[PATCH 1/4] user32/tests: Recreate the window to make it more likely to have focus.

[Rémi Bernon]

Signed-off-by: Rémi Bernon 
---

* PATCH 1/4 should help with these spurious failures:

input.c:2789: Test failed: 15: expected WM_INPUT message
  input.c:2792: Test failed: 15: expected RIM_INPUT message

The root cause is the desktop window getting foreground when we
try to change focus from the thread window back to the main window that
is created topmost. X11 focus out / focus in events receiving order
sometimes make winex11.drv decide to give the desktop window focus and
this ends up with the WM_INPUT messages not being delivered.

* PATCH 2/4 just skips the recurrent failure on Chinese:China locale.

* PATCH 3/4 and PATCH 4/4 which also (Superseding 205693) are trying to
  fix the following spurious failures:

input.c:2162: Test failed: GetRawInputBuffer returned 4
  input.c:2157: Test failed: GetRawInputBuffer returned unexpected size: 0
  input.c:2162: Test failed: GetRawInputBuffer returned 0
  input.c:2163: Test failed: GetRawInputBuffer returned unexpected size: 0
  input.c:2164: Test failed: Unexpected rawinput data: 0
  input.c:2165: Test failed: Unexpected rawinput data: 0
  input.c:2166: Test failed: Unexpected rawinput data: 0
  input.c:2224: Test failed: GetRawInputData succeeded
  input.c:2225: Test failed: GetRawInputData returned deadbeef
  input.c:2338: Test failed: Spurious WM_INPUT messages

I've got no idea if they really do, as I'm not yet able to reproduce
  them consistently when submitting the patches manually. Submitting
  patches from the M-L seem to trigger the failures more reliably. So
  if they don't I guess they could just be ignored.

dlls/user32/tests/input.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/dlls/user32/tests/input.c b/dlls/user32/tests/input.c
index 3d4f0d03d0d..8c67fc67cb3 100644
--- a/dlls/user32/tests/input.c
+++ b/dlls/user32/tests/input.c
@@ -2758,6 +2758,10 @@ static void test_rawinput(const char* argv0)
         {
         case 14:
         case 15:
+            DestroyWindow(hwnd);
+            hwnd = CreateWindowA("static", "static", WS_VISIBLE | WS_POPUP,
+                                 pt.x - 50, pt.y - 50, 100, 100, 0, NULL, NULL, NULL);
+            ok(hwnd != 0, "CreateWindow failed\n");
             SetWindowPos(hwnd, HWND_TOPMOST, 0, 0, 0, 0, SWP_NOSIZE|SWP_NOMOVE);
             SetForegroundWindow(hwnd);
             empty_message_queue();

[PATCH 2/4] user32/tests: Skip inconsistent OemKeyScan tests on non-us keyboard.

[Rémi Bernon]

Signed-off-by: Rémi Bernon 
---
 dlls/user32/tests/input.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/dlls/user32/tests/input.c b/dlls/user32/tests/input.c
index 8c67fc67cb3..9e877c9ae76 100644
--- a/dlls/user32/tests/input.c
+++ b/dlls/user32/tests/input.c
@@ -4171,6 +4171,13 @@ static void test_OemKeyScan(void)
     WCHAR oem, wchr;
     char oem_char;

+    BOOL us_kbd = (GetKeyboardLayout(0) == (HKL)(ULONG_PTR)0x04090409);
+    if (!us_kbd)
+    {
+        skip("skipping test with inconsistent results on non-us keyboard\n");
+        return;
+    }
+
     for (oem = 0; oem < 0x200; oem++)
     {
         ret = OemKeyScan( oem );

[PATCH 2/4] user32/tests: Skip inconsistent OemKeyScan tests on non-us keyboard.

[Marvin]

Hi,

While running your changed tests, I think I found new failures.
Being a bot and all I'm not very good at pattern recognition, so I might be
wrong, but could you please double-check?

Full results can be found at:
https://testbot.winehq.org/JobDetails.pl?Key=90473

Your paranoid android.

=== w7u_2qxl (32 bit report) ===

user32:
input.c:757: Test failed:  0 (a4/0): 00 from 00 -> 80 unexpected
input.c:757: Test failed:  0 (a4/0): 41 from 01 -> 00 unexpected

=== wvistau64 (64 bit report) ===

user32:
input.c:757: Test failed:  0 (a4/0): 01 from 01 -> 00 unexpected
input.c:757: Test failed:  0 (a4/0): 11 from 01 -> 00 unexpected
input.c:757: Test failed:  0 (a4/0): a2 from 01 -> 00 unexpected

=== w10pro64_ja (64 bit report) ===

user32:
input.c:1323: Test failed: GetCursorPos: (100,100)

[PATCH 3/4] user32/tests: Use empty_message_queue to peek messages.

[Rémi Bernon]

Signed-off-by: Rémi Bernon 
---
 dlls/user32/tests/input.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/dlls/user32/tests/input.c b/dlls/user32/tests/input.c
index 9e877c9ae76..a5b130c4a35 100644
--- a/dlls/user32/tests/input.c
+++ b/dlls/user32/tests/input.c
@@ -2139,7 +2139,6 @@ static LRESULT CALLBACK rawinputbuffer_wndproc(HWND hwnd, UINT msg, WPARAM wpara
     RAWINPUT ri;
     char buffer[16 * sizeof(RAWINPUT64)];
     UINT size, count, rawinput_size, iteration = rawinputbuffer_wndproc_count++;
-    MSG message;

if (is_wow64) rawinput_size = sizeof(RAWINPUT64);
     else rawinput_size = sizeof(RAWINPUT);
@@ -2182,7 +2181,7 @@ static LRESULT CALLBACK rawinputbuffer_wndproc(HWND hwnd, UINT msg, WPARAM wpara
             ok(rawinput_buffer_mouse_x(buffer, 0) == 5, "Unexpected rawinput data: %d\n", rawinput_buffer_mouse_x(buffer, 0));

/* peek the messages now, they should still arrive in the correct order */
-            while (PeekMessageA(&message, 0, WM_INPUT, WM_INPUT, PM_REMOVE)) DispatchMessageA(&message);
+            empty_message_queue();
         }

/* reading the message data now should fail on the second iteration, the data

[PATCH 3/4] user32/tests: Use empty_message_queue to peek messages.

[Marvin]

Hi,

While running your changed tests, I think I found new failures.
Being a bot and all I'm not very good at pattern recognition, so I might be
wrong, but could you please double-check?

Full results can be found at:
https://testbot.winehq.org/JobDetails.pl?Key=90474

Your paranoid android.

=== w7u_2qxl (32 bit report) ===

user32:
input.c:757: Test failed:  0 (a4/0): 00 from 00 -> 80 unexpected
input.c:757: Test failed:  0 (a4/0): 41 from 01 -> 00 unexpected

=== w7u_el (32 bit report) ===

user32:
input.c:4445: Test failed: SendInput triggered unexpected message 0xc042

=== w1064v1809 (32 bit report) ===

user32:
input.c:2787: Test failed: 0: expected WM_MOUSEMOVE message
input.c:2787: Test failed: 1: expected WM_MOUSEMOVE message
input.c:2787: Test failed: 2: expected WM_MOUSEMOVE message

=== wvistau64 (64 bit report) ===

user32:
input.c:757: Test failed:  0 (a4/0): 01 from 01 -> 00 unexpected
input.c:757: Test failed:  0 (a4/0): 11 from 01 -> 00 unexpected
input.c:757: Test failed:  0 (a4/0): a2 from 01 -> 00 unexpected

=== w1064_tsign (64 bit report) ===

user32:
input.c:3517: Test failed: expected loop with WM_NCHITTEST messages
input.c:3570: Test failed: expected WM_LBUTTONDOWN message
input.c:3571: Test failed: expected WM_LBUTTONUP message

=== w10pro64_ja (64 bit report) ===

user32:
input.c:1303: Test failed: Wrong set pos: (99,100)
input.c:1323: Test failed: GetCursorPos: (99,100)

[PATCH 3/4] user32/tests: Use empty_message_queue to peek messages.

[Rémi Bernon]

On 5/14/21 12:14 PM, Marvin wrote:
> Hi,
>
> While running your changed tests, I think I found new failures.
> Being a bot and all I'm not very good at pattern recognition, so I might be
> wrong, but could you please double-check?
>
> Full results can be found at:
> https://testbot.winehq.org/JobDetails.pl?Key=90474
>
> Your paranoid android.
>
> === w7u_2qxl (32 bit report) ===
>
> user32:
> input.c:757: Test failed:  0 (a4/0): 00 from 00 -> 80 unexpected
> input.c:757: Test failed:  0 (a4/0): 41 from 01 -> 00 unexpected
>
> === w7u_el (32 bit report) ===
>
> user32:
> input.c:4445: Test failed: SendInput triggered unexpected message 0xc042
>
> === w1064v1809 (32 bit report) ===
>
> user32:
> input.c:2787: Test failed: 0: expected WM_MOUSEMOVE message
> input.c:2787: Test failed: 1: expected WM_MOUSEMOVE message
> input.c:2787: Test failed: 2: expected WM_MOUSEMOVE message
>
> === wvistau64 (64 bit report) ===
>
> user32:
> input.c:757: Test failed:  0 (a4/0): 01 from 01 -> 00 unexpected
> input.c:757: Test failed:  0 (a4/0): 11 from 01 -> 00 unexpected
> input.c:757: Test failed:  0 (a4/0): a2 from 01 -> 00 unexpected
>
> === w1064_tsign (64 bit report) ===
>
> user32:
> input.c:3517: Test failed: expected loop with WM_NCHITTEST messages
> input.c:3570: Test failed: expected WM_LBUTTONDOWN message
> input.c:3571: Test failed: expected WM_LBUTTONUP message
>
> === w10pro64_ja (64 bit report) ===
>
> user32:
> input.c:1303: Test failed: Wrong set pos: (99,100)
> input.c:1323: Test failed: GetCursorPos: (99,100)

Seems to be passing now but I still wouldn't call it a victory. I think 
the failures are more likely to appear when all user32 tests are run at 
once (maybe caused by some left over state from previous tests).

[PATCH 3/4] user32/tests: Use empty_message_queue to peek messages.

[Rémi Bernon]

On 5/14/21 12:18 PM, Rémi Bernon wrote:
> On 5/14/21 12:14 PM, Marvin wrote:
> > Hi,
> >
> > While running your changed tests, I think I found new failures.
> > Being a bot and all I'm not very good at pattern recognition, so I 
> > might be
> > wrong, but could you please double-check?
> >
> > Full results can be found at:
> > https://testbot.winehq.org/JobDetails.pl?Key=90474
> >
> > Your paranoid android.
> >
> > === w7u_2qxl (32 bit report) ===
> >
> > user32:
> > input.c:757: Test failed:  0 (a4/0): 00 from 00 -> 80 unexpected
> > input.c:757: Test failed:  0 (a4/0): 41 from 01 -> 00 unexpected
> >
> > === w7u_el (32 bit report) ===
> >
> > user32:
> > input.c:4445: Test failed: SendInput triggered unexpected message 0xc042
> >
> > === w1064v1809 (32 bit report) ===
> >
> > user32:
> > input.c:2787: Test failed: 0: expected WM_MOUSEMOVE message
> > input.c:2787: Test failed: 1: expected WM_MOUSEMOVE message
> > input.c:2787: Test failed: 2: expected WM_MOUSEMOVE message
> >
> > === wvistau64 (64 bit report) ===
> >
> > user32:
> > input.c:757: Test failed:  0 (a4/0): 01 from 01 -> 00 unexpected
> > input.c:757: Test failed:  0 (a4/0): 11 from 01 -> 00 unexpected
> > input.c:757: Test failed:  0 (a4/0): a2 from 01 -> 00 unexpected
> >
> > === w1064_tsign (64 bit report) ===
> >
> > user32:
> > input.c:3517: Test failed: expected loop with WM_NCHITTEST messages
> > input.c:3570: Test failed: expected WM_LBUTTONDOWN message
> > input.c:3571: Test failed: expected WM_LBUTTONUP message
> >
> > === w10pro64_ja (64 bit report) ===
> >
> > user32:
> > input.c:1303: Test failed: Wrong set pos: (99,100)
> > input.c:1323: Test failed: GetCursorPos: (99,100)
>
> Seems to be passing now but I still wouldn't call it a victory. I think 
> the failures are more likely to appear when all user32 tests are run at 
> once (maybe caused by some left over state from previous tests).

I'm actually able to reproduce the failures locally when running the 
tests through winetest. These patches won't fix the problem, although 
picking them probably won't hurt either.

[PATCH 4/4] user32/tests: Count WM_INPUT messages only as rawinputbuffer iterations.

[Rémi Bernon]

Signed-off-by: Rémi Bernon 
---
 dlls/user32/tests/input.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/dlls/user32/tests/input.c b/dlls/user32/tests/input.c
index a5b130c4a35..1499e299ce7 100644
--- a/dlls/user32/tests/input.c
+++ b/dlls/user32/tests/input.c
@@ -2138,13 +2138,15 @@ static LRESULT CALLBACK rawinputbuffer_wndproc(HWND hwnd, UINT msg, WPARAM wpara
 {
     RAWINPUT ri;
     char buffer[16 * sizeof(RAWINPUT64)];
-    UINT size, count, rawinput_size, iteration = rawinputbuffer_wndproc_count++;
+    UINT size, count, rawinput_size, iteration;

if (is_wow64) rawinput_size = sizeof(RAWINPUT64);
     else rawinput_size = sizeof(RAWINPUT);

if (msg == WM_INPUT)
     {
+        iteration = rawinputbuffer_wndproc_count++;
+
         count = GetRawInputBuffer(NULL, NULL, sizeof(RAWINPUTHEADER));
         ok(count == ~0U, "GetRawInputBuffer succeeded\n");

[PATCH 4/4] user32/tests: Count WM_INPUT messages only as rawinputbuffer iterations.

[Marvin]

Hi,

While running your changed tests, I think I found new failures.
Being a bot and all I'm not very good at pattern recognition, so I might be
wrong, but could you please double-check?

Full results can be found at:
https://testbot.winehq.org/JobDetails.pl?Key=90475

Your paranoid android.

=== w7u_2qxl (32 bit report) ===

user32:
input.c:757: Test failed:  0 (a4/0): 00 from 00 -> 80 unexpected
input.c:757: Test failed:  0 (a4/0): 41 from 01 -> 00 unexpected

=== wvistau64 (64 bit report) ===

user32:
input.c:757: Test failed:  0 (a4/0): 01 from 01 -> 00 unexpected
input.c:757: Test failed:  0 (a4/0): 11 from 01 -> 00 unexpected
input.c:757: Test failed:  0 (a4/0): a2 from 01 -> 00 unexpected

[Marvin]

Hi,

While running your changed tests, I think I found new failures.
Being a bot and all I'm not very good at pattern recognition, so I might be
wrong, but could you please double-check?

Full results can be found at:
https://testbot.winehq.org/JobDetails.pl?Key=90472

Your paranoid android.

=== w7u_2qxl (32 bit report) ===

user32:
input.c:757: Test failed:  0 (a4/0): 00 from 00 -> 80 unexpected
input.c:757: Test failed:  0 (a4/0): 41 from 01 -> 00 unexpected

=== wvistau64 (64 bit report) ===

user32:
input.c:757: Test failed:  0 (a4/0): 01 from 01 -> 00 unexpected
input.c:757: Test failed:  0 (a4/0): 11 from 01 -> 00 unexpected
input.c:757: Test failed:  0 (a4/0): a2 from 01 -> 00 unexpected

=== w1064v1809 (64 bit report) ===

user32:
input.c:2788: Test failed: 0: expected WM_MOUSEMOVE message
input.c:2788: Test failed: 1: expected WM_MOUSEMOVE message
input.c:2788: Test failed: 2: expected WM_MOUSEMOVE message

=== w10pro64_ja (64 bit report) ===

user32:
input.c:3405: Test failed: expected WM_LBUTTONDOWN message
input.c:3406: Test failed: expected WM_LBUTTONUP message
input.c:3433: Test failed: expected WM_NCHITTEST message
input.c:3434: Test failed: expected WM_RBUTTONDOWN message
input.c:3435: Test failed: expected WM_RBUTTONUP message
input.c:3464: Test failed: expected WM_LBUTTONDOWN message
input.c:3465: Test failed: expected WM_LBUTTONUP message
input.c:3518: Test failed: expected loop with WM_NCHITTEST messages
input.c:3571: Test failed: expected WM_LBUTTONDOWN message
input.c:3572: Test failed: expected WM_LBUTTONUP message