The patch changes the output format in 2 ways:
- line number is displayed for all source lines (matching TUI mode)
- source locations for the hottest lines are printed
at the line end in order to preserve layout
Before:
0.00 : 405ef1: inc %r15
: tmpsd * (TD + tmpsd * TDD)));
0.01 : 405ef4: vfmadd213sd 0x2b9b3(%rip),%xmm0,%xmm3 # 4318b0 <_IO_stdin_used+0x8b0>
: tmpsd * (TC +
eff.c:1811 0.67 : 405efd: vfmadd213sd 0x2b9b2(%rip),%xmm0,%xmm3 # 4318b8 <_IO_stdin_used+0x8b8>
: TA + tmpsd * (TB +
0.35 : 405f06: vfmadd213sd 0x2b9b1(%rip),%xmm0,%xmm3 # 4318c0 <_IO_stdin_used+0x8c0>
: dumbo =
eff.c:1809 1.41 : 405f0f: vfmadd213sd 0x2b9b0(%rip),%xmm0,%xmm3 # 4318c8 <_IO_stdin_used+0x8c8>
: sumi -= sj * tmpsd * dij2i * dumbo;
eff.c:1813 2.58 : 405f18: vmulsd %xmm3,%xmm0,%xmm0
2.81 : 405f1c: vfnmadd213sd 0x30(%rsp),%xmm1,%xmm0
3.78 : 405f23: vmovsd %xmm0,0x30(%rsp)
: for (k = 0; k < lpears[i] + upears[i]; k++) {
eff.c:1761 0.90 : 405f29: cmp %r15d,%r12d
After:
0.00 : 405ef1: inc %r15
: 1812 tmpsd * (TD + tmpsd * TDD)));
0.01 : 405ef4: vfmadd213sd 0x2b9b3(%rip),%xmm0,%xmm3 # 4318b0 <_IO_stdin_used+0x8b0>
: 1811 tmpsd * (TC +
0.67 : 405efd: vfmadd213sd 0x2b9b2(%rip),%xmm0,%xmm3 # 4318b8 <_IO_stdin_used+0x8b8> // eff.c:1811
: 1810 TA + tmpsd * (TB +
0.35 : 405f06: vfmadd213sd 0x2b9b1(%rip),%xmm0,%xmm3 # 4318c0 <_IO_stdin_used+0x8c0>
: 1809 dumbo =
1.41 : 405f0f: vfmadd213sd 0x2b9b0(%rip),%xmm0,%xmm3 # 4318c8 <_IO_stdin_used+0x8c8> // eff.c:1809
: 1813 sumi -= sj * tmpsd * dij2i * dumbo;
2.58 : 405f18: vmulsd %xmm3,%xmm0,%xmm0 // eff.c:1813
2.81 : 405f1c: vfnmadd213sd 0x30(%rsp),%xmm1,%xmm0
3.78 : 405f23: vmovsd %xmm0,0x30(%rsp)
: 1761 for (k = 0; k < lpears[i] + upears[i]; k++) {
Where e.g. '// eff.c:1811' shares the same color as the percentantage
at the line beginning.
Signed-off-by: Martin Liška <mliska@suse.cz>
Link: http://lore.kernel.org/lkml/a0d53f31-f633-5013-c386-a4452391b081@suse.cz
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Right now, when Line numbers are displayed, one can't easily find a source file that the line corresponds to. When a source line is selected and 'l' is pressed, full source file location is displayed in perf UI footer line. The hotkey works only for source code lines. Signed-off-by: Martin Liška <mliska@suse.cz> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Link: http://lore.kernel.org/lkml/25a6384f-d862-5dda-4fec-8f0555599c75@suse.cz Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Committer notes: This allows setting this in from the command line: $ perf config annotate.demangle $ perf config annotate.demangle=yes $ perf config annotate.demangle annotate.demangle=yes $ cat ~/.perfconfig # this file is auto-generated. [report] sort-order = srcline [annotate] demangle = yes $ $ $ perf config annotate.demangle_kernel $ perf config annotate.demangle_kernel=yes $ perf config annotate.demangle_kernel annotate.demangle_kernel=yes $ cat ~/.perfconfig # this file is auto-generated. [report] sort-order = srcline [annotate] demangle = yes demangle_kernel = yes $ Signed-off-by: Martin Liška <mliska@suse.cz> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Link: https://lore.kernel.org/r/c96aabe7-791f-9503-295f-3147a9d19b60@suse.cz Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Considering the following testcase:
int
foo(int a, int b)
{
for (unsigned i = 0; i < 1000000000; i++)
a += b;
return a;
}
int main()
{
foo (3, 4);
return 0;
}
'perf annotate' displays:
86.52 │40055e: → ja 40056c <foo(int, int)+0x26>
13.37 │400560: mov -0x18(%rbp),%eax
│400563: add %eax,-0x14(%rbp)
│400566: addl $0x1,-0x4(%rbp)
0.11 │40056a: → jmp 400557 <foo(int, int)+0x11>
│40056c: mov -0x14(%rbp),%eax
│40056f: pop %rbp
and the 'ja 40056c' does not link to the location in the function. It's
caused by fact that comma is wrongly parsed, it's part of function
signature.
With my patch I see:
86.52 │ ┌──ja 26
13.37 │ │ mov -0x18(%rbp),%eax
│ │ add %eax,-0x14(%rbp)
│ │ addl $0x1,-0x4(%rbp)
0.11 │ │↑ jmp 11
│26:└─→mov -0x14(%rbp),%eax
and 'o' output prints:
86.52 │4005┌── ↓ ja 40056c <foo(int, int)+0x26>
13.37 │4005│0: mov -0x18(%rbp),%eax
│4005│3: add %eax,-0x14(%rbp)
│4005│6: addl $0x1,-0x4(%rbp)
0.11 │4005│a: ↑ jmp 400557 <foo(int, int)+0x11>
│4005└─→ mov -0x14(%rbp),%eax
On the contrary, compiling the very same file with gcc -x c, the parsing
is fine because function arguments are not displayed:
jmp 400543 <foo+0x1d>
Committer testing:
Before:
$ cat cpp_args_annotate.c
int
foo(int a, int b)
{
for (unsigned i = 0; i < 1000000000; i++)
a += b;
return a;
}
int main()
{
foo (3, 4);
return 0;
}
$ gcc --version |& head -1
gcc (GCC) 10.2.1 20201125 (Red Hat 10.2.1-9)
$ gcc -g cpp_args_annotate.c -o cpp_args_annotate
$ perf record ./cpp_args_annotate
[ perf record: Woken up 2 times to write data ]
[ perf record: Captured and wrote 0.275 MB perf.data (7188 samples) ]
$ perf annotate --stdio2 foo
Samples: 7K of event 'cycles:u', 4000 Hz, Event count (approx.): 7468429289, [percent: local period]
foo() /home/acme/c/cpp_args_annotate
Percent
0000000000401106 <foo>:
foo():
int
foo(int a, int b)
{
push %rbp
mov %rsp,%rbp
mov %edi,-0x14(%rbp)
mov %esi,-0x18(%rbp)
for (unsigned i = 0; i < 1000000000; i++)
movl $0x0,-0x4(%rbp)
↓ jmp 1d
a += b;
13.45 13: mov -0x18(%rbp),%eax
add %eax,-0x14(%rbp)
for (unsigned i = 0; i < 1000000000; i++)
addl $0x1,-0x4(%rbp)
0.09 1d: cmpl $0x3b9ac9ff,-0x4(%rbp)
86.46 ↑ jbe 13
return a;
mov -0x14(%rbp),%eax
}
pop %rbp
← retq
$
I.e. works for C, now lets switch to C++:
$ g++ -g cpp_args_annotate.c -o cpp_args_annotate
$ perf record ./cpp_args_annotate
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.268 MB perf.data (6976 samples) ]
$ perf annotate --stdio2 foo
Samples: 6K of event 'cycles:u', 4000 Hz, Event count (approx.): 7380681761, [percent: local period]
foo() /home/acme/c/cpp_args_annotate
Percent
0000000000401106 <foo(int, int)>:
foo(int, int):
int
foo(int a, int b)
{
push %rbp
mov %rsp,%rbp
mov %edi,-0x14(%rbp)
mov %esi,-0x18(%rbp)
for (unsigned i = 0; i < 1000000000; i++)
movl $0x0,-0x4(%rbp)
cmpl $0x3b9ac9ff,-0x4(%rbp)
86.53 → ja 40112c <foo(int, int)+0x26>
a += b;
13.32 mov -0x18(%rbp),%eax
0.00 add %eax,-0x14(%rbp)
for (unsigned i = 0; i < 1000000000; i++)
addl $0x1,-0x4(%rbp)
0.15 → jmp 401117 <foo(int, int)+0x11>
return a;
mov -0x14(%rbp),%eax
}
pop %rbp
← retq
$
Reproduced.
Now with this patch:
Reusing the C++ built binary, as we can see here:
$ readelf -wi cpp_args_annotate | grep producer
<c> DW_AT_producer : (indirect string, offset: 0x2e): GNU C++14 10.2.1 20201125 (Red Hat 10.2.1-9) -mtune=generic -march=x86-64 -g
$
And furthermore:
$ file cpp_args_annotate
cpp_args_annotate: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, BuildID[sha1]=4fe3cab260204765605ec630d0dc7a7e93c361a9, for GNU/Linux 3.2.0, with debug_info, not stripped
$ perf buildid-list -i cpp_args_annotate
4fe3cab260204765605ec630d0dc7a7e93c361a9
$ perf buildid-list | grep cpp_args_annotate
4fe3cab260204765605ec630d0dc7a7e93c361a9 /home/acme/c/cpp_args_annotate
$
It now works:
$ perf annotate --stdio2 foo
Samples: 6K of event 'cycles:u', 4000 Hz, Event count (approx.): 7380681761, [percent: local period]
foo() /home/acme/c/cpp_args_annotate
Percent
0000000000401106 <foo(int, int)>:
foo(int, int):
int
foo(int a, int b)
{
push %rbp
mov %rsp,%rbp
mov %edi,-0x14(%rbp)
mov %esi,-0x18(%rbp)
for (unsigned i = 0; i < 1000000000; i++)
movl $0x0,-0x4(%rbp)
11: cmpl $0x3b9ac9ff,-0x4(%rbp)
86.53 ↓ ja 26
a += b;
13.32 mov -0x18(%rbp),%eax
0.00 add %eax,-0x14(%rbp)
for (unsigned i = 0; i < 1000000000; i++)
addl $0x1,-0x4(%rbp)
0.15 ↑ jmp 11
return a;
26: mov -0x14(%rbp),%eax
}
pop %rbp
← retq
$
Signed-off-by: Martin Liška <mliska@suse.cz>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Slaby <jslaby@suse.cz>
Link: http://lore.kernel.org/lkml/13e1a405-edf9-e4c2-4327-a9b454353730@suse.cz
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
No need to include it otherwise. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Support the MIPS architecture using the ins_ops association method. With
this patch, perf-annotate can work well on MIPS.
Testing it with a perf.data file collected on a mips machine:
$./perf annotate -i perf.data
: Disassembly of section .text:
:
: 00000000000be6a0 <get_next_seq>:
: get_next_seq():
0.00 : be6a0: lw v0,0(a0)
0.00 : be6a4: daddiu sp,sp,-128
0.00 : be6a8: ld a7,72(a0)
0.00 : be6ac: gssq s5,s4,80(sp)
0.00 : be6b0: gssq s1,s0,48(sp)
0.00 : be6b4: gssq s8,gp,112(sp)
0.00 : be6b8: gssq s7,s6,96(sp)
0.00 : be6bc: gssq s3,s2,64(sp)
0.00 : be6c0: sd a3,0(sp)
0.00 : be6c4: move s0,a0
0.00 : be6c8: sd v0,32(sp)
0.00 : be6cc: sd a5,8(sp)
0.00 : be6d0: sd zero,8(a0)
0.00 : be6d4: sd a6,16(sp)
0.00 : be6d8: ld s2,48(a0)
8.53 : be6dc: ld s1,40(a0)
9.42 : be6e0: ld v1,32(a0)
0.00 : be6e4: nop
0.00 : be6e8: ld s4,24(a0)
0.00 : be6ec: ld s5,16(a0)
0.00 : be6f0: sd a7,40(sp)
10.11 : be6f4: ld s6,64(a0)
...
The original patch link:
https://lore.kernel.org/patchwork/patch/1180480/
Signed-off-by: Dengcheng Zhu <dzhu@wavecomp.com>
Cc: Dengcheng Zhu <dzhu@wavecomp.com>
Cc: Jiaxun Yang <jiaxun.yang@flygoat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Xuefeng Li <lixuefeng@loongson.cn>
Cc: linux-mips@vger.kernel.org
[ fanpeng@loongson.cn: Add missing "bgtzl", "bltzl", "bgezl", "blezl", "beql" and "bnel" for pre-R6processors ]
Signed-off-by: Peng Fan <fanpeng@loongson.cn>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Passing build_id object to build_id__sprintf function, so it can operate with the proper size of build id. This will create proper md5 build id readable names, like following: a50e350e97c43b4708d09bcd85ebfff7 instead of: a50e350e97c43b4708d09bcd85ebfff700000000 Signed-off-by: Jiri Olsa <jolsa@kernel.org> Acked-by: Ian Rogers <irogers@google.com> Link: https://lore.kernel.org/r/20201013192441.1299447-5-jolsa@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Replace build_id byte array with struct build_id object and all the code that references it. The objective is to carry size together with build id array, so it's better to keep both together. This is preparatory change for following patches, and there's no functional change. Signed-off-by: Jiri Olsa <jolsa@kernel.org> Acked-by: Ian Rogers <irogers@google.com> Link: https://lore.kernel.org/r/20201013192441.1299447-2-jolsa@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
# perf annotate --stdio2 acpi_processor_ffh_cstate_enter > default
# perf config annotate.disassembler_style=intel
# perf config annotate.disassembler_style
annotate.disassembler_style=intel
# perf annotate --stdio2 acpi_processor_ffh_cstate_enter > intel
# diff -u default intel
--- default 2020-09-04 13:09:26.019205732 -0300
+++ intel 2020-09-04 13:09:52.823795081 -0300
@@ -1,42 +1,42 @@
Samples: 1K of event 'cycles', 4000 Hz, Event count (approx.): 990065316, [percent: local period]
acpi_processor_ffh_cstate_enter() /lib/modules/5.9.0-rc3/build/vmlinux
-Percent → callq __fentry__
- mov cpu_number,%edx
- mov %edx,%edx
- mov cpu_cstate_entry,%rax
- add -0x7dbe9700(,%rdx,8),%rax
- movzbl 0x9(%rdi),%edx
- mov 0x4(%rax,%rdx,8),%edi
- mov (%rax,%rdx,8),%esi
- → jmpq 137ccc6
- 2d: → jmpq 137ccd8
+Percent → call __fentry__
+ mov edx,DWORD PTR gs:[rip+0x7e541d74]
+ mov edx,edx
+ mov rax,QWORD PTR [rip+0x152b8fb]
+ add rax,QWORD PTR [rdx*8-0x7dbe9700]
+ movzx edx,BYTE PTR [rdi+0x9]
+ mov edi,DWORD PTR [rax+rdx*8+0x4]
+ mov esi,DWORD PTR [rax+rdx*8]
+ → jmp 137ccc6
+ 2d: → jmp 137ccd8
mfence
- mov %gs:0x17bc0,%rax
- clflush (%rax)
+ mov rax,QWORD PTR gs:0x17bc0
+ clflush BYTE PTR [rax]
mfence
- xor %edx,%edx
- mov %rdx,%rcx
- mov %gs:0x17bc0,%rax
- 0.00 monitor %rax,%ecx,%edx
- mov (%rax),%rax
- test $0x8,%al
+ xor edx,edx
+ mov rcx,rdx
+ mov rax,QWORD PTR gs:0x17bc0
+ 0.00 monitor
+ mov rax,QWORD PTR [rax]
+ test al,0x8
↓ jne 71
- ↓ jmpq 68
- verw 0x538b08(%rip) # ffffffff82008150 <ds.0>
- 68: mov %rsi,%rax
- mov %rdi,%rcx
-100.00 mwait %eax,%ecx
- 71: mov %gs:0x17bc0,%rax
- lock andb $0xdf,0x2(%rax)
- lock addl $0x0,-0x4(%rsp)
- mov (%rax),%rax
- test $0x8,%al
+ ↓ jmp 68
+ verw WORD PTR [rip+0x538b08] # ffffffff82008150 <ds.0>
+ 68: mov rax,rsi
+ mov rcx,rdi
+100.00 mwait
+ 71: mov rax,QWORD PTR gs:0x17bc0
+ lock and BYTE PTR [rax+0x2],0xdf
+ lock add DWORD PTR [rsp-0x4],0x0
+ mov rax,QWORD PTR [rax]
+ test al,0x8
↓ je 97
- andl $0x7fffffff,__preempt_count
- 97: ← retq
- mov %gs:0x17bc0,%rax
- lock orb $0x20,0x2(%rax)
- mov (%rax),%rax
- test $0x8,%al
+ and DWORD PTR gs:[rip+0x7e548509],0x7fffffff
+ 97: ret
+ mov rax,QWORD PTR gs:0x17bc0
+ lock or BYTE PTR [rax+0x2],0x20
+ mov rax,QWORD PTR [rax]
+ test al,0x8
↑ jne 71
- ↑ jmpq 2d
+ ↑ jmp 2d
#
Requested-by: Matt P. Dziubinski <matdzb@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Our local MSAN (Memory Sanitizer) build of perf throws a warning that comes from the "dso__disassemble_filename" function in "tools/perf/util/annotate.c" when running perf record. The warning stems from the call to readlink, in which "build_id_path" was being read into "linkname". Since readlink does not null terminate, an uninitialized memory access would later occur when "linkname" is passed into the strstr function. This is simply fixed by null-terminating "linkname" after the call to readlink. To reproduce this warning, build perf by running: $ make -C tools/perf CLANG=1 CC=clang EXTRA_CFLAGS="-fsanitize=memory -fsanitize-memory-track-origins" (Additionally, llvm might have to be installed and clang might have to be specified as the compiler - export CC=/usr/bin/clang) Then running: tools/perf/perf record -o - ls / | tools/perf/perf --no-pager annotate -i - --stdio Please see the cover letter for why false positive warnings may be generated. Signed-off-by: Numfor Mbiziwo-Tiapo <nums@google.com> Acked-by: Ian Rogers <irogers@google.com> Acked-by: Jiri Olsa <jolsa@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Mark Drayton <mbd@fb.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Song Liu <songliubraving@fb.com> Cc: Stephane Eranian <eranian@google.com> Link: http://lore.kernel.org/lkml/20190729205750.193289-1-nums@google.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>