ping -c 2 localhost PING localhost(localhost (::1)) 56 data bytes 64 bytes from localhost (::1): icmp_seq=1 ttl=64 time=0.024 ms 0.000 ( 0.011 ms): ping/51357 sendto(fd: 5, buff: 0x7ffcca35e620, len: 20, addr: { .family: NETLINK }, addr_len: 0xc) = 20 0.135 ( 0.026 ms): ping/51357 sendto(fd: 4, buff: 0x5601398f7b20, len: 64, addr: { .family: INET6, port: 58, addr: ::1 }, addr_len: 0x1c) = 64 1014.929 ( 0.050 ms): ping/51357 sendto(fd: 4, buff: 0x5601398f7b20, len: 64, flags: CONFIRM, addr: { .family: INET6, port: 58, addr: ::1 }, addr_len: 0x1c) = 64 64 bytes from localhost (::1): icmp_seq=2 ttl=64 time=0.046 ms --- localhost ping statistics --- 2 packets transmitted, 2 received, 0% packet loss, time 1015ms rtt min/avg/max/mdev = 0.024/0.035/0.046/0.011 ms # Signed-off-by: -
Ian Rogers authored
New features like the BPF --filter support in perf record have made the BPF event functionality somewhat redundant. As shown by commit fcb027c1a4f6 ("perf tools: Revert enable indices setting syntax for BPF map") and commit 14e4b9f4 ("perf trace: Raw augmented syscalls fix libbpf 1.0+ compatibility") the BPF event support hasn't been well maintained and it adds considerable complexity in areas like event parsing, not least as '/' is a separator for event modifiers as well as in paths. This patch removes support in the event parser for BPF events and then the associated functions are removed. This leads to the removal of whole source files like bpf-loader.c. Removing support means that augmented syscalls in perf trace is broken, this will be fixed in a later commit adding support using BPF skeletons. The removal of BPF events causes an unused label warning from flex generated code, so update build to ignore it: ``` util/parse-events-flex.c:2704:1: error: label ‘find_rule’ defined but not used [-Werror=unused-label] 2704 | find_rule: /* we branch to this label when backing up */ ``` Committer notes: Extracted from a larger patch that was also removing the support for linking with libllvm and libclang, that were an alternative to using an external clang execution to compile the .c event source code into BPF bytecode. Testing it: # perf trace -e /home/acme/git/perf/tools/perf/examples/bpf/augmented_raw_syscalls.c event syntax error: '/home/acme/git/perf/tools/perf/examples/bpf/augmented_raw_syscalls.c' \___ Bad event or PMU Unabled to find PMU or event on a PMU of 'home' Initial error: event syntax error: '/home/acme/git/perf/tools/perf/examples/bpf/augmented_raw_syscalls.c' \___ Cannot find PMU `home'. Missing kernel support? Run 'perf list' for a list of valid events Usage: perf trace [<options>] [<command>] or: perf trace [<options>] -- <command> [<options>] or: perf trace record [<options>] [<command>] or: perf trace record [<options>] -- <command> [<options>] -e, --event <event> event/syscall selector. use 'perf list' to list available events # Signed-off-by: -
Ian Rogers authored
This never was in the default build for perf, is difficult to maintain as it uses clang/llvm internals so ditch it, keeping, for now, the external compilation of .c BPF into .o bytecode and its subsequent loading, that is also going to be removed, do it separately to help bisection and to properly document what is being removed and why. Committer notes: Extracted from a larger patch and removed some leftovers, namely deleting these now unused feature tests: tools/build/feature/test-clang.cpp tools/build/feature/test-cxx.cpp tools/build/feature/test-llvm-version.cpp tools/build/feature/test-llvm.cpp Testing the use of BPF events after applying this patch: To use the external clang/llvm toolchain to compile a .c event and then use libbpf to load it, to get the syscalls:sys_enter_open* tracepoints and read the filename pointer, putting it into the ring buffer right after the usual tracepoint payload for 'perf trace' to then print it: [root@quaco ~]# perf trace -e /home/acme/git/perf-tools-next/tools/perf/examples/bpf/augmented_raw_syscalls.c,open* --max-events=10 0.000 systemd-oomd/959 openat(dfd: CWD, filename: "/proc/meminfo", flags: RDONLY|CLOEXEC) = 12 0.083 abrt-dump-jour/1453 openat(dfd: CWD, filename: "/var/log/journal/d6a97235307247e09f13f326fb607e3c/system.journal", flags: RDONLY|CLOEXEC|NONBLOCK) = 4 0.063 abrt-dump-jour/1454 openat(dfd: CWD, filename: "/var/log/journal/d6a97235307247e09f13f326fb607e3c/system.journal", flags: RDONLY|CLOEXEC|NONBLOCK) = 4 0.082 abrt-dump-jour/1455 openat(dfd: CWD, filename: "/var/log/journal/d6a97235307247e09f13f326fb607e3c/system.journal", flags: RDONLY|CLOEXEC|NONBLOCK) = 4 250.124 systemd-oomd/959 openat(dfd: CWD, filename: "/proc/meminfo", flags: RDONLY|CLOEXEC) = 12 250.521 systemd-oomd/959 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1000.slice/user@1000.service/app.slice/memory.pressure", flags: RDONLY|CLOEXEC) = 12 251.047 systemd-oomd/959 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1000.slice/user@1000.service/app.slice/memory.current", flags: RDONLY|CLOEXEC) = 12 251.162 systemd-oomd/959 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1000.slice/user@1000.service/app.slice/memory.min", flags: RDONLY|CLOEXEC) = 12 251.242 systemd-oomd/959 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1000.slice/user@1000.service/app.slice/memory.low", flags: RDONLY|CLOEXEC) = 12 251.353 systemd-oomd/959 openat(dfd: CWD, filename: "/sys/fs/cgroup/user.slice/user-1000.slice/user@1000.service/app.slice/memory.swap.current", flags: RDONLY|CLOEXEC) = 12 [root@quaco ~]# Same thing, but with a prebuilt .o BPF bytecode: [root@quaco ~]# perf trace -e /home/acme/git/perf-tools-next/tools/perf/examples/bpf/augmented_raw_syscalls.o,open* --max-events=10 0.000 systemd-oomd/959 openat(dfd: CWD, filename: "/proc/meminfo", flags: RDONLY|CLOEXEC) = 12 0.083 abrt-dump-jour/1453 openat(dfd: CWD, filename: "/var/log/journal/d6a97235307247e09f13f326fb607e3c/system.journal", flags: RDONLY|CLOEXEC|NONBLOCK) = 4 0.083 abrt-dump-jour/1455 openat(dfd: CWD, filename: "/var/log/journal/d6a97235307247e09f13f326fb607e3c/system.journal", flags: RDONLY|CLOEXEC|NONBLOCK) = 4 0.062 abrt-dump-jour/1454 openat(dfd: CWD, filename: "/var/log/journal/d6a97235307247e09f13f326fb607e3c/system.journal", flags: RDONLY|CLOEXEC|NONBLOCK) = 4 249.985 systemd-oomd/959 openat(dfd: CWD, filename: "/proc/meminfo", flags: RDONLY|CLOEXEC) = 12 466.763 thermald/1234 openat(dfd: CWD, filename: "/sys/class/powercap/intel-rapl/intel-rapl:0/intel-rapl:0:2/energy_uj") = 13 467.145 thermald/1234 openat(dfd: CWD, filename: "/sys/class/powercap/intel-rapl/intel-rapl:0/energy_uj") = 13 467.311 thermald/1234 openat(dfd: CWD, filename: "/sys/class/thermal/thermal_zone2/temp") = 13 500.040 cgroupify/24006 openat(dfd: 4, filename: ".", flags: RDONLY|CLOEXEC|DIRECTORY|NONBLOCK) = 5 500.295 cgroupify/24006 openat(dfd: 4, filename: "24616/cgroup.procs") = 5 [root@quaco ~]# Signed-off-by:
-
