[Bug 9988] New: IOCTL_SERIAL_WAIT_ON_MASK does not always return correct answer

http://bugs.winehq.org/show_bug.cgi?id=9988

Summary: IOCTL_SERIAL_WAIT_ON_MASK does not always return correct
                    answer
           Product: Wine
           Version: 0.9.46.
          Platform: Other
        OS/Version: Linux
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: wine-misc
        AssignedTo: [e-mail]
        ReportedBy: [e-mail]

The io_control() function with IOCTL_SERIAL_WAIT_ON_MASK in dlls/ntdll/serial.c
does not seem to always return the equivalent status as windows.

Unfortunately I only have an proprietary windows application (no source) that
shows this. It is an application called FanControl shipped with a water cooling
control unit from Innovatek (germany). The application controls the unit via an
USB FTDI serial port. The application only works if there is such a device
available. Rather difficult to reproduce on the average system :(.

The fix is to remove a few lines of code from the wait_on() function:

*events = check_events(fd, commio->evtmask,
                               &commio->irq_info, &commio->irq_info,
                               commio->mstat, commio->mstat);
    if (*events) goto out_now;

where the 'if' is the essential bit. The status returned to the caller will be
zero when the check_events() returns something. This seems to confuse the
windows application. When the above lines are taken out the returned code is
always "STATUS_PENDING" and the handling through wait_for_event() will always
return the right things to my windows application.

The code mentioned above looks like a shortcut to avoid starting a workitem
with RtlQueueWorkItem(). My feeling is that this shortcut is not always valid.

I will attach a patch.

[Bug 9988] IOCTL_SERIAL_WAIT_ON_MASK does not always return correct answer

http://bugs.winehq.org/show_bug.cgi?id=9988

--- Comment #1 from Tom Brus   2007-10-11 07:47:08 ---
Created an attachment (id=8510)
 --> (http://bugs.winehq.org/attachment.cgi?id=8510)
patch for serial.c

[Bug 9988] IOCTL_SERIAL_WAIT_ON_MASK does not always return correct answer

http://bugs.winehq.org/show_bug.cgi?id=9988

--- Comment #2 from Tom Brus   2007-10-12 09:36:10 ---
I have reasons to believe that the above fix is not a proper fix. I am
investigating further, please do not act yet. Remarks are welcome though ;)

-Tom

[Bug 9988] IOCTL_SERIAL_WAIT_ON_MASK does not always return correct answer

http://bugs.winehq.org/show_bug.cgi?id=9988

--- Comment #3 from Tom Brus   2007-10-21 14:00:10 ---
I got the sources of the windows application and have determined what is going
wrong. Unfortunately I can not share the sources themselves for copyright
reasons.

I pinpointed the problem to the WaitCommEvent() function (which does the
IOCTL_SERIAL_WAIT_ON_MASK). I deduced that under windows this function will
probably never return true. It always returns false with the GetLastError()
being ERROR_IO_PENDING. This is at least seems to be true for the situation
where the rcv and tx buffers are both empty and SetCommMask() is set to
(EV_RXCHAR | EV_TXEMPTY).

In contrast the wine implementation returns true (because of the 'shortcut'
discussed previously in this bug).

The following abbreviated code is based on the windows application code. I
stripped some unimportant details:

for (;;) {
        ok = WaitCommEvent(...);
        if (ok) {
            if (chars_available_for_read==0)
                continue;
        } else {
            switch (GetLastError())  {
                case ERROR_IO_PENDING: break;
                default              : report_error();break;
            }
        }
        switch (WaitForMultipleObjects(...)) {
            ...
        }
    }

This code runs fine on windows and hangs under wine. I traced it under wine and
the following happens: WaitCommEvent() returns true because the tx buffer is
empty; the chars_available_for_read is 0 so the for loop will start over. This
will happen in an endless succession: a classic hang.

You can argue in several ways that this code can be improved but that is beside
the point: it works under windows. You can also argue that wine's
WaitCommEvent() works according to the manual, but that is again besides the
point because wine should emulate windows, not its manual, right? ;)

On second evaluation the above fix and patch are correct. Could whoever is
responsible please look at them and either discuss options or apply the patch.
Thanks.

[Bug 9988] IOCTL_SERIAL_WAIT_ON_MASK does not always return correct answer

http://bugs.winehq.org/show_bug.cgi?id=9988

Juan Lang  changed:

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

--- Comment #4 from Juan Lang   2007-12-06 13:36:29 ---
You should try sending this to wine-patches for inclusion.  If it isn't
accepted, writing a test case that shows the behavior you're seeing (1) would
go a long way toward getting it included.

(1) A test that shows WaitCommEvent returns FALSE (ERROR_IO_PENDING) when "the
rx and tx buffers are empty and SetCommMask() is set to (EV_RXCHAR |
EV_TXEMPTY)".

[Bug 9988] IOCTL_SERIAL_WAIT_ON_MASK does not always return correct answer

http://bugs.winehq.org/show_bug.cgi?id=9988

Damjan Jovanovic  changed:

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

--- Comment #5 from Damjan Jovanovic   2007-12-07 05:02:11 ---
Is this a dup of bug #9356?

[Bug 9988] IOCTL_SERIAL_WAIT_ON_MASK does not always return correct answer

http://bugs.winehq.org/show_bug.cgi?id=9988

--- Comment #6 from Juan Lang   2007-12-07 05:15:08 ---
I don't think so.  The ioctls are different, and the patch is different.

[Bug 9988] IOCTL_SERIAL_WAIT_ON_MASK does not always return correct answer

http://bugs.winehq.org/show_bug.cgi?id=9988

Vitaliy Margolen  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|-unknown                    |ntdll

[Bug 9988] IOCTL_SERIAL_WAIT_ON_MASK does not always return correct answer

http://bugs.winehq.org/show_bug.cgi?id=9988

Dan Kegel  changed:

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

[Bug 9988] IOCTL_SERIAL_WAIT_ON_MASK does not always return correct answer

http://bugs.winehq.org/show_bug.cgi?id=9988

Michael Karpukhin  changed:

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

[Bug 9988] IOCTL_SERIAL_WAIT_ON_MASK does not always return correct answer

http://bugs.winehq.org/show_bug.cgi?id=9988

--- Comment #7 from Austin English   2009-04-17 12:06:56 ---
Is this still an issue in current (1.1.19 or newer) wine?

[Bug 9988] IOCTL_SERIAL_WAIT_ON_MASK does not always return correct answer

http://bugs.winehq.org/show_bug.cgi?id=9988

Austin English  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |hardware

--- Comment #8 from Austin English   2009-10-29 13:50:09 ---
Is this still an issue in current (1.1.32 or newer) wine?

[Bug 9988] IOCTL_SERIAL_WAIT_ON_MASK does not always return correct answer

http://bugs.winehq.org/show_bug.cgi?id=9988

Tom Brus  changed:

What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |INVALID

--- Comment #9 from Tom Brus   2009-10-30 03:23:52 ---
I am currently not in the situation to try and reproduce this. I have set this
one to RESOLVED-INVALID to keep your bugsbase clean. If I stumble upon it in
the future, I will reopen or make a new bug.

[Bug 9988] IOCTL_SERIAL_WAIT_ON_MASK does not always return correct answer

http://bugs.winehq.org/show_bug.cgi?id=9988

Dmitry Timoshkov  changed:

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

--- Comment #10 from Dmitry Timoshkov   2009-10-30 03:25:22 ---
Closing invalid.