https://bugs.winehq.org/show_bug.cgi?id=41637
Bug ID: 41637 Summary: Altium Designer 16: Wineserver stuck / not responding since 1.9.2* Product: Wine Version: 1.9.22 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: wineserver Assignee: wine-bugs@winehq.org Reporter: afb@mailbox.org Distribution: ---
Altium Designer 16 stopped working since Wine 1.9.2* / Arch Linux 64.
After updating from Wine 1.9.19 to 1.9.22, neither Altium 16.1.12 in present prefixes, nor Altium installer in new prefixes starts anymore. Instead, /usr/bin/wineserver gets stuck at full cpu load and does not react to wineserver -k or SIGTERM.
What was also done: - purge and reinstall of Wine and all dependencies - switch of Wine-staging 1.9.21 - log in as different (new) user - downgrading to Wine 1.9.20 - setup of a chroot environment on a second hdd with nothing more than Xorg, drivers and Wine / dependencies ... without improvement of the situation
Link to the recent Altium installer: https://s3.amazonaws.com/altium-release-manager/Altium_Designer_16/AltiumDes...
In general, Wine seems to work, which was tested with Notepad++ from: https://notepad-plus-plus.org/download/v7.1.html
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #1 from alex freudenberg afb@mailbox.org --- Created attachment 56011 --> https://bugs.winehq.org/attachment.cgi?id=56011 Debug output / WINEDEBUG=+server
The last couple lines of debug outpug before wineserver gets stuck. Executed command was:
env WINEARCH=win32 WINEDEBUG=+server ./AltiumDesignerSetup_16_1_12.exe
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #2 from alex freudenberg afb@mailbox.org --- Building Wine 1.9.22 from source on Arch Linux 64 results in the same issue.
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #3 from alex freudenberg afb@mailbox.org --- The issue occurs from kernel 4.8.3-1-ARCH onwards. Last working kernel version is 4.8.2-1-ARCH.
https://bugs.winehq.org/show_bug.cgi?id=41637
winetest@luukku.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |winetest@luukku.com
--- Comment #4 from winetest@luukku.com --- (In reply to alex freudenberg from comment #3)
The issue occurs from kernel 4.8.3-1-ARCH onwards. Last working kernel version is 4.8.2-1-ARCH.
Wine bugs are rarely fixed updating or downgrading a kernel. The best thing you could do is regressions testing wine against the broken application.
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #5 from alex freudenberg afb@mailbox.org --- Thanks. With mentioned kernel 4.8.3-1-ARCH I already tested Wine 1.9.22 down to 1.9.19 without any improvement. Should I go backwards further?
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #6 from winetest@luukku.com --- (In reply to alex freudenberg from comment #5)
Thanks. With mentioned kernel 4.8.3-1-ARCH I already tested Wine 1.9.22 down to 1.9.19 without any improvement. Should I go backwards further?
https://wiki.winehq.org/Regression_Testing
This is how you supposed to find the commit(s) that broke the program. Knowing wine version it works isnt accurate enough. It's not that hard and you can also speed up the compile time.
https://bugs.winehq.org/show_bug.cgi?id=41637
Eugene Savelov savelov@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |savelov@gmail.com
https://bugs.winehq.org/show_bug.cgi?id=41637
m.amjadi@engineer.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |m.amjadi@engineer.com
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #7 from m.amjadi@engineer.com --- Same issue with Altium Designer 15 here
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #8 from alex freudenberg afb@mailbox.org --- In contrast to my first statement, I have to admit that this bug is definitely independent from the Wine version. In preparation of regression testing, I was not able to find a Wine release that runs Altium Designer on kernel 4.8.3-1-ARCH or later. After downgrading to kernel 4.8.2-1-ARCH, any Wine version from recent releases down to 1.6.* works!
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #9 from winetest@luukku.com --- Really weird then. I have seen only 2 kernel related bugs at this bug tracker. One was related to far cry fog and one was related to 16/32 bit programs not working anymore on certain versions of kernels since they disabled some support at kernel side.
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #10 from Eugene Savelov savelov@gmail.com --- Having examined linux patch-4.8.3, I found some changes to ptrace API, does wine use it? I found some earlier issues with wine regarding it - could this be related? Can't think of anything else that may affect wine in mostly hardware-related stuff contained in linux kernel patch.
diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index f79576a541ff..a1606eadd9ce 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -173,8 +173,8 @@ unsigned long kernel_stack_pointer(struct pt_regs *regs) return sp;
prev_esp = (u32 *)(context); - if (prev_esp) - return (unsigned long)prev_esp; + if (*prev_esp) + return (unsigned long)*prev_esp;
return (unsigned long)regs; }
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #11 from Eugene Savelov savelov@gmail.com --- according to https://outflux.net/blog/archives/2012/02/06/use-of-ptrace/ wine is a valid user of ptrace (in addition to gdb, strace, ltrace)
https://bugs.winehq.org/show_bug.cgi?id=41637
alex freudenberg afb@mailbox.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Version|1.9.22 |unspecified Summary|Altium Designer 16: |Delphi: Wineserver stuck / |Wineserver stuck / not |not responding since |responding since 1.9.2* |linux-4.8.3
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #12 from Matteo Bruni matteo.mystral@gmail.com --- (In reply to alex freudenberg from comment #3)
The issue occurs from kernel 4.8.3-1-ARCH onwards. Last working kernel version is 4.8.2-1-ARCH.
It would be nice if someone wants to give a shot at bisecting the kernel between 4.8.2 and 4.8.3 to find the commit which broke it. You need to be able and willing to build your own kernel but it's not too complicated, if a bit distro-dependent. There are a number of guides on the web which should help with that.
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #13 from alex freudenberg afb@mailbox.org --- 89eeba1594ac641a30b91942961e80fae978f839 is the commit that made the difference:
http://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable.git/diff/?h=...
https://bugs.winehq.org/show_bug.cgi?id=41637
Jeff D. Hanson jhansonxi@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |jhansonxi@gmail.com
--- Comment #14 from Jeff D. Hanson jhansonxi@gmail.com --- Encountering this also. Wine Staging 1.9.23 Xubuntu 14.04 x86_64 AltiumDesignerSetup_16_1_12.exe
Can't even SIGKILL wineserver when it occurs.
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #15 from Jeff D. Hanson jhansonxi@gmail.com --- FYI, I tested Saturn PCB Toolkit on Wine Staging 1.9.23 and it didn't have any problems. It's known to use Delphi (bug #39792). https://appdb.winehq.org/objectManager.php?sClass=application&iId=13578
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #16 from Jeff D. Hanson jhansonxi@gmail.com --- Created attachment 56335 --> https://bugs.winehq.org/attachment.cgi?id=56335 Altium 16.1.2 installer
Installer only. Requires Altium account login before it will install anything.
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #17 from Jeff D. Hanson jhansonxi@gmail.com --- My Xubuntu 14.04 kernel on the system affected by this bug:
Linux … 3.13.0-105-generic #152-Ubuntu SMP Fri Dec 2 15:37:11 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux
If it's a 4.8.3 patch that's causing it then it may have been backported to 14.04 by Ubuntu. I'll try some earlier kernels to verify.
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #18 from Jeff D. Hanson jhansonxi@gmail.com --- On Xubuntu 14.04 x86_64, kernel 3.13.0-98-generic #145-Ubuntu (linux-image-3.13.0-98-generic) is the last working version.
https://bugs.winehq.org/show_bug.cgi?id=41637
Robert Walker bob.mt.wya@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |bob.mt.wya@gmail.com
https://bugs.winehq.org/show_bug.cgi?id=41637
Erich E. Hoover erich.e.hoover@wine-staging.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |erich.e.hoover@wine-staging | |.com
--- Comment #19 from Erich E. Hoover erich.e.hoover@wine-staging.com --- (In reply to alex freudenberg from comment #13)
89eeba1594ac641a30b91942961e80fae978f839 is the commit that made the difference:
http://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable.git/diff/ ?h=v4.8. 3&id=89eeba1594ac641a30b91942961e80fae978f839&id2=0312017cbbb5f46dbf4932f65df 0c2c66f9c8f27
Has anyone posted on the kernel mailing list to ask for help? This tacitly looks like the kernel may be requiring additional permissions for the requested action.
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #20 from Matteo Bruni matteo.mystral@gmail.com --- (In reply to alex freudenberg from comment #13)
89eeba1594ac641a30b91942961e80fae978f839 is the commit that made the difference:
Cool, thank you for the regression test. Did you doublecheck that everything works fine with 4.8.3 and the patch reverted (assuming that the patch reverts cleanly?)
(In reply to Jeff D. Hanson from comment #16)
Created attachment 56335 [details] Altium 16.1.2 installer
I don't think we want copyrighted binaries here in bugzilla.
(In reply to Erich E. Hoover from comment #19)
Has anyone posted on the kernel mailing list to ask for help? This tacitly looks like the kernel may be requiring additional permissions for the requested action.
Not yet, it seems. Do you want to take charge of it?
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #21 from alex freudenberg afb@mailbox.org --- (In reply to Matteo Bruni from comment #20)
(In reply to alex freudenberg from comment #13)
89eeba1594ac641a30b91942961e80fae978f839 is the commit that made the difference:
Cool, thank you for the regression test. Did you doublecheck that everything works fine with 4.8.3 and the patch reverted (assuming that the patch reverts cleanly?)
Yes, I took kernel 4.8.3 and reverted the changes from the commit in question and - at least Altium Designer - works again.
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #22 from Erich E. Hoover erich.e.hoover@wine-staging.com --- (In reply to alex freudenberg from comment #21)
... Yes, I took kernel 4.8.3 and reverted the changes from the commit in question and - at least Altium Designer - works again.
Since you already have some familiarity with the issue, would you mind starting an LKML thread about this? I am happy to be CC'd on it to do what I can to help solve the problem.
https://bugs.winehq.org/show_bug.cgi?id=41637
Joachim Priesner joachim.priesner.bugs@web.de changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |joachim.priesner.bugs@web.d | |e
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #23 from alex freudenberg afb@mailbox.org --- (In reply to Erich E. Hoover from comment #22)
(In reply to alex freudenberg from comment #21)
... Yes, I took kernel 4.8.3 and reverted the changes from the commit in question and - at least Altium Designer - works again.
Since you already have some familiarity with the issue, would you mind starting an LKML thread about this? I am happy to be CC'd on it to do what I can to help solve the problem.
Still no response from LKML.
It is possible, though, to revert the two changes made by patch-4.8.2-3 in static struct page *follow_page_pte of mm/gup.c in recent kernel versions to make Wine work again. Currently, I'm running kernel 4.8.13 with that applied. I will prepare a patch for those who want to build a custom kernel.
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #24 from alex freudenberg afb@mailbox.org --- Patches for recent kernels are available here: https://github.com/baldrianbandit/linux-kernel-wineserver-fix
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #25 from Joachim Priesner joachim.priesner.bugs@web.de --- Note that the kernel commit in question is the fix for the "Dirty COW" vulnerability (https://en.wikipedia.org/wiki/Dirty_COW). Reverting it will probably make the system vulnerable again.
The LKML thread in question is https://lkml.org/lkml/2016/12/20/150
I am currently trying to investigate the issue as well, but don't expect quick results :)
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #26 from Joachim Priesner joachim.priesner.bugs@web.de --- I identified the root cause of the issue, turns out someone submitted a fix to LKML a few days ago: https://lkml.org/lkml/2017/1/5/13
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #27 from Jeff D. Hanson jhansonxi@gmail.com --- Ubuntu bug: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1658270
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #28 from winetest@luukku.com --- (In reply to Jeff D. Hanson from comment #27)
Ubuntu bug: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1658270
Fix was committed few distros already.
https://bugs.winehq.org/show_bug.cgi?id=41637
Matteo Bruni matteo.mystral@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Resolution|--- |NOTOURBUG
--- Comment #29 from Matteo Bruni matteo.mystral@gmail.com --- Let's RESOLVE this as NOTOURBUG for the time being. As far as I understand the referenced kernel patch is in 4.10-rc6 and was backported to 4.9.7. If someone can confirm that the affected applications work fine with one of those kernels we can also close this bug.
Thank you to all the people involved with discovering and fixing this bug!
https://bugs.winehq.org/show_bug.cgi?id=41637
--- Comment #30 from Joachim Priesner joachim.priesner.bugs@web.de --- I can confirm that Altium 16.1.2. installer works fine on kernel 4.9.7 while on the same system running kernel 4.9.6 it exhibits the symptoms described in this bug report.
https://bugs.winehq.org/show_bug.cgi?id=41637
alex freudenberg afb@mailbox.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Resolution|NOTOURBUG |FIXED
--- Comment #31 from alex freudenberg afb@mailbox.org --- And I can confirm that Altium Designer 17.0.6 runs fine with kernel 4.9.7-1-ARCH. Wineserver reacts as expected, too!
https://bugs.winehq.org/show_bug.cgi?id=41637
fjfrackiewicz@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |fjfrackiewicz@gmail.com
--- Comment #32 from fjfrackiewicz@gmail.com --- (In reply to alex freudenberg from comment #31)
And I can confirm that Altium Designer 17.0.6 runs fine with kernel 4.9.7-1-ARCH. Wineserver reacts as expected, too!
But this is still a NOTOURBUG resolution rather than FIXED since the bug was upstream (the Linux kernel). Could I ask you to change it back to "RESOLVED NOTOURBUG", please?
https://bugs.winehq.org/show_bug.cgi?id=41637
Bruno Jesus 00cpxxx@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- See Also| |https://launchpad.net/bugs/ | |1658270 Resolution|FIXED |NOTOURBUG
https://bugs.winehq.org/show_bug.cgi?id=41637
Austin English austinenglish@gmail.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED
--- Comment #33 from Austin English austinenglish@gmail.com --- (In reply to alex freudenberg from comment #31)
And I can confirm that Altium Designer 17.0.6 runs fine with kernel 4.9.7-1-ARCH. Wineserver reacts as expected, too!
Fixed upstream