ftmemsim-valgrind/glibc-2.X-helgrind.supp.in
Mark Wielaard a1364805fc Update helgrind and drd suppression libc and libpthread paths in glibc 2.34
glibc 2.34 moved all pthread functions into the main libc library.
And it changed the (in memory) path of the main libc library to
libc.so.6 (before it was libc-2.xx.so).

This breaks various standard suppressions for helgrind and drd.
Fix this by doing a configure check for whether we are using glibc
2.34 by checking whether pthread_create is in libc instead of in
libpthread. If we are using glibc then define GLIBC_LIBC_PATH and
GLIBC_LIBPTHREAD_PATH variables that point to the (regexp) path
of the library that contains all libc functions and pthread functions
(which will be the same path for glibc 2.34+).

Rename glibc-2.34567-NPTL-helgrind.supp to glibc-2.X-helgrind.supp.in
and glibc-2.X-drd.supp to glibc-2.X-drd.supp.in and replace the
GLIBC_LIBC_PATH and GLIBC_LIBPTHREAD_PATH at configure time.

The same could be done for the glibc-2.X.supp.in file, but hasn't
yet because it looks like most suppressions in that file are obsolete.
2021-07-21 18:31:26 +02:00

304 lines
6.4 KiB
Plaintext

# IMPORTANT: DO NOT EDIT glibc-2.X-helgrind.supp, as it is as a generated
# file. Instead edit glibc-2.X-helgrind.supp.in.
# FIXME 22 Jan 09: helgrind-glibc2X-005 overlaps with a lot of
# other stuff. They should be removed.
##----------------------------------------------------------------------##
# Suppressions for the Helgrind tool when using
# a glibc-2.{3,4,5,6,7,8,9} system
####################################################
# glibc-2.X specific
# These are generic cover-alls which catch a lot of stuff
# in various combinations of ld, libc and libpthread
#
# Note this is heavyhanded and not very clever:
#
# - suppress anything that has its top frame in ld.so
# That's fine, since it's mostly dynamic linking stuff,
# which has various deliberate (harmless) races
#
# - suppress anything that has its top frame in libc.so.
# This really isn't clever, since it could hide some
# legitimate races. But the problem is, if we don't do
# this, then loads of errors to do with stdio are reported, because
# H fails to see glibc's internal locking/unlocking of FILE*s
# as required by POSIX. A better solution is needed.
#
# - some of the stdio functions in newer glibc manipulate stdio
# FILE*s state through mempcpy, which we intercept, so we also need
# to suppress such manipulations.
#{
# helgrind-glibc2X-001
# Helgrind:Race
# obj:*/lib*/ld-2.*so*
#}
# helgrind-glibc2X-002 was merged into helgrind-glibc2X-001
# helgrind-glibc2X-003 was merged into helgrind-glibc2X-001
{
helgrind-glibc2X-004
Helgrind:Race
obj:@GLIBC_LIBC_PATH@
}
{
helgrind-glibc-io-xsputn-mempcpy
Helgrind:Race
fun:__GI_mempcpy
fun:_IO_*xsputn*
obj:@GLIBC_LIBC_PATH@
}
{
helgrind-glibc2X-005
Helgrind:Race
obj:@GLIBC_LIBPTHREAD_PATH@
}
# helgrind-glibc2X-006 was merged into helgrind-glibc2X-005
# helgrind-glibc2X-007 was merged into helgrind-glibc2X-001
# helgrind-glibc2X-008 was merged into helgrind-glibc2X-004
# helgrind-glibc2X-009 was merged into helgrind-glibc2X-004
# helgrind-glibc2X-010 was merged into helgrind-glibc2X-001
# helgrind-glibc2X-011 was merged into helgrind-glibc2X-004
# helgrind-glibc2X-012 was merged into helgrind-glibc2X-001
# helgrind-glibc2X-013 was merged into helgrind-glibc2X-001
# helgrind-glibc2X-014 was merged into helgrind-glibc2X-001
# helgrind-glibc2X-015 was merged into helgrind-glibc2X-004
# helgrind-glibc2X-016 was merged into helgrind-glibc2X-004
# These are very ugly. They are needed to suppress errors inside (eg)
# NPTL's pthread_cond_signal. Why only one stack frame -- at least we
# should see the wrapper calling the real functions, right?
# Unfortunately, no: the real functions are handwritten assembly (in
# the glibc-2.5 sources) and does not create a proper stack frame.
# Therefore it's only one level of unwinding before we're back out in
# user code rather than the 2 levels you'd expect.
{
helgrind-glibc2X-101
Helgrind:Race
obj:@GLIBC_LIBPTHREAD_PATH@
fun:pthread_*
}
{
helgrind-glibc2X-102
Helgrind:Race
fun:mythread_wrapper
obj:@GLIBC_LIBPTHREAD_PATH@
}
{
helgrind-glibc2X-103
Helgrind:Race
fun:pthread_cond_*@@GLIBC_2.*
}
{
helgrind-glibc2X-104
Helgrind:Race
fun:__lll_mutex_*
}
{
helgrind-glibc2X-105
Helgrind:Race
fun:pthread_rwlock_*lock*
}
{
helgrind-glibc2X-106
Helgrind:Race
fun:__lll_lock_wait
}
{
helgrind-glibc2X-107
Helgrind:Race
obj:@GLIBC_LIBPTHREAD_PATH@
fun:sem_*
}
{
helgrind-glibc2X-108
Helgrind:Race
fun:clone
}
{
helgrind-glibc2X-109
Helgrind:Race
fun:start_thread
}
{
helgrind-glibc2X-110
Helgrind:Race
obj:@GLIBC_LIBC_PATH@
fun:pthread_*
}
{
helgrind-glibc2X-111
Helgrind:Race
fun:__lll_*lock_*
}
{
helgrind-glibc2X-113
Helgrind:Race
fun:pthread_barrier_wait*
}
####################################################
# qt4 specific (GNU mangling)
#
{
helgrind-qt4---QMutex::lock()-QMutex::lock()
Helgrind:Race
...
fun:_ZN6QMutex4lockEv
fun:_ZN6QMutex4lockEv
}
{
helgrind-qt4---QMutex::unlock()-QMutex::unlock()
Helgrind:Race
...
fun:_ZN6QMutex6unlockEv
fun:_ZN6QMutex6unlockEv
}
{
helgrind-qt4---pthread_setspecific-QThreadPrivate::start(void*)
Helgrind:Race
fun:pthread_setspecific
fun:_ZN14QThreadPrivate5startEPv
}
####################################################
# Other stuff.
#
# pthread_exit apparently calls some kind of unwind
# mechanism - maybe to remove some number of frames
# from the thread's stack, so as to get back to the
# outermost frame for the thread? Anyway..
{
helgrind---*Unwind*-...-pthread_exit
Helgrind:Race
fun:*Unwind*
...
fun:pthread_exit
}
{
helgrind---...-*Unwind*-*pthread_unwind*
Helgrind:Race
...
fun:*Unwind*
fun:*pthread_unwind*
}
{
helgrind---...-*Unwind*-*pthread_unwind*
Helgrind:Race
...
fun:_Unwind*
...
fun:_Unwind_Backtrace
}
####################################################
# To do with thread stack allocation and deallocation?
#
{
helgrind---free_stacks-__deallocate_stack
Helgrind:Race
fun:free_stacks
fun:__deallocate_stack
}
{
helgrind---__deallocate_stack-start_thread-clone
Helgrind:Race
fun:__deallocate_stack
fun:start_thread
fun:clone
}
####################################################
# To do with pthread_{set,get}specific
#
{
helgrind---pthread_setspecific
Helgrind:Race
fun:pthread_setspecific
}
{
helgrind---pthread_getspecific
Helgrind:Race
fun:pthread_getspecific
}
####################################################
# To do with dynamic linking
#
# helgrind---ld.so-...-dlsym was merged into helgrind-glibc2X-001
{
helgrind---_dl_allocate_tls
Helgrind:Race
fun:mempcpy
fun:_dl_allocate_tls_init
...
fun:pthread_create@@GLIBC_2.2*
fun:pthread_create_WRK
fun:pthread_create@*
}
{
helgrind---_dl_allocate_tls2
Helgrind:Race
fun:memcpy
fun:__mempcpy_inline
fun:_dl_allocate_tls_init
...
fun:pthread_create@@GLIBC_2.2*
fun:pthread_create_WRK
fun:pthread_create@*
}
####################################################
# To do with GNU libgomp
#
{
helgrind---libgomp43-1
Helgrind:Race
fun:gomp_ordered_sync
}
{
helgrind---libgomp43-1
Helgrind:Race
fun:gomp_ordered_next
}
{
helgrind---libgomp43-1
Helgrind:Race
fun:gomp_ordered_last
}