The pSignatureSettings field was not unitilized so it got a random value from
the stack causing a later crash when it was dereferenced in Wintrust
(SoftPubloadSignature).
I was unable to write tests for this; it seems it doesn't consistently work on
Windows. However, Rayman 3 seems to rely on it; it maps the same buffer twice
immediately after creation, with DISCARD flags on both maps, and expects the
same address to be returned.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=53752
The code originally intended to achieve this by simply waiting on flushing_cs,
and having the application thread hold flushing_cs while seeking. Unfortunately,
this can result in starvation of the application thread, since the streaming
thread always reacquires flushing_cs immediately after releasing it. Avoid this
by using a separate condition variable.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=53403
Signed-off-by: Zebediah Figura <zfigura@codeweavers.com>
The code previously relied on inherent atomicity of atomic types, but atomicity
doesn't imply the right memory ordering. Be explicit about the threading model
we want.
Signed-off-by: Zebediah Figura <zfigura@codeweavers.com>
Starting with Win10, height of multiline edit controls no longer seems
to be hardwired to a multiple of the font height (when larger than
font height).
- Adapt the tests accordingly (detecting how to compute the expected
height of a control).
- Extend the ranges of the tested heights.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=52152
Signed-off-by: Eric Pouech <eric.pouech@gmail.com>
This fixes a regression introduced by ebbcc10b05.
Prior to that commit, buffers which were not CPU-accessible would not have the
WINED3D_BUFFER_USE_BO flag set, and accordingly
buffer_resource_sub_resource_map() would simply return the SYSMEM location,
ignoring the DISCARD and NOOVERWRITE flags.
However, the "accelerated" path in wined3d_cs_map_upload_bo() only checks for
the DISCARD flag, assuming that it is only set for dynamic GPU-accessible
buffers, and would subsequently try to allocate non-mappable memory and then map
it.
This commit avoids the accelerated path for such buffers, once again matching
our old behaviour. According to [1], this also matches Windows: DISCARD on
SYSTEMMEM buffers is ignored on both AMD and NVidia, and NOOVERWRITE is ignored
on NVidia (but not AMD).
[1] https://www.winehq.org/mailman3/hyperkitty/list/wine-devel@winehq.org/message/XACSAVAGYTJUZDFLX4JWO665VTB6M7LX/
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=53863
DISCARD maps on deferred contexts do not seem to offer the same synchronization
guarantees as do DISCARD maps on the immediate context, at least not with
respect to subsequent NOOVERWRITE maps on the immediate context. This is not
particularly surprising, seeing as it is probably far more usual to only map a
resource on one kind of context or the other.
The test currently fails on Windows 11 on NVidia drivers, but it also fails on
Mesa, for what I think are the same reasons.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=52563
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=54023
Although the test has never been observed to fail on Windows, I think the
failure is genuinely a race in the test.
File handles (like events) are signaled in order to mark that an I/O operation
has completed. In this case the I/O operation includes manipulating data on both
ends of the pipe, and as part of that may signal the other end. Internally,
however, the file handle must logically happen *after* all of this processing
has taken place, not least because (given the Windows I/O architecture) it is
the job of the I/O manager, not the IRP handler.
Since the purpose of the test is probably just to check that the file handle
will be signaled after a synchronous I/O operation has completed, we may as well
check it from the client thread, after we know for sure that it has.
Wine-Bug: https://bugs.winehq.org//show_bug.cgi?id=54078
