http://bugs.winehq.org/show_bug.cgi?id=26263
Summary: __wine_call_from_regs violates x86 ABI Product: Wine Version: 1.3.14 Platform: x86-64 OS/Version: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: -unknown AssignedTo: wine-bugs@winehq.org ReportedBy: bart.vanassche@gmail.com
Created an attachment (id=33470) --> (http://bugs.winehq.org/attachment.cgi?id=33470) Valgrind output
As far as I know x86 applications should not modify data below the stack pointer. Running the latest (2011-02-27) git version of Wine under Valgrind results in many "address is just below the stack pointer" complaints. See also the attachment for the output of the following command:
valgrind --suppressions=wine-memcheck.supp --trace-children=yes --trace-children-skip=/usr/bin/wineserver --trace-children-skip-by-arg=wineboot.exe,winemenubuilder.exe,services.exe,winedevice.exe,plugplay.exe,explorer.exe --vex-iropt-precise-memory-exns=yes --num-callers=32 --fullpath-after=$HOME/software --error-limit=no --gen-suppressions=all wine notepad
http://bugs.winehq.org/show_bug.cgi?id=26263
André H. nerv@dawncrow.de changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |nerv@dawncrow.de Component|-unknown |ntdll
--- Comment #1 from André H. nerv@dawncrow.de 2011-02-27 08:18:26 CST --- code is produced in winebuild, but used in ntdll, so setting ntdll here.
it's used in macro: http://source.winehq.org/git/wine.git?a=blob;f=dlls/ntdll/ntdll_misc.h#l243
http://bugs.winehq.org/show_bug.cgi?id=26263
--- Comment #2 from Alexandre Julliard julliard@winehq.org 2011-02-27 09:55:27 CST --- Just suppress the warning. It's not worth making the function more complex and slower just for that reason.
http://bugs.winehq.org/show_bug.cgi?id=26263
--- Comment #3 from Bart Van Assche bart.vanassche@gmail.com 2011-02-27 11:04:20 CST --- (In reply to comment #2)
Just suppress the warning. It's not worth making the function more complex and slower just for that reason.
Are there any x86 platforms supported by Wine that do not define SA_ONSTACK ? In that case a signal delivered by the kernel to the Wine process might change the data below the stack pointer. That could result in very strange effects.
http://bugs.winehq.org/show_bug.cgi?id=26263
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |austinenglish@gmail.com
http://bugs.winehq.org/show_bug.cgi?id=26263
--- Comment #4 from Alexandre Julliard julliard@winehq.org 2011-02-27 13:28:14 CST --- If signals are not delivered on the signal stack you'll have much more serious problems.
http://bugs.winehq.org/show_bug.cgi?id=26263
Dmitry Timoshkov dmitry@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Resolution| |WONTFIX
--- Comment #5 from Dmitry Timoshkov dmitry@codeweavers.com 2011-02-27 23:59:00 CST --- Marking as a WONTFIX.
http://bugs.winehq.org/show_bug.cgi?id=26263
Dmitry Timoshkov dmitry@codeweavers.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #6 from Dmitry Timoshkov dmitry@codeweavers.com 2011-02-27 23:59:12 CST --- Closing.
https://bugs.winehq.org/show_bug.cgi?id=26263
Zebediah Figura z.figura12@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |z.figura12@gmail.com Resolution|WONTFIX |FIXED Fixed by SHA1| |e3a236e6ce368e3d960cfbe3018 | |d3a5cb52a0add
--- Comment #7 from Zebediah Figura z.figura12@gmail.com --- Marking fixed as this function was removed in 2.14.
-
wine-bugs＠winehq.org