mirror of
https://github.com/ceph/ceph
synced 2026-08-01 22:45:39 +00:00
320a3a0529
The python binaries on some CI images and dev boxes ship stripped, so the allocator frames in an interpreter-shutdown leak come through unsymbolised (/usr/bin/python3.13+0x...) and the function-name matches above cannot apply. leak:python3.10 already handled this for 3.10, and a stale comment claimed 3.12 does not leak. Add leak:python3.12 and leak:python3.13, mirroring the 3.10 entry, so the interpreter globals are suppressed whatever CPython the build uses. Signed-off-by: Kefu Chai <k.chai@proxmox.com>
77 lines
2.6 KiB
Plaintext
77 lines
2.6 KiB
Plaintext
# leak suppressions needed to run vstart.sh with WITH_ASAN=ON
|
|
# to enable, add this to the environment:
|
|
# LSAN_OPTIONS="suppressions=../qa/lsan.supp"
|
|
# export ASAN_OPTIONS="detect_odr_violation=0"
|
|
|
|
# gperftools allocates a singleton of MallocExtension and never frees it
|
|
leak:^MallocExtension::Initialize
|
|
|
|
# from perfglue/heap_profiler.cc
|
|
# gperftools allocates a singleton and never frees it
|
|
leak:^InitModule
|
|
|
|
# from src/ceph.in
|
|
# python3.6
|
|
leak:^_PyObject_Alloc
|
|
leak:^_PyObject_GC_Resize
|
|
leak:^PyBytes_FromStringAndSize
|
|
leak:^PyType_GenericAlloc
|
|
leak:^set_table_resize
|
|
# python3.7
|
|
leak:^_PyObject_Realloc
|
|
leak:^PyObject_Malloc
|
|
# python3.8
|
|
leak:^_PyBytes_FromSize
|
|
leak:^_PyObject_GC_Alloc
|
|
leak:^PyMem_Calloc
|
|
leak:^PyUnicode_New
|
|
# python3.9, 3.10, 3.11
|
|
leak:^PyMem_Malloc
|
|
leak:^PyDict_Copy
|
|
leak:^_PyObject_GC_NewVar
|
|
# Catch unsymbolised interpreter frames in the standalone python binary used by
|
|
# the bin/ceph wrapper and in the libpython embedded by C++ unittests such as
|
|
# unittest_mgr_pyformatter / unittest_mgr_pyutil. Not all CPython internal
|
|
# helpers are exported in .dynsym, so the function-name matches above don't
|
|
# cover everything; this substring-matches the module path in the stack frame.
|
|
leak:python3.10
|
|
leak:python3.12
|
|
leak:python3.13
|
|
# python3.13
|
|
leak:^PyModule_ExecDef
|
|
|
|
# following suppression mirrors qa/valgrind.supp
|
|
# ignore ec plugin factory (FIXME SOMEDAY)
|
|
leak:^ceph::ErasureCodePluginRegistry::load
|
|
|
|
# SQLite internal allocations
|
|
# These are indirect leaks from SQLite's internal memory management
|
|
leak:libsqlite3.so
|
|
|
|
# boost.thread main-thread TLS; freed only when a boost-created thread exits, so
|
|
# still reachable at exit. Benign.
|
|
leak:boost::detail::make_external_thread_data
|
|
|
|
# OpenSSL one-time init; ForkDeathTest children _exit() without OPENSSL_cleanup.
|
|
leak:^OPENSSL_init_crypto
|
|
|
|
# Still-reachable OpenSSL / libcryptsetup state with no API to free at exit;
|
|
# seen in the librbd encryption and migration unittests under ASan.
|
|
#
|
|
# libcryptsetup OpenSSL cipher/per-thread state that survives crypt_free().
|
|
leak:libcryptsetup
|
|
# OpenSSL DRBG global state, freed only by OPENSSL_cleanup (never called here).
|
|
leak:^RAND_bytes_ex
|
|
leak:^RAND_priv_bytes_ex
|
|
# OpenSSL per-thread error stack/strings, freed only by OPENSSL_thread_stop.
|
|
leak:^ossl_err_get_state_int
|
|
leak:^ERR_add_error_vdata
|
|
|
|
# unittest-seastore stops the seastar reactor (SeastarRunner, on its own thread)
|
|
# without draining its pending tasks, so a few cached extents those tasks still
|
|
# held are leaked at shutdown. Three anchors for the three seastore subsystems
|
|
# they come from.
|
|
leak:crimson::os::seastore::Cache::
|
|
leak:crimson::os::seastore::omap_manager
|
|
leak:crimson::os::seastore::FixedKVBtree
|
