Fixes connection in Mortal Kombat 11; reported by Thomas Crider.
Fixes: a891713f48
Signed-off-by: Zebediah Figura <z.figura12@gmail.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Cygwin fork() fails in NtCreateSymbolicLinkObject(). We successfully
create the link but then fail to alloc_handle() with STATUS_ACCESS_DENIED,
because the requested access rights exceed what the owner is allowed.
Allow it more.
Thank you to Dmitry Timoshkov for debugging the security details from
alloc_handle() onwards.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=48891
Signed-off-by: Damjan Jovanovic <damjan.jov@gmail.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
The RIM_TYPEHID messages will have to carry the variable length HID
report.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=50506
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Without any HID report data for now.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=50506
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Fixes a regression introduced by 21f5597de4
Signed-off-by: Paul Gofman <pgofman@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Steam uses WSASend() with completion ports, reusing OVERLAPPED structures as
soon as they are returned from GetQueuedCompletionStatus(). Since completion is
queued during the select request in wait_async(), the I/O status block can be
reused even before the call to NtDeviceIoControl exits.
This works fine with current Wine, because WSASend() doesn't access the I/O
status block after queuing completion. However, a patch that changes it to use
wait_async() like other async requests causes NtDeviceIoControlFile to
consistently return garbage status codes.
Signed-off-by: Zebediah Figura <z.figura12@gmail.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
I was able to reverse-engineer a native CONNECT ioctl, with code 0x801; but I
was not able to find a valid set of parameters which would allow us to
implement either connect() or ConnectEx(). In particular, I could not find a
way to make the ioctl respect nonblocking mode, and I could not find a way to
specify an initial buffer to be sent.
Signed-off-by: Zebediah Figura <z.figura12@gmail.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
In particular, so that we can wait for POLLOUT on a connecting socket multiple
times, in case the first connection fails, without polling on it while it is
unconnected.
Signed-off-by: Zebediah Figura <z.figura12@gmail.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
In particular, prevent FD_CONNECT from being reported when not selected for.
Signed-off-by: Zebediah Figura <z.figura12@gmail.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Instead of clearing all messages when any match.
Signed-off-by: Zebediah Figura <z.figura12@gmail.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
HID rawinput hardware messages are sent from winedevice.exe, which is
attached to the services desktop. We need to broadcast its messages to
all (interactive) desktops instead.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=50506
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
And use it to find registered HID rawinput devices.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=50506
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>