[PATCH 0/2] MR953: Fix test failures on Gitlab CI.

[PATCH 1/2] wmvcore/tests: Reduce the number of printed todos.

[Alexandre Julliard]

From: Alexandre Julliard

---
 dlls/wmvcore/tests/wmvcore.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/dlls/wmvcore/tests/wmvcore.c b/dlls/wmvcore/tests/wmvcore.c
index 06e1852c8c0..aacff6f7bda 100644
--- a/dlls/wmvcore/tests/wmvcore.c
+++ b/dlls/wmvcore/tests/wmvcore.c
@@ -3422,6 +3422,8 @@ START_TEST(wmvcore)
     if(hr != S_OK)
         return;

+    winetest_mute_threshold = 3;  /* FIXME: thread tests print too many "got wrong thread" todos */
+
     test_wmreader_interfaces();
     test_wmsyncreader_interfaces();
     test_wmwriter_interfaces();

[PATCH 2/2] gdiplus: Get the compositing mode directly from the graphics object.

[Alexandre Julliard]

From: Alexandre Julliard

GdipGetCompositingMode() fails when the object is busy, which happens
when flushing changes in GdipReleaseDC().

Fixes a test crash on Gitlab CI.
---
 dlls/gdiplus/graphics.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/dlls/gdiplus/graphics.c b/dlls/gdiplus/graphics.c
index a9f7a464ce3..fca66c8654b 100644
--- a/dlls/gdiplus/graphics.c
+++ b/dlls/gdiplus/graphics.c
@@ -418,9 +418,7 @@ static GpStatus alpha_blend_bmp_pixels(GpGraphics *graphics, INT dst_x, INT dst_
 {
     GpBitmap *dst_bitmap = (GpBitmap*)graphics->image;
     INT x, y;
-    CompositingMode comp_mode;
-
-    GdipGetCompositingMode(graphics, &comp_mode);
+    CompositingMode comp_mode = graphics->compmode;

for (y=0; y<src_height; y++)
     {

[Rémi Bernon (＠rbernon)]

FWIW I have plans to fix these wmvcore todos but it'll have to wait after https://gitlab.winehq.org/wine/wine/-/merge_requests/936 and a few more other patches. In the meantime I think something like that makes sense.

[PATCH 0/2] MR953: Fix test failures on Gitlab CI. - approved

[Zebediah Figura (＠zfigura)]

This merge request was approved by Zebediah Figura.

[PATCH 0/2] MR953: Fix test failures on Gitlab CI. - approved

[Esme Povirk (＠madewokherd)]

This merge request was approved by Esme Povirk.

[Esme Povirk (＠madewokherd)]

The change looks correct to me, but I'm confused why using comp_mode uninitialized caused a crash.

[Alexandre Julliard (＠julliard)]

> The change looks correct to me, but I'm confused why using comp_mode uninitialized caused a crash.

It's because if the uninitialized value happens to be
CompositingModeSourceCopy we set pixels unconditionally, but the test
passes an invalid buffer to verify that only modified pixels are being
updated.