- 19 Jul, 2017 40 commits
-
-
Jin Yao authored
Show branch type in callchain entry. The branch type is printed with other LBR information (such as cycles/abort/...). For example: perf record -g -j any,save_type perf report --branch-history --stdio --no-children 38.50% div.c:45 [.] main div | ---main div.c:42 (RET CROSS_2M cycles:2) compute_flag div.c:28 (cycles:2) compute_flag div.c:27 (RET CROSS_2M cycles:1) rand rand.c:28 (cycles:1) rand rand.c:28 (RET CROSS_2M cycles:1) __random random.c:298 (cycles:1) __random random.c:297 (COND_BWD CROSS_2M cycles:1) __random random.c:295 (cycles:1) __random random.c:295 (COND_BWD CROSS_2M cycles:1) __random random.c:295 (cycles:1) __random random.c:295 (RET CROSS_2M cycles:9) Change log v6: Remove the branch_type_str() since it's moved to branch.c. v5: Rewrite the branch info print code in util/callchain.c. v4: Comparing to previous version, the major changes are: Signed-off-by:Yao Jin <yao.jin@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: Kan Liang <kan.liang@intel.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1500379995-6449-8-git-send-email-yao.jin@linux.intel.comSigned-off-by:
Arnaldo Carvalho de Melo <acme@redhat.com>
-
Jin Yao authored
Show the branch type statistics at the end of perf report --stdio. For example: perf report --stdio COND_FWD: 28.5% COND_BWD: 9.4% CROSS_4K: 0.7% CROSS_2M: 14.1% COND: 37.9% UNCOND: 0.2% IND: 6.7% CALL: 26.5% RET: 28.7% SYSRET: 0.0% The branch types are: COND_FWD: conditional forward COND_BWD: conditional backward COND: conditional branch UNCOND: unconditional branch IND: indirect CALL: function call IND_CALL: indirect function call RET: function return SYSCALL: syscall SYSRET: syscall return COND_CALL: conditional function call COND_RET: conditional function return CROSS_4K and CROSS_2M: They are the metrics checking for branches cross 4K or 2MB pages. It's an approximate computing. We don't know if the area is 4K or 2MB, so always compute both. To make the output simple, if a branch crosses 2M area, CROSS_4K will not be incremented. Change log v7: Since the common branch type definitions are changed, some tags/strings are updated accordingly. v6: Remove branch_type_stat_display() since it's moved to branch.c. v5: Remove the unnecessary sort__mode checking in hist_iter__branch_callback(). v4: Comparing to previous version, the major changes are: Add the computing of JCC forward/JCC backward and cross page checking by using the from and to addresses. Signed-off-by: -
Jin Yao authored
Create new util/branch.c and util/branch.h to contain the common branch functions. Such as: branch_type_count(): Count the numbers of branch types branch_type_name() : Return the name of branch type branch_type_stat_display(): Display branch type statistics info branch_type_str(): Construct the branch type string. The branch type is saved in branch_flags. Change log: v8: Change PERF_BR_NONE to PERF_BR_UNKNOWN. v7: Since the common branch type name is changed (e.g. JCC->COND), this patch is performed the modification accordingly. v6: Move that multiline conditional code inside {} brackets. Move branch_type_stat_display() from builtin-report.c to branch.c. Move branch_type_str() from callchain.c to branch.c. v5: It's a new patch in v5 patch series. Signed-off-by: -
Jin Yao authored
The branch info such as predicted/cycles/... are printed at the callchain entries. For example: perf report --branch-history --no-children --stdio --1.07%--main div.c:39 (predicted:52.4% cycles:1 iterations:17) main div.c:44 (predicted:52.4% cycles:1) main div.c:42 (cycles:2) compute_flag div.c:28 (cycles:2) compute_flag div.c:27 (cycles:1) rand rand.c:28 (cycles:1) rand rand.c:28 (cycles:1) __random random.c:298 (cycles:1) __random random.c:297 (cycles:1) __random random.c:295 (cycles:1) __random random.c:295 (cycles:1) __random random.c:295 (cycles:1) But the current code is difficult to maintain and extend. This patch refactors the code for easy maintenance. Change log: v6: 1. Put the multiline condition code into {} brackets in counts_str_build() 2. Keep the original display order, that is: predicted, abort, cycles, iterations v5: It's a new patch in v5 patch series. Signed-off-by: -
Jin Yao authored
The option indicates the kernel to save branch type during sampling. One example: perf record -g --branch-filter any,save_type <command> Signed-off-by:
-
Jin Yao authored
Perf already has support for disassembling the branch instruction and using the branch type for filtering. The patch just records the branch type in perf_branch_entry. Before recording, the patch converts the x86 branch type to common branch type. Change log: v10: Set the branch_map array to be static. The previous version has it on stack then makes the compiler to create it every time when the function gets called. v9: Use __ffs() to find first bit in type in common_branch_type(). It lets the code be clear. v8: Change PERF_BR_NONE to PERF_BR_UNKNOWN. v7: Just convert following x86 branch types to common branch types. X86_BR_CALL -> PERF_BR_CALL X86_BR_RET -> PERF_BR_RET X86_BR_JCC -> PERF_BR_COND X86_BR_JMP -> PERF_BR_UNCOND X86_BR_IND_CALL -> PERF_BR_IND_CALL X86_BR_ZERO_CALL -> PERF_BR_CALL X86_BR_IND_JMP -> PERF_BR_IND X86_BR_SYSCALL -> PERF_BR_SYSCALL X86_BR_SYSRET -> PERF_BR_SYSRET Others are set to PERF_BR_NONE v6: Not changed. v5: Just fix the merge error. No other update. v4: Comparing to previous version, the major changes are: 1. Uses a lookup table to convert x86 branch type to common branch type. 2. Move the JCC forward/JCC backward and cross page computing to user space. 3. Initialize branch type to 0 in intel_pmu_lbr_read_32 and intel_pmu_lbr_read_64 Signed-off-by:Peter Zijlstra <peterz@infradead.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Kan Liang <kan.liang@intel.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Link: http://lkml.kernel.org/r/1500379995-6449-3-git-send-email-yao.jin@linux.intel.comSigned-off-by:
-
Jin Yao authored
It is often useful to know the branch types while analyzing branch data. For example, a call is very different from a conditional branch. Currently we have to look it up in binary while the binary may later not be available and even the binary is available but user has to take some time. It is very useful for user to check it directly in perf report. Perf already has support for disassembling the branch instruction to get the x86 branch type. To keep consistent on kernel and userspace and make the classification more common, the patch adds the common branch type classification in perf_event.h. The patch only defines a minimum but most common set of branch types. PERF_BR_UNKNOWN : unknown PERF_BR_COND ：conditional PERF_BR_UNCOND : unconditional PERF_BR_IND : indirect PERF_BR_CALL : function call PERF_BR_IND_CALL : indirect function call PERF_BR_RET : function return PERF_BR_SYSCALL : syscall PERF_BR_SYSRET : syscall return PERF_BR_COND_CALL : conditional function call PERF_BR_COND_RET : conditional function return The patch also adds a new field type (4 bits) in perf_branch_entry to record the branch type. Since the disassembling of branch instruction needs some overhead, a new PERF_SAMPLE_BRANCH_TYPE_SAVE is introduced to indicate if it needs to disassemble the branch instruction and record the branch type. Change log: v10: Not changed. v9: Not changed. v8: Change PERF_BR_NONE to PERF_BR_UNKNOWN. No other change. v7: Just keep the most common branch types. Others are removed. v6: Not changed. v5: Not changed. The v5 patch series just change the userspace. v4: Comparing to previous version, the major changes are: 1. Remove the PERF_BR_JCC_FWD/PERF_BR_JCC_BWD, they will be computed later in userspace. 2. Remove the "cross" field in perf_branch_entry. The cross page computing will be done later in userspace. Signed-off-by:Michael Ellerman <mpe@ellerman.id.au> Acked-by:
-
David Carrillo-Cisneros authored
Add event descriptor to perf header output in pipe-mode. After this patch: $ perf record -e cycles sleep 1 | perf report --header # ======== # captured on: Mon Jun 5 22:52:13 2017 # ======== # # hostname : lphh20 # os release : 4.3.5-smp-801.43.0.0 # perf version : 4.12.rc2.g439987 # arch : x86_64 # nrcpus online : 72 # nrcpus avail : 72 # cpudesc : Intel(R) Xeon(R) CPU E5-2696 v3 @ 2.30GHz # cpuid : GenuineIntel,6,63,2 # total memory : 264134144 kB # cmdline : /root/perf record -e cycles sleep 1 # event : name = cycles, , size = 112, { sample_period, sample_freq } = 4000, sample_type = IP|TID|TIME|PERIOD, disabled = 1, inherit = 1, mmap = 1, comm = 1, freq = 1, enable_on_exec = 1, task = 1, sample_id_all = 1, exclude_guest = 1, mmap2 = 1, comm_exec = 1 # CPU_TOPOLOGY info available, use -I to display # NUMA_TOPOLOGY info available, use -I to display # pmu mappings: intel_bts = 6, cpu = 4, msr = 49, uncore_cbox_10 = 36, uncore_cbox_11 = 37, uncore_cbox_12 = 38, uncore_cbox_13 = 39, uncore_cbox_14 = 40, uncore_cbox_15 = 41, uncore_cbox_16 = 42, uncore_cbox_17 = 43, software = 1, power = 7, uncore_irp = 24, uncore_pcu = 48, tracepoint = 2, uncore_imc_0 = 16, uncore_imc_1 = 17, uncore_imc_2 = 18, uncore_imc_3 = 19, uncore_imc_4 = 20, uncore_imc_5 = 21, uncore_imc_6 = 22, uncore_imc_7 = 23, uncore_qpi_0 = 8, uncore_qpi_1 = 9, uncore_cbox_0 = 26, uncore_cbox_1 = 27, uncore_cbox_2 = 28, uncore_cbox_3 = 29, uncore_cbox_4 = 30, uncore_cbox_5 = 31, uncore_cbox_6 = 32, uncore_cbox_7 = 33, uncore_cbox_8 = 34, uncore_cbox_9 = 35, uncore_r2pcie = 13, uncore_r3qpi_0 = 10, uncore_r3qpi_1 = 11, uncore_r3qpi_2 = 12, uncore_sbox_0 = 44, uncore_sbox_1 = 45, uncore_sbox_2 = 46, uncore_sbox_3 = 47, breakpoint = 5, uncore_ha_0 = 14, uncore_ha_1 = 15, uncore_ubox = 25 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.000 MB (null) ] Prior to this patch, event was not printed. Signed-off-by:David Carrillo-Cisneros <davidcc@google.com> Acked-by:
David Ahern <dsahern@gmail.com> Acked-by:
-
David Carrillo-Cisneros authored
Add header record types to pipe-mode, reusing the functions used in file-mode and leveraging the new struct feat_fd. For alignment, check that synthesized events don't exceed pagesize. Add the perf_event__synthesize_feature event call back to process the new header records. Before this patch: $ perf record -o - -e cycles sleep 1 | perf report --stdio --header [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.000 MB - ] ... After this patch: $ perf record -o - -e cycles sleep 1 | perf report --stdio --header # ======== # captured on: Mon May 22 16:33:43 2017 # ======== # # hostname : my_hostname # os release : 4.11.0-dbx-up_perf # perf version : 4.11.rc6.g6277c80 # arch : x86_64 # nrcpus online : 72 # nrcpus avail : 72 # cpudesc : Intel(R) Xeon(R) CPU E5-2696 v3 @ 2.30GHz # cpuid : GenuineIntel,6,63,2 # total memory : 263457192 kB # cmdline : /root/perf record -o - -e cycles -c 100000 sleep 1 # HEADER_CPU_TOPOLOGY info available, use -I to display # HEADER_NUMA_TOPOLOGY info available, use -I to display # pmu mappings: intel_bts = 6, uncore_imc_4 = 22, uncore_sbox_1 = 47, uncore_cbox_5 = 33, uncore_ha_0 = 16, uncore_cbox [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.000 MB - ] ... Support added for the subcommands: report, inject, annotate and script. Signed-off-by:
-
David Carrillo-Cisneros authored
Add show_feat_hdr to control level of printed information of feature headers. Signed-off-by:
-
David Carrillo-Cisneros authored
There are three FEAT_OP* macros: - FEAT_OPA: for features without process record. - FEAT_OPP: for features with process record. - FEAT_OPF: like FEAT_OPP but to show only if show_full_info flags is set. To add pipe-mode headers we need yet another variation of the macros (one to specify whether a feature generates an auxiliar record). Instead, we redefine macros so that: - show_full_info is specified as an argument (to remove the FEAT_OPF variation) and, - it always sets "process" handler (to remove the FEAT_OPA variation). Individual process handlers can be NULLed individually. This allows to define two variations only: - FEAT_OPR: synthesizes auxiliar event record. - FEAT_OPN: doesn't synthesize an auxiliar event record. Signed-off-by: -
David Carrillo-Cisneros authored
Extend struct feat_fd to use a temporal buffer in pipe-mode, instead of perf.data's file descriptor. The header features build_id and aux_trace already have logic to print in file-mode that heavily rely on lseek the file. For now, leave such features inactive in pipe-mode and print a warning if their functions are called in pipe-mode. Signed-off-by:
-
David Carrillo-Cisneros authored
In pipe-mode, we will operate over a buffer instead of a file descriptor but write_pmu_mappings uses lseek to move over the perf.data file. Refactor write_pmu_mappings to avoid the usage of lseek and allow reusing the same logic in pipe-mode (next patch). Signed-off-by:
-
David Carrillo-Cisneros authored
As preparation for using header records in-pipe mode, replace int fd with struct feat_fd ff in read functions for all header record types. This patch does not change behavior. Signed-off-by:
-
David Carrillo-Cisneros authored
struct perf_file_section is used in process_##_feat as container for size and offset in the file descriptor. These attributes are meaninful in pipe-mode but struct perf_file_section is not. Add offset and size variables to struct feat_fd to store perf_file_section's values in file-mode. Later on, the same variables can be reused for pipe-mode. This patch does not change behavior. Signed-off-by:
-
David Carrillo-Cisneros authored
As preparation for using header records in pipe-mode, replace int fd with struct feat_fd ff in process functions for all header record types. This patch does not change behavior. Signed-off-by:
-
David Carrillo-Cisneros authored
As preparation for using header records in pipe mode, replace int fd with struct feat_fd ff in print functions for all header record types. Signed-off-by:
-
David Carrillo-Cisneros authored
Introduce struct feat_fd. This patch uses it as a wrapper around fd in write_* functions for feature headers. Next patches will extend its functionality to other feature header functions. This patch does not change behavior. Signed-off-by:
-
David Carrillo-Cisneros authored
Now that writen takes a const buffer, use it in do_write instead of duplicating its functionality. Export do_write to use it consistently in header.c and build_id.c . Signed-off-by:
-
David Carrillo-Cisneros authored
Make buf in helper function "writen" constant to simplify the life of its callers. This requires to hack a cast of buf prior to passing it to "ion" which is simpler than the alternative of reworking the "ion" function to provide a read and a write paths, the latter with constant buf argument. Signed-off-by:
-
David Carrillo-Cisneros authored
Do not proceed if write_padded() error failed. Also, add comments to remind that the return value of write_* functions in util/header.c is an errno code and not the number of bytes written. Signed-off-by:
-
David Carrillo-Cisneros authored
Simplify code by adding a macro to handle the common case of processing header features that are a simple string. Signed-off-by:
-
David Carrillo-Cisneros authored
Most callers of readn() in perf header read either a 32 or a 64 bits number, error check it and swap it, if necessary. Create do_read_u32 and do_read_u64 to simplify these use cases. Signed-off-by:
-
Jin Yao authored
Currently perf supports a mode to query inline stack. It works well for finding user space inline functions but it doesn't work for kernel ones, due to some unnecessary check. This patch removes these unnecessary checks. Now kernel inline functions can be reported. For example: perf report --inline -g func --stdio |--46.19%--do_huge_pmd_anonymous_page | do_huge_pmd_anonymous_page (inline) | __do_huge_pmd_anonymous_page (inline) | __SetPageUptodate (inline) | __set_bit (inline) The result is compared with the output of addr2line. They match. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
Removing syscall_fmt::err_msg and instead always formatting negative returns as errno values. With this we can remove a lot of entries that have no special handling besides the ones we can do by looking at the tracefs format files, i.e. the types for the fields (e.g. pid_t), well known names (e.g. fd). Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/n/tip-rg9u7a3qqdnzo37d212vnz2o@git.kernel.orgSigned-off-by:
-
Arnaldo Carvalho de Melo authored
Before: 77059.513 ( 0.005 ms): bash/6649 fcntl(fd: 1</dev/pts/12>, cmd: DUPFD, arg: 10) = 10</dev/pts/12> After: 77059.513 ( 0.005 ms): bash/6649 fcntl(fd: 1</dev/pts/12>, cmd: DUPFD, arg: 10</dev/pts/12>) = 10</dev/pts/12> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/n/tip-0k8iszng0slcuw0rc6xq1x5l@git.kernel.orgSigned-off-by:
-
Arnaldo Carvalho de Melo authored
Before: 77059.513 ( 0.005 ms): bash/6649 fcntl(fd: 1</dev/pts/12>, arg: 10) = 10</dev/pts/12> After: 77059.513 ( 0.005 ms): bash/6649 fcntl(fd: 1</dev/pts/12>, cmd: DUPFD, arg: 10) = 10</dev/pts/12> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/n/tip-woois88uwcr4xu38xx1ihiwo@git.kernel.orgSigned-off-by:
-
Arnaldo Carvalho de Melo authored
The 'perf trace' tool is suppressing args set to zero, with the exception of string tables (strarrays), which are kinda like enums, i.e. we have maps to go from numbers to strings. But the 'cmd' fcntl arg requires more specialized treatment, as its value will regulate if the next fcntl syscall arg, 'arg', should be ignored (not used) and also how to format the syscall return (fd, file flags, etc), so add a 'show_zero" bool to struct syscall_arg_fmt, to regulate this more explicitely. Will be used in a following patch with fcntl, here is just the mechanism. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/n/tip-all738jctxets8ffyizp5lzo@git.kernel.orgSigned-off-by:
-
Arnaldo Carvalho de Melo authored
Instead of having syscall_fmt.{arg_scnprintf,arg_parm}, introduce struct syscall_arg_fmt and have these two, paving the way for more state to change the formatting algorithms. For instance, in the 'fcntl' 'cmd' case it is better not to suppress it when being zero, showing instead its name "DUPFD". We had that in an ad-hoc way just for strarrays, but with more involved cases like fcntl, that can't be done with just a strarray, we'll need a ".show_zero" arg in the 'cmd' syscall_arg_fmt. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/n/tip-ch06o2j72zbjx5xww4qp67au@git.kernel.orgSigned-off-by: -
Arnaldo Carvalho de Melo authored
One more looking prettier. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/n/tip-ytr7idkese8sjtvn5g60130e@git.kernel.orgSigned-off-by:
-
Arnaldo Carvalho de Melo authored
Like will be done with fcntl(fd, F_GETLEASE, F_RDLCK|F_WRLCK|F_UNLCK) Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/n/tip-3p11bgirtntjfmbixfkz8i2m@git.kernel.orgSigned-off-by:
-
Jiri Olsa authored
Some of the stat events are quite rare to find on common machines (like front end cycles). Adding an 'optional' term to mark such events in attr tests. Event marked as optional will not fail the test case if it's not found in results. Signed-off-by:
-
Jiri Olsa authored
>From following commit: commit 4979d0c7 ("perf stat record: Add record command") we started to assign PERF_SAMPLE_IDENTIFIER to sample_type. Fixing the attr stat tests accordingly. Signed-off-by:
-
Jiri Olsa authored
We have a test to detect to highest precise possible, so test can't just predict precise_ip value. Signed-off-by:
-
Jiri Olsa authored
The final period can differ from what user specifies on command line due to the perf_event_max_sample_rate sysctl setup. Thus we can't predixt the sample_period value any more. Signed-off-by:
-
Jiri Olsa authored
The stat command creates all events disabled and enables them either manualy or via the enable_on_exec bit. Signed-off-by:
-
Jiri Olsa authored
The record command now properly returns the status of the tracee if there's any. We need to properly set the expected return value of the tracee in the attr tests. Signed-off-by:
-
Jiri Olsa authored
Following commit: commit 509051ea ("perf record: Rename --no-delay to --no-buffering") removed '-D' option and renamed --no-delay into --no-buffering. Fixing that in the attr tests. Signed-off-by:
-
Jiri Olsa authored
Following commit: commit 5c0cf224 ("perf record: Store data mmaps for dwarf unwind") have enabled address sampling for dwarf unwind, we need to reflect that in this test by adding ADDR sample_type and enabling mmap_data. Signed-off-by:
-
Jiri Olsa authored
There's an event open fallback which set exclude_kernel=1 in case use does not have enough privileges. Adding both 0|1 for this attribute, because we don't know what value it is. Signed-off-by:
-
