* [PATCH v2] perf check: Add sanitizer feature and use to avoid test failure @ 2024-10-18 5:56 Ian Rogers 2024-10-22 5:41 ` Namhyung Kim 0 siblings, 1 reply; 7+ messages in thread From: Ian Rogers @ 2024-10-18 5:56 UTC (permalink / raw) To: Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo, Namhyung Kim, Mark Rutland, Alexander Shishkin, Jiri Olsa, Ian Rogers, Adrian Hunter, Kan Liang, Thomas Richter, James Clark, linux-perf-users, linux-kernel Sanitizer builds can break expectations for test disassembly, particularly in the annotate test. Add features for the different sanitizer options seen in the source tree. Use the added sanitizer feature to skip the annotate test when sanitizers are in use. Signed-off-by: Ian Rogers <irogers@google.com> --- v2 build fix. --- tools/perf/builtin-check.c | 49 ++++++++++++++++++++++++++++++ tools/perf/tests/shell/annotate.sh | 6 ++++ 2 files changed, 55 insertions(+) diff --git a/tools/perf/builtin-check.c b/tools/perf/builtin-check.c index 0b76b6e42b78..c44444008d64 100644 --- a/tools/perf/builtin-check.c +++ b/tools/perf/builtin-check.c @@ -9,6 +9,49 @@ #include <string.h> #include <subcmd/parse-options.h> +#if defined(__has_feature) +#define HAS_COMPILER_FEATURE(feature) __has_feature(feature) +#else +#define HAS_COMPILER_FEATURE(feature) 0 +#endif + +#if defined(__SANITIZE_ADDRESS__) || defined(ADDRESS_SANITIZER) || \ + HAS_COMPILER_FEATURE(address_sanitizer) +#define HAVE_SANITIZER_ADDRESS 1 +#define HAVE_SANITIZER_LEAK 1 +#elif defined(LEAK_SANITIZER) || HAS_COMPILER_FEATURE(leak_sanitizer) +#define HAVE_SANITIZER_ADDRESS 0 +#define HAVE_SANITIZER_LEAK 1 +#else +#define HAVE_SANITIZER_ADDRESS 0 +#define HAVE_SANITIZER_LEAK 0 +#endif + +#if defined(MEMORY_SANITIZER) || HAS_COMPILER_FEATURE(memory_sanitizer) +#define HAVE_SANITIZER_MEMORY 1 +#else +#define HAVE_SANITIZER_MEMORY 0 +#endif + +#if defined(THREAD_SANITIZER) || HAS_COMPILER_FEATURE(thread_sanitizer) +#define HAVE_SANITIZER_THREAD 1 +#else +#define HAVE_SANITIZER_THREAD 0 +#endif + +#if defined(UNDEFINED_SANITIZER) || HAS_COMPILER_FEATURE(undefined_sanitizer) +#define HAVE_SANITIZER_UNDEFINED 1 +#else +#define HAVE_SANITIZER_UNDEFINED 0 +#endif + +#if HAVE_SANITIZER_ADDRESS || HAVE_SANITIZER_LEAK || HAVE_SANITIZER_MEMORY || \ + HAVE_SANITIZER_THREAD || HAVE_SANITIZER_UNDEFINED +#define HAVE_SANITIZER 1 +#else +#define HAVE_SANITIZER 0 +#endif + static const char * const check_subcommands[] = { "feature", NULL }; static struct option check_options[] = { OPT_BOOLEAN('q', "quiet", &quiet, "do not show any warnings or messages"), @@ -47,6 +90,12 @@ struct feature_status supported_features[] = { FEATURE_STATUS("libunwind", HAVE_LIBUNWIND_SUPPORT), FEATURE_STATUS("lzma", HAVE_LZMA_SUPPORT), FEATURE_STATUS("numa_num_possible_cpus", HAVE_LIBNUMA_SUPPORT), + FEATURE_STATUS("sanitizer", HAVE_SANITIZER), + FEATURE_STATUS("sanitizer_address", HAVE_SANITIZER_ADDRESS), + FEATURE_STATUS("sanitizer_leak", HAVE_SANITIZER_LEAK), + FEATURE_STATUS("sanitizer_memory", HAVE_SANITIZER_MEMORY), + FEATURE_STATUS("sanitizer_thread", HAVE_SANITIZER_THREAD), + FEATURE_STATUS("sanitizer_undefined", HAVE_SANITIZER_UNDEFINED), FEATURE_STATUS("syscall_table", HAVE_SYSCALL_TABLE_SUPPORT), FEATURE_STATUS("zlib", HAVE_ZLIB_SUPPORT), FEATURE_STATUS("zstd", HAVE_ZSTD_SUPPORT), diff --git a/tools/perf/tests/shell/annotate.sh b/tools/perf/tests/shell/annotate.sh index 1590a37363de..199f547e656d 100755 --- a/tools/perf/tests/shell/annotate.sh +++ b/tools/perf/tests/shell/annotate.sh @@ -4,6 +4,12 @@ set -e +if perf check feature -q sanitizer +then + echo "Skip test with sanitizers due to differing assembly code" + exit 2 +fi + shelldir=$(dirname "$0") # shellcheck source=lib/perf_has_symbol.sh -- 2.47.0.105.g07ac214952-goog ^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [PATCH v2] perf check: Add sanitizer feature and use to avoid test failure 2024-10-18 5:56 [PATCH v2] perf check: Add sanitizer feature and use to avoid test failure Ian Rogers @ 2024-10-22 5:41 ` Namhyung Kim 2024-10-22 17:39 ` Ian Rogers 0 siblings, 1 reply; 7+ messages in thread From: Namhyung Kim @ 2024-10-22 5:41 UTC (permalink / raw) To: Ian Rogers Cc: Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo, Mark Rutland, Alexander Shishkin, Jiri Olsa, Adrian Hunter, Kan Liang, Thomas Richter, James Clark, linux-perf-users, linux-kernel On Thu, Oct 17, 2024 at 10:56:27PM -0700, Ian Rogers wrote: > Sanitizer builds can break expectations for test disassembly, > particularly in the annotate test. Add features for the different > sanitizer options seen in the source tree. Use the added sanitizer > feature to skip the annotate test when sanitizers are in use. Can you please split the perf check changes from the test change? It's good to add an example output (of perf version --build-options) with new sanitizer features. Also it might be helpful if you share how the test fails. I don't think the disasm format is changed. Then it probably missed to find the target symbol in the first 250 lines for some reason. Thanks, Namhyung > > Signed-off-by: Ian Rogers <irogers@google.com> > --- > v2 build fix. > --- > tools/perf/builtin-check.c | 49 ++++++++++++++++++++++++++++++ > tools/perf/tests/shell/annotate.sh | 6 ++++ > 2 files changed, 55 insertions(+) > > diff --git a/tools/perf/builtin-check.c b/tools/perf/builtin-check.c > index 0b76b6e42b78..c44444008d64 100644 > --- a/tools/perf/builtin-check.c > +++ b/tools/perf/builtin-check.c > @@ -9,6 +9,49 @@ > #include <string.h> > #include <subcmd/parse-options.h> > > +#if defined(__has_feature) > +#define HAS_COMPILER_FEATURE(feature) __has_feature(feature) > +#else > +#define HAS_COMPILER_FEATURE(feature) 0 > +#endif > + > +#if defined(__SANITIZE_ADDRESS__) || defined(ADDRESS_SANITIZER) || \ > + HAS_COMPILER_FEATURE(address_sanitizer) > +#define HAVE_SANITIZER_ADDRESS 1 > +#define HAVE_SANITIZER_LEAK 1 > +#elif defined(LEAK_SANITIZER) || HAS_COMPILER_FEATURE(leak_sanitizer) > +#define HAVE_SANITIZER_ADDRESS 0 > +#define HAVE_SANITIZER_LEAK 1 > +#else > +#define HAVE_SANITIZER_ADDRESS 0 > +#define HAVE_SANITIZER_LEAK 0 > +#endif > + > +#if defined(MEMORY_SANITIZER) || HAS_COMPILER_FEATURE(memory_sanitizer) > +#define HAVE_SANITIZER_MEMORY 1 > +#else > +#define HAVE_SANITIZER_MEMORY 0 > +#endif > + > +#if defined(THREAD_SANITIZER) || HAS_COMPILER_FEATURE(thread_sanitizer) > +#define HAVE_SANITIZER_THREAD 1 > +#else > +#define HAVE_SANITIZER_THREAD 0 > +#endif > + > +#if defined(UNDEFINED_SANITIZER) || HAS_COMPILER_FEATURE(undefined_sanitizer) > +#define HAVE_SANITIZER_UNDEFINED 1 > +#else > +#define HAVE_SANITIZER_UNDEFINED 0 > +#endif > + > +#if HAVE_SANITIZER_ADDRESS || HAVE_SANITIZER_LEAK || HAVE_SANITIZER_MEMORY || \ > + HAVE_SANITIZER_THREAD || HAVE_SANITIZER_UNDEFINED > +#define HAVE_SANITIZER 1 > +#else > +#define HAVE_SANITIZER 0 > +#endif > + > static const char * const check_subcommands[] = { "feature", NULL }; > static struct option check_options[] = { > OPT_BOOLEAN('q', "quiet", &quiet, "do not show any warnings or messages"), > @@ -47,6 +90,12 @@ struct feature_status supported_features[] = { > FEATURE_STATUS("libunwind", HAVE_LIBUNWIND_SUPPORT), > FEATURE_STATUS("lzma", HAVE_LZMA_SUPPORT), > FEATURE_STATUS("numa_num_possible_cpus", HAVE_LIBNUMA_SUPPORT), > + FEATURE_STATUS("sanitizer", HAVE_SANITIZER), > + FEATURE_STATUS("sanitizer_address", HAVE_SANITIZER_ADDRESS), > + FEATURE_STATUS("sanitizer_leak", HAVE_SANITIZER_LEAK), > + FEATURE_STATUS("sanitizer_memory", HAVE_SANITIZER_MEMORY), > + FEATURE_STATUS("sanitizer_thread", HAVE_SANITIZER_THREAD), > + FEATURE_STATUS("sanitizer_undefined", HAVE_SANITIZER_UNDEFINED), > FEATURE_STATUS("syscall_table", HAVE_SYSCALL_TABLE_SUPPORT), > FEATURE_STATUS("zlib", HAVE_ZLIB_SUPPORT), > FEATURE_STATUS("zstd", HAVE_ZSTD_SUPPORT), > diff --git a/tools/perf/tests/shell/annotate.sh b/tools/perf/tests/shell/annotate.sh > index 1590a37363de..199f547e656d 100755 > --- a/tools/perf/tests/shell/annotate.sh > +++ b/tools/perf/tests/shell/annotate.sh > @@ -4,6 +4,12 @@ > > set -e > > +if perf check feature -q sanitizer > +then > + echo "Skip test with sanitizers due to differing assembly code" > + exit 2 > +fi > + > shelldir=$(dirname "$0") > > # shellcheck source=lib/perf_has_symbol.sh > -- > 2.47.0.105.g07ac214952-goog > ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v2] perf check: Add sanitizer feature and use to avoid test failure 2024-10-22 5:41 ` Namhyung Kim @ 2024-10-22 17:39 ` Ian Rogers 2024-10-23 6:18 ` Namhyung Kim 0 siblings, 1 reply; 7+ messages in thread From: Ian Rogers @ 2024-10-22 17:39 UTC (permalink / raw) To: Namhyung Kim Cc: Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo, Mark Rutland, Alexander Shishkin, Jiri Olsa, Adrian Hunter, Kan Liang, Thomas Richter, James Clark, linux-perf-users, linux-kernel On Mon, Oct 21, 2024 at 10:41 PM Namhyung Kim <namhyung@kernel.org> wrote: > > On Thu, Oct 17, 2024 at 10:56:27PM -0700, Ian Rogers wrote: > > Sanitizer builds can break expectations for test disassembly, > > particularly in the annotate test. Add features for the different > > sanitizer options seen in the source tree. Use the added sanitizer > > feature to skip the annotate test when sanitizers are in use. > > Can you please split the perf check changes from the test change? > It's good to add an example output (of perf version --build-options) > with new sanitizer features. Okay, will split/add in v3. > Also it might be helpful if you share how the test fails. I don't > think the disasm format is changed. Then it probably missed to find > the target symbol in the first 250 lines for some reason. Sure, the reproduction is trivial, just add -fsanitize=address, so I'm surprised you're not already seeing it: ``` $ perf test annotate -v --- start --- test child forked, pid 444258 68e8a0-68e96b l noploop perf does have symbol 'noploop' Basic perf annotate test : 0 0x68e8a0 <noploop>: 0.00 : 68e8a0: pushq %rbp 0.00 : 68e8a1: movq %rsp, %rbp 0.00 : 68e8a4: subq $0x30, %rsp 0.00 : 68e8a8: movq %fs:0x28, %rax 0.00 : 68e8b1: movq %rax, -8(%rbp) 0.00 : 68e8b5: movl %edi, -0xc(%rbp) 0.00 : 68e8b8: movq %rsi, -0x18(%rbp) 0.00 : 68e8bc: movl $1, -0x1c(%rbp) 0.00 : 68e8c3: cmpl $0, -0xc(%rbp) 0.00 : 68e8c7: jle 0x68e8fe 0.00 : 68e8cd: movq -0x18(%rbp), %rax 0.00 : 68e8d1: movq %rax, -0x28(%rbp) 0.00 : 68e8d5: shrq $3, %rax 0.00 : 68e8d9: cmpb $0, 0x7fff8000(%rax) 0.00 : 68e8e0: je 0x68e8ef 0.00 : 68e8e6: movq -0x28(%rbp), %rdi 0.00 : 68e8ea: callq 0x2adea0 0.00 : 68e8ef: movq -0x28(%rbp), %rax 0.00 : 68e8f3: movq (%rax), %rdi 0.00 : 68e8f6: callq 0x28fac0 0.00 : 68e8fb: movl %eax, -0x1c(%rbp) 0.00 : 68e8fe: movl $2, %edi 0.00 : 68e903: leaq 0x66(%rip), %rsi # 68e970 <sighandler> 0.00 : 68e90a: callq 0x27e650 0.00 : 68e90f: movl $0xe, %edi 0.00 : 68e914: leaq 0x55(%rip), %rsi # 68e970 <sighandler> 0.00 : 68e91b: callq 0x27e650 0.00 : 68e920: movl -0x1c(%rbp), %edi 0.00 : 68e923: callq 0x2086e0 0.03 : 68e928: movl 0x18aca72(%rip), %eax # 1f3b3a0 <buildid_dir+0xa0> 0.00 : 68e92e: cmpl $0, %eax 0.00 : 68e931: setne %al 50.02 : 68e934: xorb $0xff, %al 0.00 : 68e936: testb $1, %al 41.28 : 68e938: jne 0x68e943 0.00 : 68e93e: jmp 0x68e948 8.67 : 68e943: jmp 0x68e928 0.00 : 68e948: movq %fs:0x28, %rax 0.00 : 68e951: movq -8(%rbp), %rcx 0.00 : 68e955: cmpq %rcx, %rax 0.00 : 68e958: jne 0x68e966 0.00 : 68e95e: xorl %eax, %eax 0.00 : 68e960: addq $0x30, %rsp 0.00 : 68e964: popq %rbp 0.00 : 68e965: retq 0.00 : 68e966: callq 0x2086f0 0.00 : 2d6910: endbr64 0.00 : 2d6914: pushq %rbp 0.00 : 2d6915: pushq %r15 0.00 : 2d6917: pushq %r14 0.00 : 2d6919: pushq %r13 0.00 : 2d691b: pushq %r12 0.00 : 2d691d: pushq %rbx 0.00 : 2d691e: subq $0x48, %rsp 0.00 : 2d6922: movq %rdx, 0x20(%rsp) 0.00 : 2d6927: movq %r8, 0x30(%rsp) 0.00 : 2d692c: leal -1(%r8), %eax 0.00 : 2d6930: cmpl $2, %eax 0.00 : 2d6933: jae 0x2d6b2c 0.00 : 2d6939: movq %rsi, %r9 0.00 : 2d693c: movq %rdi, %r12 0.00 : 2d693f: cmpb $0, 0x1c5f99c(%rip) # 1f362e2 <cpu_bitmap+0x122> 0.00 : 2d6946: movl $8, %eax 0.00 : 2d694b: movl $1, %ebx 0.00 : 2d6950: cmoveq %rax, %rbx 0.00 : 2d6954: cmpb $0, 0x1c5f989(%rip) # 1f362e4 <cpu_bitmap+0x124> 0.00 : 2d695b: movq %rsi, 0x28(%rsp) 0.00 : 2d6960: je 0x2d697e 0.00 : 2d6962: leaq 0x8e9708(%rip), %rdi 0.00 : 2d6969: movq %rcx, %rsi 0.00 : 2d696c: movq %r12, %rdx 0.00 : 2d696f: movq %r9, %rcx 0.00 : 2d6972: xorl %eax, %eax 0.00 : 2d6974: callq 0x2c18c0 0.00 : 2d6979: movq 0x28(%rsp), %r9 0.00 : 2d697e: leaq -1(%rbx), %rax 0.00 : 2d6982: andq %r12, %rax 0.00 : 2d6985: leaq (%rbx, %r12), %r13 0.00 : 2d6989: subq %rax, %r13 0.00 : 2d698c: testq %rax, %rax 0.00 : 2d698f: cmoveq %r12, %r13 0.00 : 2d6993: leaq 8(%r13), %rax 0.00 : 2d6997: cmpq %r9, %rax 0.00 : 2d699a: jbe 0x2d69ab 0.00 : 2d699c: addq $0x48, %rsp 0.00 : 2d69a0: popq %rbx 0.00 : 2d69a1: popq %r12 0.00 : 2d69a3: popq %r13 0.00 : 2d69a5: popq %r14 0.00 : 2d69a7: popq %r15 0.00 : 2d69a9: popq %rbp 0.00 : 2d69aa: retq 0.00 : 2d69ab: movabsq $0x7fffffffbfff, %r15 0.00 : 2d69b5: leaq 0x40(%rsp), %rbp 0.00 : 2d69ba: movq %r12, 0x38(%rsp) 0.00 : 2d69bf: jmp 0x2d69ea 0.00 : 2d69c1: nopw %cs:(%rax, %rax) 0.00 : 2d69d0: movq 0x38(%rsp), %r12 0.00 : 2d69d5: movq 0x28(%rsp), %r9 0.00 : 2d69da: leaq (%rbx, %r13), %rax 0.00 : 2d69de: addq $8, %rax 0.00 : 2d69e2: addq %rbx, %r13 0.00 : 2d69e5: cmpq %r9, %rax 0.00 : 2d69e8: ja 0x2d699c 0.00 : 2d69ea: movq (%r13), %r14 0.00 : 2d69ee: leaq -0x4000(%r14), %rax 0.00 : 2d69f5: cmpq %r15, %rax 55.45 : 2d69f8: ja 0x2d69da 0.00 : 2d69fa: movq %r14, %rdi 44.55 : 2d69fd: callq 0x20df90 0.00 : 2d6a02: movq %rax, 0x18(%rsp) 0.00 : 2d6a07: testq %rax, %rax 0.00 : 2d6a0a: je 0x2d69d5 0.00 : 2d6a0c: cmpq %r12, %rax 0.00 : 2d6a0f: je 0x2d69d5 0.00 : 2d6a11: movq %rbp, %r12 0.00 : 2d6a14: movq %rbp, %rdi 0.00 : 2d6a17: movq %rax, %rsi 0.00 : 2d6a1a: callq 0x20e190 0.00 : 2d6a1f: movq %rbp, %rdi 0.00 : 2d6a22: callq 0x20e1d0 0.00 : 2d6a27: cmpl $2, %eax 0.00 : 2d6a2a: je 0x2d69d0 0.00 : 2d6a2c: movq %rbp, %rdi 0.00 : 2d6a2f: callq 0x20e1d0 0.00 : 2d6a34: cmpl $3, %eax 0.00 : 2d6a37: je 0x2d69d0 0.00 : 2d6a39: cmpb $0, 0x1c5f8a3(%rip) # 1f362e3 <cpu_bitmap+0x123> 0.00 : 2d6a40: jne 0x2d6aab 0.00 : 2d6a42: movq %r13, %rdi 0.00 : 2d6a45: callq 0x2a9030 0.00 : 2d6a4a: testb %al, %al 0.00 : 2d6a4c: je 0x2d6aab 0.00 : 2d6a4e: cmpb $0, 0x1c5f88f(%rip) # 1f362e4 <cpu_bitmap+0x124> 0.00 : 2d6a55: je 0x2d69d0 0.00 : 2d6a5b: movq 0x18(%rsp), %rax 0.00 : 2d6a60: movq %rax, 8(%rsp) 0.00 : 2d6a65: movq %rbp, %r12 0.00 : 2d6a68: movq %rbp, %rdi 0.00 : 2d6a6b: callq 0x20e210 0.00 : 2d6a70: movq 8(%rsp), %rcx 0.00 : 2d6a75: addq %rcx, %rax 0.00 : 2d6a78: movq %rax, 0x10(%rsp) 0.00 : 2d6a7d: movq %rbp, %rdi 0.00 : 2d6a80: callq 0x20e210 0.00 : 2d6a85: leaq 0x8e95ff(%rip), %rdi 0.00 : 2d6a8c: movq %r13, %rsi 0.00 : 2d6a8f: movq %r14, %rdx 0.00 : 2d6a92: movq 8(%rsp), %rcx 0.00 : 2d6a97: movq 0x10(%rsp), %r8 0.00 : 2d6a9c: movq %rax, %r9 0.00 : 2d6a9f: xorl %eax, %eax 0.00 : 2d6aa1: callq 0x2c18c0 0.00 : 2d6aa6: jmp 0x2d69d0 0.00 : 2d6aab: movq %rbp, %r12 0.00 : 2d6aae: movq %rbp, %rdi 0.00 : 2d6ab1: movq 0x30(%rsp), %rsi 0.00 : 2d6ab6: callq 0x20e1f0 0.00 : 2d6abb: cmpb $0, 0x1c5f822(%rip) # 1f362e4 <cpu_bitmap+0x124> 0.00 : 2d6ac2: je 0x2d6b0c 0.00 : 2d6ac4: movq 0x18(%rsp), %rax 0.00 : 2d6ac9: movq %rax, 8(%rsp) 0.00 : 2d6ace: movq %r12, %rdi 0.00 : 2d6ad1: callq 0x20e210 0.00 : 2d6ad6: movq 8(%rsp), %rcx 0.00 : 2d6adb: addq %rcx, %rax 0.00 : 2d6ade: movq %rax, 0x10(%rsp) 0.00 : 2d6ae3: movq %r12, %rdi 0.00 : 2d6ae6: callq 0x20e210 0.00 : 2d6aeb: leaq 0x8e95dd(%rip), %rdi 0.00 : 2d6af2: movq %r13, %rsi 0.00 : 2d6af5: movq %r14, %rdx 0.00 : 2d6af8: movq 8(%rsp), %rcx 0.00 : 2d6afd: movq 0x10(%rsp), %r8 0.00 : 2d6b02: movq %rax, %r9 0.00 : 2d6b05: xorl %eax, %eax 0.00 : 2d6b07: callq 0x2c18c0 0.00 : 2d6b0c: cmpq $0, 0x20(%rsp) 0.00 : 2d6b12: je 0x2d69d0 0.00 : 2d6b18: movq 0x20(%rsp), %rdi 0.00 : 2d6b1d: leaq 0x18(%rsp), %rsi 0.00 : 2d6b22: callq 0x2b45a0 0.00 : 2d6b27: jmp 0x2d69d0 0.00 : 2d6b2c: leaq 0x8e94cf(%rip), %rdi 0.00 : 2d6b33: leaq 0x8e94fe(%rip), %rdx 0.00 : 2d6b3a: movl $0x128, %esi 0.00 : 2d6b3f: xorl %ecx, %ecx 0.00 : 2d6b41: xorl %r8d, %r8d 0.00 : 2d6b44: callq 0x2c77f0 0.00 : 9200: pushq %r15 0.00 : 9202: pushq %r14 0.00 : 9204: pushq %r13 0.00 : 9206: pushq %r12 0.00 : 9208: pushq %rbp 0.00 : 9209: pushq %rbx 0.00 : 920a: subq $0x88, %rsp 0.00 : 9211: movq %rdi, 8(%rsp) 0.00 : 9216: movl 8(%r9), %edi 0.00 : 921a: movq %r8, 0x20(%rsp) 0.00 : 921f: movq 0xc0(%rsp), %rbp 0.00 : 9227: movq %rdx, 0x28(%rsp) 0.00 : 922c: movq 0xd8(%rsp), %r8 0.00 : 9234: movq %rdi, %r10 0.00 : 9237: movq %rcx, 0x10(%rsp) 0.00 : 923c: movq 0xe8(%rsp), %r11 0.00 : 9244: movq (%r9), %r13 0.00 : 9247: movl %esi, %eax 0.00 : 9249: movl 0xe0(%rsp), %r9d 0.00 : 9251: movl %esi, %r15d 0.00 : 9254: shrl $6, %eax 0.00 : 9257: movq %r8, %r12 0.00 : 925a: movl %eax, 0x18(%rsp) 0.00 : 925e: andl $2, %r9d 0.00 : 9262: jmp 0x92ff 0.00 : 9267: nopw (%rax, %rax) 0.00 : 9270: movl 0x30c(%rbx), %edi 0.00 : 9276: testl %edi, %edi 0.00 : 9278: je 0x92f2 0.00 : 927a: movq 0x70(%rbx), %rax 0.00 : 927e: movl $0, 0x74(%rsp) 0.00 : 9286: movq $0, 0x78(%rsp) 0.00 : 928f: movq 8(%rax), %rsi 0.00 : 9293: xorl %eax, %eax 0.00 : 9295: testb $0x20, 0x336(%rbx) 0.00 : 929c: je 0x92a4 0.00 : 929e: movq (%rbx), %rax 0.00 : 92a1: addq %rax, %rsi 0.00 : 92a4: movq 0x68(%rbx), %rdx 0.00 : 92a8: addq 8(%rdx), %rax 0.00 : 92ac: movq 0x318(%rbx), %rdx 0.00 : 92b3: movq %rax, %r8 0.00 : 92b6: testq %rdx, %rdx 0.00 : 92b9: je 0x93b0 0.00 : 92bf: movl 0x314(%rbx), %ecx 0.00 : 92c5: movl 0x18(%rsp), %eax 13.66 : 92c9: andl 0x310(%rbx), %eax 16.28 : 92cf: movq (%rdx, %rax, 8), %rax 0.00 : 92d3: movl %r15d, %edx 0.00 : 92d6: shrl %cl, %edx 0.00 : 92d8: movl %edx, %ecx 0.00 : 92da: movq %rax, %rdx Basic annotate [Failed: missing disasm output when specifying the target symbol] ---- end(-1) ---- 77: perf annotate basic tests : FAILED! ``` Thanks, Ian ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v2] perf check: Add sanitizer feature and use to avoid test failure 2024-10-22 17:39 ` Ian Rogers @ 2024-10-23 6:18 ` Namhyung Kim 2024-10-23 22:19 ` Namhyung Kim 0 siblings, 1 reply; 7+ messages in thread From: Namhyung Kim @ 2024-10-23 6:18 UTC (permalink / raw) To: Ian Rogers Cc: Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo, Mark Rutland, Alexander Shishkin, Jiri Olsa, Adrian Hunter, Kan Liang, Thomas Richter, James Clark, linux-perf-users, linux-kernel On Tue, Oct 22, 2024 at 10:39:36AM -0700, Ian Rogers wrote: > On Mon, Oct 21, 2024 at 10:41 PM Namhyung Kim <namhyung@kernel.org> wrote: > > > > On Thu, Oct 17, 2024 at 10:56:27PM -0700, Ian Rogers wrote: > > > Sanitizer builds can break expectations for test disassembly, > > > particularly in the annotate test. Add features for the different > > > sanitizer options seen in the source tree. Use the added sanitizer > > > feature to skip the annotate test when sanitizers are in use. > > > > Can you please split the perf check changes from the test change? > > It's good to add an example output (of perf version --build-options) > > with new sanitizer features. > > Okay, will split/add in v3. Thanks! > > > Also it might be helpful if you share how the test fails. I don't > > think the disasm format is changed. Then it probably missed to find > > the target symbol in the first 250 lines for some reason. > > Sure, the reproduction is trivial, just add -fsanitize=address, so I'm > surprised you're not already seeing it: > ``` > $ perf test annotate -v > --- start --- > test child forked, pid 444258 > 68e8a0-68e96b l noploop > perf does have symbol 'noploop' > Basic perf annotate test > : 0 0x68e8a0 <noploop>: > 0.00 : 68e8a0: pushq %rbp > 0.00 : 68e8a1: movq %rsp, %rbp > 0.00 : 68e8a4: subq $0x30, %rsp > 0.00 : 68e8a8: movq %fs:0x28, %rax > 0.00 : 68e8b1: movq %rax, -8(%rbp) > 0.00 : 68e8b5: movl %edi, -0xc(%rbp) > 0.00 : 68e8b8: movq %rsi, -0x18(%rbp) > 0.00 : 68e8bc: movl $1, -0x1c(%rbp) > 0.00 : 68e8c3: cmpl $0, -0xc(%rbp) > 0.00 : 68e8c7: jle 0x68e8fe > 0.00 : 68e8cd: movq -0x18(%rbp), %rax > 0.00 : 68e8d1: movq %rax, -0x28(%rbp) > 0.00 : 68e8d5: shrq $3, %rax > 0.00 : 68e8d9: cmpb $0, 0x7fff8000(%rax) > 0.00 : 68e8e0: je 0x68e8ef > 0.00 : 68e8e6: movq -0x28(%rbp), %rdi > 0.00 : 68e8ea: callq 0x2adea0 > 0.00 : 68e8ef: movq -0x28(%rbp), %rax > 0.00 : 68e8f3: movq (%rax), %rdi > 0.00 : 68e8f6: callq 0x28fac0 > 0.00 : 68e8fb: movl %eax, -0x1c(%rbp) > 0.00 : 68e8fe: movl $2, %edi > 0.00 : 68e903: leaq 0x66(%rip), %rsi # 68e970 <sighandler> > 0.00 : 68e90a: callq 0x27e650 > 0.00 : 68e90f: movl $0xe, %edi > 0.00 : 68e914: leaq 0x55(%rip), %rsi # 68e970 <sighandler> > 0.00 : 68e91b: callq 0x27e650 > 0.00 : 68e920: movl -0x1c(%rbp), %edi > 0.00 : 68e923: callq 0x2086e0 > 0.03 : 68e928: movl 0x18aca72(%rip), %eax # > 1f3b3a0 <buildid_dir+0xa0> > 0.00 : 68e92e: cmpl $0, %eax > 0.00 : 68e931: setne %al > 50.02 : 68e934: xorb $0xff, %al > 0.00 : 68e936: testb $1, %al > 41.28 : 68e938: jne 0x68e943 > 0.00 : 68e93e: jmp 0x68e948 > 8.67 : 68e943: jmp 0x68e928 > 0.00 : 68e948: movq %fs:0x28, %rax > 0.00 : 68e951: movq -8(%rbp), %rcx > 0.00 : 68e955: cmpq %rcx, %rax > 0.00 : 68e958: jne 0x68e966 > 0.00 : 68e95e: xorl %eax, %eax > 0.00 : 68e960: addq $0x30, %rsp > 0.00 : 68e964: popq %rbp > 0.00 : 68e965: retq > 0.00 : 68e966: callq 0x2086f0 > 0.00 : 2d6910: endbr64 > 0.00 : 2d6914: pushq %rbp > 0.00 : 2d6915: pushq %r15 > 0.00 : 2d6917: pushq %r14 > 0.00 : 2d6919: pushq %r13 > 0.00 : 2d691b: pushq %r12 > 0.00 : 2d691d: pushq %rbx > 0.00 : 2d691e: subq $0x48, %rsp > 0.00 : 2d6922: movq %rdx, 0x20(%rsp) > 0.00 : 2d6927: movq %r8, 0x30(%rsp) > 0.00 : 2d692c: leal -1(%r8), %eax > 0.00 : 2d6930: cmpl $2, %eax > 0.00 : 2d6933: jae 0x2d6b2c > 0.00 : 2d6939: movq %rsi, %r9 > 0.00 : 2d693c: movq %rdi, %r12 > 0.00 : 2d693f: cmpb $0, 0x1c5f99c(%rip) # 1f362e2 > <cpu_bitmap+0x122> > 0.00 : 2d6946: movl $8, %eax > 0.00 : 2d694b: movl $1, %ebx > 0.00 : 2d6950: cmoveq %rax, %rbx > 0.00 : 2d6954: cmpb $0, 0x1c5f989(%rip) # 1f362e4 > <cpu_bitmap+0x124> > 0.00 : 2d695b: movq %rsi, 0x28(%rsp) > 0.00 : 2d6960: je 0x2d697e > 0.00 : 2d6962: leaq 0x8e9708(%rip), %rdi > 0.00 : 2d6969: movq %rcx, %rsi > 0.00 : 2d696c: movq %r12, %rdx > 0.00 : 2d696f: movq %r9, %rcx > 0.00 : 2d6972: xorl %eax, %eax > 0.00 : 2d6974: callq 0x2c18c0 > 0.00 : 2d6979: movq 0x28(%rsp), %r9 > 0.00 : 2d697e: leaq -1(%rbx), %rax > 0.00 : 2d6982: andq %r12, %rax > 0.00 : 2d6985: leaq (%rbx, %r12), %r13 > 0.00 : 2d6989: subq %rax, %r13 > 0.00 : 2d698c: testq %rax, %rax > 0.00 : 2d698f: cmoveq %r12, %r13 > 0.00 : 2d6993: leaq 8(%r13), %rax > 0.00 : 2d6997: cmpq %r9, %rax > 0.00 : 2d699a: jbe 0x2d69ab > 0.00 : 2d699c: addq $0x48, %rsp > 0.00 : 2d69a0: popq %rbx > 0.00 : 2d69a1: popq %r12 > 0.00 : 2d69a3: popq %r13 > 0.00 : 2d69a5: popq %r14 > 0.00 : 2d69a7: popq %r15 > 0.00 : 2d69a9: popq %rbp > 0.00 : 2d69aa: retq > 0.00 : 2d69ab: movabsq $0x7fffffffbfff, %r15 > 0.00 : 2d69b5: leaq 0x40(%rsp), %rbp > 0.00 : 2d69ba: movq %r12, 0x38(%rsp) > 0.00 : 2d69bf: jmp 0x2d69ea > 0.00 : 2d69c1: nopw %cs:(%rax, %rax) > 0.00 : 2d69d0: movq 0x38(%rsp), %r12 > 0.00 : 2d69d5: movq 0x28(%rsp), %r9 > 0.00 : 2d69da: leaq (%rbx, %r13), %rax > 0.00 : 2d69de: addq $8, %rax > 0.00 : 2d69e2: addq %rbx, %r13 > 0.00 : 2d69e5: cmpq %r9, %rax > 0.00 : 2d69e8: ja 0x2d699c > 0.00 : 2d69ea: movq (%r13), %r14 > 0.00 : 2d69ee: leaq -0x4000(%r14), %rax > 0.00 : 2d69f5: cmpq %r15, %rax > 55.45 : 2d69f8: ja 0x2d69da > 0.00 : 2d69fa: movq %r14, %rdi > 44.55 : 2d69fd: callq 0x20df90 > 0.00 : 2d6a02: movq %rax, 0x18(%rsp) > 0.00 : 2d6a07: testq %rax, %rax > 0.00 : 2d6a0a: je 0x2d69d5 > 0.00 : 2d6a0c: cmpq %r12, %rax > 0.00 : 2d6a0f: je 0x2d69d5 > 0.00 : 2d6a11: movq %rbp, %r12 > 0.00 : 2d6a14: movq %rbp, %rdi > 0.00 : 2d6a17: movq %rax, %rsi > 0.00 : 2d6a1a: callq 0x20e190 > 0.00 : 2d6a1f: movq %rbp, %rdi > 0.00 : 2d6a22: callq 0x20e1d0 > 0.00 : 2d6a27: cmpl $2, %eax > 0.00 : 2d6a2a: je 0x2d69d0 > 0.00 : 2d6a2c: movq %rbp, %rdi > 0.00 : 2d6a2f: callq 0x20e1d0 > 0.00 : 2d6a34: cmpl $3, %eax > 0.00 : 2d6a37: je 0x2d69d0 > 0.00 : 2d6a39: cmpb $0, 0x1c5f8a3(%rip) # 1f362e3 > <cpu_bitmap+0x123> > 0.00 : 2d6a40: jne 0x2d6aab > 0.00 : 2d6a42: movq %r13, %rdi > 0.00 : 2d6a45: callq 0x2a9030 > 0.00 : 2d6a4a: testb %al, %al > 0.00 : 2d6a4c: je 0x2d6aab > 0.00 : 2d6a4e: cmpb $0, 0x1c5f88f(%rip) # 1f362e4 > <cpu_bitmap+0x124> > 0.00 : 2d6a55: je 0x2d69d0 > 0.00 : 2d6a5b: movq 0x18(%rsp), %rax > 0.00 : 2d6a60: movq %rax, 8(%rsp) > 0.00 : 2d6a65: movq %rbp, %r12 > 0.00 : 2d6a68: movq %rbp, %rdi > 0.00 : 2d6a6b: callq 0x20e210 > 0.00 : 2d6a70: movq 8(%rsp), %rcx > 0.00 : 2d6a75: addq %rcx, %rax > 0.00 : 2d6a78: movq %rax, 0x10(%rsp) > 0.00 : 2d6a7d: movq %rbp, %rdi > 0.00 : 2d6a80: callq 0x20e210 > 0.00 : 2d6a85: leaq 0x8e95ff(%rip), %rdi > 0.00 : 2d6a8c: movq %r13, %rsi > 0.00 : 2d6a8f: movq %r14, %rdx > 0.00 : 2d6a92: movq 8(%rsp), %rcx > 0.00 : 2d6a97: movq 0x10(%rsp), %r8 > 0.00 : 2d6a9c: movq %rax, %r9 > 0.00 : 2d6a9f: xorl %eax, %eax > 0.00 : 2d6aa1: callq 0x2c18c0 > 0.00 : 2d6aa6: jmp 0x2d69d0 > 0.00 : 2d6aab: movq %rbp, %r12 > 0.00 : 2d6aae: movq %rbp, %rdi > 0.00 : 2d6ab1: movq 0x30(%rsp), %rsi > 0.00 : 2d6ab6: callq 0x20e1f0 > 0.00 : 2d6abb: cmpb $0, 0x1c5f822(%rip) # 1f362e4 > <cpu_bitmap+0x124> > 0.00 : 2d6ac2: je 0x2d6b0c > 0.00 : 2d6ac4: movq 0x18(%rsp), %rax > 0.00 : 2d6ac9: movq %rax, 8(%rsp) > 0.00 : 2d6ace: movq %r12, %rdi > 0.00 : 2d6ad1: callq 0x20e210 > 0.00 : 2d6ad6: movq 8(%rsp), %rcx > 0.00 : 2d6adb: addq %rcx, %rax > 0.00 : 2d6ade: movq %rax, 0x10(%rsp) > 0.00 : 2d6ae3: movq %r12, %rdi > 0.00 : 2d6ae6: callq 0x20e210 > 0.00 : 2d6aeb: leaq 0x8e95dd(%rip), %rdi > 0.00 : 2d6af2: movq %r13, %rsi > 0.00 : 2d6af5: movq %r14, %rdx > 0.00 : 2d6af8: movq 8(%rsp), %rcx > 0.00 : 2d6afd: movq 0x10(%rsp), %r8 > 0.00 : 2d6b02: movq %rax, %r9 > 0.00 : 2d6b05: xorl %eax, %eax > 0.00 : 2d6b07: callq 0x2c18c0 > 0.00 : 2d6b0c: cmpq $0, 0x20(%rsp) > 0.00 : 2d6b12: je 0x2d69d0 > 0.00 : 2d6b18: movq 0x20(%rsp), %rdi > 0.00 : 2d6b1d: leaq 0x18(%rsp), %rsi > 0.00 : 2d6b22: callq 0x2b45a0 > 0.00 : 2d6b27: jmp 0x2d69d0 > 0.00 : 2d6b2c: leaq 0x8e94cf(%rip), %rdi > 0.00 : 2d6b33: leaq 0x8e94fe(%rip), %rdx > 0.00 : 2d6b3a: movl $0x128, %esi > 0.00 : 2d6b3f: xorl %ecx, %ecx > 0.00 : 2d6b41: xorl %r8d, %r8d > 0.00 : 2d6b44: callq 0x2c77f0 > 0.00 : 9200: pushq %r15 > 0.00 : 9202: pushq %r14 > 0.00 : 9204: pushq %r13 > 0.00 : 9206: pushq %r12 > 0.00 : 9208: pushq %rbp > 0.00 : 9209: pushq %rbx > 0.00 : 920a: subq $0x88, %rsp > 0.00 : 9211: movq %rdi, 8(%rsp) > 0.00 : 9216: movl 8(%r9), %edi > 0.00 : 921a: movq %r8, 0x20(%rsp) > 0.00 : 921f: movq 0xc0(%rsp), %rbp > 0.00 : 9227: movq %rdx, 0x28(%rsp) > 0.00 : 922c: movq 0xd8(%rsp), %r8 > 0.00 : 9234: movq %rdi, %r10 > 0.00 : 9237: movq %rcx, 0x10(%rsp) > 0.00 : 923c: movq 0xe8(%rsp), %r11 > 0.00 : 9244: movq (%r9), %r13 > 0.00 : 9247: movl %esi, %eax > 0.00 : 9249: movl 0xe0(%rsp), %r9d > 0.00 : 9251: movl %esi, %r15d > 0.00 : 9254: shrl $6, %eax > 0.00 : 9257: movq %r8, %r12 > 0.00 : 925a: movl %eax, 0x18(%rsp) > 0.00 : 925e: andl $2, %r9d > 0.00 : 9262: jmp 0x92ff > 0.00 : 9267: nopw (%rax, %rax) > 0.00 : 9270: movl 0x30c(%rbx), %edi > 0.00 : 9276: testl %edi, %edi > 0.00 : 9278: je 0x92f2 > 0.00 : 927a: movq 0x70(%rbx), %rax > 0.00 : 927e: movl $0, 0x74(%rsp) > 0.00 : 9286: movq $0, 0x78(%rsp) > 0.00 : 928f: movq 8(%rax), %rsi > 0.00 : 9293: xorl %eax, %eax > 0.00 : 9295: testb $0x20, 0x336(%rbx) > 0.00 : 929c: je 0x92a4 > 0.00 : 929e: movq (%rbx), %rax > 0.00 : 92a1: addq %rax, %rsi > 0.00 : 92a4: movq 0x68(%rbx), %rdx > 0.00 : 92a8: addq 8(%rdx), %rax > 0.00 : 92ac: movq 0x318(%rbx), %rdx > 0.00 : 92b3: movq %rax, %r8 > 0.00 : 92b6: testq %rdx, %rdx > 0.00 : 92b9: je 0x93b0 > 0.00 : 92bf: movl 0x314(%rbx), %ecx > 0.00 : 92c5: movl 0x18(%rsp), %eax > 13.66 : 92c9: andl 0x310(%rbx), %eax > 16.28 : 92cf: movq (%rdx, %rax, 8), %rax > 0.00 : 92d3: movl %r15d, %edx > 0.00 : 92d6: shrl %cl, %edx > 0.00 : 92d8: movl %edx, %ecx > 0.00 : 92da: movq %rax, %rdx > Basic annotate [Failed: missing disasm output when specifying the target symbol] Hmm.. this is strange. The error message says it failed when it specified the target symbol (noploop) for perf annotate. As it's the dominant symbol, it should have the same output for the first function (noploop) whether it has target symbol or not and it should match the disasm_regex. I'm curious how it can fail here. Thanks, Namhyung > ---- end(-1) ---- > 77: perf annotate basic tests : FAILED! > ``` > > Thanks, > Ian ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v2] perf check: Add sanitizer feature and use to avoid test failure 2024-10-23 6:18 ` Namhyung Kim @ 2024-10-23 22:19 ` Namhyung Kim 2024-10-23 23:24 ` Namhyung Kim 0 siblings, 1 reply; 7+ messages in thread From: Namhyung Kim @ 2024-10-23 22:19 UTC (permalink / raw) To: Ian Rogers Cc: Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo, Mark Rutland, Alexander Shishkin, Jiri Olsa, Adrian Hunter, Kan Liang, Thomas Richter, James Clark, linux-perf-users, linux-kernel On Tue, Oct 22, 2024 at 11:18:18PM -0700, Namhyung Kim wrote: > On Tue, Oct 22, 2024 at 10:39:36AM -0700, Ian Rogers wrote: > > Sure, the reproduction is trivial, just add -fsanitize=address, so I'm > > surprised you're not already seeing it: > > ``` > > $ perf test annotate -v > > --- start --- > > test child forked, pid 444258 > > 68e8a0-68e96b l noploop > > perf does have symbol 'noploop' > > Basic perf annotate test > > : 0 0x68e8a0 <noploop>: > > 0.00 : 68e8a0: pushq %rbp > > 0.00 : 68e8a1: movq %rsp, %rbp > > 0.00 : 68e8a4: subq $0x30, %rsp [...] > > 0.00 : 92d6: shrl %cl, %edx > > 0.00 : 92d8: movl %edx, %ecx > > 0.00 : 92da: movq %rax, %rdx > > Basic annotate [Failed: missing disasm output when specifying the target symbol] > > Hmm.. this is strange. The error message says it failed when it > specified the target symbol (noploop) for perf annotate. > > As it's the dominant symbol, it should have the same output for the > first function (noploop) whether it has target symbol or not and it > should match the disasm_regex. I'm curious how it can fail here. Hmm.. ok. For some reason, it wasn't failed when I add DEBUG=1. Without DEBUG, I can see it now. ================================================================= ==1053492==ERROR: LeakSanitizer: detected memory leaks Direct leak of 33 byte(s) in 1 object(s) allocated from: #0 0x7f1ad78edd20 in strdup ../../../../src/libsanitizer/asan/asan_interceptors.cpp:566 #1 0x55eda19cb76f in perf_data__open (linux/tools/perf/perf+0x65276f) (BuildId: 6fc1b7cdc123c7bd586ce55ea8b727875f42cda2) #2 0x55eda18ffafa in __perf_session__new (linux/tools/perf/perf+0x586afa) (BuildId: 6fc1b7cdc123c7bd586ce55ea8b727875f42cda2) #3 0x55eda15485d3 in cmd_annotate (linux/tools/perf/perf+0x1cf5d3) (BuildId: 6fc1b7cdc123c7bd586ce55ea8b727875f42cda2) #4 0x55eda1695467 in run_builtin (linux/tools/perf/perf+0x31c467) (BuildId: 6fc1b7cdc123c7bd586ce55ea8b727875f42cda2) #5 0x55eda1695c0e in handle_internal_command (linux/tools/perf/perf+0x31cc0e) (BuildId: 6fc1b7cdc123c7bd586ce55ea8b727875f42cda2) #6 0x55eda153ba72 in main (linux/tools/perf/perf+0x1c2a72) (BuildId: 6fc1b7cdc123c7bd586ce55ea8b727875f42cda2) #7 0x7f1acda43b89 in __libc_start_call_main ../sysdeps/nptl/libc_start_call_main.h:58 SUMMARY: AddressSanitizer: 33 byte(s) leaked in 1 allocation(s). Unexpected signal in test_basic No idea how it can leak the data->file.path (that's what I can find where strdup is used in the function). Thanks, Namhyung ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v2] perf check: Add sanitizer feature and use to avoid test failure 2024-10-23 22:19 ` Namhyung Kim @ 2024-10-23 23:24 ` Namhyung Kim 2024-10-24 2:42 ` Ian Rogers 0 siblings, 1 reply; 7+ messages in thread From: Namhyung Kim @ 2024-10-23 23:24 UTC (permalink / raw) To: Ian Rogers Cc: Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo, Mark Rutland, Alexander Shishkin, Jiri Olsa, Adrian Hunter, Kan Liang, Thomas Richter, James Clark, linux-perf-users, linux-kernel On Wed, Oct 23, 2024 at 03:19:38PM -0700, Namhyung Kim wrote: > On Tue, Oct 22, 2024 at 11:18:18PM -0700, Namhyung Kim wrote: > > On Tue, Oct 22, 2024 at 10:39:36AM -0700, Ian Rogers wrote: > > > Sure, the reproduction is trivial, just add -fsanitize=address, so I'm > > > surprised you're not already seeing it: > > > ``` > > > $ perf test annotate -v > > > --- start --- > > > test child forked, pid 444258 > > > 68e8a0-68e96b l noploop > > > perf does have symbol 'noploop' > > > Basic perf annotate test > > > : 0 0x68e8a0 <noploop>: > > > 0.00 : 68e8a0: pushq %rbp > > > 0.00 : 68e8a1: movq %rsp, %rbp > > > 0.00 : 68e8a4: subq $0x30, %rsp > [...] > > > 0.00 : 92d6: shrl %cl, %edx > > > 0.00 : 92d8: movl %edx, %ecx > > > 0.00 : 92da: movq %rax, %rdx > > > Basic annotate [Failed: missing disasm output when specifying the target symbol] > > > > Hmm.. this is strange. The error message says it failed when it > > specified the target symbol (noploop) for perf annotate. > > > > As it's the dominant symbol, it should have the same output for the > > first function (noploop) whether it has target symbol or not and it > > should match the disasm_regex. I'm curious how it can fail here. > > Hmm.. ok. For some reason, it wasn't failed when I add DEBUG=1. Oh, now I'm seeing why. We skip perf_session__delete() on !DEBUG build. :( > > Without DEBUG, I can see it now. > > ================================================================= > ==1053492==ERROR: LeakSanitizer: detected memory leaks > > Direct leak of 33 byte(s) in 1 object(s) allocated from: > #0 0x7f1ad78edd20 in strdup ../../../../src/libsanitizer/asan/asan_interceptors.cpp:566 > #1 0x55eda19cb76f in perf_data__open (linux/tools/perf/perf+0x65276f) (BuildId: 6fc1b7cdc123c7bd586ce55ea8b727875f42cda2) > #2 0x55eda18ffafa in __perf_session__new (linux/tools/perf/perf+0x586afa) (BuildId: 6fc1b7cdc123c7bd586ce55ea8b727875f42cda2) > #3 0x55eda15485d3 in cmd_annotate (linux/tools/perf/perf+0x1cf5d3) (BuildId: 6fc1b7cdc123c7bd586ce55ea8b727875f42cda2) > #4 0x55eda1695467 in run_builtin (linux/tools/perf/perf+0x31c467) (BuildId: 6fc1b7cdc123c7bd586ce55ea8b727875f42cda2) > #5 0x55eda1695c0e in handle_internal_command (linux/tools/perf/perf+0x31cc0e) (BuildId: 6fc1b7cdc123c7bd586ce55ea8b727875f42cda2) > #6 0x55eda153ba72 in main (linux/tools/perf/perf+0x1c2a72) (BuildId: 6fc1b7cdc123c7bd586ce55ea8b727875f42cda2) > #7 0x7f1acda43b89 in __libc_start_call_main ../sysdeps/nptl/libc_start_call_main.h:58 > > SUMMARY: AddressSanitizer: 33 byte(s) leaked in 1 allocation(s). > Unexpected signal in test_basic > > No idea how it can leak the data->file.path (that's what I can find > where strdup is used in the function). Maybe we need to revisit how much speed up it can give. Thanks, Namhyung ^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v2] perf check: Add sanitizer feature and use to avoid test failure 2024-10-23 23:24 ` Namhyung Kim @ 2024-10-24 2:42 ` Ian Rogers 0 siblings, 0 replies; 7+ messages in thread From: Ian Rogers @ 2024-10-24 2:42 UTC (permalink / raw) To: Namhyung Kim Cc: Peter Zijlstra, Ingo Molnar, Arnaldo Carvalho de Melo, Mark Rutland, Alexander Shishkin, Jiri Olsa, Adrian Hunter, Kan Liang, Thomas Richter, James Clark, linux-perf-users, linux-kernel On Wed, Oct 23, 2024 at 4:24 PM Namhyung Kim <namhyung@kernel.org> wrote: > > On Wed, Oct 23, 2024 at 03:19:38PM -0700, Namhyung Kim wrote: > > On Tue, Oct 22, 2024 at 11:18:18PM -0700, Namhyung Kim wrote: > > > On Tue, Oct 22, 2024 at 10:39:36AM -0700, Ian Rogers wrote: > > > > Sure, the reproduction is trivial, just add -fsanitize=address, so I'm > > > > surprised you're not already seeing it: > > > > ``` > > > > $ perf test annotate -v > > > > --- start --- > > > > test child forked, pid 444258 > > > > 68e8a0-68e96b l noploop > > > > perf does have symbol 'noploop' > > > > Basic perf annotate test > > > > : 0 0x68e8a0 <noploop>: > > > > 0.00 : 68e8a0: pushq %rbp > > > > 0.00 : 68e8a1: movq %rsp, %rbp > > > > 0.00 : 68e8a4: subq $0x30, %rsp > > [...] > > > > 0.00 : 92d6: shrl %cl, %edx > > > > 0.00 : 92d8: movl %edx, %ecx > > > > 0.00 : 92da: movq %rax, %rdx > > > > Basic annotate [Failed: missing disasm output when specifying the target symbol] > > > > > > Hmm.. this is strange. The error message says it failed when it > > > specified the target symbol (noploop) for perf annotate. > > > > > > As it's the dominant symbol, it should have the same output for the > > > first function (noploop) whether it has target symbol or not and it > > > should match the disasm_regex. I'm curious how it can fail here. > > > > Hmm.. ok. For some reason, it wasn't failed when I add DEBUG=1. > > Oh, now I'm seeing why. We skip perf_session__delete() on !DEBUG build. > :( > > > > > Without DEBUG, I can see it now. > > > > ================================================================= > > ==1053492==ERROR: LeakSanitizer: detected memory leaks > > > > Direct leak of 33 byte(s) in 1 object(s) allocated from: > > #0 0x7f1ad78edd20 in strdup ../../../../src/libsanitizer/asan/asan_interceptors.cpp:566 > > #1 0x55eda19cb76f in perf_data__open (linux/tools/perf/perf+0x65276f) (BuildId: 6fc1b7cdc123c7bd586ce55ea8b727875f42cda2) > > #2 0x55eda18ffafa in __perf_session__new (linux/tools/perf/perf+0x586afa) (BuildId: 6fc1b7cdc123c7bd586ce55ea8b727875f42cda2) > > #3 0x55eda15485d3 in cmd_annotate (linux/tools/perf/perf+0x1cf5d3) (BuildId: 6fc1b7cdc123c7bd586ce55ea8b727875f42cda2) > > #4 0x55eda1695467 in run_builtin (linux/tools/perf/perf+0x31c467) (BuildId: 6fc1b7cdc123c7bd586ce55ea8b727875f42cda2) > > #5 0x55eda1695c0e in handle_internal_command (linux/tools/perf/perf+0x31cc0e) (BuildId: 6fc1b7cdc123c7bd586ce55ea8b727875f42cda2) > > #6 0x55eda153ba72 in main (linux/tools/perf/perf+0x1c2a72) (BuildId: 6fc1b7cdc123c7bd586ce55ea8b727875f42cda2) > > #7 0x7f1acda43b89 in __libc_start_call_main ../sysdeps/nptl/libc_start_call_main.h:58 > > > > SUMMARY: AddressSanitizer: 33 byte(s) leaked in 1 allocation(s). > > Unexpected signal in test_basic > > > > No idea how it can leak the data->file.path (that's what I can find > > where strdup is used in the function). > > Maybe we need to revisit how much speed up it can give. I think this is a different issue and not sanitizer related except for the sanitizer catching the deliberate leak. My full build command line is: ``` $ make -C tools/perf O=/tmp/perf DEBUG=1 EXTRA_CFLAGS="-O0 -g -fno-omit-frame-pointer -fsanitize=address" CC=clang CXX=clang++ HOSTCC=clang ``` Of which the -O0 may be the thing that is making the difference and breaking the test with sanitizers for me. Thanks, Ian ^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2024-10-24 2:43 UTC | newest] Thread overview: 7+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2024-10-18 5:56 [PATCH v2] perf check: Add sanitizer feature and use to avoid test failure Ian Rogers 2024-10-22 5:41 ` Namhyung Kim 2024-10-22 17:39 ` Ian Rogers 2024-10-23 6:18 ` Namhyung Kim 2024-10-23 22:19 ` Namhyung Kim 2024-10-23 23:24 ` Namhyung Kim 2024-10-24 2:42 ` Ian Rogers
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).