I wanted to use ._xy element access to make it shorter, but that attempt was crushed by the reality - such indexing apparently compiles but does not produce correct element access loads. I'm going to update once this is fixed.
--
v2: vkd3d-shader/hlsl: Add determinant() function.
https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/329
On Wed Sep 6 20:27:50 2023 +0000, Helix Graziani wrote:
> how do i do that?
git rebase -i HEAD~# (# can be any number of commits that aren't merge commits)
git rebase <branch> (<branch> is any branch, use it like git merge)
in this case I would say the easiest way (that I know of) is to get a diff, reset and then commit the diff, and then force push.
--
https://gitlab.winehq.org/wine/wine/-/merge_requests/3724#note_44471
On Wed Sep 6 20:13:16 2023 +0000, Etaash Mathamsetty wrote:
> btw your pull request isn't squashed and then merged, you have to squash
> all the commits into one (since this is just a single stub).
> To avoid merge commits use git rebase rather than git merge
ah ok
--
https://gitlab.winehq.org/wine/wine/-/merge_requests/3724#note_44466
--
v2: dmusic: Get rid of the IDirectMusicInstrumentImpl typedef.
dmusic: Rename IDirectMusicInstrumentImpl method prefix to instrument.
dmusic: Simplify and cleanup IDirectMusicInstrument constructor.
dmusic: Get rid of struct collection liWavePoolTablePosition member.
dmusic: Get rid of struct collection liCollectionPosition member.
dmusic: Get rid of the IDirectMusicCollectionImpl typedef.
dmusic: Rename IDirectMusicCollectionImpl method prefix to collection.
dmusic: Move constructor parameter checks to class factory.
https://gitlab.winehq.org/wine/wine/-/merge_requests/3740
--
v4: dmsynth: Return DMUS_E_SYNTHNOTCONFIGURED when sink fails to activate.
dmsynth: Return S_FALSE if IDirectMusicSynth_Activate is no-op.
dmsynth: Forward IDirectMusicSynth_GetLatencyClock to the sink.
dmsynth: Implement latency IReferenceClock interface on the sink.
include: Use IReferenceClock interface from strmif.idl in dmusicc.h.
dmsynth: Move constructor parameter checks to class factory.
https://gitlab.winehq.org/wine/wine/-/merge_requests/3738
I wanted to use ._xy element access to make it shorter, but that attempt was crushed by the reality - such indexing apparently compiles but does not produce correct element access loads. I'm going to update once this is fixed.
--
https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/329
--
v7: tests: Add a test for fmod() with vector arguments.
vkd3d-shader: Use ternary operator in fmod() implementation.
vkd3d-shader/tpf: Use 'movc' to implement ternary operator.
https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/268
This MR replaces `vkd3d_shader_register.immconst_type` with `vkd3d_shader_register.dimension` which is intended for all register types and not just immconsts.
This dimension is parsed in tpf.c, and initialized according to the register type in d3dbc.c.
Having this field in vkd3d_shader_register allows us to avoid hardcoding special cases for the register dimensions when writing sm4 bytecode (e.g. for the sampler src register in gather instructions). Also, it allows for some fixes to d3d_asm.c, which are introduced in part 2 (https://gitlab.winehq.org/fcasas/vkd3d/-/commits/add_vkd3d_reg_dim).
--
v6: vkd3d-shader/d3dbc: Initialize register dimension for all register types.
vkd3d-shader/tpf: Parse register dimension for all register types.
vkd3d-shader: Turn vkd3d_shader_register.immconst_type into vkd3d_shader_register.dimension.
vkd3d-shader/tpf: Use vsir_register_init() in shader_sm1_parse_dst_param().
vkd3d-shader/tpf: Use vsir_register_init() in shader_sm1_parse_src_param().
vkd3d-shader/tpf: Use vsir_register_init() in shader_sm4_read_param().
vkd3d-shader/spirv: Use vsir_register_init() in spirv_compiler_emit_default_control_point_phase().
vkd3d-shader/spirv: Use vsir_register_init() in spirv_compiler_emit_resource_declaration().
vkd3d-shader/spirv: Use vsir_register_init() in spirv_compiler_emit_sampler_declaration().
vkd3d-shader/spirv: Use vsir_register_init() in spirv_compiler_emit_dcl_immediate_constant_buffer().
vkd3d-shader/spirv: Use vsir_register_init() in spirv_compiler_emit_cbv_declaration().
vkd3d-shader/spirv: Use vsir_register_init() in spirv_compiler_emit_dcl_indexable_temp().
vkd3d-shader/spirv: Use vsir_register_init() in spirv_compiler_emit_hull_shader_builtins().
vkd3d-shader/spirv: Use vsir_register_init() in spirv_compiler_get_invocation_id().
https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/319
On Wed Sep 6 17:08:48 2023 +0000, Jeffrey Smith wrote:
> Are you sure? It appears to be inconsistent with the styling for the
> rest of the if condition. All other or-separated terms are using
> parenthesis, regardless of whether they contain an &&.
You're probably right. Let's leave it as is.
--
https://gitlab.winehq.org/wine/wine/-/merge_requests/3744#note_44449
An application subclasses edit control and crashes on unexpected WM_TIMER messages. A timer was
added a long time ago and it was for auto scrolling the view when the cursor is outside of the
client area. However, with a EDIT_EM_ScrollCaret() call in EDIT_WM_MouseMove(), the view can be
scrolled to the correct position automatically. So the timer is not necessary at all. This is tested
manually with a long text in the edit control with notepad on XP, Win7 and Win10. Before this patch,
the caret will move backwards every 100ms when the clicking the left mouse button on the bottom of
the text and then dragging the cursor upwards and outside of the client area. After this, the
behavior is the same as native.
--
https://gitlab.winehq.org/wine/wine/-/merge_requests/3623
This begins the initial implementation of a unified QueryInterface on the dispex. There's a couple of hacks used to make the conversion split up easier, but they're all resolved within this MR.
--
https://gitlab.winehq.org/wine/wine/-/merge_requests/3723
This MR adds support for handling `wl_pointer` devices:
1. Handle enter/leave/(absolute) motion/button/axis events to allow users to interact with applications using such devices, e.g., a mouse.
2. Update the cursor image used for Wayland surfaces based on the Windows cursor bitmap data.
A few notes:
1. A single `wl_seat`/`wl_pointer` is supported for now.
2. Many of the input interactions require window management support that hasn't been implemented yet (e.g., popup positioning, resizing, max/fullscreen).
3. There is some potential for bad interactions between the input event dispatching mechanism and the `win32u` flushing logic for `window_surface` , which can lead to visual glitches. Although I don't think this is blocker for this MR, I would like to start a discussion about the problem and mitigation ideas in a comment below.
Thanks!
--
https://gitlab.winehq.org/wine/wine/-/merge_requests/3686
It can be unnecessary at best and unsupported at worst (e.g. no ARB_texture_multisample).
--
v3: wined3d: Don't skip ARB fragment program selection when fog state is dirty.
wined3d: Don't force going through a texture when downloading from renderbuffers.
wined3d: Handle depth textures in texture2d_read_from_framebuffer().
wined3d: Rename wined3d_context_gl_apply_fbo_state_blit() function.
wined3d: Don't setup FBO and draw buffers in wined3d_context_gl_apply_blit_state().
wined3d: Don't call wined3d_texture_load() from wined3d_context_gl_apply_blit_state().
wined3d: Don't bind the FBO to GL_READ_FRAMEBUFFER in wined3d_context_gl_apply_blit_state().
wined3d: Prepare/load the destination resource location in arbfp_blitter_blit().
wined3d: Prepare/load the destination resource location in ffp_blitter_blit().
wined3d: Don't call wined3d_context_gl_apply_blit_state() from texture2d_read_from_framebuffer().
https://gitlab.winehq.org/wine/wine/-/merge_requests/3211
Now that we can write HLSL intrinsics in HLSL, cleaning up missing
functions becomes much easier.
This commit also extends the trigonometry tests a little bit to make
sure that tan works right.
--
v4: vkd3d-shader/hlsl: Implement intrinsic tan.
https://gitlab.winehq.org/wine/vkd3d/-/merge_requests/321
If DISABLEAUTOREDIRECTS is set in BINDINFO options, urlmon does not set HTTP verb to GET when handling redirections.
Although HTTP specification is vague on the correct behaviour here, many web servers expect this. This is what's causing the "400 Bad Request" error when user tries to log into GMail accounts using Outlook.
--
https://gitlab.winehq.org/wine/wine/-/merge_requests/3725
Since dcf0bf1f, one could no longer kill a GUI application launched from
unix shell (with start.exe /exec) with ctrl-c.
(depends on https://gitlab.winehq.org/wine/wine/-/merge_requests/3310
to be fully functional).
Signed-off-by: Eric Pouech <epouech(a)codeweavers.com>
--
v5: programs/cmd: Better handle ctrl-c events.
ntdll: Unconditionally forward console events to kernelbase.
kernel32/console: Add tests for GenerateConsoleCtrlEvent().
https://gitlab.winehq.org/wine/wine/-/merge_requests/3312
Since dcf0bf1f, one could no longer kill a GUI application launched from
unix shell (with start.exe /exec) with ctrl-c.
(depends on https://gitlab.winehq.org/wine/wine/-/merge_requests/3310
to be fully functional).
Signed-off-by: Eric Pouech <epouech(a)codeweavers.com>
--
v4: programs/cmd: Better handle ctrl-c events.
ntdll,server: Terminate GUI programs launched from unix shell.
kernel32/console: Add tests for GenerateConsoleCtrlEvent().
https://gitlab.winehq.org/wine/wine/-/merge_requests/3312
--
v3: dmsynth: Return DMUS_E_SYNTHNOTCONFIGURED when sink fails to activate.
dmsynth: Return S_FALSE if IDirectMusicSynth_Activate is no-op.
dmsynth: Forward IDirectMusicSynth_GetLatencyClock to the sink.
dmsynth: Implement latency IReferenceClock interface on the sink.
include: Use IReferenceClock interface from strmif.idl in dmusicc.h.
dmsynth: Simplify and cleanup class constructors.
https://gitlab.winehq.org/wine/wine/-/merge_requests/3738