Followup to c641a9a2 and b7fa3068.
Failures of many windows tests show up in [1] since a few days.
Seems CI was not happy about it in the original MR too. [2][3]
1. https://test.winehq.org/data/patterns.html#rtworkq:rtworkq 2. https://gitlab.winehq.org/cmccarthy/wine/-/jobs/134360#L102 3. https://gitlab.winehq.org/cmccarthy/wine/-/jobs/134361#L102
CC: @cmccarthy
From: Bernhard Übelacker bernhardu@mailbox.org
Followup to c641a9a2 and b7fa3068. --- dlls/rtworkq/tests/rtworkq.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-)
diff --git a/dlls/rtworkq/tests/rtworkq.c b/dlls/rtworkq/tests/rtworkq.c index e6930083e8f..9ca6ea31fa1 100644 --- a/dlls/rtworkq/tests/rtworkq.c +++ b/dlls/rtworkq/tests/rtworkq.c @@ -271,16 +271,22 @@ static void test_undefined_queue_id(void) ok(callback_result == result, "Expected result %p, got %p.\n", result, callback_result);
hr = RtwqPutWorkItem(RTWQ_CALLBACK_QUEUE_PRIVATE_MASK, 0, result); - ok(hr == S_OK, "got %#lx\n", hr); + todo_wine + ok(hr == RTWQ_E_INVALID_WORKQUEUE, "got %#lx\n", hr); res = wait_async_callback_result(&test_callback->IRtwqAsyncCallback_iface, 100, &callback_result); - ok(res == 0, "got %#lx\n", res); - ok(callback_result == result, "Expected result %p, got %p.\n", result, callback_result); + todo_wine + ok(res == WAIT_TIMEOUT, "got %#lx\n", res); + todo_wine + ok(callback_result == NULL, "Expected result %p, got %p.\n", NULL, callback_result);
hr = RtwqPutWorkItem(RTWQ_CALLBACK_QUEUE_PRIVATE_MASK & (RTWQ_CALLBACK_QUEUE_PRIVATE_MASK - 1), 0, result); - ok(hr == S_OK, "got %#lx\n", hr); + todo_wine + ok(hr == RTWQ_E_INVALID_WORKQUEUE, "got %#lx\n", hr); res = wait_async_callback_result(&test_callback->IRtwqAsyncCallback_iface, 100, &callback_result); - ok(res == 0, "got %#lx\n", res); - ok(callback_result == result, "Expected result %p, got %p.\n", result, callback_result); + todo_wine + ok(res == WAIT_TIMEOUT, "got %#lx\n", res); + todo_wine + ok(callback_result == NULL, "Expected result %p, got %p.\n", NULL, callback_result);
IRtwqAsyncResult_Release(result);
Oops, I didn't check closely the test results on PRIVATE_MASK values after they were added to the MR.
I think the idea was that values 0-0xffff are accepted, and private queues are above that, in the range (0x0001-0xffff) << 16.
Submitted !7437 instead.
Closing, there are still some failures, but in other parts.
This merge request was closed by Nikolay Sivov.
Thanks for taking care and properly fixing it.