summaryrefslogtreecommitdiff
path: root/tools/perf
AgeCommit message (Collapse)Author
2019-07-29perf trace: Handle raw_syscalls:sys_enter just like the BPF_OUTPUT augmented ↵Arnaldo Carvalho de Melo
event So, we use a PERF_COUNT_SW_BPF_OUTPUT to output the augmented sys_enter payload, i.e. to output more than just the raw syscall args, and if something goes wrong when handling an unfiltered syscall, we bail out and just return 1 in the bpf program associated with raw_syscalls:sys_enter, meaning, don't filter that tracepoint, in which case what will appear in the perf ring buffer isn't the BPF_OUTPUT event, but the original raw_syscalls:sys_enter event with its normal payload. Now that we're switching to using a bpf_tail_call + BPF_MAP_TYPE_PROG_ARRAY we're going to use this in the common case, so a bug where raw_syscalls:sys_enter wasn't being handled by trace__sys_enter() surfaced and for that case, instead of using the strace-like augmenter (trace__sys_enter()), we continued to use the normal generic tracepoint handler: (gdb) p evsel $2 = (struct perf_evsel *) 0xc03e40 (gdb) p evsel->name $3 = 0xbc56c0 "raw_syscalls:sys_enter" (gdb) p ((struct perf_evsel *) 0xc03e40)->name $4 = 0xbc56c0 "raw_syscalls:sys_enter" (gdb) p ((struct perf_evsel *) 0xc03e40)->handler $5 = (void *) 0x495eb3 <trace__event_handler> This resulted in this: 0.027 raw_syscalls:sys_enter:NR 12 (0, 7fcfcac64c9b, 4d, 7fcfcac64c9b, 7fcfcac6ce00, 19) ... [continued]: brk()) = 0x563b88677000 I.e. only the sys_exit tracepoint was being properly handled, but since the sys_enter went to the generic trace__event_handler() we printed it using libtraceevent's formatter instead of 'perf trace's strace-like one. Fix it by setting trace__sys_enter() as the handler for raw_syscalls:sys_enter and setup the tp_field tracepoint field accessors. Now, to test it we just make raw_syscalls:sys_enter return 1 right after checking if the pid is filtered, making it not use bpf_perf_output_event() but rather ask for the tracepoint not to be filtered and the result is the expected one: brk(NULL) = 0x556f42d6e000 I.e. raw_syscalls:sys_enter returns 1, gets handled by trace__sys_enter() and gets it combined with the raw_syscalls:sys_exit in a strace-like way. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-0mkocgk31nmy0odknegcby4z@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-29perf trace: Put the per-syscall entry/exit prog_array BPF map infrastructure ↵Arnaldo Carvalho de Melo
in place I.e. look for "syscalls_sys_enter" and "syscalls_sys_exit" BPF maps of type PROG_ARRAY and populate it with the handlers as specified per syscall, for now only 'open' is wiring it to something, in time all syscalls that need to copy arguments entering a syscall or returning from one will set these to the right handlers, reusing when possible pre-existing ones. Next step is to use bpf_tail_call() into that. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-t0p4u43i9vbpzs1xtowna3gb@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-29perf trace: Allow specifying the bpf prog to augment specific syscallsArnaldo Carvalho de Melo
This is a step in the direction of being able to use a BPF_MAP_TYPE_PROG_ARRAY to handle syscalls that need to copy pointer payloads in addition to the raw tracepoint syscall args. There is a first example in tools/perf/examples/bpf/augmented_raw_syscalls.c for the 'open' syscall. Next step is to introduce the prog array map and use this 'open' augmenter, then use that augmenter in other syscalls that also only copy the first arg as a string, and then show how to use with a syscall that reads more than one filename, like 'rename', etc. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-pys4v57x5qqrybb4cery2mc8@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-29perf trace: Add BPF handler for unaugmented syscallsArnaldo Carvalho de Melo
Will be used to assign to syscalls that don't need augmentation, i.e. those with just integer args. All syscalls will be in a BPF_MAP_TYPE_PROG_ARRAY, and the bpf_tail_call() keyed by the syscall id will either find nothing in place, which means the syscall is being filtered, or a function that will either add things like filenames to the ring buffer, right after the raw syscall args, or be this unaugmented handler that will just return 1, meaning don't filter the original raw_syscalls:sys_{enter,exit} tracepoint. For now it is not really being used, this is just leg work to break the patch into smaller pieces. It introduces a trace__find_bpf_program_by_title() helper that in turn uses libbpf's bpf_object__find_program_by_title() on the BPF object with the __augmented_syscalls__ map. "title" is how libbpf calls the SEC() argument for functions, i.e. the ELF section that follows a convention to specify what BPF program (a function with this SEC() marking) should be connected to which tracepoint, kprobes, etc. In perf anything that is of the form SEC("sys:event_name") will be connected to that tracepoint by perf's BPF loader. In this case its something that will be bpf_tail_call()ed from either the "raw_syscalls:sys_enter" or "raw_syscall:sys_exit" tracepoints, so its named "!raw_syscalls:unaugmented" to convey that idea, i.e. its not going to be directly attached to a tracepoint, thus it starts with a "!". Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-meucpjx2u0slpkayx56lxqq6@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-29perf trace: Order -e syscalls tableArnaldo Carvalho de Melo
The ev_qualifier is an array with the syscall ids passed via -e on the command line, sort it as we'll search it when setting up the BPF_MAP_TYPE_PROG_ARRAY. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-c8hprylp3ai6e0z9burn2r3s@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-29perf trace: Look up maps just on the __augmented_syscalls__ BPF objectArnaldo Carvalho de Melo
We can conceivably have multiple BPF object files for other purposes, so better look just on the BPF object containing the __augmented_syscalls__ map for all things augmented_syscalls related. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-3jt8knkuae9lt705r1lns202@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-29perf trace: Add pointer to BPF object containing __augmented_syscalls__Arnaldo Carvalho de Melo
So that we can use it when looking for other components of that object file, such as other programs to add to the BPF_MAP_TYPE_PROG_ARRAY and use with bpf_tail_call(). Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-1ibmz7ouv6llqxajy7m8igtd@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-29perf evsel: Store backpointer to attached bpf_objectArnaldo Carvalho de Melo
We may want to get to this bpf_object, to search for other BPF programs, etc. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-3y8hrb6lszjfi23vjlic3cib@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-29perf bpf: Do not attach a BPF prog to a tracepoint if its name starts with !Arnaldo Carvalho de Melo
With BPF_MAP_TYPE_PROG_ARRAY + bpf_tail_call() we want to have BPF programs, i.e. functions in a object file that perf's BPF loader shouldn't try to attach to anything, i.e. "!syscalls:sys_enter_open" should just stay there, not be attached to a tracepoint with that name, it'll be used by, for instance, 'perf trace' to associate with syscalls that copy, in addition to the syscall raw args, a filename pointed by the first arg, i.e. multiple syscalls that need copying the same pointer arg in the same way, as a filename, for instance, will share the same BPF program/function. Right now when perf's BPF loader sees a function with a name "sys:name" it'll look for a tracepoint and will associate that BPF program with it, say: SEC("raw_syscalls:sys_enter") int sys_enter(struct syscall_enter_args *args) { //SNIP } Will crate a perf_evsel tracepoint event and then associate with it that BPF program. This convention at some point will switch to the one used by the BPF loader in libbpf, but to experiment with BPF_MAP_TYPE_PROG_ARRAY in 'perf trace' lets do this, that will not require changing too much stuff. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-lk6dasjr1yf9rtvl292b2hpc@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-29perf include bpf: Add bpf_tail_call() prototypeArnaldo Carvalho de Melo
Will be used together with BPF_MAP_TYPE_PROG_ARRAY in tools/perf/examples/bpf/augmented_raw_syscalls.c. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-pd1bpy8i31nta6jqwdex871g@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-29perf tools: Fix perf.data documentation units for memory sizeVince Weaver
The perf.data-file-format documentation incorrectly says the HEADER_TOTAL_MEM results are in bytes. The results are in kilobytes (perf reads the value from /proc/meminfo) Signed-off-by: Vince Weaver <vincent.weaver@maine.edu> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/alpine.DEB.2.21.1907251155500.22624@macbook-air Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-29perf header: Fix use of unitialized value warningNumfor Mbiziwo-Tiapo
When building our local version of perf with MSAN (Memory Sanitizer) and running the perf record command, MSAN throws a use of uninitialized value warning in "tools/perf/util/util.c:333:6". This warning stems from the "buf" variable being passed into "write". It originated as the variable "ev" with the type union perf_event* defined in the "perf_event__synthesize_attr" function in "tools/perf/util/header.c". In the "perf_event__synthesize_attr" function they allocate space with a malloc call using ev, then go on to only assign some of the member variables before passing "ev" on as a parameter to the "process" function therefore "ev" contains uninitialized memory. Changing the malloc call to zalloc to initialize all the members of "ev" which gets rid of the warning. To reproduce this warning, build perf by running: make -C tools/perf CLANG=1 CC=clang EXTRA_CFLAGS="-fsanitize=memory\ -fsanitize-memory-track-origins" (Additionally, llvm might have to be installed and clang might have to be specified as the compiler - export CC=/usr/bin/clang) then running: tools/perf/perf record -o - ls / | tools/perf/perf --no-pager annotate\ -i - --stdio Please see the cover letter for why false positive warnings may be generated. Signed-off-by: Numfor Mbiziwo-Tiapo <nums@google.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Mark Drayton <mbd@fb.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Song Liu <songliubraving@fb.com> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/20190724234500.253358-2-nums@google.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-29perf header: Fix divide by zero error if f_header.attr_size==0Vince Weaver
So I have been having lots of trouble with hand-crafted perf.data files causing segfaults and the like, so I have started fuzzing the perf tool. First issue found: If f_header.attr_size is 0 in the perf.data file, then perf will crash with a divide-by-zero error. Committer note: Added a pr_err() to tell the user why the command failed. Signed-off-by: Vince Weaver <vincent.weaver@maine.edu> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/alpine.DEB.2.21.1907231100440.14532@macbook-air Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-29tools perf beauty: Fix usbdevfs_ioctl table generator to handle _IOC()Arnaldo Carvalho de Melo
In addition to _IOW() and _IOR(), to handle this case: #define USBDEVFS_CONNINFO_EX(len) _IOC(_IOC_READ, 'U', 32, len) That will happen in the next sync of this header file. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-3br5e4t64e4lp0goo84che3s@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-26tools include UAPI: Sync x86's syscalls_64.tbl and generic unistd.h to pick ↵Arnaldo Carvalho de Melo
up clone3 and pidfd_open 05a70a8ec287 ("unistd: protect clone3 via __ARCH_WANT_SYS_CLONE3") 8f3220a80654 ("arch: wire-up clone3() syscall") 7615d9e1780e ("arch: wire-up pidfd_open()") Silencing the following tools/perf build warnings Warning: Kernel ABI header at 'tools/include/uapi/asm-generic/unistd.h' differs from latest version at 'include/uapi/asm-generic/unistd.h' diff -u tools/include/uapi/asm-generic/unistd.h include/uapi/asm-generic/unistd.h Warning: Kernel ABI header at 'tools/perf/arch/x86/entry/syscalls/syscall_64.tbl' differs from latest version at 'arch/x86/entry/syscalls/syscall_64.tbl' diff -u tools/perf/arch/x86/entry/syscalls/syscall_64.tbl arch/x86/entry/syscalls/syscall_64.tbl Now 'perf trace -e pidfd*,clone*' will trace those syscalls as well as the others with those prefixes. $ diff -u /tmp/build/perf/arch/x86/include/generated/asm/syscalls_64.c.before /tmp/build/perf/arch/x86/include/generated/asm/syscalls_64.c --- /tmp/build/perf/arch/x86/include/generated/asm/syscalls_64.c.before 2019-07-26 12:24:55.020944201 -0300 +++ /tmp/build/perf/arch/x86/include/generated/asm/syscalls_64.c 2019-07-26 12:25:03.919047217 -0300 @@ -344,5 +344,7 @@ [431] = "fsconfig", [432] = "fsmount", [433] = "fspick", + [434] = "pidfd_open", + [435] = "clone3", }; -#define SYSCALLTBL_x86_64_MAX_ID 433 +#define SYSCALLTBL_x86_64_MAX_ID 435 $ Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Brendan Gregg <brendan.d.gregg@gmail.com> Cc: Christian Brauner <christian@brauner.io> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-0isnnqxtr1ihz6p8wzjiy47d@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-23Merge tag 'perf-urgent-for-mingo-5.3-20190723' of ↵Ingo Molnar
git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent Pull perf/urgent fixes from Arnaldo Carvalho de Melo: perf.data: Alexey Budankov: - Fix loading of compressed data split across adjacent records Jiri Olsa: - Fix buffer size setting for processing CPU topology perf.data header. perf stat: Jiri Olsa: - Fix segfault for event group in repeat mode Cong Wang: - Always separate "stalled cycles per insn" line, it was being appended to the "instructions" line. perf script: Andi Kleen: - Fix --max-blocks man page description. - Improve man page description of metrics. - Fix off by one in brstackinsn IPC computation. perf probe: Arnaldo Carvalho de Melo: - Avoid calling freeing routine multiple times for same pointer. perf build: - Do not use -Wshadow on gcc < 4.8, avoiding too strict warnings treated as errors, breaking the build. Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ingo Molnar <mingo@kernel.org>
2019-07-23perf probe: Avoid calling freeing routine multiple times for same pointerArnaldo Carvalho de Melo
When perf_add_probe_events() we call cleanup_perf_probe_events() for the pev pointer it receives, then, as part of handling this failure the main 'perf probe' goes on and calls cleanup_params() and that will again call cleanup_perf_probe_events()for the same pointer, so just set nevents to zero when handling the failure of perf_add_probe_events() to avoid the double free. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-x8qgma4g813z96dvtw9w219q@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-23perf probe: Set pev->nargs to zero after freeing pev->args entriesArnaldo Carvalho de Melo
So that, when perf_add_probe_events() fails, like in: # perf probe icmp_rcv:64 "type=icmph->type" Failed to find 'icmph' in this function. Error: Failed to add events. Segmentation fault (core dumped) # We don't segfault. clear_perf_probe_event() was zeroing the whole pev, and since the switch to zfree() for the members in the pev, that memset() was removed, which left nargs with its original value, in the above case 1. With the memset the same pev could be passed to clear_perf_probe_event() multiple times, since all it would have would be zeroes, and free() accepts zero, the loop would not happen and we would just memset it again to zeroes. Without it we got that segfault, so zero nargs to keep it like it was, next cset will avoid calling clear_perf_probe_event() for the same pevs in case of failure. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Fixes: d8f9da240495 ("perf tools: Use zfree() where applicable") Link: https://lkml.kernel.org/n/tip-802f2jypnwqsvyavvivs8464@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-23perf session: Fix loading of compressed data split across adjacent recordsAlexey Budankov
Fix decompression failure found during the loading of compressed trace collected on larger scale systems (>48 cores). The error happened due to lack of decompression space for a mmaped buffer data chunk split across adjacent PERF_RECORD_COMPRESSED records. $ perf report -i bt.16384.data --stats failed to decompress (B): 63869 -> 0 : Destination buffer is too small user stack dump failure Can't parse sample, err = -14 0x2637e436 [0x4080]: failed to process type: 9 Error: failed to process sample $ perf test 71 71: Zstd perf.data compression/decompression : Ok Signed-off-by: Alexey Budankov <alexey.budankov@linux.intel.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/4d839e1b-9c48-89c4-9702-a12217420611@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-23perf stat: Always separate stalled cycles per insnCong Wang
The "stalled cycles per insn" is appended to "instructions" when the CPU has this hardware counter directly. We should always make it a separate line, which also aligns to the output when we hit the "if (total && avg)" branch. Before: $ sudo perf stat --all-cpus --field-separator , --log-fd 1 -einstructions,cycles -- sleep 1 4565048704,,instructions,64114578096,100.00,1.34,insn per cycle,, 3396325133,,cycles,64146628546,100.00,, After: $ sudo ./tools/perf/perf stat --all-cpus --field-separator , --log-fd 1 -einstructions,cycles -- sleep 1 6721924,,instructions,24026790339,100.00,0.22,insn per cycle ,,,,,0.00,stalled cycles per insn 30939953,,cycles,24025512526,100.00,, Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Andi Kleen <ak@linux.intel.com> Link: http://lkml.kernel.org/r/20190517221039.8975-1-xiyou.wangcong@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-23perf stat: Fix segfault for event group in repeat modeJiri Olsa
Numfor Mbiziwo-Tiapo reported segfault on stat of event group in repeat mode: # perf stat -e '{cycles,instructions}' -r 10 ls It's caused by memory corruption due to not cleaned evsel's id array and index, which needs to be rebuilt in every stat iteration. Currently the ids index grows, while the array (which is also not freed) has the same size. Fixing this by releasing id array and zeroing ids index in perf_evsel__close function. We also need to keep the evsel_list alive for stat record (which is disabled in repeat mode). Reported-by: Numfor Mbiziwo-Tiapo <nums@google.com> Signed-off-by: Jiri Olsa <jolsa@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Mark Drayton <mbd@fb.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Song Liu <songliubraving@fb.com> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/20190715142121.GC6032@krava Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-23perf tools: Fix proper buffer size for feature processingJiri Olsa
After Song Liu's segfault fix for pipe mode, Arnaldo reported following error: # perf record -o - | perf script 0x514 [0x1ac]: failed to process type: 80 It's caused by wrong buffer size setup in feature processing, which makes cpu topology feature fail, because it's using buffer size to recognize its header version. Reported-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Jiri Olsa <jolsa@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: David Carrillo-Cisneros <davidcc@google.com> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Song Liu <songliubraving@fb.com> Fixes: e9def1b2e74e ("perf tools: Add feature header record to pipe-mode") Link: http://lkml.kernel.org/r/20190715140426.32509-1-jolsa@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-23perf script: Fix off by one in brstackinsn IPC computationAndi Kleen
When we hit the end of a program block, need to count the last instruction too for the IPC computation. This caused large errors for small blocks. % perf script -b ls / > /dev/null Before: % perf script -F +brstackinsn --xed ... 00007f94c9ac70d8 jz 0x7f94c9ac70e3 # PRED 3 cycles [36] 4.33 IPC 00007f94c9ac70e3 testb $0x20, 0x31d(%rbx) 00007f94c9ac70ea jnz 0x7f94c9ac70b0 00007f94c9ac70ec testb $0x8, 0x205ad(%rip) 00007f94c9ac70f3 jz 0x7f94c9ac6ff0 # PRED 1 cycles [37] 3.00 IPC After: % perf script -F +brstackinsn --xed ... 00007f94c9ac70d8 jz 0x7f94c9ac70e3 # PRED 3 cycles [15] 4.67 IPC 00007f94c9ac70e3 testb $0x20, 0x31d(%rbx) 00007f94c9ac70ea jnz 0x7f94c9ac70b0 00007f94c9ac70ec testb $0x8, 0x205ad(%rip) 00007f94c9ac70f3 jz 0x7f94c9ac6ff0 # PRED 1 cycles [16] 4.00 IPC Suggested-by: Denis Bakhvalov <denis.bakhvalov@intel.com> Signed-off-by: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20190711181922.18765-2-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-23perf script: Improve man page description of metricsAndi Kleen
Clarify that a metric is based on events, not referring to itself. Also some improvements with the sentences. Signed-off-by: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20190711181922.18765-3-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-23perf script: Fix --max-blocks man page descriptionAndi Kleen
The --max-blocks description was using the old name brstackasm. Use brstackinsn instead. Signed-off-by: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/20190711181922.18765-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-20Merge branch 'perf-urgent-for-linus' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip Pull perf tooling updates from Thomas Gleixner: "A set of perf improvements and fixes: perf db-export: - Improvements in how COMM details are exported to databases for post processing and use in the sql-viewer.py UI. - Export switch events to the database. BPF: - Bump rlimit(MEMLOCK) for 'perf test bpf' and 'perf trace', just like selftests/bpf/bpf_rlimit.h do, which makes errors due to exhaustion of this limit, which are kinda cryptic (EPERM sometimes) less frequent. perf version: - Fix segfault due to missing OPT_END(), noticed on PowerPC. perf vendor events: - Add JSON files for IBM s/390 machine type 8561. perf cs-etm (ARM): - Fix two cases of error returns not bing done properly: Invalid ERR_PTR() use and loss of propagation error codes" * 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (28 commits) perf version: Fix segfault due to missing OPT_END() perf vendor events s390: Add JSON files for machine type 8561 perf cs-etm: Return errcode in cs_etm__process_auxtrace_info() perf cs-etm: Remove errnoeous ERR_PTR() usage in cs_etm__process_auxtrace_info perf scripts python: export-to-postgresql.py: Export switch events perf scripts python: export-to-sqlite.py: Export switch events perf db-export: Export switch events perf db-export: Factor out db_export__threads() perf script: Add scripting operation process_switch() perf scripts python: exported-sql-viewer.py: Use new 'has_calls' column perf scripts python: exported-sql-viewer.py: Remove redundant semi-colons perf scripts python: export-to-postgresql.py: Add has_calls column to comms table perf scripts python: export-to-sqlite.py: Add has_calls column to comms table perf db-export: Also export thread's current comm perf db-export: Factor out db_export__comm() perf scripts python: export-to-postgresql.py: Export comm details perf scripts python: export-to-sqlite.py: Export comm details perf db-export: Export comm details perf db-export: Fix a white space issue in db_export__sample() perf db-export: Move export__comm_thread into db_export__sample() ...
2019-07-18Merge tag 'trace-v5.3' of ↵Linus Torvalds
git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace Pull tracing updates from Steven Rostedt: "The main changes in this release include: - Add user space specific memory reading for kprobes - Allow kprobes to be executed earlier in boot The rest are mostly just various clean ups and small fixes" * tag 'trace-v5.3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace: (33 commits) tracing: Make trace_get_fields() global tracing: Let filter_assign_type() detect FILTER_PTR_STRING tracing: Pass type into tracing_generic_entry_update() ftrace/selftest: Test if set_event/ftrace_pid exists before writing ftrace/selftests: Return the skip code when tracing directory not configured in kernel tracing/kprobe: Check registered state using kprobe tracing/probe: Add trace_event_call accesses APIs tracing/probe: Add probe event name and group name accesses APIs tracing/probe: Add trace flag access APIs for trace_probe tracing/probe: Add trace_event_file access APIs for trace_probe tracing/probe: Add trace_event_call register API for trace_probe tracing/probe: Add trace_probe init and free functions tracing/uprobe: Set print format when parsing command tracing/kprobe: Set print format right after parsed command kprobes: Fix to init kprobes in subsys_initcall tracepoint: Use struct_size() in kmalloc() ring-buffer: Remove HAVE_64BIT_ALIGNED_ACCESS ftrace: Enable trampoline when rec count returns back to one tracing/kprobe: Do not run kprobe boot tests if kprobe_event is on cmdline tracing: Make a separate config for trace event self tests ...
2019-07-15perf version: Fix segfault due to missing OPT_END()Ravi Bangoria
'perf version' on powerpc segfaults when used with non-supported option: # perf version -a Segmentation fault (core dumped) Fix this. Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com> Reviewed-by: Kamalesh Babulal <kamalesh@linux.vnet.ibm.com> Tested-by: Mamatha Inamdar <mamatha4@linux.vnet.ibm.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Kamalesh Babulal <kamalesh@linux.vnet.ibm.com> Link: http://lkml.kernel.org/r/20190611030109.20228-1-ravi.bangoria@linux.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-12perf vendor events s390: Add JSON files for machine type 8561Thomas Richter
Add CPU measurement counter facility event description files (JSON) for IBM machine types 8561 and 8562. Signed-off-by: Thomas Richter <tmricht@linux.ibm.com> Reviewed-by: Vasily Gorbik <gor@linux.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Hendrik Brueckner <brueckner@linux.ibm.com> Link: http://lkml.kernel.org/r/20190712123113.100882-1-tmricht@linux.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-11perf cs-etm: Return errcode in cs_etm__process_auxtrace_info()YueHaibing
The 'err' variable is set in the error path, but it's not returned to callers. Don't always return -EINVAL, return err. Signed-off-by: YueHaibing <yuehaibing@huawei.com> Reviewed-by: Mathieu Poirier <mathieu.poirier@linaro.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: linux-arm-kernel@lists.infradead.org Fixes: cd8bfd8c973e ("perf tools: Add processing of coresight metadata") Link: http://lkml.kernel.org/r/20190321023122.21332-3-yuehaibing@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-11perf cs-etm: Remove errnoeous ERR_PTR() usage in cs_etm__process_auxtrace_infoYueHaibing
intlist__findnew() doesn't uses ERR_PTR() as a return mechanism so its callers shouldn't try to extract the error using PTR_ERR( ret) from intlist__findnew(), make cs_etm__process_auxtrace_info return -ENOMEM instead. Signed-off-by: YueHaibing <yuehaibing@huawei.com> Reviewed-by: Mathieu Poirier <mathieu.poirier@linaro.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Suzuki Poulouse <suzuki.poulose@arm.com> Cc: linux-arm-kernel@lists.infradead.org Fixes: cd8bfd8c973e ("perf tools: Add processing of coresight metadata") Link: http://lkml.kernel.org/r/20190321023122.21332-2-yuehaibing@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf scripts python: export-to-postgresql.py: Export switch eventsAdrian Hunter
Export switch events to a new table 'context_switches' and create a view 'context_switches_view'. The table and view will show automatically in the exported-sql-viewer.py script. If the table ends up empty, then it and the view are dropped. Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-22-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf scripts python: export-to-sqlite.py: Export switch eventsAdrian Hunter
Export switch events to a new table 'context_switches' and create a view 'context_switches_view'. The table and view will show automatically in the exported-sql-viewer.py script. If the table ends up empty, then it and the view are dropped. Committer testing: Use the exported-sql-viewer.py and look at "Tables" -> "context_switches": id machine_id time cpu thread_out_id comm_out_id thread_in_id comm_in_id flags 1 1 187836111885918 7 1 1 2 2 3 2 1 187836111889369 7 1 1 2 2 0 3 1 187836112464618 7 2 3 1 1 1 4 1 187836112465511 7 2 3 1 1 0 Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-21-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf db-export: Export switch eventsAdrian Hunter
Export details of switch events including the threads and their current comms. Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-20-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf db-export: Factor out db_export__threads()Adrian Hunter
In preparation for exporting switch events, factor out db_export__threads(). Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-19-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf script: Add scripting operation process_switch()Adrian Hunter
Add scripting operation process_switch() to process switch events. Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-18-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf scripts python: exported-sql-viewer.py: Use new 'has_calls' columnAdrian Hunter
If the new 'has_calls' column is present, use it with the call graph and call tree to select only comms that have calls. Committer testing: Just started the exported-sql-view.py and accessed all the reports, no backtraces. Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-17-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf scripts python: exported-sql-viewer.py: Remove redundant semi-colonsAdrian Hunter
Remove redundant semi-colons added inadvertently. Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-16-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf scripts python: export-to-postgresql.py: Add has_calls column to comms ↵Adrian Hunter
table Now that a thread's current comm is exported, it shows up in the call graph and call tree even if it has no calls. That can happen because the calls are recorded against the main thread's initial comm. Add a table column to make it easy for the exported-sql-viewer.py script to select only comms with calls. Committer testing: $ rm -f simple-retpoline.db $ sudo ~acme/bin/perf script -i simple-retpoline.perf.data --itrace=be -s ~/libexec/perf-core/scripts/python/export-to-sqlite.py simple-retpoline.db branches calls 2019-07-10 12:25:33.200529 Creating database ... 2019-07-10 12:25:33.211548 Writing records... 2019-07-10 12:25:33.549630 Adding indexes 2019-07-10 12:25:33.560715 Dropping unused tables 2019-07-10 12:25:33.580201 Done $ sha256sum tools/perf/scripts/python/export-to-sqlite.py ~/libexec/perf-core/scripts/python/export-to-sqlite.py 2922b642c392004dffa1d8789296478c85904623f5895bcb9b6cbf33e3ca999f tools/perf/scripts/python/export-to-sqlite.py 2922b642c392004dffa1d8789296478c85904623f5895bcb9b6cbf33e3ca999f /home/acme/libexec/perf-core/scripts/python/export-to-sqlite.py $ $ sqlite3 simple-retpoline.db SQLite version 3.26.0 2018-12-01 12:34:55 Enter ".help" for usage hints. sqlite> .schema comms CREATE TABLE comms (id integer NOT NULL PRIMARY KEY,comm varchar(16),c_thread_id bigint,c_time bigint,exec_flag boolean, has_calls boolean); sqlite> select id,has_calls from comms; 0|1 1|1 sqlite> select distinct comm_id from calls; 0 1 sqlite> Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-15-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf scripts python: export-to-sqlite.py: Add has_calls column to comms tableAdrian Hunter
Now that a thread's current comm is exported, it shows up in the call graph and call tree even if it has no calls. That can happen because the calls are recorded against the main thread's initial comm. Add a table column to make it easy for the exported-sql-viewer.py script to select only comms with calls. Committer notes: Running the export-to-sqlite.py worked without warnings and using the exported-sql-viewer.py worked as before. Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-14-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf db-export: Also export thread's current commAdrian Hunter
Currently, the initial comm of the main thread is exported. Export also a thread's current comm. That better supports the tracing of multi-threaded applications that set different comms for different threads to make it easier to distinguish them. Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-13-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf db-export: Factor out db_export__comm()Adrian Hunter
In preparation for exporting the current comm for a thread, factor out db_export__comm(). Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-12-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf scripts python: export-to-postgresql.py: Export comm detailsAdrian Hunter
Add table columns for thread id, comm start time and exec flag. Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-11-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf scripts python: export-to-sqlite.py: Export comm detailsAdrian Hunter
Add table columns for thread id, comm start time and exec flag. Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-10-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf db-export: Export comm detailsAdrian Hunter
In preparation for exporting the current comm for a thread, export comm thread id, start time and exec flag. Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-9-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf db-export: Fix a white space issue in db_export__sample()Adrian Hunter
Fix a white space issue in db_export__sample() Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-8-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf db-export: Move export__comm_thread into db_export__sample()Adrian Hunter
Move call to db_export__comm_thread() from db_export__thread() into db_export__sample() because it makes the code easier to understand, and add explanatory comments. Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-7-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf db-export: Export comm before exporting threadAdrian Hunter
Export comm before exporting the non-main thread because db_export__thread() also exports the comm_thread. Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-6-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf db-export: Export main_thread in db_export__sample()Adrian Hunter
Export main_thread in db_export__sample() because it makes the code easier to understand, and prepares db_export__thread() for further simplification. Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-5-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-07-10perf db-export: Pass main_thread to db_export__thread()Adrian Hunter
Calls to db_export__thread() already have main_thread so there is no reason to get it again, instead pass it as a parameter. Note that one difference in this approach is that the main thread is not created if it does not exist. It is better if it is not created because: - If main_thread is being traced it will have been created already. - If it is not being traced, there will be no other information about it, and it will never get deleted because there will be no EXIT event. Signed-off-by: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/20190710085810.1650-4-adrian.hunter@intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>