|
From: Alexander P. <gl...@go...> - 2010-08-10 08:13:49
|
Julian, Nicholas,
aren't these suppressions missing in the trunk after the merge?
On Mon, Jun 7, 2010 at 3:32 PM, <sv...@va...> wrote:
> Author: sewardj
> Date: 2010-06-07 12:32:31 +0100 (Mon, 07 Jun 2010)
> New Revision: 11154
>
> Log:
> Initial suppression files for MacOSX 10.6.
>
>
> Added:
> branches/MACOSX106/darwin10-drd.supp
> branches/MACOSX106/darwin10.supp
>
>
> Added: branches/MACOSX106/darwin10-drd.supp
> ===================================================================
>
> Added: branches/MACOSX106/darwin10.supp
> ===================================================================
> --- branches/MACOSX106/darwin10.supp (rev 0)
> +++ branches/MACOSX106/darwin10.supp 2010-06-07 11:32:31 UTC (rev 11154)
> @@ -0,0 +1,273 @@
> +
> +# Suppressions for Darwin 10.x / Mac OS X 10.6 Snow Leopard
> +
> +##----------------------------------------------------------------------##
> +# Memcheck
> +##----------------------------------------------------------------------##
> +
> +{
> + mach_msg_trap-1
> + Memcheck:Param
> + mach_msg(msg.msgh_remote_port)
> + fun:mach_msg_trap
> + obj:/System/Library/Frameworks/CoreFoundation*
> + obj:/System/Library/Frameworks/ApplicationServices*
> +}
> +
> +{
> + mach_msg_trap-2
> + Memcheck:Param
> + mach_msg(msg.msgh_remote_port)
> + fun:mach_msg_trap
> + obj:/System/Library/Frameworks/CoreFoundation*
> + obj:/System/Library/Frameworks/CoreServices*
> +}
> +
> +{
> + mach_msg_trap-3
> + Memcheck:Param
> + mach_msg(msg.msgh_remote_port)
> + fun:mach_msg_trap
> + obj:/System/Library/Frameworks/CoreFoundation*
> + obj:/System/Library/Frameworks/Carbon*
> +}
> +
> +{
> + mach_msg_trap-4
> + Memcheck:Param
> + mach_msg(msg.msgh_remote_port)
> + fun:mach_msg_trap
> + obj:/System/Library/Frameworks/CoreFoundation*
> + obj:/System/Library/Frameworks/CoreFoundation*
> +}
> +
> +{
> + mach_msg_trap-5
> + Memcheck:Param
> + mach_msg(msg.msgh_remote_port)
> + fun:mach_msg_trap
> + obj:/System/Library/Frameworks/CoreFoundation*
> + obj:/System/Library/Frameworks/AppKit*
> +}
> +
> +{
> + macos-Cond-1
> + Memcheck:Cond
> + fun:GetVariationInfoFromName
> + obj:/System/Library/Frameworks/ApplicationServices*
> + obj:/System/Library/Frameworks/ApplicationServices*
> +}
> +
> +{
> + macos-Cond-2
> + Memcheck:Cond
> + fun:*PMMutex*Lock*
> + obj:/System/Library/Frameworks/ApplicationServices*
> + obj:/System/Library/Frameworks/ApplicationServices*
> +}
> +
> +{
> + macos-Cond-3
> + Memcheck:Cond
> + fun:sseCGSBlendXXXX8888
> + obj:/System/Library/Frameworks/ApplicationServices*
> + obj:/System/Library/Frameworks/ApplicationServices*
> +}
> +
> +{
> + macos-Cond-4
> + Memcheck:Cond
> + fun:*CASettingsStorage*RefreshSettings*
> + obj:/System/Library/Frameworks/CoreAudio*
> + obj:/System/Library/Frameworks/CoreAudio*
> +}
> +
> +{
> + macos-Cond-5
> + Memcheck:Cond
> + fun:gle*
> + obj:/System/Library/Frameworks/OpenGL*
> + obj:/System/Library/Frameworks/OpenGL*
> +}
> +
> +{
> + macos-Cond-6
> + Memcheck:Cond
> + fun:pthread_rwlock_init$UNIX2003
> + fun:main
> +}
> +
> +# afaict this is legit. Might be caused by setenv("VAR=")
> +# where the value string is empty (not sure)
> +{
> + macos-Cond-7
> + Memcheck:Cond
> + fun:__setenv
> + fun:putenv*
> +}
> +
> +{
> + macos-futimes-1
> + Memcheck:Param
> + futimes(tvp[1])
> + fun:futimes
> + obj:/usr/lib/libSystem*
> + obj:/usr/lib/libSystem*
> +}
> +
> +{
> + macos-vsyslog-hole
> + Memcheck:Param
> + socketcall.sendto(msg)
> + fun:sendto$NOCANCEL$UNIX2003
> + fun:vsyslog
> +}
> +
> +# Still-reachable memory.
> +
> +# I chopped this one off at libSystem_initializer, there were more frames.
> +{
> + darwin-still-reachable-1
> + Memcheck:Leak
> + fun:calloc
> + fun:dwarf2_unwind_dyld_add_image_hook
> + fun:_ZN4dyld19registerAddCallbackEPFvPK11mach_headerlE
> + fun:_dyld_register_func_for_add_image
> + fun:__keymgr_initializer
> + fun:libSystem_initializer
> +}
> +
> +# I chopped this one off at libSystem_initializer, there were more frames.
> +{
> + darwin-still-reachable-2
> + Memcheck:Leak
> + fun:malloc
> + fun:get_or_create_key_element
> + fun:_keymgr_get_and_lock_processwide_ptr_2
> + fun:dwarf2_unwind_dyld_add_image_hook
> + fun:_ZN4dyld19registerAddCallbackEPFvPK11mach_headerlE
> + fun:_dyld_register_func_for_add_image
> + fun:__keymgr_initializer
> + fun:libSystem_initializer
> +}
> +
> +{
> + darwin-still-reachable-3
> + Memcheck:Leak
> + fun:malloc
> + fun:__smakebuf
> + fun:__swsetup
> + fun:__sfvwrite
> + fun:puts
> +}
> +
> +# Genuine leaks.
> +# See https://bugs.kde.org/show_bug.cgi?id=188572 about this; it's
> +# unavoidable due to BSD setenv() semantics.
> +{
> + macos-__setenv-leak-see-our-bug-188572
> + Memcheck:Leak
> + fun:malloc_zone_malloc
> + fun:__setenv
> + fun:setenv$UNIX2003
> +}
> +{
> + macos-localeconv-leak
> + Memcheck:Leak
> + fun:malloc
> + fun:localeconv_l
> + fun:__vfprintf
> + fun:vsnprintf
> +}
> +
> +##----------------------------------------------------------------------##
> +# Helgrind
> +##----------------------------------------------------------------------##
> +
> +# These ones were necessary to give no errors on a tiny non-threaded
> +# program. I don't know if they're real problems or false positives (njn).
> +
> +# keymgr seems to deliberately do some bogus actions, and if they are bogus,
> +# it passes the error codes back to the caller.
> +{
> + __keymgr_initializer lock failed
> + Helgrind:PthAPIerror
> + fun:pthread_mutex_lock
> + fun:_dyld_register_func_for_*_image
> + fun:__keymgr_initializer
> + fun:libSystem_initializer
> +}
> +{
> + __keymgr_initializer unlock failed
> + Helgrind:PthAPIerror
> + fun:pthread_mutex_unlock
> + fun:_dyld_register_func_for_*_image
> + fun:__keymgr_initializer
> + fun:libSystem_initializer
> +}
> +{
> + __keymgr_initializer bogus unlock
> + Helgrind:UnlockBogus
> + fun:pthread_mutex_unlock
> + fun:_dyld_register_func_for_*_image
> + fun:__keymgr_initializer
> + fun:libSystem_initializer
> +}
> +
> +# These ones were necessary to give no errors on a tiny threaded program.
> +# I don't know if they're real problems or false positives (njn).
> +
> +#{
> +# helgrind-darwinlibc-nuke-everything-in-dyld
> +# Helgrind:Race
> +# obj:/usr/lib/dyld
> +#}
> +
> +{
> + helgrind-darwinlibc-nuke-everything-in-libSystem.B.dylib
> + Helgrind:Race
> + obj:/usr/lib/libSystem.B.dylib
> +}
> +
> +# This would be better as "fun:\?\?\?" but string matching doesn't seem to
> +# allow escaping meta-chars.
> +#
> +# This is very bad .. not only will it hide races in any
> +# un-identified piece of code, the ??? also matches any 3-char
> +# function name.
> +{
> + helgrind-darwinlibc-nuke-everything-in-???-(unknown-code)
> + Helgrind:Race
> + fun:???
> +}
> +
> +{
> + helgrind-darwinlibc--mythread_wrapper-*thread*start*
> + Helgrind:Race
> + fun:mythread_wrapper
> + fun:*thread*start*
> +}
> +
> +{
> + helgrind-darwinlibc--pthread_create_WRK-pthread_create
> + Helgrind:Race
> + fun:pthread_create_WRK
> + fun:pthread_create
> +}
> +
> +
> +# Thread #9: Bug in libpthread: recursive write lock granted on
> +# mutex/wrlock which does not support recursion
> +# at 0x18696: pthread_cond_wait* (hg_intercepts.c:655)
> +# by 0x2300B8: pthread_rwlock_wrlock$UNIX2003 (in /usr/lib/libSystem.B.dylib)
> +# by 0x18F41: pthread_rwlock_wrlock* (hg_intercepts.c:1177)
> +#
> +# no idea what this is about
> +#
> +{
> + helgrind-darwin9--pthread-rwlock-kludgery
> + Helgrind:Misc
> + fun:pthread_cond_wait*
> + fun:pthread_rwlock_*lock*
> + fun:pthread_rwlock_*lock*
> +}
>
>
> ------------------------------------------------------------------------------
> ThinkGeek and WIRED's GeekDad team up for the Ultimate
> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
> lucky parental unit. See the prize list and enter to win:
> http://p.sf.net/sfu/thinkgeek-promo
> _______________________________________________
> Valgrind-developers mailing list
> Val...@li...
> https://lists.sourceforge.net/lists/listinfo/valgrind-developers
>
--
Alexander Potapenko
Software Engineer
Google Moscow
|