linux/tools/perf/util/annotate.h

424 lines
11 KiB
C
Raw Normal View History

License cleanup: add SPDX GPL-2.0 license identifier to files with no license Many source files in the tree are missing licensing information, which makes it harder for compliance tools to determine the correct license. By default all files without license information are under the default license of the kernel, which is GPL version 2. Update the files which contain no license information with the 'GPL-2.0' SPDX license identifier. The SPDX identifier is a legally binding shorthand, which can be used instead of the full boiler plate text. This patch is based on work done by Thomas Gleixner and Kate Stewart and Philippe Ombredanne. How this work was done: Patches were generated and checked against linux-4.14-rc6 for a subset of the use cases: - file had no licensing information it it. - file was a */uapi/* one with no licensing information in it, - file was a */uapi/* one with existing licensing information, Further patches will be generated in subsequent months to fix up cases where non-standard license headers were used, and references to license had to be inferred by heuristics based on keywords. The analysis to determine which SPDX License Identifier to be applied to a file was done in a spreadsheet of side by side results from of the output of two independent scanners (ScanCode & Windriver) producing SPDX tag:value files created by Philippe Ombredanne. Philippe prepared the base worksheet, and did an initial spot review of a few 1000 files. The 4.13 kernel was the starting point of the analysis with 60,537 files assessed. Kate Stewart did a file by file comparison of the scanner results in the spreadsheet to determine which SPDX license identifier(s) to be applied to the file. She confirmed any determination that was not immediately clear with lawyers working with the Linux Foundation. Criteria used to select files for SPDX license identifier tagging was: - Files considered eligible had to be source code files. - Make and config files were included as candidates if they contained >5 lines of source - File already had some variant of a license header in it (even if <5 lines). All documentation files were explicitly excluded. The following heuristics were used to determine which SPDX license identifiers to apply. - when both scanners couldn't find any license traces, file was considered to have no license information in it, and the top level COPYING file license applied. For non */uapi/* files that summary was: SPDX license identifier # files ---------------------------------------------------|------- GPL-2.0 11139 and resulted in the first patch in this series. If that file was a */uapi/* path one, it was "GPL-2.0 WITH Linux-syscall-note" otherwise it was "GPL-2.0". Results of that was: SPDX license identifier # files ---------------------------------------------------|------- GPL-2.0 WITH Linux-syscall-note 930 and resulted in the second patch in this series. - if a file had some form of licensing information in it, and was one of the */uapi/* ones, it was denoted with the Linux-syscall-note if any GPL family license was found in the file or had no licensing in it (per prior point). Results summary: SPDX license identifier # files ---------------------------------------------------|------ GPL-2.0 WITH Linux-syscall-note 270 GPL-2.0+ WITH Linux-syscall-note 169 ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) 21 ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) 17 LGPL-2.1+ WITH Linux-syscall-note 15 GPL-1.0+ WITH Linux-syscall-note 14 ((GPL-2.0+ WITH Linux-syscall-note) OR BSD-3-Clause) 5 LGPL-2.0+ WITH Linux-syscall-note 4 LGPL-2.1 WITH Linux-syscall-note 3 ((GPL-2.0 WITH Linux-syscall-note) OR MIT) 3 ((GPL-2.0 WITH Linux-syscall-note) AND MIT) 1 and that resulted in the third patch in this series. - when the two scanners agreed on the detected license(s), that became the concluded license(s). - when there was disagreement between the two scanners (one detected a license but the other didn't, or they both detected different licenses) a manual inspection of the file occurred. - In most cases a manual inspection of the information in the file resulted in a clear resolution of the license that should apply (and which scanner probably needed to revisit its heuristics). - When it was not immediately clear, the license identifier was confirmed with lawyers working with the Linux Foundation. - If there was any question as to the appropriate license identifier, the file was flagged for further research and to be revisited later in time. In total, over 70 hours of logged manual review was done on the spreadsheet to determine the SPDX license identifiers to apply to the source files by Kate, Philippe, Thomas and, in some cases, confirmation by lawyers working with the Linux Foundation. Kate also obtained a third independent scan of the 4.13 code base from FOSSology, and compared selected files where the other two scanners disagreed against that SPDX file, to see if there was new insights. The Windriver scanner is based on an older version of FOSSology in part, so they are related. Thomas did random spot checks in about 500 files from the spreadsheets for the uapi headers and agreed with SPDX license identifier in the files he inspected. For the non-uapi files Thomas did random spot checks in about 15000 files. In initial set of patches against 4.14-rc6, 3 files were found to have copy/paste license identifier errors, and have been fixed to reflect the correct identifier. Additionally Philippe spent 10 hours this week doing a detailed manual inspection and review of the 12,461 patched files from the initial patch version early this week with: - a full scancode scan run, collecting the matched texts, detected license ids and scores - reviewing anything where there was a license detected (about 500+ files) to ensure that the applied SPDX license was correct - reviewing anything where there was no detection but the patch license was not GPL-2.0 WITH Linux-syscall-note to ensure that the applied SPDX license was correct This produced a worksheet with 20 files needing minor correction. This worksheet was then exported into 3 different .csv files for the different types of files to be modified. These .csv files were then reviewed by Greg. Thomas wrote a script to parse the csv files and add the proper SPDX tag to the file, in the format that the file expected. This script was further refined by Greg based on the output to detect more types of files automatically and to distinguish between header and source .c files (which need different comment types.) Finally Greg ran the script using the .csv files to generate the patches. Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org> Reviewed-by: Philippe Ombredanne <pombredanne@nexb.com> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-11-01 22:07:57 +08:00
/* SPDX-License-Identifier: GPL-2.0 */
#ifndef __PERF_ANNOTATE_H
#define __PERF_ANNOTATE_H
#include <stdbool.h>
#include <stdint.h>
#include <stdio.h>
#include <linux/types.h>
#include <linux/list.h>
#include <linux/rbtree.h>
#include <pthread.h>
#include <asm/bug.h>
#include "symbol_conf.h"
perf diff: Report noisy for cycles diff This patch prints the stddev and hist for the cycles diff of program block. It can help us to understand if the cycles is noisy or not. This patch is inspired by Andi Kleen's patch: https://lwn.net/Articles/600471/ We create new option '--cycles-hist'. Example: perf record -b ./div perf record -b ./div perf diff -c cycles # Baseline [Program Block Range] Cycles Diff Shared Object Symbol # ........ .......................................................... .... ................. ............................ # 46.72% [div.c:40 -> div.c:40] 0 div [.] main 46.72% [div.c:42 -> div.c:44] 0 div [.] main 46.72% [div.c:42 -> div.c:39] 0 div [.] main 20.54% [random_r.c:357 -> random_r.c:394] 1 libc-2.27.so [.] __random_r 20.54% [random_r.c:357 -> random_r.c:380] 0 libc-2.27.so [.] __random_r 20.54% [random_r.c:388 -> random_r.c:388] 0 libc-2.27.so [.] __random_r 20.54% [random_r.c:388 -> random_r.c:391] 0 libc-2.27.so [.] __random_r 17.04% [random.c:288 -> random.c:291] 0 libc-2.27.so [.] __random 17.04% [random.c:291 -> random.c:291] 0 libc-2.27.so [.] __random 17.04% [random.c:293 -> random.c:293] 0 libc-2.27.so [.] __random 17.04% [random.c:295 -> random.c:295] 0 libc-2.27.so [.] __random 17.04% [random.c:295 -> random.c:295] 0 libc-2.27.so [.] __random 17.04% [random.c:298 -> random.c:298] 0 libc-2.27.so [.] __random 8.40% [div.c:22 -> div.c:25] 0 div [.] compute_flag 8.40% [div.c:27 -> div.c:28] 0 div [.] compute_flag 5.14% [rand.c:26 -> rand.c:27] 0 libc-2.27.so [.] rand 5.14% [rand.c:28 -> rand.c:28] 0 libc-2.27.so [.] rand 2.15% [rand@plt+0 -> rand@plt+0] 0 div [.] rand@plt 0.00% [kernel.kallsyms] [k] __x86_indirect_thunk_rax 0.00% [do_mmap+714 -> do_mmap+732] -10 [kernel.kallsyms] [k] do_mmap 0.00% [do_mmap+737 -> do_mmap+765] 1 [kernel.kallsyms] [k] do_mmap 0.00% [do_mmap+262 -> do_mmap+299] 0 [kernel.kallsyms] [k] do_mmap 0.00% [__x86_indirect_thunk_r15+0 -> __x86_indirect_thunk_r15+0] 7 [kernel.kallsyms] [k] __x86_indirect_thunk_r15 0.00% [native_sched_clock+0 -> native_sched_clock+119] -1 [kernel.kallsyms] [k] native_sched_clock 0.00% [native_write_msr+0 -> native_write_msr+16] -13 [kernel.kallsyms] [k] native_write_msr When we enable the option '--cycles-hist', the output is perf diff -c cycles --cycles-hist # Baseline [Program Block Range] Cycles Diff stddev/Hist Shared Object Symbol # ........ .......................................................... .... ................. ................. ............................ # 46.72% [div.c:40 -> div.c:40] 0 ± 37.8% ▁█▁▁██▁█ div [.] main 46.72% [div.c:42 -> div.c:44] 0 ± 49.4% ▁▁▂█▂▂▂▂ div [.] main 46.72% [div.c:42 -> div.c:39] 0 ± 24.1% ▃█▂▄▁▃▂▁ div [.] main 20.54% [random_r.c:357 -> random_r.c:394] 1 ± 33.5% ▅▂▁█▃▁▂▁ libc-2.27.so [.] __random_r 20.54% [random_r.c:357 -> random_r.c:380] 0 ± 39.4% ▁▁█▁██▅▁ libc-2.27.so [.] __random_r 20.54% [random_r.c:388 -> random_r.c:388] 0 libc-2.27.so [.] __random_r 20.54% [random_r.c:388 -> random_r.c:391] 0 ± 41.2% ▁▃▁▂█▄▃▁ libc-2.27.so [.] __random_r 17.04% [random.c:288 -> random.c:291] 0 ± 48.8% ▁▁▁▁███▁ libc-2.27.so [.] __random 17.04% [random.c:291 -> random.c:291] 0 ±100.0% ▁█▁▁▁▁▁▁ libc-2.27.so [.] __random 17.04% [random.c:293 -> random.c:293] 0 ±100.0% ▁█▁▁▁▁▁▁ libc-2.27.so [.] __random 17.04% [random.c:295 -> random.c:295] 0 ±100.0% ▁█▁▁▁▁▁▁ libc-2.27.so [.] __random 17.04% [random.c:295 -> random.c:295] 0 libc-2.27.so [.] __random 17.04% [random.c:298 -> random.c:298] 0 ± 75.6% ▃█▁▁▁▁▁▁ libc-2.27.so [.] __random 8.40% [div.c:22 -> div.c:25] 0 ± 42.1% ▁▃▁▁███▁ div [.] compute_flag 8.40% [div.c:27 -> div.c:28] 0 ± 41.8% ██▁▁▄▁▁▄ div [.] compute_flag 5.14% [rand.c:26 -> rand.c:27] 0 ± 37.8% ▁▁▁████▁ libc-2.27.so [.] rand 5.14% [rand.c:28 -> rand.c:28] 0 libc-2.27.so [.] rand 2.15% [rand@plt+0 -> rand@plt+0] 0 div [.] rand@plt 0.00% [kernel.kallsyms] [k] __x86_indirect_thunk_rax 0.00% [do_mmap+714 -> do_mmap+732] -10 [kernel.kallsyms] [k] do_mmap 0.00% [do_mmap+737 -> do_mmap+765] 1 [kernel.kallsyms] [k] do_mmap 0.00% [do_mmap+262 -> do_mmap+299] 0 [kernel.kallsyms] [k] do_mmap 0.00% [__x86_indirect_thunk_r15+0 -> __x86_indirect_thunk_r15+0] 7 [kernel.kallsyms] [k] __x86_indirect_thunk_r15 0.00% [native_sched_clock+0 -> native_sched_clock+119] -1 ± 38.5% ▄█▁ [kernel.kallsyms] [k] native_sched_clock 0.00% [native_write_msr+0 -> native_write_msr+16] -13 ± 47.1% ▁█▇▃▁▁ [kernel.kallsyms] [k] native_write_msr v8: --- Rebase to perf/core branch v7: --- 1. v6 got Jiri's ACK. 2. Rebase to latest perf/core branch. v6: --- 1. Jiri provides better code for using data__hpp_register() in ui_init(). Use this code in v6. v5: --- 1. Refine the use of data__hpp_register() in ui_init() according to Jiri's suggestion. v4: --- 1. Rename the new option from '--noisy' to '--cycles-hist' 2. Remove the option '-n'. 3. Only update the spark value and stats when '--cycles-hist' is enabled. 4. Remove the code of printing '..'. v3: --- 1. Move the histogram to a separate column 2. Move the svals[] out of struct stats v2: --- Jiri got a compile error, CC builtin-diff.o builtin-diff.c: In function ‘compute_cycles_diff’: builtin-diff.c:712:10: error: taking the absolute value of unsigned type ‘u64’ {aka ‘long unsigned int’} has no effect [-Werror=absolute-value] 712 | labs(pair->block_info->cycles_spark[i] - | ^~~~ Because the result of u64 - u64 is still u64. Now we change the type of cycles_spark[] to s64. Signed-off-by: Jin Yao <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@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lore.kernel.org/lkml/20190925011446.30678-1-yao.jin@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-09-25 09:14:46 +08:00
#include "spark.h"
struct hist_browser_timer;
struct hist_entry;
perf annotate: Remove duplicate 'name' field from disasm_line The disasm_line::name field is always equal to ins::name, being used just to locate the instruction's ins_ops from the per-arch instructions table. Eliminate this duplication, nuking that field and instead make ins__find() return an ins_ops, store it in disasm_line::ins.ops, and keep just in disasm_line::ins.name what was in disasm_line::name, this way we end up not keeping a reference to entries in the per-arch instructions table. This in turn will help supporting multiple ways to manage the per-arch instructions table, allowing resorting that array, for instance, when the entries will move after references to its addresses were made. The same problem is avoided when one grows the array with realloc. So architectures simply keeping a constant array will work as well as architectures building the table using regular expressions or other logic that involves resorting the table. Reviewed-by: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Chris Riyder <chris.ryder@arm.com> Cc: David Ahern <dsahern@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Cc: Markus Trippelsdorf <markus@trippelsdorf.de> Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com> Cc: Pawel Moll <pawel.moll@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Taeung Song <treeze.taeung@gmail.com> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/n/tip-vr899azvabnw9gtuepuqfd9t@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-11-24 22:16:06 +08:00
struct ins_ops;
struct map;
struct map_symbol;
struct addr_map_symbol;
struct option;
struct perf_sample;
struct evsel;
struct symbol;
perf annotate: Remove duplicate 'name' field from disasm_line The disasm_line::name field is always equal to ins::name, being used just to locate the instruction's ins_ops from the per-arch instructions table. Eliminate this duplication, nuking that field and instead make ins__find() return an ins_ops, store it in disasm_line::ins.ops, and keep just in disasm_line::ins.name what was in disasm_line::name, this way we end up not keeping a reference to entries in the per-arch instructions table. This in turn will help supporting multiple ways to manage the per-arch instructions table, allowing resorting that array, for instance, when the entries will move after references to its addresses were made. The same problem is avoided when one grows the array with realloc. So architectures simply keeping a constant array will work as well as architectures building the table using regular expressions or other logic that involves resorting the table. Reviewed-by: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Chris Riyder <chris.ryder@arm.com> Cc: David Ahern <dsahern@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Cc: Markus Trippelsdorf <markus@trippelsdorf.de> Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com> Cc: Pawel Moll <pawel.moll@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Taeung Song <treeze.taeung@gmail.com> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/n/tip-vr899azvabnw9gtuepuqfd9t@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-11-24 22:16:06 +08:00
struct ins {
const char *name;
struct ins_ops *ops;
};
struct ins_operands {
char *raw;
perf annotate: Fix parsing aarch64 branch instructions after objdump update Starting with binutils 2.28, aarch64 objdump adds comments to the disassembly output to show the alternative names of a condition code [1]. It is assumed that commas in objdump comments could occur in other arches now or in the future, so this fix is arch-independent. The fix could have been done with arm64 specific jump__parse and jump__scnprintf functions, but the jump__scnprintf instruction would have to have its comment character be a literal, since the scnprintf functions cannot receive a struct arch easily. This inconvenience also applies to the generic jump__scnprintf, which is why we add a raw_comment pointer to struct ins_operands, so the __parse function assigns it to be re-used by its corresponding __scnprintf function. Example differences in 'perf annotate --stdio2' output on an aarch64 perf.data file: BEFORE: → b.cs ffff200008133d1c <unwind_frame+0x18c> // b.hs, dffff7ecc47b AFTER : ↓ b.cs 18c BEFORE: → b.cc ffff200008d8d9cc <get_alloc_profile+0x31c> // b.lo, b.ul, dffff727295b AFTER : ↓ b.cc 31c The branch target labels 18c and 31c also now appear in the output: BEFORE: add x26, x29, #0x80 AFTER : 18c: add x26, x29, #0x80 BEFORE: add x21, x21, #0x8 AFTER : 31c: add x21, x21, #0x8 The Fixes: tag below is added so stable branches will get the update; it doesn't necessarily mean that commit was broken at the time, rather it didn't withstand the aarch64 objdump update. Tested no difference in output for sample x86_64, power arch perf.data files. [1] https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;a=commit;h=bb7eff5206e4795ac79c177a80fe9f4630aaf730 Signed-off-by: Kim Phillips <kim.phillips@arm.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Anton Blanchard <anton@samba.org> Cc: Christian Borntraeger <borntraeger@de.ibm.com> Cc: Jiri Olsa <jolsa@redhat.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@linux.vnet.ibm.com> Cc: Robin Murphy <robin.murphy@arm.com> Cc: Taeung Song <treeze.taeung@gmail.com> Cc: linux-arm-kernel@lists.infradead.org Fixes: b13bbeee5ee6 ("perf annotate: Fix branch instruction with multiple operands") Link: http://lkml.kernel.org/r/20180827125340.a2f7e291901d17cea05daba4@arm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-08-28 01:53:40 +08:00
char *raw_comment;
struct {
char *raw;
char *name;
struct symbol *sym;
u64 addr;
perf annotate: Fix jump target outside of function address range If jump target is outside of function range, perf is not handling it correctly. Especially when target address is lesser than function start address, target offset will be negative. But, target address declared to be unsigned, converts negative number into 2's complement. See below example. Here target of 'jumpq' instruction at 34cf8 is 34ac0 which is lesser than function start address(34cf0). 34ac0 - 34cf0 = -0x230 = 0xfffffffffffffdd0 Objdump output: 0000000000034cf0 <__sigaction>: __GI___sigaction(): 34cf0: lea -0x20(%rdi),%eax 34cf3: cmp -bashx1,%eax 34cf6: jbe 34d00 <__sigaction+0x10> 34cf8: jmpq 34ac0 <__GI___libc_sigaction> 34cfd: nopl (%rax) 34d00: mov 0x386161(%rip),%rax # 3bae68 <_DYNAMIC+0x2e8> 34d07: movl -bashx16,%fs:(%rax) 34d0e: mov -bashxffffffff,%eax 34d13: retq perf annotate before applying patch: __GI___sigaction /usr/lib64/libc-2.22.so lea -0x20(%rdi),%eax cmp -bashx1,%eax v jbe 10 v jmpq fffffffffffffdd0 nop 10: mov _DYNAMIC+0x2e8,%rax movl -bashx16,%fs:(%rax) mov -bashxffffffff,%eax retq perf annotate after applying patch: __GI___sigaction /usr/lib64/libc-2.22.so lea -0x20(%rdi),%eax cmp -bashx1,%eax v jbe 10 ^ jmpq 34ac0 <__GI___libc_sigaction> nop 10: mov _DYNAMIC+0x2e8,%rax movl -bashx16,%fs:(%rax) mov -bashxffffffff,%eax retq Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Chris Riyder <chris.ryder@arm.com> Cc: Kim Phillips <kim.phillips@arm.com> Cc: Markus Trippelsdorf <markus@trippelsdorf.de> Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Taeung Song <treeze.taeung@gmail.com> Cc: linuxppc-dev@lists.ozlabs.org Link: http://lkml.kernel.org/r/1480953407-7605-3-git-send-email-ravi.bangoria@linux.vnet.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-12-05 23:56:47 +08:00
s64 offset;
bool offset_avail;
bool outside;
} target;
union {
struct {
char *raw;
char *name;
u64 addr;
} source;
struct {
perf annotate: Remove duplicate 'name' field from disasm_line The disasm_line::name field is always equal to ins::name, being used just to locate the instruction's ins_ops from the per-arch instructions table. Eliminate this duplication, nuking that field and instead make ins__find() return an ins_ops, store it in disasm_line::ins.ops, and keep just in disasm_line::ins.name what was in disasm_line::name, this way we end up not keeping a reference to entries in the per-arch instructions table. This in turn will help supporting multiple ways to manage the per-arch instructions table, allowing resorting that array, for instance, when the entries will move after references to its addresses were made. The same problem is avoided when one grows the array with realloc. So architectures simply keeping a constant array will work as well as architectures building the table using regular expressions or other logic that involves resorting the table. Reviewed-by: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Chris Riyder <chris.ryder@arm.com> Cc: David Ahern <dsahern@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kim Phillips <kim.phillips@arm.com> Cc: Markus Trippelsdorf <markus@trippelsdorf.de> Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com> Cc: Pawel Moll <pawel.moll@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Taeung Song <treeze.taeung@gmail.com> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/n/tip-vr899azvabnw9gtuepuqfd9t@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-11-24 22:16:06 +08:00
struct ins ins;
struct ins_operands *ops;
} locked;
};
};
struct arch;
struct ins_ops {
void (*free)(struct ins_operands *ops);
int (*parse)(struct arch *arch, struct ins_operands *ops, struct map_symbol *ms);
int (*scnprintf)(struct ins *ins, char *bf, size_t size,
perf annotate: Calculate the max instruction name, align column to that We were hardcoding '6' as the max instruction name, and we have lots that are longer than that, see the diff from two 'P' printed TUI annotations for a libc function that uses instructions with long names, such as 'vpmovmskb' with its 9 chars: --- __strcmp_avx2.annotation.before 2019-03-06 16:31:39.368020425 -0300 +++ __strcmp_avx2.annotation 2019-03-06 16:32:12.079450508 -0300 @@ -2,284 +2,284 @@ Event: cycles:ppp Percent endbr64 - 0.10 mov %edi,%eax + 0.10 mov %edi,%eax - xor %edx,%edx + xor %edx,%edx - 3.54 vpxor %ymm7,%ymm7,%ymm7 + 3.54 vpxor %ymm7,%ymm7,%ymm7 - or %esi,%eax + or %esi,%eax - and $0xfff,%eax + and $0xfff,%eax - cmp $0xf80,%eax + cmp $0xf80,%eax - ↓ jg 370 + ↓ jg 370 - 27.07 vmovdqu (%rdi),%ymm1 + 27.07 vmovdqu (%rdi),%ymm1 - 7.97 vpcmpeqb (%rsi),%ymm1,%ymm0 + 7.97 vpcmpeqb (%rsi),%ymm1,%ymm0 - 2.15 vpminub %ymm1,%ymm0,%ymm0 + 2.15 vpminub %ymm1,%ymm0,%ymm0 - 4.09 vpcmpeqb %ymm7,%ymm0,%ymm0 + 4.09 vpcmpeqb %ymm7,%ymm0,%ymm0 - 0.43 vpmovmskb %ymm0,%ecx + 0.43 vpmovmskb %ymm0,%ecx - 1.53 test %ecx,%ecx + 1.53 test %ecx,%ecx - ↓ je b0 + ↓ je b0 - 5.26 tzcnt %ecx,%edx + 5.26 tzcnt %ecx,%edx - 18.40 movzbl (%rdi,%rdx,1),%eax + 18.40 movzbl (%rdi,%rdx,1),%eax - 7.09 movzbl (%rsi,%rdx,1),%edx + 7.09 movzbl (%rsi,%rdx,1),%edx - 3.34 sub %edx,%eax + 3.34 sub %edx,%eax 2.37 vzeroupper ← retq nop - 50: tzcnt %ecx,%edx + 50: tzcnt %ecx,%edx - movzbl 0x20(%rdi,%rdx,1),%eax + movzbl 0x20(%rdi,%rdx,1),%eax - movzbl 0x20(%rsi,%rdx,1),%edx + movzbl 0x20(%rsi,%rdx,1),%edx - sub %edx,%eax + sub %edx,%eax vzeroupper ← retq - data16 nopw %cs:0x0(%rax,%rax,1) + data16 nopw %cs:0x0(%rax,%rax,1) Reported-by: Travis Downs <travis.downs@gmail.com> LPU-Reference: CAOBGo4z1KfmWeOm6Et0cnX5Z6DWsG2PQbAvRn1MhVPJmXHrc5g@mail.gmail.com Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-89wsdd9h9g6bvq52sgp6d0u4@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-03-07 03:40:15 +08:00
struct ins_operands *ops, int max_ins_name);
};
bool ins__is_jump(const struct ins *ins);
bool ins__is_call(const struct ins *ins);
bool ins__is_ret(const struct ins *ins);
bool ins__is_lock(const struct ins *ins);
perf annotate: Calculate the max instruction name, align column to that We were hardcoding '6' as the max instruction name, and we have lots that are longer than that, see the diff from two 'P' printed TUI annotations for a libc function that uses instructions with long names, such as 'vpmovmskb' with its 9 chars: --- __strcmp_avx2.annotation.before 2019-03-06 16:31:39.368020425 -0300 +++ __strcmp_avx2.annotation 2019-03-06 16:32:12.079450508 -0300 @@ -2,284 +2,284 @@ Event: cycles:ppp Percent endbr64 - 0.10 mov %edi,%eax + 0.10 mov %edi,%eax - xor %edx,%edx + xor %edx,%edx - 3.54 vpxor %ymm7,%ymm7,%ymm7 + 3.54 vpxor %ymm7,%ymm7,%ymm7 - or %esi,%eax + or %esi,%eax - and $0xfff,%eax + and $0xfff,%eax - cmp $0xf80,%eax + cmp $0xf80,%eax - ↓ jg 370 + ↓ jg 370 - 27.07 vmovdqu (%rdi),%ymm1 + 27.07 vmovdqu (%rdi),%ymm1 - 7.97 vpcmpeqb (%rsi),%ymm1,%ymm0 + 7.97 vpcmpeqb (%rsi),%ymm1,%ymm0 - 2.15 vpminub %ymm1,%ymm0,%ymm0 + 2.15 vpminub %ymm1,%ymm0,%ymm0 - 4.09 vpcmpeqb %ymm7,%ymm0,%ymm0 + 4.09 vpcmpeqb %ymm7,%ymm0,%ymm0 - 0.43 vpmovmskb %ymm0,%ecx + 0.43 vpmovmskb %ymm0,%ecx - 1.53 test %ecx,%ecx + 1.53 test %ecx,%ecx - ↓ je b0 + ↓ je b0 - 5.26 tzcnt %ecx,%edx + 5.26 tzcnt %ecx,%edx - 18.40 movzbl (%rdi,%rdx,1),%eax + 18.40 movzbl (%rdi,%rdx,1),%eax - 7.09 movzbl (%rsi,%rdx,1),%edx + 7.09 movzbl (%rsi,%rdx,1),%edx - 3.34 sub %edx,%eax + 3.34 sub %edx,%eax 2.37 vzeroupper ← retq nop - 50: tzcnt %ecx,%edx + 50: tzcnt %ecx,%edx - movzbl 0x20(%rdi,%rdx,1),%eax + movzbl 0x20(%rdi,%rdx,1),%eax - movzbl 0x20(%rsi,%rdx,1),%edx + movzbl 0x20(%rsi,%rdx,1),%edx - sub %edx,%eax + sub %edx,%eax vzeroupper ← retq - data16 nopw %cs:0x0(%rax,%rax,1) + data16 nopw %cs:0x0(%rax,%rax,1) Reported-by: Travis Downs <travis.downs@gmail.com> LPU-Reference: CAOBGo4z1KfmWeOm6Et0cnX5Z6DWsG2PQbAvRn1MhVPJmXHrc5g@mail.gmail.com Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-89wsdd9h9g6bvq52sgp6d0u4@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-03-07 03:40:15 +08:00
int ins__scnprintf(struct ins *ins, char *bf, size_t size, struct ins_operands *ops, int max_ins_name);
perf annotate: Check for fused instructions Macro fusion merges two instructions to a single micro-op. Intel core platform performs this hardware optimization under limited circumstances. For example, CMP + JCC can be "fused" and executed /retired together. While with sampling this can result in the sample sometimes being on the JCC and sometimes on the CMP. So for the fused instruction pair, they could be considered together. On Nehalem, fused instruction pairs: cmp/test + jcc. On other new CPU: cmp/test/add/sub/and/inc/dec + jcc. This patch adds an x86-specific function which checks if 2 instructions are in a "fused" pair. For non-x86 arch, the function is just NULL. Changelog: v4: Move the CPU model checking to symbol__disassemble and save the CPU family/model in arch structure. It avoids checking every time when jump arrow printed. v3: Add checking for Nehalem (CMP, TEST). For other newer Intel CPUs just check it by default (CMP, TEST, ADD, SUB, AND, INC, DEC). v2: Remove the original weak function. Arnaldo points out that doing it as a weak function that will be overridden by the host arch doesn't work. So now it's implemented as an arch-specific function. Committer fix: Do not access evsel->evlist->env->cpuid, ->env can be null, introduce perf_evsel__env_cpuid(), just like perf_evsel__env_arch(), also used in this function call. The original patch was segfaulting 'perf top' + annotation. But this essentially disables this fused instructions augmentation in 'perf top', the right thing is to get the cpuid from the running kernel, left for a later patch tho. Signed-off-by: Yao Jin <yao.jin@linux.intel.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1499403995-19857-2-git-send-email-yao.jin@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2017-07-07 13:06:34 +08:00
bool ins__is_fused(struct arch *arch, const char *ins1, const char *ins2);
#define ANNOTATION__IPC_WIDTH 6
#define ANNOTATION__CYCLES_WIDTH 6
perf annotate: Create hotkey 'c' to show min/max cycles In the 'perf annotate' view, a new hotkey 'c' is created for showing the min/max cycles. For example, when press 'c', the annotate view is: Percent│ IPC Cycle(min/max) │ │ │ Disassembly of section .text: │ │ 000000000003aab0 <random@@GLIBC_2.2.5>: 8.22 │3.92 sub $0x18,%rsp │3.92 mov $0x1,%esi │3.92 xor %eax,%eax │3.92 cmpl $0x0,argp_program_version_hook@@G │3.92 1(2/1) ↓ je 20 │ lock cmpxchg %esi,__abort_msg@@GLIBC_P │ ↓ jne 29 │ ↓ jmp 43 │1.10 20: cmpxchg %esi,__abort_msg@@GLIBC_PRIVATE+ 8.93 │1.10 1(5/1) ↓ je 43 When press 'c' again, the annotate view is switched back: Percent│ IPC Cycle │ │ │ Disassembly of section .text: │ │ 000000000003aab0 <random@@GLIBC_2.2.5>: 8.22 │3.92 sub $0x18,%rsp │3.92 mov $0x1,%esi │3.92 xor %eax,%eax │3.92 cmpl $0x0,argp_program_version_hook@@GLIBC_2.2.5+0x │3.92 1 ↓ je 20 │ lock cmpxchg %esi,__abort_msg@@GLIBC_PRIVATE+0x8a0 │ ↓ jne 29 │ ↓ jmp 43 │1.10 20: cmpxchg %esi,__abort_msg@@GLIBC_PRIVATE+0x8a0 8.93 │1.10 1 ↓ je 43 Signed-off-by: Jin Yao <yao.jin@linux.intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1526569118-14217-3-git-send-email-yao.jin@linux.intel.com [ Rename all maxmin to minmax ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-05-17 22:58:38 +08:00
#define ANNOTATION__MINMAX_CYCLES_WIDTH 19
perf annotate: Compute average IPC and IPC coverage per symbol Add support to 'perf report' annotate view or 'perf annotate --stdio2' to aggregate the IPC derived from timed LBRs per symbol. We compute the average IPC and the IPC coverage percentage. For example: $ perf annotate --stdio2 Percent IPC Cycle (Average IPC: 2.30, IPC Coverage: 54.8%) Disassembly of section .text: 000000000003aac0 <random@@GLIBC_2.2.5>: 8.32 3.28 sub $0x18,%rsp 3.28 mov $0x1,%esi 3.28 xor %eax,%eax 3.28 cmpl $0x0,argp_program_version_hook@@GLIBC_2.2.5+0x1e0 11.57 3.28 1 ↓ je 20 lock cmpxchg %esi,__abort_msg@@GLIBC_PRIVATE+0x8a0 ↓ jne 29 ↓ jmp 43 11.57 1.10 20: cmpxchg %esi,__abort_msg@@GLIBC_PRIVATE+0x8a0 0.00 1.10 1 ↓ je 43 29: lea __abort_msg@@GLIBC_PRIVATE+0x8a0,%rdi sub $0x80,%rsp → callq __lll_lock_wait_private add $0x80,%rsp 0.00 3.00 43: lea __ctype_b@GLIBC_2.2.5+0x38,%rdi 3.00 lea 0xc(%rsp),%rsi 8.49 3.00 1 → callq __random_r 7.91 1.94 cmpl $0x0,argp_program_version_hook@@GLIBC_2.2.5+0x1e0 0.00 1.94 1 ↓ je 68 lock decl __abort_msg@@GLIBC_PRIVATE+0x8a0 ↓ jne 70 ↓ jmp 8a 0.00 2.00 68: decl __abort_msg@@GLIBC_PRIVATE+0x8a0 21.56 2.00 1 ↓ je 8a 70: lea __abort_msg@@GLIBC_PRIVATE+0x8a0,%rdi sub $0x80,%rsp → callq __lll_unlock_wake_private add $0x80,%rsp 21.56 2.90 8a: movslq 0xc(%rsp),%rax 2.90 add $0x18,%rsp 9.03 2.90 1 ← retq It shows for this symbol the average IPC is 2.30 and the IPC coverage is 54.8%. Signed-off-by: Jin Yao <yao.jin@linux.intel.com> Reviewed-by: Ingo Molnar <mingo@kernel.org> Reviewed-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@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1543586097-27632-2-git-send-email-yao.jin@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-11-30 21:54:54 +08:00
#define ANNOTATION__AVG_IPC_WIDTH 36
struct annotation_options {
bool hide_src_code,
use_offset,
jump_arrows,
print_lines,
full_path,
show_linenr,
show_nr_jumps,
show_nr_samples,
perf annotate: Create hotkey 'c' to show min/max cycles In the 'perf annotate' view, a new hotkey 'c' is created for showing the min/max cycles. For example, when press 'c', the annotate view is: Percent│ IPC Cycle(min/max) │ │ │ Disassembly of section .text: │ │ 000000000003aab0 <random@@GLIBC_2.2.5>: 8.22 │3.92 sub $0x18,%rsp │3.92 mov $0x1,%esi │3.92 xor %eax,%eax │3.92 cmpl $0x0,argp_program_version_hook@@G │3.92 1(2/1) ↓ je 20 │ lock cmpxchg %esi,__abort_msg@@GLIBC_P │ ↓ jne 29 │ ↓ jmp 43 │1.10 20: cmpxchg %esi,__abort_msg@@GLIBC_PRIVATE+ 8.93 │1.10 1(5/1) ↓ je 43 When press 'c' again, the annotate view is switched back: Percent│ IPC Cycle │ │ │ Disassembly of section .text: │ │ 000000000003aab0 <random@@GLIBC_2.2.5>: 8.22 │3.92 sub $0x18,%rsp │3.92 mov $0x1,%esi │3.92 xor %eax,%eax │3.92 cmpl $0x0,argp_program_version_hook@@GLIBC_2.2.5+0x │3.92 1 ↓ je 20 │ lock cmpxchg %esi,__abort_msg@@GLIBC_PRIVATE+0x8a0 │ ↓ jne 29 │ ↓ jmp 43 │1.10 20: cmpxchg %esi,__abort_msg@@GLIBC_PRIVATE+0x8a0 8.93 │1.10 1 ↓ je 43 Signed-off-by: Jin Yao <yao.jin@linux.intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1526569118-14217-3-git-send-email-yao.jin@linux.intel.com [ Rename all maxmin to minmax ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-05-17 22:58:38 +08:00
show_total_period,
show_minmax_cycle,
show_asm_raw,
annotate_src;
u8 offset_level;
int min_pcnt;
int max_lines;
int context;
const char *objdump_path;
const char *disassembler_style;
const char *prefix;
const char *prefix_strip;
unsigned int percent_type;
};
enum {
ANNOTATION__OFFSET_JUMP_TARGETS = 1,
ANNOTATION__OFFSET_CALL,
ANNOTATION__MAX_OFFSET_LEVEL,
};
#define ANNOTATION__MIN_OFFSET_LEVEL ANNOTATION__OFFSET_JUMP_TARGETS
extern struct annotation_options annotation__default_options;
struct annotation;
struct sym_hist_entry {
u64 nr_samples;
u64 period;
};
enum {
PERCENT_HITS_LOCAL,
PERCENT_HITS_GLOBAL,
PERCENT_PERIOD_LOCAL,
PERCENT_PERIOD_GLOBAL,
PERCENT_MAX,
};
struct annotation_data {
double percent[PERCENT_MAX];
double percent_sum;
struct sym_hist_entry he;
};
struct annotation_line {
struct list_head node;
struct rb_node rb_node;
s64 offset;
char *line;
int line_nr;
int jump_sources;
float ipc;
u64 cycles;
u64 cycles_max;
u64 cycles_min;
size_t privsize;
char *path;
u32 idx;
int idx_asm;
int data_nr;
struct annotation_data data[0];
};
struct disasm_line {
struct ins ins;
struct ins_operands ops;
/* This needs to be at the end. */
struct annotation_line al;
};
static inline double annotation_data__percent(struct annotation_data *data,
unsigned int which)
{
return which < PERCENT_MAX ? data->percent[which] : -1;
}
static inline const char *percent_type_str(unsigned int type)
{
static const char *str[PERCENT_MAX] = {
"local hits",
"global hits",
"local period",
"global period",
};
if (WARN_ON(type >= PERCENT_MAX))
return "N/A";
return str[type];
}
static inline struct disasm_line *disasm_line(struct annotation_line *al)
{
return al ? container_of(al, struct disasm_line, al) : NULL;
}
/*
* Is this offset in the same function as the line it is used?
* asm functions jump to other functions, for instance.
*/
static inline bool disasm_line__has_local_offset(const struct disasm_line *dl)
{
return dl->ops.target.offset_avail && !dl->ops.target.outside;
}
/*
* Can we draw an arrow from the jump to its target, for instance? I.e.
* is the jump and its target in the same function?
*/
bool disasm_line__is_valid_local_jump(struct disasm_line *dl, struct symbol *sym);
void disasm_line__free(struct disasm_line *dl);
struct annotation_line *
annotation_line__next(struct annotation_line *pos, struct list_head *head);
struct annotation_write_ops {
bool first_line, current_entry, change_color;
int width;
void *obj;
int (*set_color)(void *obj, int color);
void (*set_percent_color)(void *obj, double percent, bool current);
int (*set_jumps_percent_color)(void *obj, int nr, bool current);
void (*printf)(void *obj, const char *fmt, ...);
void (*write_graph)(void *obj, int graph);
};
void annotation_line__write(struct annotation_line *al, struct annotation *notes,
struct annotation_write_ops *ops,
struct annotation_options *opts);
int __annotation__scnprintf_samples_period(struct annotation *notes,
char *bf, size_t size,
struct evsel *evsel,
bool show_freq);
perf annotate: Calculate the max instruction name, align column to that We were hardcoding '6' as the max instruction name, and we have lots that are longer than that, see the diff from two 'P' printed TUI annotations for a libc function that uses instructions with long names, such as 'vpmovmskb' with its 9 chars: --- __strcmp_avx2.annotation.before 2019-03-06 16:31:39.368020425 -0300 +++ __strcmp_avx2.annotation 2019-03-06 16:32:12.079450508 -0300 @@ -2,284 +2,284 @@ Event: cycles:ppp Percent endbr64 - 0.10 mov %edi,%eax + 0.10 mov %edi,%eax - xor %edx,%edx + xor %edx,%edx - 3.54 vpxor %ymm7,%ymm7,%ymm7 + 3.54 vpxor %ymm7,%ymm7,%ymm7 - or %esi,%eax + or %esi,%eax - and $0xfff,%eax + and $0xfff,%eax - cmp $0xf80,%eax + cmp $0xf80,%eax - ↓ jg 370 + ↓ jg 370 - 27.07 vmovdqu (%rdi),%ymm1 + 27.07 vmovdqu (%rdi),%ymm1 - 7.97 vpcmpeqb (%rsi),%ymm1,%ymm0 + 7.97 vpcmpeqb (%rsi),%ymm1,%ymm0 - 2.15 vpminub %ymm1,%ymm0,%ymm0 + 2.15 vpminub %ymm1,%ymm0,%ymm0 - 4.09 vpcmpeqb %ymm7,%ymm0,%ymm0 + 4.09 vpcmpeqb %ymm7,%ymm0,%ymm0 - 0.43 vpmovmskb %ymm0,%ecx + 0.43 vpmovmskb %ymm0,%ecx - 1.53 test %ecx,%ecx + 1.53 test %ecx,%ecx - ↓ je b0 + ↓ je b0 - 5.26 tzcnt %ecx,%edx + 5.26 tzcnt %ecx,%edx - 18.40 movzbl (%rdi,%rdx,1),%eax + 18.40 movzbl (%rdi,%rdx,1),%eax - 7.09 movzbl (%rsi,%rdx,1),%edx + 7.09 movzbl (%rsi,%rdx,1),%edx - 3.34 sub %edx,%eax + 3.34 sub %edx,%eax 2.37 vzeroupper ← retq nop - 50: tzcnt %ecx,%edx + 50: tzcnt %ecx,%edx - movzbl 0x20(%rdi,%rdx,1),%eax + movzbl 0x20(%rdi,%rdx,1),%eax - movzbl 0x20(%rsi,%rdx,1),%edx + movzbl 0x20(%rsi,%rdx,1),%edx - sub %edx,%eax + sub %edx,%eax vzeroupper ← retq - data16 nopw %cs:0x0(%rax,%rax,1) + data16 nopw %cs:0x0(%rax,%rax,1) Reported-by: Travis Downs <travis.downs@gmail.com> LPU-Reference: CAOBGo4z1KfmWeOm6Et0cnX5Z6DWsG2PQbAvRn1MhVPJmXHrc5g@mail.gmail.com Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-89wsdd9h9g6bvq52sgp6d0u4@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-03-07 03:40:15 +08:00
int disasm_line__scnprintf(struct disasm_line *dl, char *bf, size_t size, bool raw, int max_ins_name);
size_t disasm__fprintf(struct list_head *head, FILE *fp);
void symbol__calc_percent(struct symbol *sym, struct evsel *evsel);
struct sym_hist {
u64 nr_samples;
u64 period;
struct sym_hist_entry addr[0];
};
struct cyc_hist {
u64 start;
u64 cycles;
u64 cycles_aggr;
u64 cycles_max;
u64 cycles_min;
perf diff: Report noisy for cycles diff This patch prints the stddev and hist for the cycles diff of program block. It can help us to understand if the cycles is noisy or not. This patch is inspired by Andi Kleen's patch: https://lwn.net/Articles/600471/ We create new option '--cycles-hist'. Example: perf record -b ./div perf record -b ./div perf diff -c cycles # Baseline [Program Block Range] Cycles Diff Shared Object Symbol # ........ .......................................................... .... ................. ............................ # 46.72% [div.c:40 -> div.c:40] 0 div [.] main 46.72% [div.c:42 -> div.c:44] 0 div [.] main 46.72% [div.c:42 -> div.c:39] 0 div [.] main 20.54% [random_r.c:357 -> random_r.c:394] 1 libc-2.27.so [.] __random_r 20.54% [random_r.c:357 -> random_r.c:380] 0 libc-2.27.so [.] __random_r 20.54% [random_r.c:388 -> random_r.c:388] 0 libc-2.27.so [.] __random_r 20.54% [random_r.c:388 -> random_r.c:391] 0 libc-2.27.so [.] __random_r 17.04% [random.c:288 -> random.c:291] 0 libc-2.27.so [.] __random 17.04% [random.c:291 -> random.c:291] 0 libc-2.27.so [.] __random 17.04% [random.c:293 -> random.c:293] 0 libc-2.27.so [.] __random 17.04% [random.c:295 -> random.c:295] 0 libc-2.27.so [.] __random 17.04% [random.c:295 -> random.c:295] 0 libc-2.27.so [.] __random 17.04% [random.c:298 -> random.c:298] 0 libc-2.27.so [.] __random 8.40% [div.c:22 -> div.c:25] 0 div [.] compute_flag 8.40% [div.c:27 -> div.c:28] 0 div [.] compute_flag 5.14% [rand.c:26 -> rand.c:27] 0 libc-2.27.so [.] rand 5.14% [rand.c:28 -> rand.c:28] 0 libc-2.27.so [.] rand 2.15% [rand@plt+0 -> rand@plt+0] 0 div [.] rand@plt 0.00% [kernel.kallsyms] [k] __x86_indirect_thunk_rax 0.00% [do_mmap+714 -> do_mmap+732] -10 [kernel.kallsyms] [k] do_mmap 0.00% [do_mmap+737 -> do_mmap+765] 1 [kernel.kallsyms] [k] do_mmap 0.00% [do_mmap+262 -> do_mmap+299] 0 [kernel.kallsyms] [k] do_mmap 0.00% [__x86_indirect_thunk_r15+0 -> __x86_indirect_thunk_r15+0] 7 [kernel.kallsyms] [k] __x86_indirect_thunk_r15 0.00% [native_sched_clock+0 -> native_sched_clock+119] -1 [kernel.kallsyms] [k] native_sched_clock 0.00% [native_write_msr+0 -> native_write_msr+16] -13 [kernel.kallsyms] [k] native_write_msr When we enable the option '--cycles-hist', the output is perf diff -c cycles --cycles-hist # Baseline [Program Block Range] Cycles Diff stddev/Hist Shared Object Symbol # ........ .......................................................... .... ................. ................. ............................ # 46.72% [div.c:40 -> div.c:40] 0 ± 37.8% ▁█▁▁██▁█ div [.] main 46.72% [div.c:42 -> div.c:44] 0 ± 49.4% ▁▁▂█▂▂▂▂ div [.] main 46.72% [div.c:42 -> div.c:39] 0 ± 24.1% ▃█▂▄▁▃▂▁ div [.] main 20.54% [random_r.c:357 -> random_r.c:394] 1 ± 33.5% ▅▂▁█▃▁▂▁ libc-2.27.so [.] __random_r 20.54% [random_r.c:357 -> random_r.c:380] 0 ± 39.4% ▁▁█▁██▅▁ libc-2.27.so [.] __random_r 20.54% [random_r.c:388 -> random_r.c:388] 0 libc-2.27.so [.] __random_r 20.54% [random_r.c:388 -> random_r.c:391] 0 ± 41.2% ▁▃▁▂█▄▃▁ libc-2.27.so [.] __random_r 17.04% [random.c:288 -> random.c:291] 0 ± 48.8% ▁▁▁▁███▁ libc-2.27.so [.] __random 17.04% [random.c:291 -> random.c:291] 0 ±100.0% ▁█▁▁▁▁▁▁ libc-2.27.so [.] __random 17.04% [random.c:293 -> random.c:293] 0 ±100.0% ▁█▁▁▁▁▁▁ libc-2.27.so [.] __random 17.04% [random.c:295 -> random.c:295] 0 ±100.0% ▁█▁▁▁▁▁▁ libc-2.27.so [.] __random 17.04% [random.c:295 -> random.c:295] 0 libc-2.27.so [.] __random 17.04% [random.c:298 -> random.c:298] 0 ± 75.6% ▃█▁▁▁▁▁▁ libc-2.27.so [.] __random 8.40% [div.c:22 -> div.c:25] 0 ± 42.1% ▁▃▁▁███▁ div [.] compute_flag 8.40% [div.c:27 -> div.c:28] 0 ± 41.8% ██▁▁▄▁▁▄ div [.] compute_flag 5.14% [rand.c:26 -> rand.c:27] 0 ± 37.8% ▁▁▁████▁ libc-2.27.so [.] rand 5.14% [rand.c:28 -> rand.c:28] 0 libc-2.27.so [.] rand 2.15% [rand@plt+0 -> rand@plt+0] 0 div [.] rand@plt 0.00% [kernel.kallsyms] [k] __x86_indirect_thunk_rax 0.00% [do_mmap+714 -> do_mmap+732] -10 [kernel.kallsyms] [k] do_mmap 0.00% [do_mmap+737 -> do_mmap+765] 1 [kernel.kallsyms] [k] do_mmap 0.00% [do_mmap+262 -> do_mmap+299] 0 [kernel.kallsyms] [k] do_mmap 0.00% [__x86_indirect_thunk_r15+0 -> __x86_indirect_thunk_r15+0] 7 [kernel.kallsyms] [k] __x86_indirect_thunk_r15 0.00% [native_sched_clock+0 -> native_sched_clock+119] -1 ± 38.5% ▄█▁ [kernel.kallsyms] [k] native_sched_clock 0.00% [native_write_msr+0 -> native_write_msr+16] -13 ± 47.1% ▁█▇▃▁▁ [kernel.kallsyms] [k] native_write_msr v8: --- Rebase to perf/core branch v7: --- 1. v6 got Jiri's ACK. 2. Rebase to latest perf/core branch. v6: --- 1. Jiri provides better code for using data__hpp_register() in ui_init(). Use this code in v6. v5: --- 1. Refine the use of data__hpp_register() in ui_init() according to Jiri's suggestion. v4: --- 1. Rename the new option from '--noisy' to '--cycles-hist' 2. Remove the option '-n'. 3. Only update the spark value and stats when '--cycles-hist' is enabled. 4. Remove the code of printing '..'. v3: --- 1. Move the histogram to a separate column 2. Move the svals[] out of struct stats v2: --- Jiri got a compile error, CC builtin-diff.o builtin-diff.c: In function ‘compute_cycles_diff’: builtin-diff.c:712:10: error: taking the absolute value of unsigned type ‘u64’ {aka ‘long unsigned int’} has no effect [-Werror=absolute-value] 712 | labs(pair->block_info->cycles_spark[i] - | ^~~~ Because the result of u64 - u64 is still u64. Now we change the type of cycles_spark[] to s64. Signed-off-by: Jin Yao <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@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lore.kernel.org/lkml/20190925011446.30678-1-yao.jin@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-09-25 09:14:46 +08:00
s64 cycles_spark[NUM_SPARKS];
u32 num;
u32 num_aggr;
u8 have_start;
/* 1 byte padding */
u16 reset;
};
/** struct annotated_source - symbols with hits have this attached as in sannotation
*
* @histograms: Array of addr hit histograms per event being monitored
* nr_histograms: This may not be the same as evsel->evlist->core.nr_entries if
* we have more than a group in a evlist, where we will want
* to see each group separately, that is why symbol__annotate2()
* sets src->nr_histograms to evsel->nr_members.
* @lines: If 'print_lines' is specified, per source code line percentages
* @source: source parsed from a disassembler like objdump -dS
* @cyc_hist: Average cycles per basic block
*
* lines is allocated, percentages calculated and all sorted by percentage
* when the annotation is about to be presented, so the percentages are for
* one of the entries in the histogram array, i.e. for the event/counter being
* presented. It is deallocated right after symbol__{tui,tty,etc}_annotate
* returns.
*/
struct annotated_source {
struct list_head source;
int nr_histograms;
size_t sizeof_sym_hist;
struct cyc_hist *cycles_hist;
struct sym_hist *histograms;
};
struct annotation {
pthread_mutex_t lock;
perf annotate: Add branch stack / basic block I wanted to know the hottest path through a function and figured the branch-stack (LBR) information should be able to help out with that. The below uses the branch-stack to create basic blocks and generate statistics from them. from to branch_i * ----> * | | block v * ----> * from to branch_i+1 The blocks are broken down into non-overlapping ranges, while tracking if the start of each range is an entry point and/or the end of a range is a branch. Each block iterates all ranges it covers (while splitting where required to exactly match the block) and increments the 'coverage' count. For the range including the branch we increment the taken counter, as well as the pred counter if flags.predicted. Using these number we can find if an instruction: - had coverage; given by: br->coverage / br->sym->max_coverage This metric ensures each symbol has a 100% spot, which reflects the observation that each symbol must have a most covered/hottest block. - is a branch target: br->is_target && br->start == add - for targets, how much of a branch's coverages comes from it: target->entry / branch->coverage - is a branch: br->is_branch && br->end == addr - for branches, how often it was taken: br->taken / br->coverage after all, all execution that didn't take the branch would have incremented the coverage and continued onward to a later branch. - for branches, how often it was predicted: br->pred / br->taken The coverage percentage is used to color the address and asm sections; for low (<1%) coverage we use NORMAL (uncolored), indicating that these instructions are not 'important'. For high coverage (>75%) we color the address RED. For each branch, we add an asm comment after the instruction with information on how often it was taken and predicted. Output looks like (sans color, which does loose a lot of the information :/) $ perf record --branch-filter u,any -e cycles:p ./branches 27 $ perf annotate branches Percent | Source code & Disassembly of branches for cycles:pu (217 samples) --------------------------------------------------------------------------------- : branches(): 0.00 : 40057a: push %rbp 0.00 : 40057b: mov %rsp,%rbp 0.00 : 40057e: sub $0x20,%rsp 0.00 : 400582: mov %rdi,-0x18(%rbp) 0.00 : 400586: mov %rsi,-0x20(%rbp) 0.00 : 40058a: mov -0x18(%rbp),%rax 0.00 : 40058e: mov %rax,-0x10(%rbp) 0.00 : 400592: movq $0x0,-0x8(%rbp) 0.00 : 40059a: jmpq 400656 <branches+0xdc> 1.84 : 40059f: mov -0x10(%rbp),%rax # +100.00% 3.23 : 4005a3: and $0x1,%eax 1.84 : 4005a6: test %rax,%rax 0.00 : 4005a9: je 4005bf <branches+0x45> # -54.50% (p:42.00%) 0.46 : 4005ab: mov 0x200bbe(%rip),%rax # 601170 <acc> 12.90 : 4005b2: add $0x1,%rax 2.30 : 4005b6: mov %rax,0x200bb3(%rip) # 601170 <acc> 0.46 : 4005bd: jmp 4005d1 <branches+0x57> # -100.00% (p:100.00%) 0.92 : 4005bf: mov 0x200baa(%rip),%rax # 601170 <acc> # +49.54% 13.82 : 4005c6: sub $0x1,%rax 0.46 : 4005ca: mov %rax,0x200b9f(%rip) # 601170 <acc> 2.30 : 4005d1: mov -0x10(%rbp),%rax # +50.46% 0.46 : 4005d5: mov %rax,%rdi 0.46 : 4005d8: callq 400526 <lfsr> # -100.00% (p:100.00%) 0.00 : 4005dd: mov %rax,-0x10(%rbp) # +100.00% 0.92 : 4005e1: mov -0x18(%rbp),%rax 0.00 : 4005e5: and $0x1,%eax 0.00 : 4005e8: test %rax,%rax 0.00 : 4005eb: je 4005ff <branches+0x85> # -100.00% (p:100.00%) 0.00 : 4005ed: mov 0x200b7c(%rip),%rax # 601170 <acc> 0.00 : 4005f4: shr $0x2,%rax 0.00 : 4005f8: mov %rax,0x200b71(%rip) # 601170 <acc> 0.00 : 4005ff: mov -0x10(%rbp),%rax # +100.00% 7.37 : 400603: and $0x1,%eax 3.69 : 400606: test %rax,%rax 0.00 : 400609: jne 400612 <branches+0x98> # -59.25% (p:42.99%) 1.84 : 40060b: mov $0x1,%eax 14.29 : 400610: jmp 400617 <branches+0x9d> # -100.00% (p:100.00%) 1.38 : 400612: mov $0x0,%eax # +57.65% 10.14 : 400617: test %al,%al # +42.35% 0.00 : 400619: je 40062f <branches+0xb5> # -57.65% (p:100.00%) 0.46 : 40061b: mov 0x200b4e(%rip),%rax # 601170 <acc> 2.76 : 400622: sub $0x1,%rax 0.00 : 400626: mov %rax,0x200b43(%rip) # 601170 <acc> 0.46 : 40062d: jmp 400641 <branches+0xc7> # -100.00% (p:100.00%) 0.92 : 40062f: mov 0x200b3a(%rip),%rax # 601170 <acc> # +56.13% 2.30 : 400636: add $0x1,%rax 0.92 : 40063a: mov %rax,0x200b2f(%rip) # 601170 <acc> 0.92 : 400641: mov -0x10(%rbp),%rax # +43.87% 2.30 : 400645: mov %rax,%rdi 0.00 : 400648: callq 400526 <lfsr> # -100.00% (p:100.00%) 0.00 : 40064d: mov %rax,-0x10(%rbp) # +100.00% 1.84 : 400651: addq $0x1,-0x8(%rbp) 0.92 : 400656: mov -0x8(%rbp),%rax 5.07 : 40065a: cmp -0x20(%rbp),%rax 0.00 : 40065e: jb 40059f <branches+0x25> # -100.00% (p:100.00%) 0.00 : 400664: nop 0.00 : 400665: leaveq 0.00 : 400666: retq (Note: the --branch-filter u,any was used to avoid spurious target and branch points due to interrupts/faults, they show up as very small -/+ annotations on 'weird' locations) Committer note: Please take a look at: http://vger.kernel.org/~acme/perf/annotate_basic_blocks.png To see the colors. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <andi@firstfloor.org> Cc: Anshuman Khandual <khandual@linux.vnet.ibm.com> Cc: David Carrillo-Cisneros <davidcc@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@intel.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> [ Moved sym->max_coverage to 'struct annotate', aka symbol__annotate(sym) ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-09-06 03:08:12 +08:00
u64 max_coverage;
u64 start;
perf annotate: Compute average IPC and IPC coverage per symbol Add support to 'perf report' annotate view or 'perf annotate --stdio2' to aggregate the IPC derived from timed LBRs per symbol. We compute the average IPC and the IPC coverage percentage. For example: $ perf annotate --stdio2 Percent IPC Cycle (Average IPC: 2.30, IPC Coverage: 54.8%) Disassembly of section .text: 000000000003aac0 <random@@GLIBC_2.2.5>: 8.32 3.28 sub $0x18,%rsp 3.28 mov $0x1,%esi 3.28 xor %eax,%eax 3.28 cmpl $0x0,argp_program_version_hook@@GLIBC_2.2.5+0x1e0 11.57 3.28 1 ↓ je 20 lock cmpxchg %esi,__abort_msg@@GLIBC_PRIVATE+0x8a0 ↓ jne 29 ↓ jmp 43 11.57 1.10 20: cmpxchg %esi,__abort_msg@@GLIBC_PRIVATE+0x8a0 0.00 1.10 1 ↓ je 43 29: lea __abort_msg@@GLIBC_PRIVATE+0x8a0,%rdi sub $0x80,%rsp → callq __lll_lock_wait_private add $0x80,%rsp 0.00 3.00 43: lea __ctype_b@GLIBC_2.2.5+0x38,%rdi 3.00 lea 0xc(%rsp),%rsi 8.49 3.00 1 → callq __random_r 7.91 1.94 cmpl $0x0,argp_program_version_hook@@GLIBC_2.2.5+0x1e0 0.00 1.94 1 ↓ je 68 lock decl __abort_msg@@GLIBC_PRIVATE+0x8a0 ↓ jne 70 ↓ jmp 8a 0.00 2.00 68: decl __abort_msg@@GLIBC_PRIVATE+0x8a0 21.56 2.00 1 ↓ je 8a 70: lea __abort_msg@@GLIBC_PRIVATE+0x8a0,%rdi sub $0x80,%rsp → callq __lll_unlock_wake_private add $0x80,%rsp 21.56 2.90 8a: movslq 0xc(%rsp),%rax 2.90 add $0x18,%rsp 9.03 2.90 1 ← retq It shows for this symbol the average IPC is 2.30 and the IPC coverage is 54.8%. Signed-off-by: Jin Yao <yao.jin@linux.intel.com> Reviewed-by: Ingo Molnar <mingo@kernel.org> Reviewed-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@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1543586097-27632-2-git-send-email-yao.jin@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-11-30 21:54:54 +08:00
u64 hit_cycles;
u64 hit_insn;
unsigned int total_insn;
unsigned int cover_insn;
struct annotation_options *options;
struct annotation_line **offsets;
int nr_events;
int nr_jumps;
int max_jump_sources;
int nr_entries;
int nr_asm_entries;
u16 max_line_len;
struct {
u8 addr;
u8 jumps;
u8 target;
u8 min_addr;
u8 max_addr;
perf annotate: Calculate the max instruction name, align column to that We were hardcoding '6' as the max instruction name, and we have lots that are longer than that, see the diff from two 'P' printed TUI annotations for a libc function that uses instructions with long names, such as 'vpmovmskb' with its 9 chars: --- __strcmp_avx2.annotation.before 2019-03-06 16:31:39.368020425 -0300 +++ __strcmp_avx2.annotation 2019-03-06 16:32:12.079450508 -0300 @@ -2,284 +2,284 @@ Event: cycles:ppp Percent endbr64 - 0.10 mov %edi,%eax + 0.10 mov %edi,%eax - xor %edx,%edx + xor %edx,%edx - 3.54 vpxor %ymm7,%ymm7,%ymm7 + 3.54 vpxor %ymm7,%ymm7,%ymm7 - or %esi,%eax + or %esi,%eax - and $0xfff,%eax + and $0xfff,%eax - cmp $0xf80,%eax + cmp $0xf80,%eax - ↓ jg 370 + ↓ jg 370 - 27.07 vmovdqu (%rdi),%ymm1 + 27.07 vmovdqu (%rdi),%ymm1 - 7.97 vpcmpeqb (%rsi),%ymm1,%ymm0 + 7.97 vpcmpeqb (%rsi),%ymm1,%ymm0 - 2.15 vpminub %ymm1,%ymm0,%ymm0 + 2.15 vpminub %ymm1,%ymm0,%ymm0 - 4.09 vpcmpeqb %ymm7,%ymm0,%ymm0 + 4.09 vpcmpeqb %ymm7,%ymm0,%ymm0 - 0.43 vpmovmskb %ymm0,%ecx + 0.43 vpmovmskb %ymm0,%ecx - 1.53 test %ecx,%ecx + 1.53 test %ecx,%ecx - ↓ je b0 + ↓ je b0 - 5.26 tzcnt %ecx,%edx + 5.26 tzcnt %ecx,%edx - 18.40 movzbl (%rdi,%rdx,1),%eax + 18.40 movzbl (%rdi,%rdx,1),%eax - 7.09 movzbl (%rsi,%rdx,1),%edx + 7.09 movzbl (%rsi,%rdx,1),%edx - 3.34 sub %edx,%eax + 3.34 sub %edx,%eax 2.37 vzeroupper ← retq nop - 50: tzcnt %ecx,%edx + 50: tzcnt %ecx,%edx - movzbl 0x20(%rdi,%rdx,1),%eax + movzbl 0x20(%rdi,%rdx,1),%eax - movzbl 0x20(%rsi,%rdx,1),%edx + movzbl 0x20(%rsi,%rdx,1),%edx - sub %edx,%eax + sub %edx,%eax vzeroupper ← retq - data16 nopw %cs:0x0(%rax,%rax,1) + data16 nopw %cs:0x0(%rax,%rax,1) Reported-by: Travis Downs <travis.downs@gmail.com> LPU-Reference: CAOBGo4z1KfmWeOm6Et0cnX5Z6DWsG2PQbAvRn1MhVPJmXHrc5g@mail.gmail.com Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lkml.kernel.org/n/tip-89wsdd9h9g6bvq52sgp6d0u4@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-03-07 03:40:15 +08:00
u8 max_ins_name;
} widths;
bool have_cycles;
struct annotated_source *src;
};
static inline int annotation__cycles_width(struct annotation *notes)
{
perf annotate: Create hotkey 'c' to show min/max cycles In the 'perf annotate' view, a new hotkey 'c' is created for showing the min/max cycles. For example, when press 'c', the annotate view is: Percent│ IPC Cycle(min/max) │ │ │ Disassembly of section .text: │ │ 000000000003aab0 <random@@GLIBC_2.2.5>: 8.22 │3.92 sub $0x18,%rsp │3.92 mov $0x1,%esi │3.92 xor %eax,%eax │3.92 cmpl $0x0,argp_program_version_hook@@G │3.92 1(2/1) ↓ je 20 │ lock cmpxchg %esi,__abort_msg@@GLIBC_P │ ↓ jne 29 │ ↓ jmp 43 │1.10 20: cmpxchg %esi,__abort_msg@@GLIBC_PRIVATE+ 8.93 │1.10 1(5/1) ↓ je 43 When press 'c' again, the annotate view is switched back: Percent│ IPC Cycle │ │ │ Disassembly of section .text: │ │ 000000000003aab0 <random@@GLIBC_2.2.5>: 8.22 │3.92 sub $0x18,%rsp │3.92 mov $0x1,%esi │3.92 xor %eax,%eax │3.92 cmpl $0x0,argp_program_version_hook@@GLIBC_2.2.5+0x │3.92 1 ↓ je 20 │ lock cmpxchg %esi,__abort_msg@@GLIBC_PRIVATE+0x8a0 │ ↓ jne 29 │ ↓ jmp 43 │1.10 20: cmpxchg %esi,__abort_msg@@GLIBC_PRIVATE+0x8a0 8.93 │1.10 1 ↓ je 43 Signed-off-by: Jin Yao <yao.jin@linux.intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/r/1526569118-14217-3-git-send-email-yao.jin@linux.intel.com [ Rename all maxmin to minmax ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2018-05-17 22:58:38 +08:00
if (notes->have_cycles && notes->options->show_minmax_cycle)
return ANNOTATION__IPC_WIDTH + ANNOTATION__MINMAX_CYCLES_WIDTH;
return notes->have_cycles ? ANNOTATION__IPC_WIDTH + ANNOTATION__CYCLES_WIDTH : 0;
}
static inline int annotation__pcnt_width(struct annotation *notes)
{
perf annotate: Fix --show-total-period for tui/stdio2 perf annotate --show-total-period does not really show total period. The reason is we have two separate variables for the same purpose. One is in symbol_conf.show_total_period and another is annotation_options.show_total_period. We save command line option in symbol_conf.show_total_period but uses annotation_option.show_total_period while rendering tui/stdio2 browser. Though, we copy symbol_conf.show_total_period to annotation__default_options.show_total_period but that is not really effective as we don't use annotation__default_options once we copy default options to dynamic variable annotate.opts in cmd_annotate(). Instead of all these complication, keep only one variable and use it all over. symbol_conf.show_total_period is used by perf report/top as well. So let's kill annotation_options.show_total_period. On a side note, I've kept annotation_options.show_total_period definition because it's still used by perf-config code. Follow up patch to fix perf-config for annotate will remove annotation_options.show_total_period. Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexey Budankov <alexey.budankov@linux.intel.com> Cc: Changbin Du <changbin.du@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Jin Yao <yao.jin@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Leo Yan <leo.yan@linaro.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Song Liu <songliubraving@fb.com> Cc: Taeung Song <treeze.taeung@gmail.com> Cc: Thomas Richter <tmricht@linux.ibm.com> Cc: Yisheng Xie <xieyisheng1@huawei.com> Link: http://lore.kernel.org/lkml/20200213064306.160480-3-ravi.bangoria@linux.ibm.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2020-02-13 14:43:00 +08:00
return (symbol_conf.show_total_period ? 12 : 7) * notes->nr_events;
}
static inline bool annotation_line__filter(struct annotation_line *al, struct annotation *notes)
{
return notes->options->hide_src_code && al->offset == -1;
}
void annotation__set_offsets(struct annotation *notes, s64 size);
void annotation__compute_ipc(struct annotation *notes, size_t size);
void annotation__mark_jump_targets(struct annotation *notes, struct symbol *sym);
void annotation__update_column_widths(struct annotation *notes);
void annotation__init_column_widths(struct annotation *notes, struct symbol *sym);
static inline struct sym_hist *annotated_source__histogram(struct annotated_source *src, int idx)
{
return ((void *)src->histograms) + (src->sizeof_sym_hist * idx);
}
static inline struct sym_hist *annotation__histogram(struct annotation *notes, int idx)
{
return annotated_source__histogram(notes->src, idx);
}
static inline struct annotation *symbol__annotation(struct symbol *sym)
{
return (void *)sym - symbol_conf.priv_size;
}
int addr_map_symbol__inc_samples(struct addr_map_symbol *ams, struct perf_sample *sample,
struct evsel *evsel);
int addr_map_symbol__account_cycles(struct addr_map_symbol *ams,
struct addr_map_symbol *start,
unsigned cycles);
int hist_entry__inc_addr_samples(struct hist_entry *he, struct perf_sample *sample,
struct evsel *evsel, u64 addr);
struct annotated_source *symbol__hists(struct symbol *sym, int nr_hists);
void symbol__annotate_zero_histograms(struct symbol *sym);
int symbol__annotate(struct map_symbol *ms,
struct evsel *evsel, size_t privsize,
struct annotation_options *options,
struct arch **parch);
int symbol__annotate2(struct map_symbol *ms,
struct evsel *evsel,
struct annotation_options *options,
struct arch **parch);
enum symbol_disassemble_errno {
SYMBOL_ANNOTATE_ERRNO__SUCCESS = 0,
/*
* Choose an arbitrary negative big number not to clash with standard
* errno since SUS requires the errno has distinct positive values.
* See 'Issue 6' in the link below.
*
* http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/errno.h.html
*/
__SYMBOL_ANNOTATE_ERRNO__START = -10000,
SYMBOL_ANNOTATE_ERRNO__NO_VMLINUX = __SYMBOL_ANNOTATE_ERRNO__START,
perf annotate: Enable annotation of BPF programs In symbol__disassemble(), DSO_BINARY_TYPE__BPF_PROG_INFO dso calls into a new function symbol__disassemble_bpf(), where annotation line information is filled based on the bpf_prog_info and btf data saved in given perf_env. symbol__disassemble_bpf() uses binutils's libopcodes to disassemble bpf programs. Committer testing: After fixing this: - u64 *addrs = (u64 *)(info_linear->info.jited_ksyms); + u64 *addrs = (u64 *)(uintptr_t)(info_linear->info.jited_ksyms); Detected when crossbuilding to a 32-bit arch. And making all this dependent on HAVE_LIBBFD_SUPPORT and HAVE_LIBBPF_SUPPORT: 1) Have a BPF program running, one that has BTF info, etc, I used the tools/perf/examples/bpf/augmented_raw_syscalls.c put in place by 'perf trace'. # grep -B1 augmented_raw ~/.perfconfig [trace] add_events = /home/acme/git/perf/tools/perf/examples/bpf/augmented_raw_syscalls.c # # perf trace -e *mmsg dnf/6245 sendmmsg(20, 0x7f5485a88030, 2, MSG_NOSIGNAL) = 2 NetworkManager/10055 sendmmsg(22<socket:[1056822]>, 0x7f8126ad1bb0, 2, MSG_NOSIGNAL) = 2 2) Then do a 'perf record' system wide for a while: # perf record -a ^C[ perf record: Woken up 68 times to write data ] [ perf record: Captured and wrote 19.427 MB perf.data (366891 samples) ] # 3) Check that we captured BPF and BTF info in the perf.data file: # perf report --header-only | grep 'b[pt]f' # event : name = cycles:ppp, , id = { 294789, 294790, 294791, 294792, 294793, 294794, 294795, 294796 }, size = 112, { sample_period, sample_freq } = 4000, sample_type = IP|TID|TIME|CPU|PERIOD, read_format = ID, disabled = 1, inherit = 1, mmap = 1, comm = 1, freq = 1, task = 1, precise_ip = 3, sample_id_all = 1, exclude_guest = 1, mmap2 = 1, comm_exec = 1, ksymbol = 1, bpf_event = 1 # bpf_prog_info of id 13 # bpf_prog_info of id 14 # bpf_prog_info of id 15 # bpf_prog_info of id 16 # bpf_prog_info of id 17 # bpf_prog_info of id 18 # bpf_prog_info of id 21 # bpf_prog_info of id 22 # bpf_prog_info of id 41 # bpf_prog_info of id 42 # btf info of id 2 # 4) Check which programs got recorded: # perf report | grep bpf_prog | head 0.16% exe bpf_prog_819967866022f1e1_sys_enter [k] bpf_prog_819967866022f1e1_sys_enter 0.14% exe bpf_prog_c1bd85c092d6e4aa_sys_exit [k] bpf_prog_c1bd85c092d6e4aa_sys_exit 0.08% fuse-overlayfs bpf_prog_819967866022f1e1_sys_enter [k] bpf_prog_819967866022f1e1_sys_enter 0.07% fuse-overlayfs bpf_prog_c1bd85c092d6e4aa_sys_exit [k] bpf_prog_c1bd85c092d6e4aa_sys_exit 0.01% clang-4.0 bpf_prog_c1bd85c092d6e4aa_sys_exit [k] bpf_prog_c1bd85c092d6e4aa_sys_exit 0.01% clang-4.0 bpf_prog_819967866022f1e1_sys_enter [k] bpf_prog_819967866022f1e1_sys_enter 0.00% clang bpf_prog_c1bd85c092d6e4aa_sys_exit [k] bpf_prog_c1bd85c092d6e4aa_sys_exit 0.00% runc bpf_prog_819967866022f1e1_sys_enter [k] bpf_prog_819967866022f1e1_sys_enter 0.00% clang bpf_prog_819967866022f1e1_sys_enter [k] bpf_prog_819967866022f1e1_sys_enter 0.00% sh bpf_prog_c1bd85c092d6e4aa_sys_exit [k] bpf_prog_c1bd85c092d6e4aa_sys_exit # This was with the default --sort order for 'perf report', which is: --sort comm,dso,symbol If we just look for the symbol, for instance: # perf report --sort symbol | grep bpf_prog | head 0.26% [k] bpf_prog_819967866022f1e1_sys_enter - - 0.24% [k] bpf_prog_c1bd85c092d6e4aa_sys_exit - - # or the DSO: # perf report --sort dso | grep bpf_prog | head 0.26% bpf_prog_819967866022f1e1_sys_enter 0.24% bpf_prog_c1bd85c092d6e4aa_sys_exit # We'll see the two BPF programs that augmented_raw_syscalls.o puts in place, one attached to the raw_syscalls:sys_enter and another to the raw_syscalls:sys_exit tracepoints, as expected. Now we can finally do, from the command line, annotation for one of those two symbols, with the original BPF program source coude intermixed with the disassembled JITed code: # perf annotate --stdio2 bpf_prog_819967866022f1e1_sys_enter Samples: 950 of event 'cycles:ppp', 4000 Hz, Event count (approx.): 553756947, [percent: local period] bpf_prog_819967866022f1e1_sys_enter() bpf_prog_819967866022f1e1_sys_enter Percent int sys_enter(struct syscall_enter_args *args) 53.41 push %rbp 0.63 mov %rsp,%rbp 0.31 sub $0x170,%rsp 1.93 sub $0x28,%rbp 7.02 mov %rbx,0x0(%rbp) 3.20 mov %r13,0x8(%rbp) 1.07 mov %r14,0x10(%rbp) 0.61 mov %r15,0x18(%rbp) 0.11 xor %eax,%eax 1.29 mov %rax,0x20(%rbp) 0.11 mov %rdi,%rbx return bpf_get_current_pid_tgid(); 2.02 → callq *ffffffffda6776d9 2.76 mov %eax,-0x148(%rbp) mov %rbp,%rsi int sys_enter(struct syscall_enter_args *args) add $0xfffffffffffffeb8,%rsi return bpf_map_lookup_elem(pids, &pid) != NULL; movabs $0xffff975ac2607800,%rdi 1.26 → callq *ffffffffda6789e9 cmp $0x0,%rax 2.43 → je 0 add $0x38,%rax 0.21 xor %r13d,%r13d if (pid_filter__has(&pids_filtered, getpid())) 0.81 cmp $0x0,%rax → jne 0 mov %rbp,%rdi probe_read(&augmented_args.args, sizeof(augmented_args.args), args); 2.22 add $0xfffffffffffffeb8,%rdi 0.11 mov $0x40,%esi 0.32 mov %rbx,%rdx 2.74 → callq *ffffffffda658409 syscall = bpf_map_lookup_elem(&syscalls, &augmented_args.args.syscall_nr); 0.22 mov %rbp,%rsi 1.69 add $0xfffffffffffffec0,%rsi syscall = bpf_map_lookup_elem(&syscalls, &augmented_args.args.syscall_nr); movabs $0xffff975bfcd36000,%rdi add $0xd0,%rdi 0.21 mov 0x0(%rsi),%eax 0.93 cmp $0x200,%rax → jae 0 0.10 shl $0x3,%rax 0.11 add %rdi,%rax 0.11 → jmp 0 xor %eax,%eax if (syscall == NULL || !syscall->enabled) 1.07 cmp $0x0,%rax → je 0 if (syscall == NULL || !syscall->enabled) 6.57 movzbq 0x0(%rax),%rdi if (syscall == NULL || !syscall->enabled) cmp $0x0,%rdi 0.95 → je 0 mov $0x40,%r8d switch (augmented_args.args.syscall_nr) { mov -0x140(%rbp),%rdi switch (augmented_args.args.syscall_nr) { cmp $0x2,%rdi → je 0 cmp $0x101,%rdi → je 0 cmp $0x15,%rdi → jne 0 case SYS_OPEN: filename_arg = (const void *)args->args[0]; mov 0x10(%rbx),%rdx → jmp 0 case SYS_OPENAT: filename_arg = (const void *)args->args[1]; mov 0x18(%rbx),%rdx if (filename_arg != NULL) { cmp $0x0,%rdx → je 0 xor %edi,%edi augmented_args.filename.reserved = 0; mov %edi,-0x104(%rbp) augmented_args.filename.size = probe_read_str(&augmented_args.filename.value, mov %rbp,%rdi add $0xffffffffffffff00,%rdi augmented_args.filename.size = probe_read_str(&augmented_args.filename.value, mov $0x100,%esi → callq *ffffffffda658499 mov $0x148,%r8d augmented_args.filename.size = probe_read_str(&augmented_args.filename.value, mov %eax,-0x108(%rbp) augmented_args.filename.size = probe_read_str(&augmented_args.filename.value, mov %rax,%rdi shl $0x20,%rdi shr $0x20,%rdi if (augmented_args.filename.size < sizeof(augmented_args.filename.value)) { cmp $0xff,%rdi → ja 0 len -= sizeof(augmented_args.filename.value) - augmented_args.filename.size; add $0x48,%rax len &= sizeof(augmented_args.filename.value) - 1; and $0xff,%rax mov %rax,%r8 mov %rbp,%rcx return perf_event_output(args, &__augmented_syscalls__, BPF_F_CURRENT_CPU, &augmented_args, len); add $0xfffffffffffffeb8,%rcx mov %rbx,%rdi movabs $0xffff975fbd72d800,%rsi mov $0xffffffff,%edx → callq *ffffffffda658ad9 mov %rax,%r13 } mov %r13,%rax 0.72 mov 0x0(%rbp),%rbx mov 0x8(%rbp),%r13 1.16 mov 0x10(%rbp),%r14 0.10 mov 0x18(%rbp),%r15 0.42 add $0x28,%rbp 0.54 leaveq 0.54 ← retq # Please see 'man perf-config' to see how to control what should be seen, via ~/.perfconfig [annotate] section, for instance, one can suppress the source code and see just the disassembly, etc. Alternatively, use the TUI bu just using 'perf annotate', press '/bpf_prog' to see the bpf symbols, press enter and do the interactive annotation, which allows for dumping to a file after selecting the the various output tunables, for instance, the above without source code intermixed, plus showing all the instruction offsets: # perf annotate bpf_prog_819967866022f1e1_sys_enter Then press: 's' to hide the source code + 'O' twice to show all instruction offsets, then 'P' to print to the bpf_prog_819967866022f1e1_sys_enter.annotation file, which will have: # cat bpf_prog_819967866022f1e1_sys_enter.annotation bpf_prog_819967866022f1e1_sys_enter() bpf_prog_819967866022f1e1_sys_enter Event: cycles:ppp 53.41 0: push %rbp 0.63 1: mov %rsp,%rbp 0.31 4: sub $0x170,%rsp 1.93 b: sub $0x28,%rbp 7.02 f: mov %rbx,0x0(%rbp) 3.20 13: mov %r13,0x8(%rbp) 1.07 17: mov %r14,0x10(%rbp) 0.61 1b: mov %r15,0x18(%rbp) 0.11 1f: xor %eax,%eax 1.29 21: mov %rax,0x20(%rbp) 0.11 25: mov %rdi,%rbx 2.02 28: → callq *ffffffffda6776d9 2.76 2d: mov %eax,-0x148(%rbp) 33: mov %rbp,%rsi 36: add $0xfffffffffffffeb8,%rsi 3d: movabs $0xffff975ac2607800,%rdi 1.26 47: → callq *ffffffffda6789e9 4c: cmp $0x0,%rax 2.43 50: → je 0 52: add $0x38,%rax 0.21 56: xor %r13d,%r13d 0.81 59: cmp $0x0,%rax 5d: → jne 0 63: mov %rbp,%rdi 2.22 66: add $0xfffffffffffffeb8,%rdi 0.11 6d: mov $0x40,%esi 0.32 72: mov %rbx,%rdx 2.74 75: → callq *ffffffffda658409 0.22 7a: mov %rbp,%rsi 1.69 7d: add $0xfffffffffffffec0,%rsi 84: movabs $0xffff975bfcd36000,%rdi 8e: add $0xd0,%rdi 0.21 95: mov 0x0(%rsi),%eax 0.93 98: cmp $0x200,%rax 9f: → jae 0 0.10 a1: shl $0x3,%rax 0.11 a5: add %rdi,%rax 0.11 a8: → jmp 0 aa: xor %eax,%eax 1.07 ac: cmp $0x0,%rax b0: → je 0 6.57 b6: movzbq 0x0(%rax),%rdi bb: cmp $0x0,%rdi 0.95 bf: → je 0 c5: mov $0x40,%r8d cb: mov -0x140(%rbp),%rdi d2: cmp $0x2,%rdi d6: → je 0 d8: cmp $0x101,%rdi df: → je 0 e1: cmp $0x15,%rdi e5: → jne 0 e7: mov 0x10(%rbx),%rdx eb: → jmp 0 ed: mov 0x18(%rbx),%rdx f1: cmp $0x0,%rdx f5: → je 0 f7: xor %edi,%edi f9: mov %edi,-0x104(%rbp) ff: mov %rbp,%rdi 102: add $0xffffffffffffff00,%rdi 109: mov $0x100,%esi 10e: → callq *ffffffffda658499 113: mov $0x148,%r8d 119: mov %eax,-0x108(%rbp) 11f: mov %rax,%rdi 122: shl $0x20,%rdi 126: shr $0x20,%rdi 12a: cmp $0xff,%rdi 131: → ja 0 133: add $0x48,%rax 137: and $0xff,%rax 13d: mov %rax,%r8 140: mov %rbp,%rcx 143: add $0xfffffffffffffeb8,%rcx 14a: mov %rbx,%rdi 14d: movabs $0xffff975fbd72d800,%rsi 157: mov $0xffffffff,%edx 15c: → callq *ffffffffda658ad9 161: mov %rax,%r13 164: mov %r13,%rax 0.72 167: mov 0x0(%rbp),%rbx 16b: mov 0x8(%rbp),%r13 1.16 16f: mov 0x10(%rbp),%r14 0.10 173: mov 0x18(%rbp),%r15 0.42 177: add $0x28,%rbp 0.54 17b: leaveq 0.54 17c: ← retq Another cool way to test all this is to symple use 'perf top' look for those symbols, go there and press enter, annotate it live :-) Signed-off-by: Song Liu <songliubraving@fb.com> Reviewed-by: Jiri Olsa <jolsa@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexei Starovoitov <ast@kernel.org> Cc: Daniel Borkmann <daniel@iogearbox.net> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stanislav Fomichev <sdf@google.com> Link: http://lkml.kernel.org/r/20190312053051.2690567-13-songliubraving@fb.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2019-03-12 13:30:48 +08:00
SYMBOL_ANNOTATE_ERRNO__NO_LIBOPCODES_FOR_BPF,
SYMBOL_ANNOTATE_ERRNO__ARCH_INIT_CPUID_PARSING,
SYMBOL_ANNOTATE_ERRNO__ARCH_INIT_REGEXP,
SYMBOL_ANNOTATE_ERRNO__BPF_INVALID_FILE,
SYMBOL_ANNOTATE_ERRNO__BPF_MISSING_BTF,
__SYMBOL_ANNOTATE_ERRNO__END,
};
int symbol__strerror_disassemble(struct map_symbol *ms, int errnum, char *buf, size_t buflen);
int symbol__annotate_printf(struct map_symbol *ms, struct evsel *evsel,
struct annotation_options *options);
void symbol__annotate_zero_histogram(struct symbol *sym, int evidx);
void symbol__annotate_decay_histogram(struct symbol *sym, int evidx);
void annotated_source__purge(struct annotated_source *as);
int map_symbol__annotation_dump(struct map_symbol *ms, struct evsel *evsel,
struct annotation_options *opts);
bool ui__has_annotation(void);
int symbol__tty_annotate(struct map_symbol *ms, struct evsel *evsel, struct annotation_options *opts);
int symbol__tty_annotate2(struct map_symbol *ms, struct evsel *evsel, struct annotation_options *opts);
#ifdef HAVE_SLANG_SUPPORT
int symbol__tui_annotate(struct map_symbol *ms, struct evsel *evsel,
struct hist_browser_timer *hbt,
struct annotation_options *opts);
#else
static inline int symbol__tui_annotate(struct map_symbol *ms __maybe_unused,
struct evsel *evsel __maybe_unused,
struct hist_browser_timer *hbt __maybe_unused,
struct annotation_options *opts __maybe_unused)
{
return 0;
}
#endif
void annotation_config__init(void);
int annotate_parse_percent_type(const struct option *opt, const char *_str,
int unset);
int annotate_check_args(struct annotation_options *args);
#endif /* __PERF_ANNOTATE_H */