[Bug 44816] New: Cygwin/ MSYS2 `script -e` exit status forwarding randomly returns zero for non zero child process

https://bugs.winehq.org/show_bug.cgi?id=44816

Bug ID: 44816
           Summary: Cygwin/MSYS2 `script -e` exit status forwarding
                    randomly returns zero for non zero child process
           Product: Wine
           Version: 3.4
          Hardware: x86-64
                OS: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: ntdll
          Assignee: [e-mail]
          Reporter: [e-mail]
      Distribution: ---

Hello folks,

to track
https://github.com/wine-staging/wine-staging/tree/master/patches/ntdll-Deall...

Mentioned here:

https://wine-staging.com/news/2015-08-23-release-1.7.50.html

--- quote ---
Do not allow to deallocate thread stack for current thread (MSYS2, Wine Staging
Bug #241)
--- quote ---

After some archaeology I found the Cygwin mailing list discussion here:

https://cygwin.com/ml/cygwin/2015-09/msg00114.html

--- quote ---
> Ah, ok.  What OS does Wine emulate here?  Have a look at
> https://sourceware.org/git/?p=newlib-cygwin.git;a=blob;f=winsup/cygwin/cygth...
>
> The terminate_thread_frees_stack flag is set to false for XP/2003 and to
> true for any newer OS.  I guess this is a double-free because Wine's
> TerminateThread already freed the stack and Cygwin got the info it's
> supposedly running under XP/2003, so it tries to workaround the fact
> that TerminateThread on these systems didn't free the stack by themselves.

Wine emulates Windows XP here, I double checked Wine source code and I
can confirm Wine doesn't free the stack:
NtTerminateThread() -> abort_thread() -> terminate_thread()
https://github.com/wine-compholio/wine-patched/blob/8b3a785e97a7e28ff58731b5...

Cygwin reports wincap as wincap_xpsp2 which is also correct here.
> we need to know what address ret=61005767 is refering to.  addr2line would help.

As you point out, ret=61005767 is the call to VirtualFree() inside
cygthread::terminate_thread ()
https://sourceware.org/git/?p=newlib-cygwin.git;a=blob;f=winsup/cygwin/cygth...

After combine the information we have so far, I think the problem is
clear now. It's not Cygwin's fault, it's Wine's fault. Wine's
WaitForSingleObject() needs to wait long enough after the
TerminateThread() call, so the target thread has enough time to finish
the system cleanup and return correct exit_code before Cygwin call
VirtualFree to free the stack. But in our current implementation
Wine's WaitForSingleObject returns too early, it shouldn't return
before the system thread cleanup done, as a result there is a race
here. We are thinking of a solution in Wine. Thanks again for the
great help!
--- quote ---

$ wine --version
wine-3.4

Regards

[Bug 44816] Cygwin/ MSYS2 `script -e` exit status forwarding randomly returns zero for non zero child process

https://bugs.winehq.org/show_bug.cgi?id=44816

Anastasius Focht  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
                URL|                            |https://cygwin.com/install.
                   |                            |html
             Status|NEW                         |STAGED
           Keywords|                            |download, patch, source
    Staged patchset|                            |https://github.com/wine-sta
                   |                            |ging/wine-staging/tree/mast
                   |                            |er/patches/ntdll-Dealloc_Th
                   |                            |read_Stack

[Bug 44816] Cygwin/ MSYS2 `script -e` exit status forwarding randomly returns zero for non zero child process

https://bugs.winehq.org/show_bug.cgi?id=44816

tokktokk  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[e-mail]

[Bug 44816] Cygwin/ MSYS2 `script -e` exit status forwarding randomly returns zero for non zero child process

https://bugs.winehq.org/show_bug.cgi?id=44816

Andrey  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[e-mail]

[Bug 44816] Cygwin/MSYS2 `script -e` exit status forwarding randomly returns zero for non zero child process

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=44816

Zebediah Figura  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[e-mail]

--- Comment #1 from Zebediah Figura  ---
I'm guessing the double-free results from the kernel freeing the thread's stack
once it's terminated (I'm not sure, and it's hard to be sure just from skimming
the kernel code, but it seems like the only sane thing to do?) since we
certainly don't free it ourselves.

[Bug 44816] Cygwin/MSYS2 `script -e` exit status forwarding randomly returns zero for non zero child process

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=44816

--- Comment #2 from Zebediah Figura  ---
(In reply to Zebediah Figura from comment #1)
> I'm guessing the double-free results from the kernel freeing the thread's
> stack once it's terminated (I'm not sure, and it's hard to be sure just from
> skimming the kernel code, but it seems like the only sane thing to do?)
> since we certainly don't free it ourselves.

Actually, on further testing, it seems the kernel doesn't free anything. So
what gives? How could there possibly be a double free here?

[Bug 44816] Cygwin/MSYS2 `script -e` exit status forwarding randomly returns zero for non zero child process

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=44816

Joel Holdsworth  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[e-mail]

[Bug 44816] Cygwin/MSYS2 `script -e` exit status forwarding randomly returns zero for non zero child process

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=44816

Bernhard Übelacker  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |[e-mail]

--- Comment #3 from Bernhard Übelacker  ---
This patch got removed from the wine-staging patch set here:

https://gitlab.winehq.org/wine/wine-staging/-/commit/e3cca687befffb4baee144d...

On top of current wine git plus a few patches needed for msys2 installation,
I tried to reproduce this issue with a current 64-bit msys2 and the latest
available 32-bit cygwin installation, but following command returned
always the same expected value to me.

while true; do script -e -q -c 'bash.exe -c "exit 123"' -a tmp.txt; echo $?
$(date); done

As staging patch got removed and the issue seems no longer visible,
should this bug be closed?

[Bug 44816] Cygwin/MSYS2 `script -e` exit status forwarding randomly returns zero for non zero child process

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=44816

Zeb Figura  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
      Fixed by SHA1|                            |25b093f3845a0ae1b2e2fe1c070
                   |                            |1e98064f8e8d6
             Status|STAGED                      |RESOLVED
         Resolution|---                         |FIXED

--- Comment #4 from Zeb Figura  ---
(In reply to Bernhard Übelacker from comment #3)
> This patch got removed from the wine-staging patch set here:
>
> https://gitlab.winehq.org/wine/wine-staging/-/commit/
> e3cca687befffb4baee144dcd55f01a3204176eb
>
> On top of current wine git plus a few patches needed for msys2 installation,
> I tried to reproduce this issue with a current 64-bit msys2 and the latest
> available 32-bit cygwin installation, but following command returned
> always the same expected value to me.
>
> while true; do script -e -q -c 'bash.exe -c "exit 123"' -a tmp.txt; echo $?
> $(date); done
>
> As staging patch got removed and the issue seems no longer visible,
> should this bug be closed?

I think it's fair to assume so, yes. Should be fixed by
https://source.winehq.org/git/wine.git/commitdiff/25b093f3845a0ae1b2e2fe1c0701e98064f8e8d6
or the i386 version.

[Bug 44816] Cygwin/MSYS2 `script -e` exit status forwarding randomly returns zero for non zero child process

[WineHQ Bugzilla]

https://bugs.winehq.org/show_bug.cgi?id=44816

Alexandre Julliard  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |CLOSED

--- Comment #5 from Alexandre Julliard  ---
Closing bugs fixed in 8.14.