Repository navigation
Update the docs and add some more information - #4
Merged
Merged
Conversation
andreasgal
added a commit
that referenced
this pull request
May 8, 2013
Update the docs and add some more information
Closed
Closed
This was referenced Oct 13, 2016
ghost
mentioned this pull request
bernhardu
added a commit
to bernhardu/rr
that referenced
this pull request
Jun 23, 2021
These two tests are stuck: nested_detach-no-syscallbuf nested_detach-32-no-syscallbuf For some reason just the no-syscallbuf variants get stuck, the regular ones succeed. $ bin/rr record -n bin/rr record --nested=detach bin/simple PID 1305400 uses 100% CPU, until timeout kills all. 1305387 └─ bin/rr record -n bin/rr record --nested=detach bin/simple 12 1305399 ├─ bin/rr record --nested=detach bin/simple 1 1305400 └─ bin/rr record --nested=detach bin/simple 1 $ gdb -q --pid 1305400 (gdb) add-symbol-file /usr/lib/x86_64-linux-gnu/libasan.so.6.0.0 0x7f818ad37a60 (gdb) add-symbol-file .../x86_64_asan/obj/bin/rr 0x564caa04e830 (gdb) add-symbol-file .../x86_64_asan/obj/lib/rr/librrpreload.so 0x7f818ad030d0 (gdb) stepi 0x00007f818ad50339 4150 int res = REAL(pthread_mutex_lock)(m); 1: x/i $pc => 0x7f818ad50339 <__interceptor_pthread_mutex_lock(void*)+25>: jmp *0xf4311(%rip) # 0x7f818ae44650 (gdb) bt #0 0x00007f818ad50339 in __interceptor_pthread_mutex_lock (m=<optimized out>) at ../../../../src/libsanitizer/sanitizer_common/sanitizer_common_interceptors.inc:4150 rr-debugger#1 __interceptor_pthread_mutex_lock (m=<optimized out>) at ../../../../src/libsanitizer/sanitizer_common/sanitizer_common_interceptors.inc:4146 rr-debugger#2 0x0000564caa0e2ddc in rr::CompressedWriter::CompressedWriter (this=0x6110000002c0, filename=..., block_size=1048576, num_threads=1) at .../rr/src/CompressedWriter.cc:62 rr-debugger#3 0x0000564caa5a9453 in rr::TraceWriter::TraceWriter (this=0x619000003380, file_name=..., output_trace_dir=..., ticks_semantics_=rr::TICKS_RETIRED_CONDITIONAL_BRANCHES) at .../rr/src/TraceStream.cc:1307 rr-debugger#4 0x0000564caa2b9ed2 in rr::RecordSession::RecordSession (this=0x619000003280, exe_path=..., argv=..., envp=..., disable_cpuid_features=..., syscallbuf=rr::RecordSession::ENABLE_SYSCALL_BUF, syscallbuf_desched_sig=30, bind_cpu=rr::BIND_CPU, output_trace_dir=..., trace_id=0x0, use_audit=false, unmap_vdso=false) at .../rr/src/RecordSession.cc:2296 rr-debugger#5 0x0000564caa2b92ea in rr::RecordSession::create (argv=..., extra_env=..., disable_cpuid_features=..., syscallbuf=rr::RecordSession::ENABLE_SYSCALL_BUF, syscallbuf_desched_sig=30 '\036', bind_cpu=rr::BIND_CPU, output_trace_dir=..., trace_id=0x0, use_audit=false, unmap_vdso=false, force_asan_active=false) at .../rr/src/RecordSession.cc:2261 rr-debugger#6 0x0000564caa29cd01 in rr::record (args=..., flags=...) at .../rr/src/RecordCommand.cc:632 rr-debugger#7 0x0000564caa29e90f in rr::RecordCommand::run (this=0x564caa956fe0, args=...) at .../rr/src/RecordCommand.cc:791 rr-debugger#8 0x0000564caa614909 in main (argc=4, argv=0x7ffd2a91dbe8) at .../rr/src/main.cc:271 ... (gdb) stepi pthread_mutex_lock (mutex=0x6110000002c8) at .../rr/src/preload/overrides.c:24 24 mutex->__data.__kind &= ~PTHREAD_MUTEX_PRIO_INHERIT_NP; 1: x/i $pc => 0x7f818ad08660 <pthread_mutex_lock>: mov 0x4ba9(%rip),%rax # 0x7f818ad0d210 (gdb) bt #0 pthread_mutex_lock (mutex=0x6110000002c8) at .../rr/src/preload/overrides.c:24 rr-debugger#1 0x0000564caa0e2ddc in rr::CompressedWriter::CompressedWriter (this=0x6110000002c0, filename=..., block_size=1048576, num_threads=1) at .../rr/src/CompressedWriter.cc:62 rr-debugger#2 0x0000564caa5a9453 in rr::TraceWriter::TraceWriter (this=0x619000003380, file_name=..., output_trace_dir=..., ticks_semantics_=rr::TICKS_RETIRED_CONDITIONAL_BRANCHES) at .../rr/src/TraceStream.cc:1307 ... (gdb) stepi 74 return __pthread_mutex_lock(mutex); 1: x/i $pc => 0x7f818ad08678 <pthread_mutex_lock+24>: jmp 0x7f818ad03070 (gdb) bt #0 pthread_mutex_lock (mutex=0x6110000002c8) at .../rr/src/preload/overrides.c:74 rr-debugger#1 0x0000564caa0e2ddc in rr::CompressedWriter::CompressedWriter (this=0x6110000002c0, filename=..., block_size=1048576, num_threads=1) at .../rr/src/CompressedWriter.cc:62 rr-debugger#2 0x0000564caa5a9453 in rr::TraceWriter::TraceWriter (this=0x619000003380, file_name=..., output_trace_dir=..., ticks_semantics_=rr::TICKS_RETIRED_CONDITIONAL_BRANCHES) at .../rr/src/TraceStream.cc:1307 ... (gdb) print environ $1 = (char **) 0x0 $ cat /proc/1305400/environ | tr '\0' '\n' | grep LD_PRELOAD LD_PRELOAD=libasan.so.6:.../x86_64_asan/obj/bin/../lib/rr/librrpreload.so:::
bernhardu
added a commit
to bernhardu/rr
that referenced
this pull request
Jun 24, 2021
These two tests are stuck: nested_detach-no-syscallbuf nested_detach-32-no-syscallbuf For some reason just the no-syscallbuf variants get stuck, the regular ones succeed. $ bin/rr record -n bin/rr record --nested=detach bin/simple PID 1305400 uses 100% CPU, until timeout kills all. 1305387 └─ bin/rr record -n bin/rr record --nested=detach bin/simple 12 1305399 ├─ bin/rr record --nested=detach bin/simple 1 1305400 └─ bin/rr record --nested=detach bin/simple 1 $ gdb -q --pid 1305400 (gdb) add-symbol-file /usr/lib/x86_64-linux-gnu/libasan.so.6.0.0 0x7f818ad37a60 (gdb) add-symbol-file .../x86_64_asan/obj/bin/rr 0x564caa04e830 (gdb) add-symbol-file .../x86_64_asan/obj/lib/rr/librrpreload.so 0x7f818ad030d0 (gdb) stepi 0x00007f818ad50339 4150 int res = REAL(pthread_mutex_lock)(m); 1: x/i $pc => 0x7f818ad50339 <__interceptor_pthread_mutex_lock(void*)+25>: jmp *0xf4311(%rip) # 0x7f818ae44650 (gdb) bt #0 0x00007f818ad50339 in __interceptor_pthread_mutex_lock (m=<optimized out>) at ../../../../src/libsanitizer/sanitizer_common/sanitizer_common_interceptors.inc:4150 rr-debugger#1 __interceptor_pthread_mutex_lock (m=<optimized out>) at ../../../../src/libsanitizer/sanitizer_common/sanitizer_common_interceptors.inc:4146 rr-debugger#2 0x0000564caa0e2ddc in rr::CompressedWriter::CompressedWriter (this=0x6110000002c0, filename=..., block_size=1048576, num_threads=1) at .../rr/src/CompressedWriter.cc:62 rr-debugger#3 0x0000564caa5a9453 in rr::TraceWriter::TraceWriter (this=0x619000003380, file_name=..., output_trace_dir=..., ticks_semantics_=rr::TICKS_RETIRED_CONDITIONAL_BRANCHES) at .../rr/src/TraceStream.cc:1307 rr-debugger#4 0x0000564caa2b9ed2 in rr::RecordSession::RecordSession (this=0x619000003280, exe_path=..., argv=..., envp=..., disable_cpuid_features=..., syscallbuf=rr::RecordSession::ENABLE_SYSCALL_BUF, syscallbuf_desched_sig=30, bind_cpu=rr::BIND_CPU, output_trace_dir=..., trace_id=0x0, use_audit=false, unmap_vdso=false) at .../rr/src/RecordSession.cc:2296 rr-debugger#5 0x0000564caa2b92ea in rr::RecordSession::create (argv=..., extra_env=..., disable_cpuid_features=..., syscallbuf=rr::RecordSession::ENABLE_SYSCALL_BUF, syscallbuf_desched_sig=30 '\036', bind_cpu=rr::BIND_CPU, output_trace_dir=..., trace_id=0x0, use_audit=false, unmap_vdso=false, force_asan_active=false) at .../rr/src/RecordSession.cc:2261 rr-debugger#6 0x0000564caa29cd01 in rr::record (args=..., flags=...) at .../rr/src/RecordCommand.cc:632 rr-debugger#7 0x0000564caa29e90f in rr::RecordCommand::run (this=0x564caa956fe0, args=...) at .../rr/src/RecordCommand.cc:791 rr-debugger#8 0x0000564caa614909 in main (argc=4, argv=0x7ffd2a91dbe8) at .../rr/src/main.cc:271 ... (gdb) stepi pthread_mutex_lock (mutex=0x6110000002c8) at .../rr/src/preload/overrides.c:24 24 mutex->__data.__kind &= ~PTHREAD_MUTEX_PRIO_INHERIT_NP; 1: x/i $pc => 0x7f818ad08660 <pthread_mutex_lock>: mov 0x4ba9(%rip),%rax # 0x7f818ad0d210 (gdb) bt #0 pthread_mutex_lock (mutex=0x6110000002c8) at .../rr/src/preload/overrides.c:24 rr-debugger#1 0x0000564caa0e2ddc in rr::CompressedWriter::CompressedWriter (this=0x6110000002c0, filename=..., block_size=1048576, num_threads=1) at .../rr/src/CompressedWriter.cc:62 rr-debugger#2 0x0000564caa5a9453 in rr::TraceWriter::TraceWriter (this=0x619000003380, file_name=..., output_trace_dir=..., ticks_semantics_=rr::TICKS_RETIRED_CONDITIONAL_BRANCHES) at .../rr/src/TraceStream.cc:1307 ... (gdb) stepi 74 return __pthread_mutex_lock(mutex); 1: x/i $pc => 0x7f818ad08678 <pthread_mutex_lock+24>: jmp 0x7f818ad03070 (gdb) bt #0 pthread_mutex_lock (mutex=0x6110000002c8) at .../rr/src/preload/overrides.c:74 rr-debugger#1 0x0000564caa0e2ddc in rr::CompressedWriter::CompressedWriter (this=0x6110000002c0, filename=..., block_size=1048576, num_threads=1) at .../rr/src/CompressedWriter.cc:62 rr-debugger#2 0x0000564caa5a9453 in rr::TraceWriter::TraceWriter (this=0x619000003380, file_name=..., output_trace_dir=..., ticks_semantics_=rr::TICKS_RETIRED_CONDITIONAL_BRANCHES) at .../rr/src/TraceStream.cc:1307 ... (gdb) print environ $1 = (char **) 0x0 $ cat /proc/1305400/environ | tr '\0' '\n' | grep LD_PRELOAD LD_PRELOAD=libasan.so.6:.../x86_64_asan/obj/bin/../lib/rr/librrpreload.so:::
bernhardu
added a commit
to bernhardu/rr
that referenced
this pull request
Jun 25, 2021
These two tests are stuck: nested_detach-no-syscallbuf nested_detach-32-no-syscallbuf For some reason just the no-syscallbuf variants get stuck, the regular ones succeed. $ bin/rr record -n bin/rr record --nested=detach bin/simple PID 1305400 uses 100% CPU, until timeout kills all. 1305387 └─ bin/rr record -n bin/rr record --nested=detach bin/simple 12 1305399 ├─ bin/rr record --nested=detach bin/simple 1 1305400 └─ bin/rr record --nested=detach bin/simple 1 $ gdb -q --pid 1305400 (gdb) add-symbol-file /usr/lib/x86_64-linux-gnu/libasan.so.6.0.0 0x7f818ad37a60 (gdb) add-symbol-file .../x86_64_asan/obj/bin/rr 0x564caa04e830 (gdb) add-symbol-file .../x86_64_asan/obj/lib/rr/librrpreload.so 0x7f818ad030d0 (gdb) stepi 0x00007f818ad50339 4150 int res = REAL(pthread_mutex_lock)(m); 1: x/i $pc => 0x7f818ad50339 <__interceptor_pthread_mutex_lock(void*)+25>: jmp *0xf4311(%rip) # 0x7f818ae44650 (gdb) bt #0 0x00007f818ad50339 in __interceptor_pthread_mutex_lock (m=<optimized out>) at ../../../../src/libsanitizer/sanitizer_common/sanitizer_common_interceptors.inc:4150 rr-debugger#1 __interceptor_pthread_mutex_lock (m=<optimized out>) at ../../../../src/libsanitizer/sanitizer_common/sanitizer_common_interceptors.inc:4146 rr-debugger#2 0x0000564caa0e2ddc in rr::CompressedWriter::CompressedWriter (this=0x6110000002c0, filename=..., block_size=1048576, num_threads=1) at .../rr/src/CompressedWriter.cc:62 rr-debugger#3 0x0000564caa5a9453 in rr::TraceWriter::TraceWriter (this=0x619000003380, file_name=..., output_trace_dir=..., ticks_semantics_=rr::TICKS_RETIRED_CONDITIONAL_BRANCHES) at .../rr/src/TraceStream.cc:1307 rr-debugger#4 0x0000564caa2b9ed2 in rr::RecordSession::RecordSession (this=0x619000003280, exe_path=..., argv=..., envp=..., disable_cpuid_features=..., syscallbuf=rr::RecordSession::ENABLE_SYSCALL_BUF, syscallbuf_desched_sig=30, bind_cpu=rr::BIND_CPU, output_trace_dir=..., trace_id=0x0, use_audit=false, unmap_vdso=false) at .../rr/src/RecordSession.cc:2296 rr-debugger#5 0x0000564caa2b92ea in rr::RecordSession::create (argv=..., extra_env=..., disable_cpuid_features=..., syscallbuf=rr::RecordSession::ENABLE_SYSCALL_BUF, syscallbuf_desched_sig=30 '\036', bind_cpu=rr::BIND_CPU, output_trace_dir=..., trace_id=0x0, use_audit=false, unmap_vdso=false, force_asan_active=false) at .../rr/src/RecordSession.cc:2261 rr-debugger#6 0x0000564caa29cd01 in rr::record (args=..., flags=...) at .../rr/src/RecordCommand.cc:632 rr-debugger#7 0x0000564caa29e90f in rr::RecordCommand::run (this=0x564caa956fe0, args=...) at .../rr/src/RecordCommand.cc:791 rr-debugger#8 0x0000564caa614909 in main (argc=4, argv=0x7ffd2a91dbe8) at .../rr/src/main.cc:271 ... (gdb) stepi pthread_mutex_lock (mutex=0x6110000002c8) at .../rr/src/preload/overrides.c:24 24 mutex->__data.__kind &= ~PTHREAD_MUTEX_PRIO_INHERIT_NP; 1: x/i $pc => 0x7f818ad08660 <pthread_mutex_lock>: mov 0x4ba9(%rip),%rax # 0x7f818ad0d210 (gdb) bt #0 pthread_mutex_lock (mutex=0x6110000002c8) at .../rr/src/preload/overrides.c:24 rr-debugger#1 0x0000564caa0e2ddc in rr::CompressedWriter::CompressedWriter (this=0x6110000002c0, filename=..., block_size=1048576, num_threads=1) at .../rr/src/CompressedWriter.cc:62 rr-debugger#2 0x0000564caa5a9453 in rr::TraceWriter::TraceWriter (this=0x619000003380, file_name=..., output_trace_dir=..., ticks_semantics_=rr::TICKS_RETIRED_CONDITIONAL_BRANCHES) at .../rr/src/TraceStream.cc:1307 ... (gdb) stepi 74 return __pthread_mutex_lock(mutex); 1: x/i $pc => 0x7f818ad08678 <pthread_mutex_lock+24>: jmp 0x7f818ad03070 (gdb) bt #0 pthread_mutex_lock (mutex=0x6110000002c8) at .../rr/src/preload/overrides.c:74 rr-debugger#1 0x0000564caa0e2ddc in rr::CompressedWriter::CompressedWriter (this=0x6110000002c0, filename=..., block_size=1048576, num_threads=1) at .../rr/src/CompressedWriter.cc:62 rr-debugger#2 0x0000564caa5a9453 in rr::TraceWriter::TraceWriter (this=0x619000003380, file_name=..., output_trace_dir=..., ticks_semantics_=rr::TICKS_RETIRED_CONDITIONAL_BRANCHES) at .../rr/src/TraceStream.cc:1307 ... (gdb) print environ $1 = (char **) 0x0 $ cat /proc/1305400/environ | tr '\0' '\n' | grep LD_PRELOAD LD_PRELOAD=libasan.so.6:.../x86_64_asan/obj/bin/../lib/rr/librrpreload.so:::
bernhardu
added a commit
to bernhardu/rr
that referenced
this pull request
Jun 25, 2021
These two tests are stuck: nested_detach-no-syscallbuf nested_detach-32-no-syscallbuf For some reason just the no-syscallbuf variants get stuck, the regular ones succeed. $ bin/rr record -n bin/rr record --nested=detach bin/simple PID 1305400 uses 100% CPU, until timeout kills all. 1305387 └─ bin/rr record -n bin/rr record --nested=detach bin/simple 12 1305399 ├─ bin/rr record --nested=detach bin/simple 1 1305400 └─ bin/rr record --nested=detach bin/simple 1 $ gdb -q --pid 1305400 (gdb) add-symbol-file /usr/lib/x86_64-linux-gnu/libasan.so.6.0.0 0x7f818ad37a60 (gdb) add-symbol-file .../x86_64_asan/obj/bin/rr 0x564caa04e830 (gdb) add-symbol-file .../x86_64_asan/obj/lib/rr/librrpreload.so 0x7f818ad030d0 (gdb) stepi 0x00007f818ad50339 4150 int res = REAL(pthread_mutex_lock)(m); 1: x/i $pc => 0x7f818ad50339 <__interceptor_pthread_mutex_lock(void*)+25>: jmp *0xf4311(%rip) # 0x7f818ae44650 (gdb) bt #0 0x00007f818ad50339 in __interceptor_pthread_mutex_lock (m=<optimized out>) at ../../../../src/libsanitizer/sanitizer_common/sanitizer_common_interceptors.inc:4150 rr-debugger#1 __interceptor_pthread_mutex_lock (m=<optimized out>) at ../../../../src/libsanitizer/sanitizer_common/sanitizer_common_interceptors.inc:4146 rr-debugger#2 0x0000564caa0e2ddc in rr::CompressedWriter::CompressedWriter (this=0x6110000002c0, filename=..., block_size=1048576, num_threads=1) at .../rr/src/CompressedWriter.cc:62 rr-debugger#3 0x0000564caa5a9453 in rr::TraceWriter::TraceWriter (this=0x619000003380, file_name=..., output_trace_dir=..., ticks_semantics_=rr::TICKS_RETIRED_CONDITIONAL_BRANCHES) at .../rr/src/TraceStream.cc:1307 rr-debugger#4 0x0000564caa2b9ed2 in rr::RecordSession::RecordSession (this=0x619000003280, exe_path=..., argv=..., envp=..., disable_cpuid_features=..., syscallbuf=rr::RecordSession::ENABLE_SYSCALL_BUF, syscallbuf_desched_sig=30, bind_cpu=rr::BIND_CPU, output_trace_dir=..., trace_id=0x0, use_audit=false, unmap_vdso=false) at .../rr/src/RecordSession.cc:2296 rr-debugger#5 0x0000564caa2b92ea in rr::RecordSession::create (argv=..., extra_env=..., disable_cpuid_features=..., syscallbuf=rr::RecordSession::ENABLE_SYSCALL_BUF, syscallbuf_desched_sig=30 '\036', bind_cpu=rr::BIND_CPU, output_trace_dir=..., trace_id=0x0, use_audit=false, unmap_vdso=false, force_asan_active=false) at .../rr/src/RecordSession.cc:2261 rr-debugger#6 0x0000564caa29cd01 in rr::record (args=..., flags=...) at .../rr/src/RecordCommand.cc:632 rr-debugger#7 0x0000564caa29e90f in rr::RecordCommand::run (this=0x564caa956fe0, args=...) at .../rr/src/RecordCommand.cc:791 rr-debugger#8 0x0000564caa614909 in main (argc=4, argv=0x7ffd2a91dbe8) at .../rr/src/main.cc:271 ... (gdb) stepi pthread_mutex_lock (mutex=0x6110000002c8) at .../rr/src/preload/overrides.c:24 24 mutex->__data.__kind &= ~PTHREAD_MUTEX_PRIO_INHERIT_NP; 1: x/i $pc => 0x7f818ad08660 <pthread_mutex_lock>: mov 0x4ba9(%rip),%rax # 0x7f818ad0d210 (gdb) bt #0 pthread_mutex_lock (mutex=0x6110000002c8) at .../rr/src/preload/overrides.c:24 rr-debugger#1 0x0000564caa0e2ddc in rr::CompressedWriter::CompressedWriter (this=0x6110000002c0, filename=..., block_size=1048576, num_threads=1) at .../rr/src/CompressedWriter.cc:62 rr-debugger#2 0x0000564caa5a9453 in rr::TraceWriter::TraceWriter (this=0x619000003380, file_name=..., output_trace_dir=..., ticks_semantics_=rr::TICKS_RETIRED_CONDITIONAL_BRANCHES) at .../rr/src/TraceStream.cc:1307 ... (gdb) stepi 74 return __pthread_mutex_lock(mutex); 1: x/i $pc => 0x7f818ad08678 <pthread_mutex_lock+24>: jmp 0x7f818ad03070 (gdb) bt #0 pthread_mutex_lock (mutex=0x6110000002c8) at .../rr/src/preload/overrides.c:74 rr-debugger#1 0x0000564caa0e2ddc in rr::CompressedWriter::CompressedWriter (this=0x6110000002c0, filename=..., block_size=1048576, num_threads=1) at .../rr/src/CompressedWriter.cc:62 rr-debugger#2 0x0000564caa5a9453 in rr::TraceWriter::TraceWriter (this=0x619000003380, file_name=..., output_trace_dir=..., ticks_semantics_=rr::TICKS_RETIRED_CONDITIONAL_BRANCHES) at .../rr/src/TraceStream.cc:1307 ... (gdb) print environ $1 = (char **) 0x0 $ cat /proc/1305400/environ | tr '\0' '\n' | grep LD_PRELOAD LD_PRELOAD=libasan.so.6:.../x86_64_asan/obj/bin/../lib/rr/librrpreload.so:::
bernhardu
added a commit
to bernhardu/rr
that referenced
this pull request
Jul 5, 2021
These two tests are stuck: nested_detach-no-syscallbuf nested_detach-32-no-syscallbuf For some reason just the no-syscallbuf variants get stuck, the regular ones succeed. $ bin/rr record -n bin/rr record --nested=detach bin/simple PID 1305400 uses 100% CPU, until timeout kills all. 1305387 └─ bin/rr record -n bin/rr record --nested=detach bin/simple 12 1305399 ├─ bin/rr record --nested=detach bin/simple 1 1305400 └─ bin/rr record --nested=detach bin/simple 1 $ gdb -q --pid 1305400 (gdb) add-symbol-file /usr/lib/x86_64-linux-gnu/libasan.so.6.0.0 0x7f818ad37a60 (gdb) add-symbol-file .../x86_64_asan/obj/bin/rr 0x564caa04e830 (gdb) add-symbol-file .../x86_64_asan/obj/lib/rr/librrpreload.so 0x7f818ad030d0 (gdb) stepi 0x00007f818ad50339 4150 int res = REAL(pthread_mutex_lock)(m); 1: x/i $pc => 0x7f818ad50339 <__interceptor_pthread_mutex_lock(void*)+25>: jmp *0xf4311(%rip) # 0x7f818ae44650 (gdb) bt #0 0x00007f818ad50339 in __interceptor_pthread_mutex_lock (m=<optimized out>) at ../../../../src/libsanitizer/sanitizer_common/sanitizer_common_interceptors.inc:4150 rr-debugger#1 __interceptor_pthread_mutex_lock (m=<optimized out>) at ../../../../src/libsanitizer/sanitizer_common/sanitizer_common_interceptors.inc:4146 rr-debugger#2 0x0000564caa0e2ddc in rr::CompressedWriter::CompressedWriter (this=0x6110000002c0, filename=..., block_size=1048576, num_threads=1) at .../rr/src/CompressedWriter.cc:62 rr-debugger#3 0x0000564caa5a9453 in rr::TraceWriter::TraceWriter (this=0x619000003380, file_name=..., output_trace_dir=..., ticks_semantics_=rr::TICKS_RETIRED_CONDITIONAL_BRANCHES) at .../rr/src/TraceStream.cc:1307 rr-debugger#4 0x0000564caa2b9ed2 in rr::RecordSession::RecordSession (this=0x619000003280, exe_path=..., argv=..., envp=..., disable_cpuid_features=..., syscallbuf=rr::RecordSession::ENABLE_SYSCALL_BUF, syscallbuf_desched_sig=30, bind_cpu=rr::BIND_CPU, output_trace_dir=..., trace_id=0x0, use_audit=false, unmap_vdso=false) at .../rr/src/RecordSession.cc:2296 rr-debugger#5 0x0000564caa2b92ea in rr::RecordSession::create (argv=..., extra_env=..., disable_cpuid_features=..., syscallbuf=rr::RecordSession::ENABLE_SYSCALL_BUF, syscallbuf_desched_sig=30 '\036', bind_cpu=rr::BIND_CPU, output_trace_dir=..., trace_id=0x0, use_audit=false, unmap_vdso=false, force_asan_active=false) at .../rr/src/RecordSession.cc:2261 rr-debugger#6 0x0000564caa29cd01 in rr::record (args=..., flags=...) at .../rr/src/RecordCommand.cc:632 rr-debugger#7 0x0000564caa29e90f in rr::RecordCommand::run (this=0x564caa956fe0, args=...) at .../rr/src/RecordCommand.cc:791 rr-debugger#8 0x0000564caa614909 in main (argc=4, argv=0x7ffd2a91dbe8) at .../rr/src/main.cc:271 ... (gdb) stepi pthread_mutex_lock (mutex=0x6110000002c8) at .../rr/src/preload/overrides.c:24 24 mutex->__data.__kind &= ~PTHREAD_MUTEX_PRIO_INHERIT_NP; 1: x/i $pc => 0x7f818ad08660 <pthread_mutex_lock>: mov 0x4ba9(%rip),%rax # 0x7f818ad0d210 (gdb) bt #0 pthread_mutex_lock (mutex=0x6110000002c8) at .../rr/src/preload/overrides.c:24 rr-debugger#1 0x0000564caa0e2ddc in rr::CompressedWriter::CompressedWriter (this=0x6110000002c0, filename=..., block_size=1048576, num_threads=1) at .../rr/src/CompressedWriter.cc:62 rr-debugger#2 0x0000564caa5a9453 in rr::TraceWriter::TraceWriter (this=0x619000003380, file_name=..., output_trace_dir=..., ticks_semantics_=rr::TICKS_RETIRED_CONDITIONAL_BRANCHES) at .../rr/src/TraceStream.cc:1307 ... (gdb) stepi 74 return __pthread_mutex_lock(mutex); 1: x/i $pc => 0x7f818ad08678 <pthread_mutex_lock+24>: jmp 0x7f818ad03070 (gdb) bt #0 pthread_mutex_lock (mutex=0x6110000002c8) at .../rr/src/preload/overrides.c:74 rr-debugger#1 0x0000564caa0e2ddc in rr::CompressedWriter::CompressedWriter (this=0x6110000002c0, filename=..., block_size=1048576, num_threads=1) at .../rr/src/CompressedWriter.cc:62 rr-debugger#2 0x0000564caa5a9453 in rr::TraceWriter::TraceWriter (this=0x619000003380, file_name=..., output_trace_dir=..., ticks_semantics_=rr::TICKS_RETIRED_CONDITIONAL_BRANCHES) at .../rr/src/TraceStream.cc:1307 ... (gdb) print environ $1 = (char **) 0x0 $ cat /proc/1305400/environ | tr '\0' '\n' | grep LD_PRELOAD LD_PRELOAD=libasan.so.6:.../x86_64_asan/obj/bin/../lib/rr/librrpreload.so:::
bernhardu
added a commit
to bernhardu/rr
that referenced
this pull request
Apr 7, 2023
…aks=0.
==2156196==Running thread 2156188 was not suspended. False leaks are possible.
=================================================================
==2156193==ERROR: LeakSanitizer: detected memory leaks
Direct leak of 10525163 byte(s) in 11 object(s) allocated from:
#0 0x7f41d28b94c8 in operator new(unsigned long) ../../../../src/libsanitizer/asan/asan_new_delete.cpp:95
rr-debugger#1 0x55c53c6587a4 in std::__new_allocator<unsigned char>::allocate(unsigned long, void const*) /usr/include/c++/12/bits/new_allocator.h:137
rr-debugger#2 0x55c53c6587a4 in std::allocator_traits<std::allocator<unsigned char> >::allocate(std::allocator<unsigned char>&, unsigned long) /usr/include/c++/12/bits/alloc_traits.h:464
rr-debugger#3 0x55c53c6587a4 in std::_Vector_base<unsigned char, std::allocator<unsigned char> >::_M_allocate(unsigned long) /usr/include/c++/12/bits/stl_vector.h:378
rr-debugger#4 0x55c53c6587a4 in std::vector<unsigned char, std::allocator<unsigned char> >::_M_default_append(unsigned long) /usr/include/c++/12/bits/vector.tcc:650
rr-debugger#5 0x55c53c69370b in std::vector<unsigned char, std::allocator<unsigned char> >::resize(unsigned long) /usr/include/c++/12/bits/stl_vector.h:1011
rr-debugger#6 0x55c53c69370b in rr::CompressedWriter::compression_thread() /home/bernhard/data/entwicklung/2023/rr/2022-09-11/rr/src/CompressedWriter.cc:162
rr-debugger#7 0x55c53c693908 in rr::CompressedWriter::compression_thread_callback(void*) /home/bernhard/data/entwicklung/2023/rr/2022-09-11/rr/src/CompressedWriter.cc:29
rr-debugger#8 0x7f41d22a7fd3 in start_thread nptl/pthread_create.c:442
SUMMARY: AddressSanitizer: 10525163 byte(s) leaked in 11 allocation(s).
Keno
added a commit
to ChronitonAI/rr
that referenced
this pull request
Oct 6, 2026
When rr removes a range from its model of an address space, it also unmaps its local copies of the mappings in the range (rr maps syscall buffers into its own address space too). The range passed to the callback starts where the previous mapping ended, so it can start before the mapping. Then the local address computed for the munmap was below the mapping's local copy, and the size was too large, so rr unmapped other memory of its own below that copy. 1be3973 limited the end of the local munmap to the end of the mapping, but not its start. This happens in replay: when a vfork child execs, its syscall buffer and scratch memory stay behind in the address space it shared with its parent. rr unmaps them while recording (since a6a80fd), but not in replay, so they stay in replay's model with their local copies. When the tracee later maps memory over a range that starts below the dead syscall buffer, which is free in the recording, rr unmapped the local copy of a live syscall buffer that it had mapped right below the dead one in its own address space, and then crashed when it read that buffer: Program received signal SIGSEGV #0 __memmove_avx_unaligned_erms () #1 rr::Task::read_bytes_fallible (...) at src/Task.cc:3176 ... rr-debugger#4 rr::ReplaySession::setup_replay_one_trace_frame (...) A program whose threads spawn processes with posix_spawn() hit this in replay of 1 of 40 recordings with --chaos: replay maps a new task's scratch memory at the address it had in the recording, over 1 MiB of free memory and a dead syscall buffer. So unmap the local copy of only the part of the range that the mapping covers. Test: in vfork_exec_map_over_buffers, a vfork child sets up its syscall buffer and execs a process that sets up its own and then waits. Then the parent maps memory over 16 MiB of free memory below where the child's syscall buffer was, and over the buffer itself, and the exec'd process uses its syscall buffer again. To put free memory below the child's syscall buffer, the parent first fills the memory after its own syscall buffer, where rr puts the next one, and frees it after the child execs. This is deterministic. Without the syscall buffer the test does nothing. Results on x86_64 (Linux 7.0, glibc 2.31): - Without this change, the test fails in every run with the syscall buffer (20 of 20 per bitness: rr crashes in replay), and passes without it. - With this change, it passes in 50 of 50 runs in each of the four variants (64- and 32-bit, with and without the syscall buffer), and natively in 500 of 500 runs per bitness. - The recording that crashed in replay now replays to the end, with the same output as the recording. - The full test suite shows no new failures. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Keno
added a commit
to ChronitonAI/rr
that referenced
this pull request
Oct 6, 2026
…e starts before it When rr removes a range from its model of an address space, it also unmaps its local copies of the mappings in the range (rr maps syscall buffers into its own address space too). The range passed to the callback is the part of the range not yet processed: it starts at the start of the range, or where the previous mapping in it ended, so it can start before the mapping. Then the local address computed for the munmap was below the mapping's local copy, and the size was too large, so rr unmapped the whole local copy and as much of its own memory below it as the range extends below the mapping, whether the range covers the mapping fully or partly. 1be3973 limited the end of the local munmap to the end of the mapping, but not its start. This happens in replay: when a vfork child execs, its syscall buffer and scratch memory stay behind in the address space it shared with its parent. rr unmaps them while recording (since a6a80fd), but not in replay (the next commit fixes that), so they stay in replay's model with their local copies. When the tracee later maps memory over a range that starts below the dead syscall buffer, which is free in the recording, rr unmapped the local copy of a live syscall buffer that it had mapped right below the dead one in its own address space, and then crashed when it read that buffer: Program received signal SIGSEGV #0 __memmove_avx_unaligned_erms () #1 rr::Task::read_bytes_fallible (...) at src/Task.cc:3176 ... rr-debugger#4 rr::ReplaySession::setup_replay_one_trace_frame (...) A program whose threads spawn processes with posix_spawn() hit this in replay of 1 of 40 recordings with --chaos: replay maps a new task's scratch memory at the address it had in the recording, over 1 MiB of free memory and a dead syscall buffer. So unmap the local copy of only the part of the range that the mapping covers. Test: in vfork_exec_map_over_buffers, a vfork child sets up its syscall buffer and execs a process that sets up its own and then waits. Then the parent maps memory over 16 MiB of free memory below where the child's syscall buffer was, and over the buffer itself, and the exec'd process uses its syscall buffer again. To put free memory below the child's syscall buffer, the parent first fills the memory after its own syscall buffer, where rr puts the next one, and frees it after the child execs. This is deterministic. Without the syscall buffer the test does nothing. After the next commit, replay no longer has the dead buffer in its model, so the test doesn't reach this path anymore; the fix still applies to any range that starts below a local mapping. Results on x86_64 (Linux 7.0, glibc 2.31): - Without this change, the test fails in every run with the syscall buffer (20 of 20 per bitness: rr crashes in replay), and passes without it. - With this change, it passes in 50 of 50 runs in each of the four variants (64- and 32-bit, with and without the syscall buffer), and natively in 500 of 500 runs per bitness. - The recording that crashed in replay now replays to the end, with the same output as the recording. - The full test suite shows no new failures. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Mostly style/formatting cleanup. Added some information in a few places too.