* Re: [RFC PATCH v1] perf tests: Test annotate with data type profiling and rust
[not found] <20260127083030.5909-1-9erthalion6@gmail.com>
@ 2026-01-27 17:00 ` Arnaldo Carvalho de Melo
2026-01-30 8:44 ` Dmitry Dolgov
0 siblings, 1 reply; 7+ messages in thread
From: Arnaldo Carvalho de Melo @ 2026-01-27 17:00 UTC (permalink / raw)
To: Dmitrii Dolgov; +Cc: linux-perf-users, Namhyung Kim, Linux Kernel Mailing List
On Tue, Jan 27, 2026 at 09:30:27AM +0100, Dmitrii Dolgov wrote:
> Exercise the annotate command with data type profiling feature on the
> rust runtime:
> * Add a new type of workload code_with_type, which has a dummy rust
> library linked with it. It's recommended that for such scenarios rust
> library is statically linked [1].
> * Add a new shell test, which will profile the new workload, then
> annotate the result expecting to see some data structures from the
> rust code.
I think this should be split into at least two patchs, one introducing
the new workload and then the new perf test.
Trying it now it is failing for me on both an AMD 9950X3D with Fedora 43
and on a Gen Intel(R) Core(TM) i7-1365U Lenovo X1 Carbon notebook also
with Fedora 43, both with:
⬢ [acme@toolbx perf-tools-next]$ rustc --version
rustc 1.92.0 (ded5c06cf 2025-12-08) (Fedora 1.92.0-1.fc43)
⬢ [acme@toolbx perf-tools-next]$
Isn't that 'perf record' just using cycles? Wouldn't it be better to use
'perf mem record'?
Like:
acme@x1:~/git/perf-tools-next$ tools/perf/perf mem record perf test -w code_with_type
Error:
Failure to open event 'cpu_core/mem-loads,ldlat=30/u' on PMU 'cpu_core' which will be removed.
No fallback found for 'cpu_core/mem-loads,ldlat=30/u' for error 61
[ perf record: Woken up 2 times to write data ]
[ perf record: Captured and wrote 0.351 MB perf.data (4672 samples) ]
acme@x1:~/git/perf-tools-next$ perf evlist -v
cpu_atom/mem-loads,ldlat=30/Pu: type: 10 (cpu_atom), size: 144, config: 0x5d0 (mem_uops_retired.load_latency_gt_16), { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|ADDR|PERIOD|IDENTIFIER|DATA_SRC|WEIGHT_STRUCT, read_format: ID|LOST, disabled: 1, inherit: 1, exclude_kernel: 1, exclude_hv: 1, freq: 1, enable_on_exec: 1, precise_ip: 3, sample_id_all: 1, { bp_addr, config1 }: 0x1f
cpu_atom/mem-stores/Pu: type: 10 (cpu_atom), size: 144, config: 0x6d0 (mem_uops_retired.store_latency), { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|ADDR|PERIOD|IDENTIFIER|DATA_SRC|WEIGHT_STRUCT, read_format: ID|LOST, disabled: 1, inherit: 1, exclude_kernel: 1, exclude_hv: 1, freq: 1, enable_on_exec: 1, precise_ip: 3, sample_id_all: 1
cpu_core/mem-loads-aux/u: type: 4 (cpu_core), size: 144, config: 0x8203 (mem-loads-aux), { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|ADDR|PERIOD|IDENTIFIER|DATA_SRC|WEIGHT_STRUCT, read_format: ID|LOST, disabled: 1, inherit: 1, exclude_kernel: 1, exclude_hv: 1, freq: 1, enable_on_exec: 1, precise_ip: 3, sample_id_all: 1
cpu_core/mem-stores/Pu: type: 4 (cpu_core), size: 144, config: 0x2cd (mem-stores), { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|ADDR|PERIOD|IDENTIFIER|DATA_SRC|WEIGHT_STRUCT, read_format: ID|LOST, disabled: 1, inherit: 1, exclude_kernel: 1, exclude_hv: 1, freq: 1, enable_on_exec: 1, precise_ip: 3, sample_id_all: 1
dummy:u: type: 1 (PERF_TYPE_SOFTWARE), size: 144, config: 0x9 (PERF_COUNT_SW_DUMMY), { sample_period, sample_freq }: 1, sample_type: IP|TID|TIME|ADDR|IDENTIFIER|DATA_SRC|WEIGHT_STRUCT, read_format: ID|LOST, inherit: 1, exclude_kernel: 1, exclude_hv: 1, mmap: 1, comm: 1, task: 1, mmap_data: 1, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1, ksymbol: 1, bpf_event: 1, build_id: 1
# Tip: use 'perf evlist -g' to show group information
acme@x1:~/git/perf-tools-next$
Then:
perf annotate --code-with-type
And one of the functions on the TUI is:
Samples: 4K of events 'anon group { cpu_core/mem-loads-aux/u, cpu_core/mem-stores/Pu }', 4000 Hz, Event count (approx.): 2746033426
alloc::raw_vec::RawVecInner<A>::with_capacity_in /home/acme/git/perf-tools-next/tools/perf/perf [Percent: local period] [Type]
[Type] is there, matching --code-with-type, but I see no types in the
annotation output., even with { cpu_core/mem-loads-aux/u, cpu_core/mem-stores/Pu }
Now on the 9950X3D, AMD, using IBS:
acme@number:~/git/perf-tools-next$ tools/perf/perf mem record perf test -w code_with_type
Error:
Failure to open event 'ibs_op/ldlat=0/u' on PMU 'ibs_op' which will be removed.
No fallback found for 'ibs_op/ldlat=0/u' for error 22
Error:
Failure to open any events for recording.
acme@number:~/git/perf-tools-next$ sudo tools/perf/perf mem record perf test -w code_with_type
[ perf record: Woken up 2 times to write data ]
[ perf record: Captured and wrote 0.307 MB perf.data (4005 samples) ]
acme@number:~/git/perf-tools-next$ sudo tools/perf/perf annotate --code-with-type
File perf.data not owned by current user or root (use -f to override)
acme@number:~/git/perf-tools-next$ sudo tools/perf/perf annotate -f --code-with-type
Again, [Type] is there, but no types:
Samples: 4K of event 'ibs_op/ldlat=0/', 4000 Hz, Event count (approx.): 5414001952
test_rs /home/acme/git/perf-tools-next/tools/perf/perf [Percent: local period] [Type]
0.14 │ sub $0x98,%rsp ◆
│ mov %edi,0x2c(%rsp) ▒
│ mov %edi,0x84(%rsp) ▒
│ lea _IO_stdin_used+0x20,%rsi ▒
0.07 │ lea 0x60(%rsp),%rdi ▒
0.07 │ mov $0x4,%edx ▒
│ → call <alloc::string::String as core::convert::From<&str>>::f
I'm not managing to reproduce your results, can you please elaborate
some more?
- Arnaldo
The tests with your patch as-is:
⬢ [acme@toolbx perf-tools-next]$ m
make: Entering directory '/home/acme/git/perf-tools-next/tools/perf'
BUILD: Doing 'make -j32' parallel build
Warning: Kernel ABI header differences:
diff -u tools/arch/arm64/include/asm/cputype.h arch/arm64/include/asm/cputype.h
diff -u tools/perf/arch/s390/entry/syscalls/syscall.tbl arch/s390/kernel/syscalls/syscall.tbl
Auto-detecting system features:
... libdw: [ on ]
... glibc: [ on ]
... libelf: [ on ]
... libnuma: [ on ]
... numa_num_possible_cpus: [ on ]
... libpython: [ on ]
... libcapstone: [ on ]
... llvm-perf: [ on ]
... zlib: [ on ]
... lzma: [ on ]
... bpf: [ on ]
... libaio: [ on ]
... libzstd: [ on ]
... libopenssl: [ on ]
INSTALL libsubcmd_headers
INSTALL libperf_headers
INSTALL libapi_headers
INSTALL libsymbol_headers
INSTALL libbpf_headers
AR /tmp/build/perf-tools-next/libpmu-events.a
make[5]: *** No rule to make target '/tmp/build/perf-tools-next/tests/workloads/libcode_with_type.a', needed by '/tmp/build/perf-tools-next/tests/workloads/perf-test-in.o'. Stop.
make[4]: *** [/home/acme/git/perf-tools-next/tools/build/Makefile.build:156: workloads] Error 2
make[3]: *** [/home/acme/git/perf-tools-next/tools/build/Makefile.build:156: tests] Error 2
make[2]: *** [Makefile.perf:784: /tmp/build/perf-tools-next/perf-test-in.o] Error 2
make[2]: *** Waiting for unfinished jobs....
make[1]: *** [Makefile.perf:288: sub-make] Error 2
make: *** [Makefile:119: install-bin] Error 2
make: Leaving directory '/home/acme/git/perf-tools-next/tools/perf'
⬢ [acme@toolbx perf-tools-next]$
I don't have rust installed in this container, in this case we need some
sort of capability query to tell the user that rust is needed for this
feature to be present in perf.
Lemme try installing it:
$ sudo dnf install rust
Not enough.
I see, it doesn't work if I use 'make O=/some/directory/ -C tools/perf'
Worked when I did:
make -C tools/perf
Where it ended up doing:
CC util/namespaces.o
CC util/comm.o
CC util/thread.o
CC util/threads.o
CC util/thread_map.o
rustc tests/workloads/libcode_with_type.a
CC util/parse-events-bison.o
BISON util/pmu-bison.c
⬢ [acme@toolbx perf-tools-next]$ tools/perf/perf test "data type profiling"
83: perf data type profiling tests : FAILED!
⬢ [acme@toolbx perf-tools-next]$ tools/perf/perf test -v "data type profiling"
Couldn't bump rlimit(MEMLOCK), failures may take place when creating BPF maps, etc
--- start ---
test child forked, pid 1455079
Basic perf annotate test
Basic annotate [Failed: missing target data type]
Percent | Source code & Disassembly of perf for cpu/cycles/Pu (1718 samples, percent: local period)
---------------------------------------------------------------------------------------------------------
: 0 0x55bcb0 <test_rs>:
0.00 : 55bcb0: subq $0x98, %rsp
0.24 : 55bcb7: movl %edi, 0x2c(%rsp)
0.12 : 55bcbb: movl %edi, 0x84(%rsp)
0.00 : 55bcc2: leaq 0x22b557(%rip), %rsi # 0x787220
0.00 : 55bcc9: leaq 0x60(%rsp), %rdi
0.00 : 55bcce: movl $0x4, %edx
0.00 : 55bcd3: callq 0x55baf0 <<alloc::string::String as core::convert::From<&str>>::from>
0.06 : 55bcd8: jmp 0x55bce0 <test_rs+0x30>
0.00 : 55bcda: callq *0xa493b0(%rip) # 0xfa5090 <_DYNAMIC+0x650>
0.00 : 55bce0: movl 0x2c(%rsp), %esi
0.00 : 55bce4: movq $0x0, 0x48(%rsp)
0.00 : 55bced: movq 0x70(%rsp), %rax
0.06 : 55bcf2: movq %rax, 0x40(%rsp)
0.00 : 55bcf7: movups 0x60(%rsp), %xmm0
8.01 : 55bcfc: movaps %xmm0, 0x30(%rsp)
0.00 : 55bd01: movq $0x0, 0x50(%rsp)
0.00 : 55bd0a: movl $0x1, %edi
0.00 : 55bd0f: callq 0x55bae0 <<I as core::iter::traits::collect::IntoIterator>::into_iter>
0.00 : 55bd14: movl %edx, 0x24(%rsp)
0.00 : 55bd18: movl %eax, 0x28(%rsp)
0.00 : 55bd1c: jmp 0x55bd40 <test_rs+0x90>
0.00 : 55bd1e: leaq 0x30(%rsp), %rdi
0.00 : 55bd23: callq 0x55b6f0 <core::ptr::drop_in_place<code_with_type::Buf>>
0.00 : 55bd28: jmp 0x55bd95 <test_rs+0xe5>
0.00 : 55bd2a: movq %rax, %rcx
0.00 : 55bd2d: movl %edx, %eax
0.00 : 55bd2f: movq %rcx, 0x88(%rsp)
0.00 : 55bd37: movl %eax, 0x90(%rsp)
0.00 : 55bd3e: jmp 0x55bd1e <test_rs+0x6e>
0.00 : 55bd40: movl 0x24(%rsp), %eax
0.00 : 55bd44: movl 0x28(%rsp), %ecx
0.00 : 55bd48: movl %ecx, 0x7c(%rsp)
0.00 : 55bd4c: movl %eax, 0x80(%rsp)
0.22 : 55bd53: leaq 0x7c(%rsp), %rdi
70.17 : 55bd58: callq 0x55b820 <core::iter::range::<impl core::iter::traits::iterator::Iterator for core::ops::range::Range<A>>::next>
0.00 : 55bd5d: movl %eax, 0x20(%rsp)
19.94 : 55bd61: jmp 0x55bd63 <test_rs+0xb3>
0.00 : 55bd63: movl 0x20(%rsp), %eax
0.62 : 55bd67: movl %eax, %eax
0.00 : 55bd69: testq $0x1, %rax
0.00 : 55bd6f: je 0x55bd89 <test_rs+0xd9>
0.00 : 55bd71: movq 0x48(%rsp), %rcx
0.00 : 55bd76: movq %rcx, %rax
0.12 : 55bd79: addq $0x1, %rax
0.00 : 55bd7d: movq %rax, 0x18(%rsp)
0.00 : 55bd82: cmpq %rcx, %rax
0.00 : 55bd85: jb 0x55bdcd <test_rs+0x11d>
0.00 : 55bd87: jmp 0x55bdb9 <test_rs+0x109>
0.00 : 55bd89: leaq 0x30(%rsp), %rdi
0.06 : 55bd8e: callq 0x55b6f0 <core::ptr::drop_in_place<code_with_type::Buf>>
0.06 : 55bd93: jmp 0x55bdb1 <test_rs+0x101>
0.00 : 55bd95: callq *0xa492f5(%rip) # 0xfa5090 <_DYNAMIC+0x650>
0.00 : 55bd9b: movq %rax, %rcx
0.00 : 55bd9e: movl %edx, %eax
0.00 : 55bda0: movq %rcx, 0x88(%rsp)
0.00 : 55bda8: movl %eax, 0x90(%rsp)
0.00 : 55bdaf: jmp 0x55bd95 <test_rs+0xe5>
0.00 : 55bdb1: addq $0x98, %rsp
0.00 : 55bdb8: retq
0.00 : 55bdb9: movq 0x18(%rsp), %rax
0.00 : 55bdbe: movq %rax, 0x48(%rsp)
0.00 : 55bdc3: cmpq $0x7b, 0x48(%rsp)
0.05 : 55bdc9: je 0x55bde1 <test_rs+0x131>
0.00 : 55bdcb: jmp 0x55bdf9 <test_rs+0x149>
0.00 : 55bdcd: leaq 0xa3fc44(%rip), %rdi # 0xf9ba18
0.00 : 55bdd4: leaq -0x150ffb(%rip), %rax # 0x40ade0 <core::panicking::panic_const::panic_const_add_overflow>
0.00 : 55bddb: callq *%rax
0.00 : 55bddd: jmp 0x55bddf <test_rs+0x12f>
0.00 : 55bddf: ud2
0.00 : 55bde1: movq 0x48(%rsp), %rcx
0.00 : 55bde6: movq %rcx, %rax
0.00 : 55bde9: addq $0x1, %rax
0.00 : 55bded: movq %rax, 0x10(%rsp)
0.00 : 55bdf2: cmpq %rcx, %rax
0.00 : 55bdf5: jb 0x55be21 <test_rs+0x171>
0.00 : 55bdf7: jmp 0x55be15 <test_rs+0x165>
0.00 : 55bdf9: movq 0x48(%rsp), %rdx
0.06 : 55bdfe: movq 0x50(%rsp), %rcx
0.00 : 55be03: movq %rcx, %rax
0.00 : 55be06: addq %rdx, %rax
0.00 : 55be09: movq %rax, 0x8(%rsp)
0.06 : 55be0e: cmpq %rcx, %rax
0.00 : 55be11: jb 0x55be42 <test_rs+0x192>
0.00 : 55be13: jmp 0x55be33 <test_rs+0x183>
0.00 : 55be15: movq 0x10(%rsp), %rax
0.00 : 55be1a: movq %rax, 0x48(%rsp)
0.00 : 55be1f: jmp 0x55bdf9 <test_rs+0x149>
0.00 : 55be21: leaq 0xa3fc08(%rip), %rdi # 0xf9ba30
0.00 : 55be28: leaq -0x15104f(%rip), %rax # 0x40ade0 <core::panicking::panic_const::panic_const_add_overflow>
0.00 : 55be2f: callq *%rax
0.00 : 55be31: jmp 0x55bddf <test_rs+0x12f>
0.00 : 55be33: movq 0x8(%rsp), %rax
0.16 : 55be38: movq %rax, 0x50(%rsp)
0.00 : 55be3d: jmp 0x55bd53 <test_rs+0xa3>
0.00 : 55be42: leaq 0xa3fbff(%rip), %rdi # 0xf9ba48
0.00 : 55be49: leaq -0x151070(%rip), %rax # 0x40ade0 <core::panicking::panic_const::panic_const_add_overflow>
0.00 : 55be50: callq *%rax
0.00 : 55be52: jmp 0x55bddf <test_rs+0x12f>
0.00 : 55be54: callq *0xa49c2e(%rip) # 0xfa5a88 <_DYNAMIC+0x1048>
Percent | Source code & Disassembly of perf for cpu/cycles/Pu (1050 samples, percent: local period)
---------------------------------------------------------------------------------------------------------
: 0 0x55bc40 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next>:
0.00 : 55bc40: subq $0x38, %rsp
17.94 : 55bc44: movq %rdi, 0x8(%rsp)
0.00 : 55bc49: movq %rdi, 0x18(%rsp)
3.18 : 55bc4e: movq %rdi, %rax
0.00 : 55bc51: addq $0x4, %rax
0.00 : 55bc55: movq %rdi, 0x28(%rsp)
0.00 : 55bc5a: movq %rax, 0x30(%rsp)
0.08 : 55bc5f: movl (%rdi), %eax
0.92 : 55bc61: cmpl 0x4(%rdi), %eax
0.00 : 55bc64: jb 0x55bc70 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next+0x30>
0.00 : 55bc66: movl $0x0, 0x10(%rsp)
0.00 : 55bc6e: jmp 0x55bca2 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next+0x62>
0.00 : 55bc70: movq 0x8(%rsp), %rax
1.82 : 55bc75: movl (%rax), %edi
0.08 : 55bc77: movl %edi, 0x4(%rsp)
0.00 : 55bc7b: movl %edi, 0x24(%rsp)
0.00 : 55bc7f: movl $0x1, %esi
0.00 : 55bc84: callq 0x55b140 <<u32 as core::iter::range::Step>::forward_unchecked>
0.08 : 55bc89: movq 0x8(%rsp), %rcx
73.65 : 55bc8e: movl %eax, %edx
0.00 : 55bc90: movl 0x4(%rsp), %eax
0.00 : 55bc94: movl %edx, (%rcx)
0.27 : 55bc96: movl %eax, 0x14(%rsp)
1.58 : 55bc9a: movl $0x1, 0x10(%rsp)
0.00 : 55bca2: movl 0x10(%rsp), %eax
0.08 : 55bca6: movl 0x14(%rsp), %edx
0.33 : 55bcaa: addq $0x38, %rsp
0.00 : 55bcae: retq
Percent | Source code & Disassembly of perf for cpu/cycles/Pu (708 samples, percent: local period)
--------------------------------------------------------------------------------------------------------
: 0 0x55b140 <<u32 as core::iter::range::Step>::forward_unchecked>:
25.56 : 55b140: subq $0x28, %rsp
64.05 : 55b144: movl %edi, 0xc(%rsp)
0.12 : 55b148: movl %edi, 0x14(%rsp)
0.00 : 55b14c: movq %rsi, 0x18(%rsp)
0.00 : 55b151: movl %esi, %eax
0.00 : 55b153: movl %eax, 0x10(%rsp)
0.43 : 55b157: movl %eax, 0x24(%rsp)
0.00 : 55b15b: movl 0x10(%rsp), %esi
0.52 : 55b15f: movl 0xc(%rsp), %edi
0.00 : 55b163: leaq 0xa40826(%rip), %rdx # 0xf9b990
0.00 : 55b16a: callq 0x55b190 <core::num::<impl u32>::unchecked_add::precondition_check>
0.00 : 55b16f: movl 0x10(%rsp), %ecx
0.00 : 55b173: movl 0xc(%rsp), %eax
0.00 : 55b177: addl %ecx, %eax
0.00 : 55b179: addq $0x28, %rsp
9.32 : 55b17d: retq
Pipe perf annotate test
Pipe annotate [Failed: missing target data type]
Percent | Source code & Disassembly of perf for cpu/cycles/Pu (1824 samples, percent: local period)
---------------------------------------------------------------------------------------------------------
: 0 0x55bc40 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next>:
0.00 : 55bc40: subq $0x38, %rsp
15.47 : 55bc44: movq %rdi, 0x8(%rsp)
0.00 : 55bc49: movq %rdi, 0x18(%rsp)
3.89 : 55bc4e: movq %rdi, %rax
0.00 : 55bc51: addq $0x4, %rax
0.00 : 55bc55: movq %rdi, 0x28(%rsp)
0.00 : 55bc5a: movq %rax, 0x30(%rsp)
0.00 : 55bc5f: movl (%rdi), %eax
7.62 : 55bc61: cmpl 0x4(%rdi), %eax
0.00 : 55bc64: jb 0x55bc70 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next+0x30>
0.00 : 55bc66: movl $0x0, 0x10(%rsp)
0.00 : 55bc6e: jmp 0x55bca2 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next+0x62>
0.00 : 55bc70: movq 0x8(%rsp), %rax
28.40 : 55bc75: movl (%rax), %edi
0.00 : 55bc77: movl %edi, 0x4(%rsp)
0.00 : 55bc7b: movl %edi, 0x24(%rsp)
0.00 : 55bc7f: movl $0x1, %esi
0.00 : 55bc84: callq 0x55b140 <<u32 as core::iter::range::Step>::forward_unchecked>
0.05 : 55bc89: movq 0x8(%rsp), %rcx
40.94 : 55bc8e: movl %eax, %edx
0.00 : 55bc90: movl 0x4(%rsp), %eax
0.00 : 55bc94: movl %edx, (%rcx)
0.00 : 55bc96: movl %eax, 0x14(%rsp)
3.62 : 55bc9a: movl $0x1, 0x10(%rsp)
0.00 : 55bca2: movl 0x10(%rsp), %eax
0.00 : 55bca6: movl 0x14(%rsp), %edx
0.00 : 55bcaa: addq $0x38, %rsp
0.00 : 55bcae: retq
Percent | Source code & Disassembly of perf for cpu/cycles/Pu (871 samples, percent: local period)
--------------------------------------------------------------------------------------------------------
: 0 0x55b190 <core::num::<impl u32>::unchecked_add::precondition_check>:
0.00 : 55b190: subq $0x68, %rsp
21.36 : 55b194: movl %edi, %eax
78.52 : 55b196: movq %rdx, (%rsp)
0.00 : 55b19a: movl %eax, 0x48(%rsp)
0.00 : 55b19e: movl %esi, 0x4c(%rsp)
0.00 : 55b1a2: leaq 0x2922af(%rip), %rcx # 0x7ed458
0.00 : 55b1a9: movq %rcx, 0x50(%rsp)
0.00 : 55b1ae: movq $0xb8, 0x58(%rsp)
0.00 : 55b1b7: movl %eax, %edi
0.00 : 55b1b9: addl %esi, %edi
0.00 : 55b1bb: cmpl %eax, %edi
0.00 : 55b1bd: jb 0x55b1c4 <core::num::<impl u32>::unchecked_add::precondition_check+0x34>
0.00 : 55b1bf: addq $0x68, %rsp
0.11 : 55b1c3: retq
0.00 : 55b1c4: movq (%rsp), %rdx
0.00 : 55b1c8: leaq 0x292289(%rip), %rax # 0x7ed458
0.00 : 55b1cf: movq %rax, 0x38(%rsp)
0.00 : 55b1d4: movq $0xb8, 0x40(%rsp)
0.00 : 55b1dd: leaq 0x38(%rsp), %rax
0.00 : 55b1e2: movq %rax, 0x60(%rsp)
0.00 : 55b1e7: leaq 0x38(%rsp), %rax
0.00 : 55b1ec: movq %rax, 0x8(%rsp)
0.00 : 55b1f1: movq $0x1, 0x10(%rsp)
0.00 : 55b1fa: movq 0x27efdf(%rip), %rcx # 0x7da1e0
0.00 : 55b201: movq 0x27efe0(%rip), %rax # 0x7da1e8
0.00 : 55b208: movq %rcx, 0x28(%rsp)
0.00 : 55b20d: movq %rax, 0x30(%rsp)
0.00 : 55b212: movl $0x8, %eax
0.00 : 55b217: movq %rax, 0x18(%rsp)
0.00 : 55b21c: movq $0x0, 0x20(%rsp)
0.00 : 55b225: leaq 0x8(%rsp), %rdi
0.00 : 55b22a: xorl %esi, %esi
0.00 : 55b22c: callq *0xa4ad96(%rip) # 0xfa5fc8 <_DYNAMIC+0x1588>
Percent | Source code & Disassembly of perf for cpu/cycles/Pu (708 samples, percent: local period)
--------------------------------------------------------------------------------------------------------
: 0 0x55b140 <<u32 as core::iter::range::Step>::forward_unchecked>:
18.36 : 55b140: subq $0x28, %rsp
78.53 : 55b144: movl %edi, 0xc(%rsp)
0.00 : 55b148: movl %edi, 0x14(%rsp)
0.00 : 55b14c: movq %rsi, 0x18(%rsp)
0.00 : 55b151: movl %esi, %eax
0.00 : 55b153: movl %eax, 0x10(%rsp)
3.11 : 55b157: movl %eax, 0x24(%rsp)
0.00 : 55b15b: movl 0x10(%rsp), %esi
0.00 : 55b15f: movl 0xc(%rsp), %edi
0.00 : 55b163: leaq 0xa40826(%rip), %rdx # 0xf9b990
0.00 : 55b16a: callq 0x55b190 <core::num::<impl u32>::unchecked_add::precondition_check>
0.00 : 55b16f: movl 0x10(%rsp), %ecx
0.00 : 55b173: movl 0xc(%rsp), %eax
0.00 : 55b177: addl %ecx, %eax
0.00 : 55b179: addq $0x28, %rsp
0.00 : 55b17d: retq
---- end(-1) ----
83: perf data type profiling tests : FAILED!
⬢ [acme@toolbx perf-tools-next]$
Failed as root as well:
root@number:/home/acme/git/perf-tools-next# tools/perf/perf test 'data type profiling'
83: perf data type profiling tests : FAILED!
root@number:/home/acme/git/perf-tools-next# tools/perf/perf test -v 'data type profiling'
--- start ---
test child forked, pid 1455516
Basic perf annotate test
Basic annotate [Failed: missing target data type]
Percent | Source code & Disassembly of perf for cpu/cycles/P (1359 samples, percent: local period)
--------------------------------------------------------------------------------------------------------
:
:
:
: 3 Disassembly of section .text:
:
: 5 000000000055bcb0 <test_rs>:
: 6 data2: String,
: 7 data3: u64,
: 8 }
:
: 10 #[no_mangle]
: 11 pub extern "C" fn test_rs(count: u32) {
0.07 : 55bcb0: sub $0x98,%rsp
0.07 : 55bcb7: mov %edi,0x2c(%rsp)
0.07 : 55bcbb: mov %edi,0x84(%rsp)
: 15 let mut b = Buf { data1: 0, data2: String::from("data"), data3: 0};
0.00 : 55bcc2: lea 0x22b557(%rip),%rsi # 787220 <_IO_stdin_used+0x20>
0.00 : 55bcc9: lea 0x60(%rsp),%rdi
0.22 : 55bcce: mov $0x4,%edx
0.07 : 55bcd3: call 55baf0 <<alloc::string::String as core::convert::From<&str>>::from>
0.07 : 55bcd8: jmp 55bce0 <test_rs+0x30>
: 21 pub extern "C" fn test_rs(count: u32) {
0.00 : 55bcda: call *0xa493b0(%rip) # fa5090 <_DYNAMIC+0x650>
0.07 : 55bce0: mov 0x2c(%rsp),%esi
: 24 let mut b = Buf { data1: 0, data2: String::from("data"), data3: 0};
0.29 : 55bce4: movq $0x0,0x48(%rsp)
0.15 : 55bced: mov 0x70(%rsp),%rax
0.00 : 55bcf2: mov %rax,0x40(%rsp)
0.07 : 55bcf7: movups 0x60(%rsp),%xmm0
0.00 : 55bcfc: movaps %xmm0,0x30(%rsp)
0.07 : 55bd01: movq $0x0,0x50(%rsp)
0.00 : 55bd0a: mov $0x1,%edi
:
: 33 for _ in 1..count {
0.00 : 55bd0f: call 55bae0 <<I as core::iter::traits::collect::IntoIterator>::into_iter>
0.07 : 55bd14: mov %edx,0x24(%rsp)
0.37 : 55bd18: mov %eax,0x28(%rsp)
0.00 : 55bd1c: jmp 55bd40 <test_rs+0x90>
0.00 : 55bd1e: lea 0x30(%rsp),%rdi
: 39 b.data1 += 1;
: 40 }
:
: 42 b.data3 += b.data1;
: 43 }
: 44 }
0.00 : 55bd23: call 55b6f0 <core::ptr::drop_in_place<code_with_type::Buf>>
0.00 : 55bd28: jmp 55bd95 <test_rs+0xe5>
0.00 : 55bd2a: mov %rax,%rcx
0.00 : 55bd2d: mov %edx,%eax
0.00 : 55bd2f: mov %rcx,0x88(%rsp)
0.00 : 55bd37: mov %eax,0x90(%rsp)
0.00 : 55bd3e: jmp 55bd1e <test_rs+0x6e>
0.00 : 55bd40: mov 0x24(%rsp),%eax
0.00 : 55bd44: mov 0x28(%rsp),%ecx
: 54 for _ in 1..count {
0.00 : 55bd48: mov %ecx,0x7c(%rsp)
0.07 : 55bd4c: mov %eax,0x80(%rsp)
2.08 : 55bd53: lea 0x7c(%rsp),%rdi
4.92 : 55bd58: call 55b820 <core::iter::range::<impl core::iter::traits::iterator::Iterator for core::ops::range::Range<A>>::next>
19.10 : 55bd5d: mov %eax,0x20(%rsp)
17.78 : 55bd61: jmp 55bd63 <test_rs+0xb3>
1.55 : 55bd63: mov 0x20(%rsp),%eax
2.06 : 55bd67: mov %eax,%eax
1.99 : 55bd69: test $0x1,%rax
0.00 : 55bd6f: je 55bd89 <test_rs+0xd9>
: 65 b.data1 += 1;
1.83 : 55bd71: mov 0x48(%rsp),%rcx
1.84 : 55bd76: mov %rcx,%rax
0.00 : 55bd79: add $0x1,%rax
1.91 : 55bd7d: mov %rax,0x18(%rsp)
2.72 : 55bd82: cmp %rcx,%rax
0.00 : 55bd85: jb 55bdcd <test_rs+0x11d>
2.65 : 55bd87: jmp 55bdb9 <test_rs+0x109>
0.15 : 55bd89: lea 0x30(%rsp),%rdi
: 74 }
0.00 : 55bd8e: call 55b6f0 <core::ptr::drop_in_place<code_with_type::Buf>>
0.44 : 55bd93: jmp 55bdb1 <test_rs+0x101>
: 77 pub extern "C" fn test_rs(count: u32) {
0.00 : 55bd95: call *0xa492f5(%rip) # fa5090 <_DYNAMIC+0x650>
0.00 : 55bd9b: mov %rax,%rcx
0.00 : 55bd9e: mov %edx,%eax
0.00 : 55bda0: mov %rcx,0x88(%rsp)
0.00 : 55bda8: mov %eax,0x90(%rsp)
0.00 : 55bdaf: jmp 55bd95 <test_rs+0xe5>
: 84 }
0.07 : 55bdb1: add $0x98,%rsp
0.07 : 55bdb8: ret
1.32 : 55bdb9: mov 0x18(%rsp),%rax
: 88 b.data1 += 1;
4.28 : 55bdbe: mov %rax,0x48(%rsp)
: 90 if b.data1 == 123 {
4.50 : 55bdc3: cmpq $0x7b,0x48(%rsp)
0.00 : 55bdc9: je 55bde1 <test_rs+0x131>
1.84 : 55bdcb: jmp 55bdf9 <test_rs+0x149>
: 94 b.data1 += 1;
0.00 : 55bdcd: lea 0xa3fc44(%rip),%rdi # f9ba18 <workloads+0xd8>
0.00 : 55bdd4: lea -0x150ffb(%rip),%rax # 40ade0 <core::panicking::panic_const::panic_const_add_overflow>
0.00 : 55bddb: call *%rax
0.00 : 55bddd: jmp 55bddf <test_rs+0x12f>
0.00 : 55bddf: ud2
: 100 b.data1 += 1;
0.00 : 55bde1: mov 0x48(%rsp),%rcx
0.00 : 55bde6: mov %rcx,%rax
0.00 : 55bde9: add $0x1,%rax
0.00 : 55bded: mov %rax,0x10(%rsp)
0.00 : 55bdf2: cmp %rcx,%rax
0.00 : 55bdf5: jb 55be21 <test_rs+0x171>
0.00 : 55bdf7: jmp 55be15 <test_rs+0x165>
: 108 b.data3 += b.data1;
1.40 : 55bdf9: mov 0x48(%rsp),%rdx
1.25 : 55bdfe: mov 0x50(%rsp),%rcx
1.17 : 55be03: mov %rcx,%rax
0.00 : 55be06: add %rdx,%rax
3.30 : 55be09: mov %rax,0x8(%rsp)
3.91 : 55be0e: cmp %rcx,%rax
0.00 : 55be11: jb 55be42 <test_rs+0x192>
1.48 : 55be13: jmp 55be33 <test_rs+0x183>
0.00 : 55be15: mov 0x10(%rsp),%rax
: 118 b.data1 += 1;
0.00 : 55be1a: mov %rax,0x48(%rsp)
: 120 if b.data1 == 123 {
0.00 : 55be1f: jmp 55bdf9 <test_rs+0x149>
: 122 b.data1 += 1;
0.00 : 55be21: lea 0xa3fc08(%rip),%rdi # f9ba30 <workloads+0xf0>
0.00 : 55be28: lea -0x15104f(%rip),%rax # 40ade0 <core::panicking::panic_const::panic_const_add_overflow>
0.00 : 55be2f: call *%rax
0.00 : 55be31: jmp 55bddf <test_rs+0x12f>
4.27 : 55be33: mov 0x8(%rsp),%rax
: 128 b.data3 += b.data1;
4.87 : 55be38: mov %rax,0x50(%rsp)
: 130 for _ in 1..count {
3.47 : 55be3d: jmp 55bd53 <test_rs+0xa3>
: 132 b.data3 += b.data1;
0.00 : 55be42: lea 0xa3fbff(%rip),%rdi # f9ba48 <workloads+0x108>
0.00 : 55be49: lea -0x151070(%rip),%rax # 40ade0 <core::panicking::panic_const::panic_const_add_overflow>
0.00 : 55be50: call *%rax
0.00 : 55be52: jmp 55bddf <test_rs+0x12f>
: 137 pub extern "C" fn test_rs(count: u32) {
0.00 : 55be54: call *0xa49c2e(%rip) # fa5a88 <_DYNAMIC+0x1048>
Percent | Source code & Disassembly of perf for cpu/cycles/P (1145 samples, percent: local period)
--------------------------------------------------------------------------------------------------------
:
:
:
: 3 Disassembly of section .text:
:
: 5 000000000055bc40 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next>:
5.22 : 55bc40: sub $0x38,%rsp
1.83 : 55bc44: mov %rdi,0x8(%rsp)
2.01 : 55bc49: mov %rdi,0x18(%rsp)
1.92 : 55bc4e: mov %rdi,%rax
0.00 : 55bc51: add $0x4,%rax
4.17 : 55bc55: mov %rdi,0x28(%rsp)
2.54 : 55bc5a: mov %rax,0x30(%rsp)
6.28 : 55bc5f: mov (%rdi),%eax
5.77 : 55bc61: cmp 0x4(%rdi),%eax
0.00 : 55bc64: jb 55bc70 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next+0x30>
0.96 : 55bc66: movl $0x0,0x10(%rsp)
1.03 : 55bc6e: jmp 55bca2 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next+0x62>
1.57 : 55bc70: mov 0x8(%rsp),%rax
2.73 : 55bc75: mov (%rax),%edi
2.54 : 55bc77: mov %edi,0x4(%rsp)
2.01 : 55bc7b: mov %edi,0x24(%rsp)
1.75 : 55bc7f: mov $0x1,%esi
5.73 : 55bc84: call 55b140 <<u32 as core::iter::range::Step>::forward_unchecked>
3.67 : 55bc89: mov 0x8(%rsp),%rcx
3.41 : 55bc8e: mov %eax,%edx
3.57 : 55bc90: mov 0x4(%rsp),%eax
3.78 : 55bc94: mov %edx,(%rcx)
2.27 : 55bc96: mov %eax,0x14(%rsp)
6.12 : 55bc9a: movl $0x1,0x10(%rsp)
2.79 : 55bca2: mov 0x10(%rsp),%eax
12.41 : 55bca6: mov 0x14(%rsp),%edx
10.04 : 55bcaa: add $0x38,%rsp
3.87 : 55bcae: ret
Percent | Source code & Disassembly of perf for cpu/cycles/P (636 samples, percent: local period)
-------------------------------------------------------------------------------------------------------
:
:
:
: 3 Disassembly of section .text:
:
: 5 000000000055b140 <<u32 as core::iter::range::Step>::forward_unchecked>:
8.04 : 55b140: sub $0x28,%rsp
4.23 : 55b144: mov %edi,0xc(%rsp)
3.30 : 55b148: mov %edi,0x14(%rsp)
5.36 : 55b14c: mov %rsi,0x18(%rsp)
7.37 : 55b151: mov %esi,%eax
4.72 : 55b153: mov %eax,0x10(%rsp)
3.32 : 55b157: mov %eax,0x24(%rsp)
3.59 : 55b15b: mov 0x10(%rsp),%esi
2.99 : 55b15f: mov 0xc(%rsp),%edi
3.92 : 55b163: lea 0xa40826(%rip),%rdx # f9b990 <workloads+0x50>
8.49 : 55b16a: call 55b190 <core::num::<impl u32>::unchecked_add::precondition_check>
4.84 : 55b16f: mov 0x10(%rsp),%ecx
6.62 : 55b173: mov 0xc(%rsp),%eax
5.65 : 55b177: add %ecx,%eax
21.09 : 55b179: add $0x28,%rsp
6.47 : 55b17d: ret
Pipe perf annotate test
Pipe annotate [Failed: missing target data type]
Percent | Source code & Disassembly of perf for cpu/cycles/P (1415 samples, percent: local period)
--------------------------------------------------------------------------------------------------------
:
:
:
: 3 Disassembly of section .text:
:
: 5 000000000055bcb0 <test_rs>:
: 6 data2: String,
: 7 data3: u64,
: 8 }
:
: 10 #[no_mangle]
: 11 pub extern "C" fn test_rs(count: u32) {
0.07 : 55bcb0: sub $0x98,%rsp
0.00 : 55bcb7: mov %edi,0x2c(%rsp)
0.06 : 55bcbb: mov %edi,0x84(%rsp)
: 15 let mut b = Buf { data1: 0, data2: String::from("data"), data3: 0};
0.14 : 55bcc2: lea 0x22b557(%rip),%rsi # 787220 <_IO_stdin_used+0x20>
0.07 : 55bcc9: lea 0x60(%rsp),%rdi
0.00 : 55bcce: mov $0x4,%edx
0.00 : 55bcd3: call 55baf0 <<alloc::string::String as core::convert::From<&str>>::from>
0.00 : 55bcd8: jmp 55bce0 <test_rs+0x30>
: 21 pub extern "C" fn test_rs(count: u32) {
0.00 : 55bcda: call *0xa493b0(%rip) # fa5090 <_DYNAMIC+0x650>
0.07 : 55bce0: mov 0x2c(%rsp),%esi
: 24 let mut b = Buf { data1: 0, data2: String::from("data"), data3: 0};
0.07 : 55bce4: movq $0x0,0x48(%rsp)
0.21 : 55bced: mov 0x70(%rsp),%rax
0.14 : 55bcf2: mov %rax,0x40(%rsp)
0.00 : 55bcf7: movups 0x60(%rsp),%xmm0
0.00 : 55bcfc: movaps %xmm0,0x30(%rsp)
0.00 : 55bd01: movq $0x0,0x50(%rsp)
0.00 : 55bd0a: mov $0x1,%edi
:
: 33 for _ in 1..count {
0.07 : 55bd0f: call 55bae0 <<I as core::iter::traits::collect::IntoIterator>::into_iter>
0.42 : 55bd14: mov %edx,0x24(%rsp)
0.07 : 55bd18: mov %eax,0x28(%rsp)
0.07 : 55bd1c: jmp 55bd40 <test_rs+0x90>
0.00 : 55bd1e: lea 0x30(%rsp),%rdi
: 39 b.data1 += 1;
: 40 }
:
: 42 b.data3 += b.data1;
: 43 }
: 44 }
0.00 : 55bd23: call 55b6f0 <core::ptr::drop_in_place<code_with_type::Buf>>
0.00 : 55bd28: jmp 55bd95 <test_rs+0xe5>
0.00 : 55bd2a: mov %rax,%rcx
0.00 : 55bd2d: mov %edx,%eax
0.00 : 55bd2f: mov %rcx,0x88(%rsp)
0.00 : 55bd37: mov %eax,0x90(%rsp)
0.00 : 55bd3e: jmp 55bd1e <test_rs+0x6e>
0.00 : 55bd40: mov 0x24(%rsp),%eax
0.00 : 55bd44: mov 0x28(%rsp),%ecx
: 54 for _ in 1..count {
0.00 : 55bd48: mov %ecx,0x7c(%rsp)
0.00 : 55bd4c: mov %eax,0x80(%rsp)
3.18 : 55bd53: lea 0x7c(%rsp),%rdi
3.46 : 55bd58: call 55b820 <core::iter::range::<impl core::iter::traits::iterator::Iterator for core::ops::range::Range<A>>::next>
18.28 : 55bd5d: mov %eax,0x20(%rsp)
18.43 : 55bd61: jmp 55bd63 <test_rs+0xb3>
1.84 : 55bd63: mov 0x20(%rsp),%eax
1.34 : 55bd67: mov %eax,%eax
1.55 : 55bd69: test $0x1,%rax
0.00 : 55bd6f: je 55bd89 <test_rs+0xd9>
: 65 b.data1 += 1;
2.62 : 55bd71: mov 0x48(%rsp),%rcx
2.13 : 55bd76: mov %rcx,%rax
0.00 : 55bd79: add $0x1,%rax
1.64 : 55bd7d: mov %rax,0x18(%rsp)
3.11 : 55bd82: cmp %rcx,%rax
0.00 : 55bd85: jb 55bdcd <test_rs+0x11d>
2.26 : 55bd87: jmp 55bdb9 <test_rs+0x109>
0.00 : 55bd89: lea 0x30(%rsp),%rdi
: 74 }
0.21 : 55bd8e: call 55b6f0 <core::ptr::drop_in_place<code_with_type::Buf>>
0.28 : 55bd93: jmp 55bdb1 <test_rs+0x101>
: 77 pub extern "C" fn test_rs(count: u32) {
0.00 : 55bd95: call *0xa492f5(%rip) # fa5090 <_DYNAMIC+0x650>
0.00 : 55bd9b: mov %rax,%rcx
0.00 : 55bd9e: mov %edx,%eax
0.00 : 55bda0: mov %rcx,0x88(%rsp)
0.00 : 55bda8: mov %eax,0x90(%rsp)
0.00 : 55bdaf: jmp 55bd95 <test_rs+0xe5>
: 84 }
0.00 : 55bdb1: add $0x98,%rsp
0.00 : 55bdb8: ret
1.90 : 55bdb9: mov 0x18(%rsp),%rax
: 88 b.data1 += 1;
4.70 : 55bdbe: mov %rax,0x48(%rsp)
: 90 if b.data1 == 123 {
4.95 : 55bdc3: cmpq $0x7b,0x48(%rsp)
0.00 : 55bdc9: je 55bde1 <test_rs+0x131>
1.63 : 55bdcb: jmp 55bdf9 <test_rs+0x149>
: 94 b.data1 += 1;
0.00 : 55bdcd: lea 0xa3fc44(%rip),%rdi # f9ba18 <workloads+0xd8>
0.00 : 55bdd4: lea -0x150ffb(%rip),%rax # 40ade0 <core::panicking::panic_const::panic_const_add_overflow>
0.00 : 55bddb: call *%rax
0.00 : 55bddd: jmp 55bddf <test_rs+0x12f>
0.00 : 55bddf: ud2
: 100 b.data1 += 1;
0.00 : 55bde1: mov 0x48(%rsp),%rcx
0.00 : 55bde6: mov %rcx,%rax
0.00 : 55bde9: add $0x1,%rax
0.00 : 55bded: mov %rax,0x10(%rsp)
0.00 : 55bdf2: cmp %rcx,%rax
0.00 : 55bdf5: jb 55be21 <test_rs+0x171>
0.00 : 55bdf7: jmp 55be15 <test_rs+0x165>
: 108 b.data3 += b.data1;
1.20 : 55bdf9: mov 0x48(%rsp),%rdx
1.63 : 55bdfe: mov 0x50(%rsp),%rcx
1.36 : 55be03: mov %rcx,%rax
0.00 : 55be06: add %rdx,%rax
3.11 : 55be09: mov %rax,0x8(%rsp)
2.89 : 55be0e: cmp %rcx,%rax
0.00 : 55be11: jb 55be42 <test_rs+0x192>
1.90 : 55be13: jmp 55be33 <test_rs+0x183>
0.00 : 55be15: mov 0x10(%rsp),%rax
: 118 b.data1 += 1;
0.00 : 55be1a: mov %rax,0x48(%rsp)
: 120 if b.data1 == 123 {
0.00 : 55be1f: jmp 55bdf9 <test_rs+0x149>
: 122 b.data1 += 1;
0.00 : 55be21: lea 0xa3fc08(%rip),%rdi # f9ba30 <workloads+0xf0>
0.00 : 55be28: lea -0x15104f(%rip),%rax # 40ade0 <core::panicking::panic_const::panic_const_add_overflow>
0.00 : 55be2f: call *%rax
0.00 : 55be31: jmp 55bddf <test_rs+0x12f>
4.86 : 55be33: mov 0x8(%rsp),%rax
: 128 b.data3 += b.data1;
4.60 : 55be38: mov %rax,0x50(%rsp)
: 130 for _ in 1..count {
3.46 : 55be3d: jmp 55bd53 <test_rs+0xa3>
: 132 b.data3 += b.data1;
0.00 : 55be42: lea 0xa3fbff(%rip),%rdi # f9ba48 <workloads+0x108>
0.00 : 55be49: lea -0x151070(%rip),%rax # 40ade0 <core::panicking::panic_const::panic_const_add_overflow>
0.00 : 55be50: call *%rax
0.00 : 55be52: jmp 55bddf <test_rs+0x12f>
: 137 pub extern "C" fn test_rs(count: u32) {
0.00 : 55be54: call *0xa49c2e(%rip) # fa5a88 <_DYNAMIC+0x1048>
Percent | Source code & Disassembly of perf for cpu/cycles/P (1142 samples, percent: local period)
--------------------------------------------------------------------------------------------------------
:
:
:
: 3 Disassembly of section .text:
:
: 5 000000000055bc40 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next>:
4.12 : 55bc40: sub $0x38,%rsp
2.04 : 55bc44: mov %rdi,0x8(%rsp)
2.11 : 55bc49: mov %rdi,0x18(%rsp)
2.80 : 55bc4e: mov %rdi,%rax
0.00 : 55bc51: add $0x4,%rax
3.95 : 55bc55: mov %rdi,0x28(%rsp)
2.28 : 55bc5a: mov %rax,0x30(%rsp)
6.73 : 55bc5f: mov (%rdi),%eax
5.17 : 55bc61: cmp 0x4(%rdi),%eax
0.00 : 55bc64: jb 55bc70 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next+0x30>
1.14 : 55bc66: movl $0x0,0x10(%rsp)
0.70 : 55bc6e: jmp 55bca2 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next+0x62>
1.67 : 55bc70: mov 0x8(%rsp),%rax
2.20 : 55bc75: mov (%rax),%edi
1.75 : 55bc77: mov %edi,0x4(%rsp)
2.63 : 55bc7b: mov %edi,0x24(%rsp)
2.18 : 55bc7f: mov $0x1,%esi
5.15 : 55bc84: call 55b140 <<u32 as core::iter::range::Step>::forward_unchecked>
2.71 : 55bc89: mov 0x8(%rsp),%rcx
3.43 : 55bc8e: mov %eax,%edx
2.89 : 55bc90: mov 0x4(%rsp),%eax
2.56 : 55bc94: mov %edx,(%rcx)
2.44 : 55bc96: mov %eax,0x14(%rsp)
6.22 : 55bc9a: movl $0x1,0x10(%rsp)
3.06 : 55bca2: mov 0x10(%rsp),%eax
14.55 : 55bca6: mov 0x14(%rsp),%edx
11.64 : 55bcaa: add $0x38,%rsp
3.87 : 55bcae: ret
Percent | Source code & Disassembly of perf for cpu/cycles/P (590 samples, percent: local period)
-------------------------------------------------------------------------------------------------------
:
:
:
: 3 Disassembly of section .text:
:
: 5 000000000055b140 <<u32 as core::iter::range::Step>::forward_unchecked>:
8.72 : 55b140: sub $0x28,%rsp
2.54 : 55b144: mov %edi,0xc(%rsp)
3.90 : 55b148: mov %edi,0x14(%rsp)
4.61 : 55b14c: mov %rsi,0x18(%rsp)
6.76 : 55b151: mov %esi,%eax
3.74 : 55b153: mov %eax,0x10(%rsp)
3.22 : 55b157: mov %eax,0x24(%rsp)
4.10 : 55b15b: mov 0x10(%rsp),%esi
3.37 : 55b15f: mov 0xc(%rsp),%edi
4.57 : 55b163: lea 0xa40826(%rip),%rdx # f9b990 <workloads+0x50>
8.30 : 55b16a: call 55b190 <core::num::<impl u32>::unchecked_add::precondition_check>
6.43 : 55b16f: mov 0x10(%rsp),%ecx
7.13 : 55b173: mov 0xc(%rsp),%eax
3.85 : 55b177: add %ecx,%eax
21.24 : 55b179: add $0x28,%rsp
7.51 : 55b17d: ret
---- end(-1) ----
83: perf data type profiling tests : FAILED!
root@number:/home/acme/git/perf-tools-next#
Ok, this is on a:
root@five:~# grep "model name" -m1 /proc/cpuinfo
model name : AMD Ryzen 9 5950X 16-Core Processor
root@five:~#
Now trying on a:
acme@x1:~/git/perf-tools-next$ grep -m1 "model name" /proc/cpuinfo
model name : 13th Gen Intel(R) Core(TM) i7-1365U
acme@x1:~/git/perf-tools-next$
⬢ [acme@toolbx perf-tools-next]$ tools/perf/perf test "data type profiling"
83: perf data type profiling tests : FAILED!
⬢ [acme@toolbx perf-tools-next]$ tools/perf/perf test -v "data type profiling"
Couldn't bump rlimit(MEMLOCK), failures may take place when creating BPF maps, etc
--- start ---
test child forked, pid 615260
Basic perf annotate test
Basic annotate [Failed: missing target data type]
Percent | Source code & Disassembly of perf for cpu_atom/cycles/Pu (108 samples, percent: local period)
-------------------------------------------------------------------------------------------------------------
: 0 0x55bcb0 <test_rs>:
0.00 : 55bcb0: subq $0x98, %rsp
0.00 : 55bcb7: movl %edi, 0x2c(%rsp)
0.00 : 55bcbb: movl %edi, 0x84(%rsp)
0.00 : 55bcc2: leaq 0x22b557(%rip), %rsi # 0x787220
0.00 : 55bcc9: leaq 0x60(%rsp), %rdi
0.00 : 55bcce: movl $0x4, %edx
0.00 : 55bcd3: callq 0x55baf0 <<alloc::string::String as core::convert::From<&str>>::from>
0.00 : 55bcd8: jmp 0x55bce0 <test_rs+0x30>
0.00 : 55bcda: callq *0xa493b0(%rip) # 0xfa5090 <_DYNAMIC+0x650>
0.00 : 55bce0: movl 0x2c(%rsp), %esi
0.00 : 55bce4: movq $0x0, 0x48(%rsp)
0.00 : 55bced: movq 0x70(%rsp), %rax
0.00 : 55bcf2: movq %rax, 0x40(%rsp)
1.92 : 55bcf7: movups 0x60(%rsp), %xmm0
0.00 : 55bcfc: movaps %xmm0, 0x30(%rsp)
0.00 : 55bd01: movq $0x0, 0x50(%rsp)
0.00 : 55bd0a: movl $0x1, %edi
0.00 : 55bd0f: callq 0x55bae0 <<I as core::iter::traits::collect::IntoIterator>::into_iter>
0.00 : 55bd14: movl %edx, 0x24(%rsp)
0.00 : 55bd18: movl %eax, 0x28(%rsp)
0.00 : 55bd1c: jmp 0x55bd40 <test_rs+0x90>
0.00 : 55bd1e: leaq 0x30(%rsp), %rdi
0.00 : 55bd23: callq 0x55b6f0 <core::ptr::drop_in_place<code_with_type::Buf>>
0.00 : 55bd28: jmp 0x55bd95 <test_rs+0xe5>
0.00 : 55bd2a: movq %rax, %rcx
0.00 : 55bd2d: movl %edx, %eax
0.00 : 55bd2f: movq %rcx, 0x88(%rsp)
0.00 : 55bd37: movl %eax, 0x90(%rsp)
0.00 : 55bd3e: jmp 0x55bd1e <test_rs+0x6e>
0.96 : 55bd40: movl 0x24(%rsp), %eax
0.00 : 55bd44: movl 0x28(%rsp), %ecx
0.00 : 55bd48: movl %ecx, 0x7c(%rsp)
0.00 : 55bd4c: movl %eax, 0x80(%rsp)
1.92 : 55bd53: leaq 0x7c(%rsp), %rdi
0.96 : 55bd58: callq 0x55b820 <core::iter::range::<impl core::iter::traits::iterator::Iterator for core::ops::range::Range<A>>::next>
7.56 : 55bd5d: movl %eax, 0x20(%rsp)
4.10 : 55bd61: jmp 0x55bd63 <test_rs+0xb3>
3.83 : 55bd63: movl 0x20(%rsp), %eax
2.88 : 55bd67: movl %eax, %eax
1.92 : 55bd69: testq $0x1, %rax
1.92 : 55bd6f: je 0x55bd89 <test_rs+0xd9>
2.88 : 55bd71: movq 0x48(%rsp), %rcx
5.71 : 55bd76: movq %rcx, %rax
3.84 : 55bd79: addq $0x1, %rax
1.93 : 55bd7d: movq %rax, 0x18(%rsp)
3.66 : 55bd82: cmpq %rcx, %rax
1.48 : 55bd85: jb 0x55bdcd <test_rs+0x11d>
0.96 : 55bd87: jmp 0x55bdb9 <test_rs+0x109>
0.00 : 55bd89: leaq 0x30(%rsp), %rdi
0.00 : 55bd8e: callq 0x55b6f0 <core::ptr::drop_in_place<code_with_type::Buf>>
0.00 : 55bd93: jmp 0x55bdb1 <test_rs+0x101>
0.00 : 55bd95: callq *0xa492f5(%rip) # 0xfa5090 <_DYNAMIC+0x650>
0.00 : 55bd9b: movq %rax, %rcx
0.00 : 55bd9e: movl %edx, %eax
0.00 : 55bda0: movq %rcx, 0x88(%rsp)
0.00 : 55bda8: movl %eax, 0x90(%rsp)
0.00 : 55bdaf: jmp 0x55bd95 <test_rs+0xe5>
0.00 : 55bdb1: addq $0x98, %rsp
0.00 : 55bdb8: retq
3.81 : 55bdb9: movq 0x18(%rsp), %rax
1.91 : 55bdbe: movq %rax, 0x48(%rsp)
3.44 : 55bdc3: cmpq $0x7b, 0x48(%rsp)
1.92 : 55bdc9: je 0x55bde1 <test_rs+0x131>
3.84 : 55bdcb: jmp 0x55bdf9 <test_rs+0x149>
0.00 : 55bdcd: leaq 0xa3fc44(%rip), %rdi # 0xf9ba18
0.00 : 55bdd4: leaq -0x150ffb(%rip), %rax # 0x40ade0 <core::panicking::panic_const::panic_const_add_overflow>
0.00 : 55bddb: callq *%rax
0.00 : 55bddd: jmp 0x55bddf <test_rs+0x12f>
0.00 : 55bddf: ud2
0.00 : 55bde1: movq 0x48(%rsp), %rcx
0.00 : 55bde6: movq %rcx, %rax
0.00 : 55bde9: addq $0x1, %rax
0.00 : 55bded: movq %rax, 0x10(%rsp)
0.00 : 55bdf2: cmpq %rcx, %rax
0.00 : 55bdf5: jb 0x55be21 <test_rs+0x171>
0.00 : 55bdf7: jmp 0x55be15 <test_rs+0x165>
1.92 : 55bdf9: movq 0x48(%rsp), %rdx
2.88 : 55bdfe: movq 0x50(%rsp), %rcx
5.76 : 55be03: movq %rcx, %rax
0.96 : 55be06: addq %rdx, %rax
4.80 : 55be09: movq %rax, 0x8(%rsp)
1.92 : 55be0e: cmpq %rcx, %rax
3.20 : 55be11: jb 0x55be42 <test_rs+0x192>
3.84 : 55be13: jmp 0x55be33 <test_rs+0x183>
0.00 : 55be15: movq 0x10(%rsp), %rax
0.00 : 55be1a: movq %rax, 0x48(%rsp)
0.00 : 55be1f: jmp 0x55bdf9 <test_rs+0x149>
0.00 : 55be21: leaq 0xa3fc08(%rip), %rdi # 0xf9ba30
0.00 : 55be28: leaq -0x15104f(%rip), %rax # 0x40ade0 <core::panicking::panic_const::panic_const_add_overflow>
0.00 : 55be2f: callq *%rax
0.00 : 55be31: jmp 0x55bddf <test_rs+0x12f>
2.35 : 55be33: movq 0x8(%rsp), %rax
9.00 : 55be38: movq %rax, 0x50(%rsp)
0.00 : 55be3d: jmp 0x55bd53 <test_rs+0xa3>
0.00 : 55be42: leaq 0xa3fbff(%rip), %rdi # 0xf9ba48
0.00 : 55be49: leaq -0x151070(%rip), %rax # 0x40ade0 <core::panicking::panic_const::panic_const_add_overflow>
0.00 : 55be50: callq *%rax
0.00 : 55be52: jmp 0x55bddf <test_rs+0x12f>
0.00 : 55be54: callq *0xa49c2e(%rip) # 0xfa5a88 <_DYNAMIC+0x1048>
Percent | Source code & Disassembly of perf for cpu_atom/cycles/Pu (99 samples, percent: local period)
------------------------------------------------------------------------------------------------------------
: 0 0x55bc40 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next>:
3.13 : 55bc40: subq $0x38, %rsp
6.23 : 55bc44: movq %rdi, 0x8(%rsp)
4.63 : 55bc49: movq %rdi, 0x18(%rsp)
5.15 : 55bc4e: movq %rdi, %rax
5.22 : 55bc51: addq $0x4, %rax
1.04 : 55bc55: movq %rdi, 0x28(%rsp)
3.13 : 55bc5a: movq %rax, 0x30(%rsp)
2.09 : 55bc5f: movl (%rdi), %eax
1.04 : 55bc61: cmpl 0x4(%rdi), %eax
1.04 : 55bc64: jb 0x55bc70 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next+0x30>
3.13 : 55bc66: movl $0x0, 0x10(%rsp)
0.00 : 55bc6e: jmp 0x55bca2 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next+0x62>
4.01 : 55bc70: movq 0x8(%rsp), %rax
5.22 : 55bc75: movl (%rax), %edi
4.98 : 55bc77: movl %edi, 0x4(%rsp)
3.06 : 55bc7b: movl %edi, 0x24(%rsp)
4.18 : 55bc7f: movl $0x1, %esi
2.08 : 55bc84: callq 0x55b140 <<u32 as core::iter::range::Step>::forward_unchecked>
4.55 : 55bc89: movq 0x8(%rsp), %rcx
7.23 : 55bc8e: movl %eax, %edx
3.61 : 55bc90: movl 0x4(%rsp), %eax
1.49 : 55bc94: movl %edx, (%rcx)
4.14 : 55bc96: movl %eax, 0x14(%rsp)
6.22 : 55bc9a: movl $0x1, 0x10(%rsp)
4.04 : 55bca2: movl 0x10(%rsp), %eax
4.09 : 55bca6: movl 0x14(%rsp), %edx
3.13 : 55bcaa: addq $0x38, %rsp
2.09 : 55bcae: retq
Percent | Source code & Disassembly of perf for cpu_atom/cycles/Pu (69 samples, percent: local period)
------------------------------------------------------------------------------------------------------------
: 0 0x55b140 <<u32 as core::iter::range::Step>::forward_unchecked>:
12.97 : 55b140: subq $0x28, %rsp
3.00 : 55b144: movl %edi, 0xc(%rsp)
3.00 : 55b148: movl %edi, 0x14(%rsp)
10.49 : 55b14c: movq %rsi, 0x18(%rsp)
8.98 : 55b151: movl %esi, %eax
7.11 : 55b153: movl %eax, 0x10(%rsp)
5.99 : 55b157: movl %eax, 0x24(%rsp)
8.67 : 55b15b: movl 0x10(%rsp), %esi
0.00 : 55b15f: movl 0xc(%rsp), %edi
1.50 : 55b163: leaq 0xa40826(%rip), %rdx # 0xf9b990
4.50 : 55b16a: callq 0x55b190 <core::num::<impl u32>::unchecked_add::precondition_check>
21.79 : 55b16f: movl 0x10(%rsp), %ecx
4.49 : 55b173: movl 0xc(%rsp), %eax
3.00 : 55b177: addl %ecx, %eax
1.50 : 55b179: addq $0x28, %rsp
3.00 : 55b17d: retq
Percent | Source code & Disassembly of perf for cpu_atom/cycles/Pu (65 samples, percent: local period)
------------------------------------------------------------------------------------------------------------
: 0 0x55b190 <core::num::<impl u32>::unchecked_add::precondition_check>:
12.87 : 55b190: subq $0x68, %rsp
8.03 : 55b194: movl %edi, %eax
17.01 : 55b196: movq %rdx, (%rsp)
7.53 : 55b19a: movl %eax, 0x48(%rsp)
2.92 : 55b19e: movl %esi, 0x4c(%rsp)
6.44 : 55b1a2: leaq 0x2922af(%rip), %rcx # 0x7ed458
1.60 : 55b1a9: movq %rcx, 0x50(%rsp)
3.22 : 55b1ae: movq $0xb8, 0x58(%rsp)
3.67 : 55b1b7: movl %eax, %edi
9.65 : 55b1b9: addl %esi, %edi
6.43 : 55b1bb: cmpl %eax, %edi
9.63 : 55b1bd: jb 0x55b1c4 <core::num::<impl u32>::unchecked_add::precondition_check+0x34>
4.56 : 55b1bf: addq $0x68, %rsp
6.44 : 55b1c3: retq
0.00 : 55b1c4: movq (%rsp), %rdx
0.00 : 55b1c8: leaq 0x292289(%rip), %rax # 0x7ed458
0.00 : 55b1cf: movq %rax, 0x38(%rsp)
0.00 : 55b1d4: movq $0xb8, 0x40(%rsp)
0.00 : 55b1dd: leaq 0x38(%rsp), %rax
0.00 : 55b1e2: movq %rax, 0x60(%rsp)
0.00 : 55b1e7: leaq 0x38(%rsp), %rax
0.00 : 55b1ec: movq %rax, 0x8(%rsp)
0.00 : 55b1f1: movq $0x1, 0x10(%rsp)
0.00 : 55b1fa: movq 0x27efdf(%rip), %rcx # 0x7da1e0
0.00 : 55b201: movq 0x27efe0(%rip), %rax # 0x7da1e8
0.00 : 55b208: movq %rcx, 0x28(%rsp)
0.00 : 55b20d: movq %rax, 0x30(%rsp)
0.00 : 55b212: movl $0x8, %eax
0.00 : 55b217: movq %rax, 0x18(%rsp)
0.00 : 55b21c: movq $0x0, 0x20(%rsp)
0.00 : 55b225: leaq 0x8(%rsp), %rdi
0.00 : 55b22a: xorl %esi, %esi
0.00 : 55b22c: callq *0xa4ad96(%rip) # 0xfa5fc8 <_DYNAMIC+0x1588>
Percent | Source code & Disassembly of perf for cpu_core/cycles/Pu (1046 samples, percent: local period)
--------------------------------------------------------------------------------------------------------------
: 0 0x55bc40 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next>:
0.00 : 55bc40: subq $0x38, %rsp
0.00 : 55bc44: movq %rdi, 0x8(%rsp)
0.00 : 55bc49: movq %rdi, 0x18(%rsp)
16.36 : 55bc4e: movq %rdi, %rax
0.00 : 55bc51: addq $0x4, %rax
0.00 : 55bc55: movq %rdi, 0x28(%rsp)
0.00 : 55bc5a: movq %rax, 0x30(%rsp)
7.25 : 55bc5f: movl (%rdi), %eax
0.00 : 55bc61: cmpl 0x4(%rdi), %eax
15.21 : 55bc64: jb 0x55bc70 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next+0x30>
0.00 : 55bc66: movl $0x0, 0x10(%rsp)
0.10 : 55bc6e: jmp 0x55bca2 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next+0x62>
0.00 : 55bc70: movq 0x8(%rsp), %rax
0.10 : 55bc75: movl (%rax), %edi
0.00 : 55bc77: movl %edi, 0x4(%rsp)
14.93 : 55bc7b: movl %edi, 0x24(%rsp)
0.00 : 55bc7f: movl $0x1, %esi
1.05 : 55bc84: callq 0x55b140 <<u32 as core::iter::range::Step>::forward_unchecked>
0.00 : 55bc89: movq 0x8(%rsp), %rcx
0.00 : 55bc8e: movl %eax, %edx
0.00 : 55bc90: movl 0x4(%rsp), %eax
15.51 : 55bc94: movl %edx, (%rcx)
0.00 : 55bc96: movl %eax, 0x14(%rsp)
0.00 : 55bc9a: movl $0x1, 0x10(%rsp)
0.00 : 55bca2: movl 0x10(%rsp), %eax
14.21 : 55bca6: movl 0x14(%rsp), %edx
0.00 : 55bcaa: addq $0x38, %rsp
15.28 : 55bcae: retq
Percent | Source code & Disassembly of perf for cpu_core/cycles/Pu (954 samples, percent: local period)
-------------------------------------------------------------------------------------------------------------
: 0 0x55bcb0 <test_rs>:
0.00 : 55bcb0: subq $0x98, %rsp
0.00 : 55bcb7: movl %edi, 0x2c(%rsp)
0.00 : 55bcbb: movl %edi, 0x84(%rsp)
0.10 : 55bcc2: leaq 0x22b557(%rip), %rsi # 0x787220
0.00 : 55bcc9: leaq 0x60(%rsp), %rdi
0.00 : 55bcce: movl $0x4, %edx
0.00 : 55bcd3: callq 0x55baf0 <<alloc::string::String as core::convert::From<&str>>::from>
0.10 : 55bcd8: jmp 0x55bce0 <test_rs+0x30>
0.00 : 55bcda: callq *0xa493b0(%rip) # 0xfa5090 <_DYNAMIC+0x650>
0.00 : 55bce0: movl 0x2c(%rsp), %esi
0.00 : 55bce4: movq $0x0, 0x48(%rsp)
0.11 : 55bced: movq 0x70(%rsp), %rax
0.21 : 55bcf2: movq %rax, 0x40(%rsp)
2.20 : 55bcf7: movups 0x60(%rsp), %xmm0
0.73 : 55bcfc: movaps %xmm0, 0x30(%rsp)
0.00 : 55bd01: movq $0x0, 0x50(%rsp)
0.00 : 55bd0a: movl $0x1, %edi
0.00 : 55bd0f: callq 0x55bae0 <<I as core::iter::traits::collect::IntoIterator>::into_iter>
0.00 : 55bd14: movl %edx, 0x24(%rsp)
0.00 : 55bd18: movl %eax, 0x28(%rsp)
0.10 : 55bd1c: jmp 0x55bd40 <test_rs+0x90>
0.00 : 55bd1e: leaq 0x30(%rsp), %rdi
0.00 : 55bd23: callq 0x55b6f0 <core::ptr::drop_in_place<code_with_type::Buf>>
0.00 : 55bd28: jmp 0x55bd95 <test_rs+0xe5>
0.00 : 55bd2a: movq %rax, %rcx
0.00 : 55bd2d: movl %edx, %eax
0.00 : 55bd2f: movq %rcx, 0x88(%rsp)
0.00 : 55bd37: movl %eax, 0x90(%rsp)
0.00 : 55bd3e: jmp 0x55bd1e <test_rs+0x6e>
0.00 : 55bd40: movl 0x24(%rsp), %eax
0.00 : 55bd44: movl 0x28(%rsp), %ecx
0.00 : 55bd48: movl %ecx, 0x7c(%rsp)
0.21 : 55bd4c: movl %eax, 0x80(%rsp)
0.00 : 55bd53: leaq 0x7c(%rsp), %rdi
11.96 : 55bd58: callq 0x55b820 <core::iter::range::<impl core::iter::traits::iterator::Iterator for core::ops::range::Range<A>>::next>
0.00 : 55bd5d: movl %eax, 0x20(%rsp)
1.36 : 55bd61: jmp 0x55bd63 <test_rs+0xb3>
0.00 : 55bd63: movl 0x20(%rsp), %eax
0.00 : 55bd67: movl %eax, %eax
0.00 : 55bd69: testq $0x1, %rax
0.10 : 55bd6f: je 0x55bd89 <test_rs+0xd9>
16.76 : 55bd71: movq 0x48(%rsp), %rcx
0.00 : 55bd76: movq %rcx, %rax
0.00 : 55bd79: addq $0x1, %rax
0.00 : 55bd7d: movq %rax, 0x18(%rsp)
2.20 : 55bd82: cmpq %rcx, %rax
0.00 : 55bd85: jb 0x55bdcd <test_rs+0x11d>
3.04 : 55bd87: jmp 0x55bdb9 <test_rs+0x109>
0.00 : 55bd89: leaq 0x30(%rsp), %rdi
0.00 : 55bd8e: callq 0x55b6f0 <core::ptr::drop_in_place<code_with_type::Buf>>
0.00 : 55bd93: jmp 0x55bdb1 <test_rs+0x101>
0.00 : 55bd95: callq *0xa492f5(%rip) # 0xfa5090 <_DYNAMIC+0x650>
0.00 : 55bd9b: movq %rax, %rcx
0.00 : 55bd9e: movl %edx, %eax
0.00 : 55bda0: movq %rcx, 0x88(%rsp)
0.00 : 55bda8: movl %eax, 0x90(%rsp)
0.00 : 55bdaf: jmp 0x55bd95 <test_rs+0xe5>
0.00 : 55bdb1: addq $0x98, %rsp
0.21 : 55bdb8: retq
0.10 : 55bdb9: movq 0x18(%rsp), %rax
0.00 : 55bdbe: movq %rax, 0x48(%rsp)
0.00 : 55bdc3: cmpq $0x7b, 0x48(%rsp)
17.41 : 55bdc9: je 0x55bde1 <test_rs+0x131>
6.19 : 55bdcb: jmp 0x55bdf9 <test_rs+0x149>
0.00 : 55bdcd: leaq 0xa3fc44(%rip), %rdi # 0xf9ba18
0.00 : 55bdd4: leaq -0x150ffb(%rip), %rax # 0x40ade0 <core::panicking::panic_const::panic_const_add_overflow>
0.00 : 55bddb: callq *%rax
0.00 : 55bddd: jmp 0x55bddf <test_rs+0x12f>
0.00 : 55bddf: ud2
0.00 : 55bde1: movq 0x48(%rsp), %rcx
0.00 : 55bde6: movq %rcx, %rax
0.00 : 55bde9: addq $0x1, %rax
0.00 : 55bded: movq %rax, 0x10(%rsp)
0.00 : 55bdf2: cmpq %rcx, %rax
0.00 : 55bdf5: jb 0x55be21 <test_rs+0x171>
0.00 : 55bdf7: jmp 0x55be15 <test_rs+0x165>
0.00 : 55bdf9: movq 0x48(%rsp), %rdx
0.00 : 55bdfe: movq 0x50(%rsp), %rcx
0.00 : 55be03: movq %rcx, %rax
16.66 : 55be06: addq %rdx, %rax
0.00 : 55be09: movq %rax, 0x8(%rsp)
0.00 : 55be0e: cmpq %rcx, %rax
0.00 : 55be11: jb 0x55be42 <test_rs+0x192>
19.06 : 55be13: jmp 0x55be33 <test_rs+0x183>
0.00 : 55be15: movq 0x10(%rsp), %rax
0.00 : 55be1a: movq %rax, 0x48(%rsp)
0.00 : 55be1f: jmp 0x55bdf9 <test_rs+0x149>
0.00 : 55be21: leaq 0xa3fc08(%rip), %rdi # 0xf9ba30
0.00 : 55be28: leaq -0x15104f(%rip), %rax # 0x40ade0 <core::panicking::panic_const::panic_const_add_overflow>
0.00 : 55be2f: callq *%rax
0.00 : 55be31: jmp 0x55bddf <test_rs+0x12f>
0.00 : 55be33: movq 0x8(%rsp), %rax
0.00 : 55be38: movq %rax, 0x50(%rsp)
1.15 : 55be3d: jmp 0x55bd53 <test_rs+0xa3>
0.00 : 55be42: leaq 0xa3fbff(%rip), %rdi # 0xf9ba48
0.00 : 55be49: leaq -0x151070(%rip), %rax # 0x40ade0 <core::panicking::panic_const::panic_const_add_overflow>
0.00 : 55be50: callq *%rax
0.00 : 55be52: jmp 0x55bddf <test_rs+0x12f>
0.00 : 55be54: callq *0xa49c2e(%rip) # 0xfa5a88 <_DYNAMIC+0x1048>
Percent | Source code & Disassembly of perf for cpu_core/cycles/Pu (522 samples, percent: local period)
-------------------------------------------------------------------------------------------------------------
: 0 0x55b190 <core::num::<impl u32>::unchecked_add::precondition_check>:
0.00 : 55b190: subq $0x68, %rsp
0.00 : 55b194: movl %edi, %eax
0.00 : 55b196: movq %rdx, (%rsp)
33.53 : 55b19a: movl %eax, 0x48(%rsp)
0.00 : 55b19e: movl %esi, 0x4c(%rsp)
0.00 : 55b1a2: leaq 0x2922af(%rip), %rcx # 0x7ed458
0.00 : 55b1a9: movq %rcx, 0x50(%rsp)
33.92 : 55b1ae: movq $0xb8, 0x58(%rsp)
0.00 : 55b1b7: movl %eax, %edi
0.00 : 55b1b9: addl %esi, %edi
0.00 : 55b1bb: cmpl %eax, %edi
0.00 : 55b1bd: jb 0x55b1c4 <core::num::<impl u32>::unchecked_add::precondition_check+0x34>
3.43 : 55b1bf: addq $0x68, %rsp
29.12 : 55b1c3: retq
0.00 : 55b1c4: movq (%rsp), %rdx
0.00 : 55b1c8: leaq 0x292289(%rip), %rax # 0x7ed458
0.00 : 55b1cf: movq %rax, 0x38(%rsp)
0.00 : 55b1d4: movq $0xb8, 0x40(%rsp)
0.00 : 55b1dd: leaq 0x38(%rsp), %rax
0.00 : 55b1e2: movq %rax, 0x60(%rsp)
0.00 : 55b1e7: leaq 0x38(%rsp), %rax
0.00 : 55b1ec: movq %rax, 0x8(%rsp)
0.00 : 55b1f1: movq $0x1, 0x10(%rsp)
0.00 : 55b1fa: movq 0x27efdf(%rip), %rcx # 0x7da1e0
0.00 : 55b201: movq 0x27efe0(%rip), %rax # 0x7da1e8
0.00 : 55b208: movq %rcx, 0x28(%rsp)
0.00 : 55b20d: movq %rax, 0x30(%rsp)
0.00 : 55b212: movl $0x8, %eax
0.00 : 55b217: movq %rax, 0x18(%rsp)
0.00 : 55b21c: movq $0x0, 0x20(%rsp)
0.00 : 55b225: leaq 0x8(%rsp), %rdi
0.00 : 55b22a: xorl %esi, %esi
0.00 : 55b22c: callq *0xa4ad96(%rip) # 0xfa5fc8 <_DYNAMIC+0x1588>
Percent | Source code & Disassembly of perf for cpu_core/cycles/Pu (471 samples, percent: local period)
-------------------------------------------------------------------------------------------------------------
: 0 0x55b140 <<u32 as core::iter::range::Step>::forward_unchecked>:
0.00 : 55b140: subq $0x28, %rsp
0.00 : 55b144: movl %edi, 0xc(%rsp)
0.00 : 55b148: movl %edi, 0x14(%rsp)
24.00 : 55b14c: movq %rsi, 0x18(%rsp)
0.00 : 55b151: movl %esi, %eax
0.00 : 55b153: movl %eax, 0x10(%rsp)
0.00 : 55b157: movl %eax, 0x24(%rsp)
13.42 : 55b15b: movl 0x10(%rsp), %esi
0.00 : 55b15f: movl 0xc(%rsp), %edi
0.00 : 55b163: leaq 0xa40826(%rip), %rdx # 0xf9b990
2.33 : 55b16a: callq 0x55b190 <core::num::<impl u32>::unchecked_add::precondition_check>
0.00 : 55b16f: movl 0x10(%rsp), %ecx
0.00 : 55b173: movl 0xc(%rsp), %eax
0.00 : 55b177: addl %ecx, %eax
30.61 : 55b179: addq $0x28, %rsp
29.63 : 55b17d: retq
Percent | Source code & Disassembly of perf for cpu_core/cycles/Pu (405 samples, percent: local period)
-------------------------------------------------------------------------------------------------------------
: 0 0x55b820 <core::iter::range::<impl core::iter::traits::iterator::Iterator for core::ops::range::Range<A>>::next>:
0.00 : 55b820: pushq %rax
0.00 : 55b821: movq %rdi, (%rsp)
58.32 : 55b825: callq 0x55bc40 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next>
0.00 : 55b82a: popq %rcx
41.68 : 55b82b: retq
Pipe perf annotate test
Pipe annotate [Failed: missing target data type]
Percent | Source code & Disassembly of perf for cpu_core/cycles/Pu (1159 samples, percent: local period)
--------------------------------------------------------------------------------------------------------------
: 0 0x55bc40 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next>:
0.00 : 55bc40: subq $0x38, %rsp
0.00 : 55bc44: movq %rdi, 0x8(%rsp)
0.00 : 55bc49: movq %rdi, 0x18(%rsp)
13.73 : 55bc4e: movq %rdi, %rax
0.00 : 55bc51: addq $0x4, %rax
0.00 : 55bc55: movq %rdi, 0x28(%rsp)
0.00 : 55bc5a: movq %rax, 0x30(%rsp)
8.03 : 55bc5f: movl (%rdi), %eax
0.00 : 55bc61: cmpl 0x4(%rdi), %eax
13.37 : 55bc64: jb 0x55bc70 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next+0x30>
0.00 : 55bc66: movl $0x0, 0x10(%rsp)
0.35 : 55bc6e: jmp 0x55bca2 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next+0x62>
0.00 : 55bc70: movq 0x8(%rsp), %rax
0.00 : 55bc75: movl (%rax), %edi
0.00 : 55bc77: movl %edi, 0x4(%rsp)
13.69 : 55bc7b: movl %edi, 0x24(%rsp)
0.00 : 55bc7f: movl $0x1, %esi
1.47 : 55bc84: callq 0x55b140 <<u32 as core::iter::range::Step>::forward_unchecked>
0.00 : 55bc89: movq 0x8(%rsp), %rcx
0.00 : 55bc8e: movl %eax, %edx
0.00 : 55bc90: movl 0x4(%rsp), %eax
16.56 : 55bc94: movl %edx, (%rcx)
0.00 : 55bc96: movl %eax, 0x14(%rsp)
0.00 : 55bc9a: movl $0x1, 0x10(%rsp)
0.00 : 55bca2: movl 0x10(%rsp), %eax
16.32 : 55bca6: movl 0x14(%rsp), %edx
0.00 : 55bcaa: addq $0x38, %rsp
16.48 : 55bcae: retq
Percent | Source code & Disassembly of perf for cpu_core/cycles/Pu (1034 samples, percent: local period)
--------------------------------------------------------------------------------------------------------------
: 0 0x55bcb0 <test_rs>:
0.00 : 55bcb0: subq $0x98, %rsp
0.00 : 55bcb7: movl %edi, 0x2c(%rsp)
0.00 : 55bcbb: movl %edi, 0x84(%rsp)
0.29 : 55bcc2: leaq 0x22b557(%rip), %rsi # 0x787220
0.00 : 55bcc9: leaq 0x60(%rsp), %rdi
0.00 : 55bcce: movl $0x4, %edx
0.29 : 55bcd3: callq 0x55baf0 <<alloc::string::String as core::convert::From<&str>>::from>
0.10 : 55bcd8: jmp 0x55bce0 <test_rs+0x30>
0.00 : 55bcda: callq *0xa493b0(%rip) # 0xfa5090 <_DYNAMIC+0x650>
0.00 : 55bce0: movl 0x2c(%rsp), %esi
0.00 : 55bce4: movq $0x0, 0x48(%rsp)
0.00 : 55bced: movq 0x70(%rsp), %rax
0.00 : 55bcf2: movq %rax, 0x40(%rsp)
2.23 : 55bcf7: movups 0x60(%rsp), %xmm0
0.58 : 55bcfc: movaps %xmm0, 0x30(%rsp)
0.00 : 55bd01: movq $0x0, 0x50(%rsp)
0.00 : 55bd0a: movl $0x1, %edi
0.19 : 55bd0f: callq 0x55bae0 <<I as core::iter::traits::collect::IntoIterator>::into_iter>
0.00 : 55bd14: movl %edx, 0x24(%rsp)
0.00 : 55bd18: movl %eax, 0x28(%rsp)
0.10 : 55bd1c: jmp 0x55bd40 <test_rs+0x90>
0.00 : 55bd1e: leaq 0x30(%rsp), %rdi
0.00 : 55bd23: callq 0x55b6f0 <core::ptr::drop_in_place<code_with_type::Buf>>
0.00 : 55bd28: jmp 0x55bd95 <test_rs+0xe5>
0.00 : 55bd2a: movq %rax, %rcx
0.00 : 55bd2d: movl %edx, %eax
0.00 : 55bd2f: movq %rcx, 0x88(%rsp)
0.00 : 55bd37: movl %eax, 0x90(%rsp)
0.00 : 55bd3e: jmp 0x55bd1e <test_rs+0x6e>
0.00 : 55bd40: movl 0x24(%rsp), %eax
0.00 : 55bd44: movl 0x28(%rsp), %ecx
0.00 : 55bd48: movl %ecx, 0x7c(%rsp)
0.10 : 55bd4c: movl %eax, 0x80(%rsp)
0.00 : 55bd53: leaq 0x7c(%rsp), %rdi
11.89 : 55bd58: callq 0x55b820 <core::iter::range::<impl core::iter::traits::iterator::Iterator for core::ops::range::Range<A>>::next>
0.00 : 55bd5d: movl %eax, 0x20(%rsp)
1.84 : 55bd61: jmp 0x55bd63 <test_rs+0xb3>
0.00 : 55bd63: movl 0x20(%rsp), %eax
0.00 : 55bd67: movl %eax, %eax
0.00 : 55bd69: testq $0x1, %rax
0.19 : 55bd6f: je 0x55bd89 <test_rs+0xd9>
16.73 : 55bd71: movq 0x48(%rsp), %rcx
0.00 : 55bd76: movq %rcx, %rax
0.00 : 55bd79: addq $0x1, %rax
0.00 : 55bd7d: movq %rax, 0x18(%rsp)
2.03 : 55bd82: cmpq %rcx, %rax
0.00 : 55bd85: jb 0x55bdcd <test_rs+0x11d>
3.48 : 55bd87: jmp 0x55bdb9 <test_rs+0x109>
0.00 : 55bd89: leaq 0x30(%rsp), %rdi
0.48 : 55bd8e: callq 0x55b6f0 <core::ptr::drop_in_place<code_with_type::Buf>>
0.10 : 55bd93: jmp 0x55bdb1 <test_rs+0x101>
0.00 : 55bd95: callq *0xa492f5(%rip) # 0xfa5090 <_DYNAMIC+0x650>
0.00 : 55bd9b: movq %rax, %rcx
0.00 : 55bd9e: movl %edx, %eax
0.00 : 55bda0: movq %rcx, 0x88(%rsp)
0.00 : 55bda8: movl %eax, 0x90(%rsp)
0.00 : 55bdaf: jmp 0x55bd95 <test_rs+0xe5>
0.00 : 55bdb1: addq $0x98, %rsp
0.19 : 55bdb8: retq
0.00 : 55bdb9: movq 0x18(%rsp), %rax
0.00 : 55bdbe: movq %rax, 0x48(%rsp)
0.00 : 55bdc3: cmpq $0x7b, 0x48(%rsp)
14.19 : 55bdc9: je 0x55bde1 <test_rs+0x131>
6.68 : 55bdcb: jmp 0x55bdf9 <test_rs+0x149>
0.00 : 55bdcd: leaq 0xa3fc44(%rip), %rdi # 0xf9ba18
0.00 : 55bdd4: leaq -0x150ffb(%rip), %rax # 0x40ade0 <core::panicking::panic_const::panic_const_add_overflow>
0.00 : 55bddb: callq *%rax
0.00 : 55bddd: jmp 0x55bddf <test_rs+0x12f>
0.00 : 55bddf: ud2
0.00 : 55bde1: movq 0x48(%rsp), %rcx
0.00 : 55bde6: movq %rcx, %rax
0.00 : 55bde9: addq $0x1, %rax
0.00 : 55bded: movq %rax, 0x10(%rsp)
0.00 : 55bdf2: cmpq %rcx, %rax
0.00 : 55bdf5: jb 0x55be21 <test_rs+0x171>
0.00 : 55bdf7: jmp 0x55be15 <test_rs+0x165>
0.00 : 55bdf9: movq 0x48(%rsp), %rdx
0.00 : 55bdfe: movq 0x50(%rsp), %rcx
0.00 : 55be03: movq %rcx, %rax
17.03 : 55be06: addq %rdx, %rax
0.00 : 55be09: movq %rax, 0x8(%rsp)
0.00 : 55be0e: cmpq %rcx, %rax
0.00 : 55be11: jb 0x55be42 <test_rs+0x192>
19.93 : 55be13: jmp 0x55be33 <test_rs+0x183>
0.00 : 55be15: movq 0x10(%rsp), %rax
0.00 : 55be1a: movq %rax, 0x48(%rsp)
0.00 : 55be1f: jmp 0x55bdf9 <test_rs+0x149>
0.00 : 55be21: leaq 0xa3fc08(%rip), %rdi # 0xf9ba30
0.00 : 55be28: leaq -0x15104f(%rip), %rax # 0x40ade0 <core::panicking::panic_const::panic_const_add_overflow>
0.00 : 55be2f: callq *%rax
0.00 : 55be31: jmp 0x55bddf <test_rs+0x12f>
0.00 : 55be33: movq 0x8(%rsp), %rax
0.00 : 55be38: movq %rax, 0x50(%rsp)
1.35 : 55be3d: jmp 0x55bd53 <test_rs+0xa3>
0.00 : 55be42: leaq 0xa3fbff(%rip), %rdi # 0xf9ba48
0.00 : 55be49: leaq -0x151070(%rip), %rax # 0x40ade0 <core::panicking::panic_const::panic_const_add_overflow>
0.00 : 55be50: callq *%rax
0.00 : 55be52: jmp 0x55bddf <test_rs+0x12f>
0.00 : 55be54: callq *0xa49c2e(%rip) # 0xfa5a88 <_DYNAMIC+0x1048>
Percent | Source code & Disassembly of perf for cpu_core/cycles/Pu (576 samples, percent: local period)
-------------------------------------------------------------------------------------------------------------
: 0 0x55b190 <core::num::<impl u32>::unchecked_add::precondition_check>:
0.00 : 55b190: subq $0x68, %rsp
0.00 : 55b194: movl %edi, %eax
0.00 : 55b196: movq %rdx, (%rsp)
34.20 : 55b19a: movl %eax, 0x48(%rsp)
0.00 : 55b19e: movl %esi, 0x4c(%rsp)
0.00 : 55b1a2: leaq 0x2922af(%rip), %rcx # 0x7ed458
0.00 : 55b1a9: movq %rcx, 0x50(%rsp)
31.25 : 55b1ae: movq $0xb8, 0x58(%rsp)
0.00 : 55b1b7: movl %eax, %edi
0.00 : 55b1b9: addl %esi, %edi
0.00 : 55b1bb: cmpl %eax, %edi
0.00 : 55b1bd: jb 0x55b1c4 <core::num::<impl u32>::unchecked_add::precondition_check+0x34>
3.82 : 55b1bf: addq $0x68, %rsp
30.73 : 55b1c3: retq
0.00 : 55b1c4: movq (%rsp), %rdx
0.00 : 55b1c8: leaq 0x292289(%rip), %rax # 0x7ed458
0.00 : 55b1cf: movq %rax, 0x38(%rsp)
0.00 : 55b1d4: movq $0xb8, 0x40(%rsp)
0.00 : 55b1dd: leaq 0x38(%rsp), %rax
0.00 : 55b1e2: movq %rax, 0x60(%rsp)
0.00 : 55b1e7: leaq 0x38(%rsp), %rax
0.00 : 55b1ec: movq %rax, 0x8(%rsp)
0.00 : 55b1f1: movq $0x1, 0x10(%rsp)
0.00 : 55b1fa: movq 0x27efdf(%rip), %rcx # 0x7da1e0
0.00 : 55b201: movq 0x27efe0(%rip), %rax # 0x7da1e8
0.00 : 55b208: movq %rcx, 0x28(%rsp)
0.00 : 55b20d: movq %rax, 0x30(%rsp)
0.00 : 55b212: movl $0x8, %eax
0.00 : 55b217: movq %rax, 0x18(%rsp)
0.00 : 55b21c: movq $0x0, 0x20(%rsp)
0.00 : 55b225: leaq 0x8(%rsp), %rdi
0.00 : 55b22a: xorl %esi, %esi
0.00 : 55b22c: callq *0xa4ad96(%rip) # 0xfa5fc8 <_DYNAMIC+0x1588>
Percent | Source code & Disassembly of perf for cpu_core/cycles/Pu (554 samples, percent: local period)
-------------------------------------------------------------------------------------------------------------
: 0 0x55b140 <<u32 as core::iter::range::Step>::forward_unchecked>:
0.00 : 55b140: subq $0x28, %rsp
0.00 : 55b144: movl %edi, 0xc(%rsp)
0.00 : 55b148: movl %edi, 0x14(%rsp)
22.19 : 55b14c: movq %rsi, 0x18(%rsp)
0.00 : 55b151: movl %esi, %eax
0.00 : 55b153: movl %eax, 0x10(%rsp)
0.00 : 55b157: movl %eax, 0x24(%rsp)
14.28 : 55b15b: movl 0x10(%rsp), %esi
0.00 : 55b15f: movl 0xc(%rsp), %edi
0.00 : 55b163: leaq 0xa40826(%rip), %rdx # 0xf9b990
3.98 : 55b16a: callq 0x55b190 <core::num::<impl u32>::unchecked_add::precondition_check>
0.00 : 55b16f: movl 0x10(%rsp), %ecx
0.00 : 55b173: movl 0xc(%rsp), %eax
0.00 : 55b177: addl %ecx, %eax
30.49 : 55b179: addq $0x28, %rsp
29.07 : 55b17d: retq
Percent | Source code & Disassembly of perf for cpu_core/cycles/Pu (429 samples, percent: local period)
-------------------------------------------------------------------------------------------------------------
: 0 0x55b820 <core::iter::range::<impl core::iter::traits::iterator::Iterator for core::ops::range::Range<A>>::next>:
0.00 : 55b820: pushq %rax
0.00 : 55b821: movq %rdi, (%rsp)
60.17 : 55b825: callq 0x55bc40 <<core::ops::range::Range<T> as core::iter::range::RangeIteratorImpl>::spec_next>
0.23 : 55b82a: popq %rcx
39.60 : 55b82b: retq
---- end(-1) ----
83: perf data type profiling tests : FAILED!
⬢ [acme@toolbx perf-tools-next]$
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [RFC PATCH v1] perf tests: Test annotate with data type profiling and rust
2026-01-27 17:00 ` [RFC PATCH v1] perf tests: Test annotate with data type profiling and rust Arnaldo Carvalho de Melo
@ 2026-01-30 8:44 ` Dmitry Dolgov
2026-02-01 10:33 ` Dmitry Dolgov
0 siblings, 1 reply; 7+ messages in thread
From: Dmitry Dolgov @ 2026-01-30 8:44 UTC (permalink / raw)
To: Arnaldo Carvalho de Melo
Cc: linux-perf-users, Namhyung Kim, Linux Kernel Mailing List
> On Tue, Jan 27, 2026 at 02:00:07PM -0300, Arnaldo Carvalho de Melo wrote:
> I think this should be split into at least two patchs, one introducing
> the new workload and then the new perf test.
Makes sense, will do.
> Isn't that 'perf record' just using cycles? Wouldn't it be better to use
> 'perf mem record'?
'perf record' was an example I've picked up from LPC 2025 talk about
this, but I can switch it to 'perf mem record' as well.
> I don't have rust installed in this container, in this case we need some
> sort of capability query to tell the user that rust is needed for this
> feature to be present in perf.
Right, after some experimenting I think it would be a good idea to add a
new feature into tools/build/feature called "rust" and hide the workload
and the test behind it.
> I see, it doesn't work if I use 'make O=/some/directory/ -C tools/perf'
Ouch, I have to admit I haven't tested it that way, need to figure out
what's different in this case.
> acme@number:~/git/perf-tools-next$ sudo tools/perf/perf annotate -f --code-with-type
>
> Again, [Type] is there, but no types:
>
> Samples: 4K of event 'ibs_op/ldlat=0/', 4000 Hz, Event count (approx.): 5414001952
> test_rs /home/acme/git/perf-tools-next/tools/perf/perf [Percent: local period] [Type]
> 0.14 │ sub $0x98,%rsp
> │ mov %edi,0x2c(%rsp)
> │ mov %edi,0x84(%rsp)
> │ lea _IO_stdin_used+0x20,%rsi
> 0.07 │ lea 0x60(%rsp),%rdi
> 0.07 │ mov $0x4,%edx
> │ → call <alloc::string::String as core::convert::From<&str>>::f
>
> I'm not managing to reproduce your results, can you please elaborate
> some more?
That's most unexpected for me, the test output I see looks like this:
: 54 for _ in 1..count {
0.00 : 5423b8: mov %ecx,0x7c(%rsp)
0.00 : 5423bc: mov %eax,0x80(%rsp)
0.00 : 5423c3: lea 0x7c(%rsp),%rdi
12.28 : 5423c8: call 541f00 <core::iter::range::<impl core::iter::traits::iterator::Iterator for core::ops::range::Range<A>>::next>
0.00 : 5423cd: mov %eax,0x20(%rsp)
10.21 : 5423d1: jmp 5423d3 <test_rs+0xb3>
0.00 : 5423d3: mov 0x20(%rsp),%eax
0.00 : 5423d7: mov %eax,%eax
0.00 : 5423d9: test $0x1,%rax
0.29 : 5423df: je 5423f6 <test_rs+0xd6>
: 65 b.data1 += 1;
13.69 : 5423e1: mov 0x48(%rsp),%rax # data-type: struct Buf +0x18 (data1)
0.00 : 5423e6: add $0x1,%rax
0.00 : 5423ea: mov %rax,0x18(%rsp) # data-type: u64 +0
0.00 : 5423ef: setb %al
11.17 : 5423f2: jb 54243a <test_rs+0x11a>
0.00 : 5423f4: jmp 542426 <test_rs+0x106>
0.00 : 5423f6: lea 0x30(%rsp),%rdi
: 73 }
Where '# data-type: struct Buf' is as far as I understand a manifestation of
data type profiling succeeding. But my environment has slightly different
version of rust -- I need to figure out what's going on here, thanks.
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [RFC PATCH v1] perf tests: Test annotate with data type profiling and rust
2026-01-30 8:44 ` Dmitry Dolgov
@ 2026-02-01 10:33 ` Dmitry Dolgov
2026-02-03 14:45 ` Arnaldo Carvalho de Melo
0 siblings, 1 reply; 7+ messages in thread
From: Dmitry Dolgov @ 2026-02-01 10:33 UTC (permalink / raw)
To: Arnaldo Carvalho de Melo
Cc: linux-perf-users, Namhyung Kim, Linux Kernel Mailing List
> On Fri, Jan 30, 2026 at 09:44:47AM +0100, Dmitry Dolgov wrote:
> > I'm not managing to reproduce your results, can you please elaborate
> > some more?
>
> That's most unexpected for me, the test output I see looks like this:
>
> : 54 for _ in 1..count {
> 0.00 : 5423b8: mov %ecx,0x7c(%rsp)
> 0.00 : 5423bc: mov %eax,0x80(%rsp)
> 0.00 : 5423c3: lea 0x7c(%rsp),%rdi
> 12.28 : 5423c8: call 541f00 <core::iter::range::<impl core::iter::traits::iterator::Iterator for core::ops::range::Range<A>>::next>
> 0.00 : 5423cd: mov %eax,0x20(%rsp)
> 10.21 : 5423d1: jmp 5423d3 <test_rs+0xb3>
> 0.00 : 5423d3: mov 0x20(%rsp),%eax
> 0.00 : 5423d7: mov %eax,%eax
> 0.00 : 5423d9: test $0x1,%rax
> 0.29 : 5423df: je 5423f6 <test_rs+0xd6>
> : 65 b.data1 += 1;
> 13.69 : 5423e1: mov 0x48(%rsp),%rax # data-type: struct Buf +0x18 (data1)
> 0.00 : 5423e6: add $0x1,%rax
> 0.00 : 5423ea: mov %rax,0x18(%rsp) # data-type: u64 +0
> 0.00 : 5423ef: setb %al
> 11.17 : 5423f2: jb 54243a <test_rs+0x11a>
> 0.00 : 5423f4: jmp 542426 <test_rs+0x106>
> 0.00 : 5423f6: lea 0x30(%rsp),%rdi
> : 73 }
>
> Where '# data-type: struct Buf' is as far as I understand a manifestation of
> data type profiling succeeding. But my environment has slightly different
> version of rust -- I need to figure out what's going on here, thanks.
I've being testing this in the same way, but without any rust code --
just on a simple datasym workload. Surprisingly, it seems that
'data-type: buf' marks are disappearing even in this case starting from:
c31040085914f1188720073baa43d1483693c0a3 (perf dwarf-regs: Clean up x86
dwarf_regnum code)
I'm not sure yet why this is happening, but don't see anything in the
commit mentioning that -- since there are no code-with-type tests yet,
maybe an unexpected side effect?
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [RFC PATCH v1] perf tests: Test annotate with data type profiling and rust
2026-02-01 10:33 ` Dmitry Dolgov
@ 2026-02-03 14:45 ` Arnaldo Carvalho de Melo
2026-02-03 18:15 ` Ian Rogers
0 siblings, 1 reply; 7+ messages in thread
From: Arnaldo Carvalho de Melo @ 2026-02-03 14:45 UTC (permalink / raw)
To: Ian Rogers, Dmitry Dolgov
Cc: linux-perf-users, Namhyung Kim, Linux Kernel Mailing List
On Sun, Feb 01, 2026 at 11:33:54AM +0100, Dmitry Dolgov wrote:
> > On Fri, Jan 30, 2026 at 09:44:47AM +0100, Dmitry Dolgov wrote:
> > > I'm not managing to reproduce your results, can you please elaborate
> > > some more?
> >
> > That's most unexpected for me, the test output I see looks like this:
> >
> > : 54 for _ in 1..count {
> > 0.00 : 5423b8: mov %ecx,0x7c(%rsp)
> > 0.00 : 5423bc: mov %eax,0x80(%rsp)
> > 0.00 : 5423c3: lea 0x7c(%rsp),%rdi
> > 12.28 : 5423c8: call 541f00 <core::iter::range::<impl core::iter::traits::iterator::Iterator for core::ops::range::Range<A>>::next>
> > 0.00 : 5423cd: mov %eax,0x20(%rsp)
> > 10.21 : 5423d1: jmp 5423d3 <test_rs+0xb3>
> > 0.00 : 5423d3: mov 0x20(%rsp),%eax
> > 0.00 : 5423d7: mov %eax,%eax
> > 0.00 : 5423d9: test $0x1,%rax
> > 0.29 : 5423df: je 5423f6 <test_rs+0xd6>
> > : 65 b.data1 += 1;
> > 13.69 : 5423e1: mov 0x48(%rsp),%rax # data-type: struct Buf +0x18 (data1)
> > 0.00 : 5423e6: add $0x1,%rax
> > 0.00 : 5423ea: mov %rax,0x18(%rsp) # data-type: u64 +0
> > 0.00 : 5423ef: setb %al
> > 11.17 : 5423f2: jb 54243a <test_rs+0x11a>
> > 0.00 : 5423f4: jmp 542426 <test_rs+0x106>
> > 0.00 : 5423f6: lea 0x30(%rsp),%rdi
> > : 73 }
> >
> > Where '# data-type: struct Buf' is as far as I understand a manifestation of
> > data type profiling succeeding. But my environment has slightly different
> > version of rust -- I need to figure out what's going on here, thanks.
>
> I've being testing this in the same way, but without any rust code --
> just on a simple datasym workload. Surprisingly, it seems that
> 'data-type: buf' marks are disappearing even in this case starting from:
>
> c31040085914f1188720073baa43d1483693c0a3 (perf dwarf-regs: Clean up x86
> dwarf_regnum code)
> I'm not sure yet why this is happening, but don't see anything in the
> commit mentioning that -- since there are no code-with-type tests yet,
> maybe an unexpected side effect?
Ian, can you please take a look at this?
Cheers,
- Arnaldo
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [RFC PATCH v1] perf tests: Test annotate with data type profiling and rust
2026-02-03 14:45 ` Arnaldo Carvalho de Melo
@ 2026-02-03 18:15 ` Ian Rogers
2026-02-04 21:52 ` Namhyung Kim
0 siblings, 1 reply; 7+ messages in thread
From: Ian Rogers @ 2026-02-03 18:15 UTC (permalink / raw)
To: Arnaldo Carvalho de Melo, Namhyung Kim, Dmitry Dolgov
Cc: linux-perf-users, Linux Kernel Mailing List
On Tue, Feb 3, 2026 at 6:45 AM Arnaldo Carvalho de Melo <acme@kernel.org> wrote:
>
> On Sun, Feb 01, 2026 at 11:33:54AM +0100, Dmitry Dolgov wrote:
> > > On Fri, Jan 30, 2026 at 09:44:47AM +0100, Dmitry Dolgov wrote:
> > > > I'm not managing to reproduce your results, can you please elaborate
> > > > some more?
> > >
> > > That's most unexpected for me, the test output I see looks like this:
> > >
> > > : 54 for _ in 1..count {
> > > 0.00 : 5423b8: mov %ecx,0x7c(%rsp)
> > > 0.00 : 5423bc: mov %eax,0x80(%rsp)
> > > 0.00 : 5423c3: lea 0x7c(%rsp),%rdi
> > > 12.28 : 5423c8: call 541f00 <core::iter::range::<impl core::iter::traits::iterator::Iterator for core::ops::range::Range<A>>::next>
> > > 0.00 : 5423cd: mov %eax,0x20(%rsp)
> > > 10.21 : 5423d1: jmp 5423d3 <test_rs+0xb3>
> > > 0.00 : 5423d3: mov 0x20(%rsp),%eax
> > > 0.00 : 5423d7: mov %eax,%eax
> > > 0.00 : 5423d9: test $0x1,%rax
> > > 0.29 : 5423df: je 5423f6 <test_rs+0xd6>
> > > : 65 b.data1 += 1;
> > > 13.69 : 5423e1: mov 0x48(%rsp),%rax # data-type: struct Buf +0x18 (data1)
> > > 0.00 : 5423e6: add $0x1,%rax
> > > 0.00 : 5423ea: mov %rax,0x18(%rsp) # data-type: u64 +0
> > > 0.00 : 5423ef: setb %al
> > > 11.17 : 5423f2: jb 54243a <test_rs+0x11a>
> > > 0.00 : 5423f4: jmp 542426 <test_rs+0x106>
> > > 0.00 : 5423f6: lea 0x30(%rsp),%rdi
> > > : 73 }
> > >
> > > Where '# data-type: struct Buf' is as far as I understand a manifestation of
> > > data type profiling succeeding. But my environment has slightly different
> > > version of rust -- I need to figure out what's going on here, thanks.
> >
> > I've being testing this in the same way, but without any rust code --
> > just on a simple datasym workload. Surprisingly, it seems that
> > 'data-type: buf' marks are disappearing even in this case starting from:
> >
> > c31040085914f1188720073baa43d1483693c0a3 (perf dwarf-regs: Clean up x86
> > dwarf_regnum code)
>
> > I'm not sure yet why this is happening, but don't see anything in the
> > commit mentioning that -- since there are no code-with-type tests yet,
> > maybe an unexpected side effect?
>
> Ian, can you please take a look at this?
That particular commit is largely adding an i386 to dwarf register
number mapping table but it also expands the x86-64 mapping table:
https://web.git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git/commit/?h=perf-tools-next&id=c31040085914f1188720073baa43d1483693c0a3
The code to search the table is largely unchanged so my guess would be
that the new x86-64 entries are causing a dwarf register to be
returned in cases where previously this was -ENOENT. This may have
broken the type profiling, but Namhyung would be the expert on that.
Thanks,
Ian
> Cheers,
>
> - Arnaldo
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [RFC PATCH v1] perf tests: Test annotate with data type profiling and rust
2026-02-03 18:15 ` Ian Rogers
@ 2026-02-04 21:52 ` Namhyung Kim
2026-02-04 22:18 ` Ian Rogers
0 siblings, 1 reply; 7+ messages in thread
From: Namhyung Kim @ 2026-02-04 21:52 UTC (permalink / raw)
To: Ian Rogers
Cc: Arnaldo Carvalho de Melo, Dmitry Dolgov, linux-perf-users,
Linux Kernel Mailing List
On Tue, Feb 03, 2026 at 10:15:30AM -0800, Ian Rogers wrote:
> On Tue, Feb 3, 2026 at 6:45 AM Arnaldo Carvalho de Melo <acme@kernel.org> wrote:
> >
> > On Sun, Feb 01, 2026 at 11:33:54AM +0100, Dmitry Dolgov wrote:
> > > > On Fri, Jan 30, 2026 at 09:44:47AM +0100, Dmitry Dolgov wrote:
> > > > > I'm not managing to reproduce your results, can you please elaborate
> > > > > some more?
> > > >
> > > > That's most unexpected for me, the test output I see looks like this:
> > > >
> > > > : 54 for _ in 1..count {
> > > > 0.00 : 5423b8: mov %ecx,0x7c(%rsp)
> > > > 0.00 : 5423bc: mov %eax,0x80(%rsp)
> > > > 0.00 : 5423c3: lea 0x7c(%rsp),%rdi
> > > > 12.28 : 5423c8: call 541f00 <core::iter::range::<impl core::iter::traits::iterator::Iterator for core::ops::range::Range<A>>::next>
> > > > 0.00 : 5423cd: mov %eax,0x20(%rsp)
> > > > 10.21 : 5423d1: jmp 5423d3 <test_rs+0xb3>
> > > > 0.00 : 5423d3: mov 0x20(%rsp),%eax
> > > > 0.00 : 5423d7: mov %eax,%eax
> > > > 0.00 : 5423d9: test $0x1,%rax
> > > > 0.29 : 5423df: je 5423f6 <test_rs+0xd6>
> > > > : 65 b.data1 += 1;
> > > > 13.69 : 5423e1: mov 0x48(%rsp),%rax # data-type: struct Buf +0x18 (data1)
> > > > 0.00 : 5423e6: add $0x1,%rax
> > > > 0.00 : 5423ea: mov %rax,0x18(%rsp) # data-type: u64 +0
> > > > 0.00 : 5423ef: setb %al
> > > > 11.17 : 5423f2: jb 54243a <test_rs+0x11a>
> > > > 0.00 : 5423f4: jmp 542426 <test_rs+0x106>
> > > > 0.00 : 5423f6: lea 0x30(%rsp),%rdi
> > > > : 73 }
> > > >
> > > > Where '# data-type: struct Buf' is as far as I understand a manifestation of
> > > > data type profiling succeeding. But my environment has slightly different
> > > > version of rust -- I need to figure out what's going on here, thanks.
> > >
> > > I've being testing this in the same way, but without any rust code --
> > > just on a simple datasym workload. Surprisingly, it seems that
> > > 'data-type: buf' marks are disappearing even in this case starting from:
> > >
> > > c31040085914f1188720073baa43d1483693c0a3 (perf dwarf-regs: Clean up x86
> > > dwarf_regnum code)
> >
> > > I'm not sure yet why this is happening, but don't see anything in the
> > > commit mentioning that -- since there are no code-with-type tests yet,
> > > maybe an unexpected side effect?
> >
> > Ian, can you please take a look at this?
>
> That particular commit is largely adding an i386 to dwarf register
> number mapping table but it also expands the x86-64 mapping table:
> https://web.git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git/commit/?h=perf-tools-next&id=c31040085914f1188720073baa43d1483693c0a3
> The code to search the table is largely unchanged so my guess would be
> that the new x86-64 entries are causing a dwarf register to be
> returned in cases where previously this was -ENOENT. This may have
> broken the type profiling, but Namhyung would be the expert on that.
Any chance it can get an invalid e_machine? I'm not sure if it handles
EM_HOST properly.
Thanks,
Namhyung
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [RFC PATCH v1] perf tests: Test annotate with data type profiling and rust
2026-02-04 21:52 ` Namhyung Kim
@ 2026-02-04 22:18 ` Ian Rogers
0 siblings, 0 replies; 7+ messages in thread
From: Ian Rogers @ 2026-02-04 22:18 UTC (permalink / raw)
To: Namhyung Kim
Cc: Arnaldo Carvalho de Melo, Dmitry Dolgov, linux-perf-users,
Linux Kernel Mailing List
On Wed, Feb 4, 2026 at 1:52 PM Namhyung Kim <namhyung@kernel.org> wrote:
>
> On Tue, Feb 03, 2026 at 10:15:30AM -0800, Ian Rogers wrote:
> > On Tue, Feb 3, 2026 at 6:45 AM Arnaldo Carvalho de Melo <acme@kernel.org> wrote:
> > >
> > > On Sun, Feb 01, 2026 at 11:33:54AM +0100, Dmitry Dolgov wrote:
> > > > > On Fri, Jan 30, 2026 at 09:44:47AM +0100, Dmitry Dolgov wrote:
> > > > > > I'm not managing to reproduce your results, can you please elaborate
> > > > > > some more?
> > > > >
> > > > > That's most unexpected for me, the test output I see looks like this:
> > > > >
> > > > > : 54 for _ in 1..count {
> > > > > 0.00 : 5423b8: mov %ecx,0x7c(%rsp)
> > > > > 0.00 : 5423bc: mov %eax,0x80(%rsp)
> > > > > 0.00 : 5423c3: lea 0x7c(%rsp),%rdi
> > > > > 12.28 : 5423c8: call 541f00 <core::iter::range::<impl core::iter::traits::iterator::Iterator for core::ops::range::Range<A>>::next>
> > > > > 0.00 : 5423cd: mov %eax,0x20(%rsp)
> > > > > 10.21 : 5423d1: jmp 5423d3 <test_rs+0xb3>
> > > > > 0.00 : 5423d3: mov 0x20(%rsp),%eax
> > > > > 0.00 : 5423d7: mov %eax,%eax
> > > > > 0.00 : 5423d9: test $0x1,%rax
> > > > > 0.29 : 5423df: je 5423f6 <test_rs+0xd6>
> > > > > : 65 b.data1 += 1;
> > > > > 13.69 : 5423e1: mov 0x48(%rsp),%rax # data-type: struct Buf +0x18 (data1)
> > > > > 0.00 : 5423e6: add $0x1,%rax
> > > > > 0.00 : 5423ea: mov %rax,0x18(%rsp) # data-type: u64 +0
> > > > > 0.00 : 5423ef: setb %al
> > > > > 11.17 : 5423f2: jb 54243a <test_rs+0x11a>
> > > > > 0.00 : 5423f4: jmp 542426 <test_rs+0x106>
> > > > > 0.00 : 5423f6: lea 0x30(%rsp),%rdi
> > > > > : 73 }
> > > > >
> > > > > Where '# data-type: struct Buf' is as far as I understand a manifestation of
> > > > > data type profiling succeeding. But my environment has slightly different
> > > > > version of rust -- I need to figure out what's going on here, thanks.
> > > >
> > > > I've being testing this in the same way, but without any rust code --
> > > > just on a simple datasym workload. Surprisingly, it seems that
> > > > 'data-type: buf' marks are disappearing even in this case starting from:
> > > >
> > > > c31040085914f1188720073baa43d1483693c0a3 (perf dwarf-regs: Clean up x86
> > > > dwarf_regnum code)
> > >
> > > > I'm not sure yet why this is happening, but don't see anything in the
> > > > commit mentioning that -- since there are no code-with-type tests yet,
> > > > maybe an unexpected side effect?
> > >
> > > Ian, can you please take a look at this?
> >
> > That particular commit is largely adding an i386 to dwarf register
> > number mapping table but it also expands the x86-64 mapping table:
> > https://web.git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git/commit/?h=perf-tools-next&id=c31040085914f1188720073baa43d1483693c0a3
> > The code to search the table is largely unchanged so my guess would be
> > that the new x86-64 entries are causing a dwarf register to be
> > returned in cases where previously this was -ENOENT. This may have
> > broken the type profiling, but Namhyung would be the expert on that.
>
> Any chance it can get an invalid e_machine? I'm not sure if it handles
> EM_HOST properly.
EM_HOST should just be EM_X86_64:
https://web.git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git/tree/tools/perf/util/include/dwarf-regs.h?h=perf-tools-next#n27
or it will be EM_NONE and everything pretty broken.
It was possible previously for a non-dso to report a broken e_machine
but that was fixed in:
https://web.git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools-next.git/commit/tools/perf/util/dso.c?h=perf-tools-next&id=7d0ebeb6c0f735d4eddc679283a1de1dea2ae878
Thanks,
Ian
> Thanks,
> Namhyung
>
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2026-02-04 22:18 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <20260127083030.5909-1-9erthalion6@gmail.com>
2026-01-27 17:00 ` [RFC PATCH v1] perf tests: Test annotate with data type profiling and rust Arnaldo Carvalho de Melo
2026-01-30 8:44 ` Dmitry Dolgov
2026-02-01 10:33 ` Dmitry Dolgov
2026-02-03 14:45 ` Arnaldo Carvalho de Melo
2026-02-03 18:15 ` Ian Rogers
2026-02-04 21:52 ` Namhyung Kim
2026-02-04 22:18 ` Ian Rogers
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox