This makes event results available sooner, i.e. after they are executed
by the GPU, and also moves the polling from CS to application's thread.
Signed-off-by: Jan Sikorski <jsikorski@codeweavers.com>
Signed-off-by: Henri Verbeet <hverbeet@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
We want to avoid using the poll list, for occlusion queries especially,
as in some games the number of occlusion queries can reach the hundreds.
OpenGL backend already does this by using ARB_QUERY_BUFFER_OBJECT.
We could mimic this in Vulkan too using compute shaders to accumulate
pending queries, however, unlike with OpenGL, we can also just directly
call vkGetQueryPoolResults() from the application's thread.
A downside of the approach taken here is that Vulkan queries allocated
to a wined3d_query are only released after the query is reused or
destroyed. In principle, if this ever becomes a problem, it could be
avoided by protecting the query pool with a mutex and releasing queries
after reading the results.
Signed-off-by: Jan Sikorski <jsikorski@codeweavers.com>
Signed-off-by: Henri Verbeet <hverbeet@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
This is mostly to enable polling queries from the application thread by
removing the need to access the shared context.
Signed-off-by: Jan Sikorski <jsikorski@codeweavers.com>
Signed-off-by: Henri Verbeet <hverbeet@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Its eventual purpose is to allow for skipping the poll list mechanism
by calling query_poll() in the application thread. In this patch it is
only used for when we don't have a separate CS thread.
Signed-off-by: Jan Sikorski <jsikorski@codeweavers.com>
Signed-off-by: Henri Verbeet <hverbeet@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
And use it in WIN_CreateWindowEx to trigger class registration.
Signed-off-by: Jacek Caban <jacek@codeweavers.com>
Signed-off-by: Huw Davies <huw@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
This is more likely some specialized interface of a parameterized type,
used to create mixin types, but we have no way and no clue where it
comes from or how it should be declared.
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
When COM aggregation is involved it is important to add a reference to
the returned iface and not to the inner object.
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
llvm-mingw isn't strictly required - plain clang, llvm-dlltool and
lld also suffice. Those tools are commonly available in distribution
packages.
Signed-off-by: Martin Storsjö <martin@martin.st>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Fixes a regression introduced by 0e7fd41af9
causing missing audio elements in the 64k intros elysian by logicoma
and clean state by conspiracy.
Signed-off-by: Christopher Egert <cme3000@gmail.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
The imported algorithm is modified to match the tests.
Fixes a regression introduced by commit
671cf16f77.
Windows inet_addr() behaves basically the same as Unix
inet_addr() which is different from inet_pton().
Signed-off-by: Paul Gofman <pgofman@codeweavers.com>
Signed-off-by: Zebediah Figura <zfigura@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
So that we can access all the controller devices conveniently through
DInput without having to reimplement the whole HID client, or having
to enumerate the devices every time a new device needs to be open,
and being able to workaround the limitations of the device the XInput
controllers are usually exposing.
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
This fixes an inconsistency since
fed7e7a78b (ntdll: Simulate a syscall
return when starting a thread); in that commit, on i386 and
x86_64, the syscall_frame used for starting the thread was
zeroed with a memset, while it wasn't on arm and arm64.
This had a noticable effect on float maths, where the
uninitialized frame->restore_flags field could have the
CONTEXT_FLOATING_POINT bit set. (call_init_thunk only ORs in
CONTEXT_INTEGER into the restore_flags field, on all architectures).
If that bit was set, it would restore uninitialized data into the
fpscr register, possibly setting the FPU in a nondefault rounding
mode.
Signed-off-by: Martin Storsjö <martin@martin.st>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Like we do in wineserver.
Signed-off-by: Jacek Caban <jacek@codeweavers.com>
Signed-off-by: Huw Davies <huw@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
On non-AVX CPUs, CONTEXT_EX is not being initialized.
In WOW64 mode, this results in invalid exception records when
dispatch_wow_exception() uses RtlCopyContext().
Signed-off-by: Brendan Shanks <bshanks@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Jan Sikorski