Mostly minor expected/filter changes.
Helgrind did detect a race condition in a Valgrind file static
variable. Teh DRD dlopen test needs to be linked with the thread library.
In a VirtualBox environment I'm getting 4 hanging testcases
(3 in gdbserver_tests and none/freebsd/bug452274).
condvar_SOURCES = condvar.cpp
condvar_CXXFLAGS = $(AM_CXXFLAGS) -std=c++0x
concurrent_close_SOURCES = concurrent_close.cpp
-if !VGCONF_OS_IS_FREEBSD
+if VGCONF_OS_IS_FREEBSD
+dlopen_main_LDADD = ${LDADD}
+else
dlopen_main_LDADD = -ldl
endif
dlopen_lib_so_SOURCES = dlopen_lib.c
Helgrind:Race
fun:thr_exit
}
-
+# sigh, this is our own variable in coregrind/m_replacemalloc/vg_replace_malloc.c
+{
+ FREEBSD15-AMD64-MALLOC
+ Helgrind:Race
+ fun:malloc
+}
filter_stderr_solaris \
filter_helgrind \
filter_xml \
+ filter_fiw \
filter_freebsd.awk \
filter_stderr_freebsd \
filter_bug392331
--- /dev/null
+#! /bin/sh
+
+# free_is_write may have different arena sizes
+
+./filter_stderr "$@" |
+
+sed 's/size [^ ]* in arena/size ... in arena/'
+
This conflicts with a previous read of size 1 by thread #x
Locks held: none
at 0x........: main (free_is_write.c:36)
- Address 0x........ is 5 bytes inside an unallocated block of size 16 in arena "client"
+ Address 0x........ is 5 bytes inside an unallocated block of size ... in arena "client"
Done.
prog: free_is_write
vgopts: --free-is-write=yes
+stderr_filter: filter_fiw
filter_varinfo3 \
filter_memcheck \
filter_malloc_free \
+ filter_sendmsg \
filter_size_t \
filter_stanza \
filter_stanza.awk \
sem.stderr.exp sem.vgtest \
sendmsg.stderr.exp sendmsg.stderr.exp-solaris sendmsg.vgtest \
sendmsg.stderr.exp-freebsd \
- sendmsg.stderr.exp-freebsd-x86 \
sh-mem.stderr.exp sh-mem.vgtest \
sh-mem-random.stderr.exp sh-mem-random.stdout.exp64 \
sh-mem-random.stdout.exp sh-mem-random.vgtest \
--- /dev/null
+#! /bin/sh
+
+./filter_stderr "$@" |
+sed -e "s/frame #./frame #.../" \
+ -e "/by 0x........: sendmsg (in \/...libc...)/d"
at 0x........: sendmsg (in /...libc...)
by 0x........: main (sendmsg.c:46)
Address 0x........ is on thread 1's stack
- in frame #1, created by main (sendmsg.c:13)
+ in frame #..., created by main (sendmsg.c:13)
sendmsg: 6
...
by 0x........: main (sendmsg.c:29)
Address 0x........ is on thread 1's stack
- in frame #1, created by main (sendmsg.c:13)
+ in frame #..., created by main (sendmsg.c:13)
Syscall param sendmsg(sendmsg) points to uninitialised byte(s)
at 0x........: _sendmsg (in /...libc...)
- by 0x........: sendmsg (in /...libc...)
by 0x........: main (sendmsg.c:46)
Address 0x........ is on thread 1's stack
- in frame #2, created by main (sendmsg.c:13)
+ in frame #..., created by main (sendmsg.c:13)
Syscall param socketcall.connect(serv_addr.sa_len) points to uninitialised byte(s)
...
by 0x........: main (sendmsg.c:59)
Address 0x........ is on thread 1's stack
- in frame #1, created by main (sendmsg.c:13)
+ in frame #..., created by main (sendmsg.c:13)
sendmsg: 6
+++ /dev/null
-Syscall param socketcall.connect(serv_addr.sa_len) points to uninitialised byte(s)
- ...
- by 0x........: main (sendmsg.c:29)
- Address 0x........ is on thread 1's stack
- in frame #2, created by main (sendmsg.c:13)
-
-Syscall param sendmsg(sendmsg) points to uninitialised byte(s)
- at 0x........: _sendmsg (in /...libc...)
- by 0x........: sendmsg (in /...libc...)
- by 0x........: main (sendmsg.c:46)
- Address 0x........ is on thread 1's stack
- in frame #2, created by main (sendmsg.c:13)
-
-Syscall param socketcall.connect(serv_addr.sa_len) points to uninitialised byte(s)
- ...
- by 0x........: main (sendmsg.c:59)
- Address 0x........ is on thread 1's stack
- in frame #2, created by main (sendmsg.c:13)
-
-sendmsg: 6
by 0x........: __xnet_sendmsg (in /...libc...)
by 0x........: main (sendmsg.c:46)
Address 0x........ is on thread 1's stack
- in frame #2, created by main (sendmsg.c:13)
+ in frame #..., created by main (sendmsg.c:13)
sendmsg: 6
prog: sendmsg
vgopts: -q
+stderr_filter: filter_sendmsg
sup5
CoreError:FdNotClosed
...
- fun:recvmsg
+ fun:*recvmsg
fun:client
fun:main
}