September 2016 - Wine-Devel

Re: [PATCH] ws2_32: accept addrlelen32 parameter is optional
by Bruno Jesus Sept. 7, 2016

Sept. 7, 2016

On Tue, Sep 6, 2016 at 10:38 PM, Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> wrote: > Signed-off-by: Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> > --- > dlls/ws2_32/socket.c | 2 +- > dlls/ws2_32/tests/sock.c | 9 +++++++++ > 2 files changed, 10 insertions(+), 1 deletion(-) > > diff --git a/dlls/ws2_32/socket.c b/dlls/ws2_32/socket.c > index 79ccfc2..38fa62f 100644 > --- a/dlls/ws2_32/socket.c > +++ b/dlls/ws2_32/socket.c > @@ -2653,7 +2653,7 @@ SOCKET WINAPI WS_accept(SOCKET s, struct WS_sockaddr *addr, int *addrlen32) > SERVER_END_REQ; > if (!status) > { > - if (addr && WS_getpeername(as, addr, addrlen32)) > + if (addr && addrlen32 && WS_getpeername(as, addr, addrlen32)) > { > WS_closesocket(as); > return SOCKET_ERROR; > diff --git a/dlls/ws2_32/tests/sock.c b/dlls/ws2_32/tests/sock.c > index 2640dd4..8c6ca29 100644 > --- a/dlls/ws2_32/tests/sock.c > +++ b/dlls/ws2_32/tests/sock.c > @@ -4200,6 +4200,15 @@ static void test_accept(void) > ok(socklen != sizeof(ss), "unexpected length\n"); > ok(ss.ss_family, "family not set\n"); Hi, I have the impression that you are leaking the accepted and connector socket from previous tests, can you please check? > + socklen = sizeof(address); > + connector = setup_connector_socket(&address, socklen, FALSE); > + if (connector == INVALID_SOCKET) goto done; > + > + memset(&ss, 0, sizeof(ss)); > + accepted = accept(server_socket, (struct sockaddr *)&ss, NULL); > + ok(accepted != INVALID_SOCKET, "Failed to accept connection, %d\n", WSAGetLastError()); > + ok(!ss.ss_family, "family not set\n"); I believe the comment should be "family set\n", right? But actually your change suggests that the whole structure should be left unmodified so I believe it's better to test the whole structure to ensure the change is fully tested. You could add ss_empty variable, memset it to zero and memcmp both. Please fix the subject ;-) Best wishes, Bruno

1 0

Re: setupapi/tests: Also check for ERROR_CALL_NOT_IMPLEMENTED at the second call to SetupDiCreateDeviceInfoListExW
by Marvin Sept. 6, 2016

Sept. 6, 2016

Hi, While running your changed tests on Windows, 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=25601 Your paranoid android. === w1064 (64 bit devinst) === devinst.c:1137: Test failed: Unexpected path \\?\root#legacy_bogus#0001#{6a55b5a4-3f65-11db-b704-0011955c2bdb} devinst.c:1259: Test failed: Expected key to not exist devinst.c:1365: Test failed: Expected key to not exist

1 0

Re: [PATCH 2/2] user32/tests: On Win 8 and later, moving a window off-screen doesn't crop its update region.
by Marvin Sept. 6, 2016

Sept. 6, 2016

Hi, While running your changed tests on Windows, 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=25560 Your paranoid android. === w864 (32 bit msg) === msg.c:13939: Test failed: WM_NCRBUTTONDOWN on caption: 3: the msg sequence is not complete: expected 007b - actual 0000 === w1064 (32 bit msg) === msg.c:7444: Test failed: destroy child on thread exit: 1: the msg 0x000f was expected, but got msg 0x001a instead msg.c:7444: Test failed: destroy child on thread exit: 2: the msg 0x0014 was expected, but got msg 0x000f instead msg.c:7444: Test failed: destroy child on thread exit: 3: the msg sequence is not complete: expected 0000 - actual 0014 msg.c:13939: Test failed: WM_NCRBUTTONDOWN on caption: 3: the msg sequence is not complete: expected 007b - actual 0000 === w864 (64 bit msg) === msg.c:13939: Test failed: WM_NCRBUTTONDOWN on caption: 3: the msg sequence is not complete: expected 007b - actual 0000 === w1064 (64 bit msg) === msg.c:7089: Test failed: SetWindowPos:FrameChangedDeferErase: 10: the msg sequence is not complete: expected 0000 - actual 001a msg.c:13939: Test failed: WM_NCRBUTTONDOWN on caption: 3: the msg sequence is not complete: expected 007b - actual 0000

3 2

Re: [PATCH v3 07/20] winebus.sys: Add Linux udev bus
by Thomas Faber Sept. 6, 2016

Sept. 6, 2016

> +NTSTATUS WINAPI common_pnp_dispatch(DEVICE_OBJECT *device, IRP *irp) > +{ > + NTSTATUS rc = STATUS_NOT_SUPPORTED; You generally want to initialize this with irp->IoStatus.Status. For requests the bus driver doesn't support, it needs to return whatever the drivers higher up the stack decided. E.g. IRP_MN_QUERY_DEVICE_RELATIONS might be handled by a bus driver FDO above, in which case it will likely set the status to success -- and lower-level drivers need to propagate this status unless they have specific reason to override it. This is the reason for the requirement to initialize IoStatus to STATUS_NOT_SUPPORTED when sending the IRP. > + IO_STACK_LOCATION *irpsp = IoGetCurrentIrpStackLocation(irp); > + > + switch(irpsp->MinorFunction) > + { > + case IRP_MN_QUERY_DEVICE_RELATIONS: > + TRACE("IRP_MN_QUERY_DEVICE_RELATIONS\n"); > + break; > + case IRP_MN_QUERY_ID: > + { > + TRACE("IRP_MN_QUERY_ID\n"); > + break; > + } > + case IRP_MN_QUERY_CAPABILITIES: > + TRACE("IRP_MN_QUERY_CAPABILITIES\n"); > + break; > + } > + IoCompleteRequest( irp, IO_NO_INCREMENT ); > + return rc; > +} > + > +DWORD check_bus_disabled(UNICODE_STRING *registry_path) > +{ > + OBJECT_ATTRIBUTES attr; > + HANDLE key; > + DWORD disabled = 0; > + static const WCHAR disabledW[] = {'D','i','s','a','b','l','e','d',0}; > + static const UNICODE_STRING disabled_str = {sizeof(disabledW) - sizeof(WCHAR), sizeof(disabledW), (WCHAR*)disabledW}; > + > + InitializeObjectAttributes(&attr, registry_path, OBJ_CASE_INSENSITIVE, NULL, NULL); This is another case for OBJ_KERNEL_HANDLE. You basically want this by default in kernel code, unless you specifically intend to pass this handle to an application. > + if (ZwOpenKey(&key, KEY_ALL_ACCESS, &attr) == STATUS_SUCCESS) > + { > + DWORD size; > + KEY_VALUE_PARTIAL_INFORMATION info; > + > + if (ZwQueryValueKey(key, &disabled_str, KeyValuePartialInformation, &info, sizeof(info), &size) == STATUS_SUCCESS) Your 'info' structure isn't large enough to hold DWORD data, since the 'Data' member is a one byte array. > + { > + disabled = *(DWORD*)&info.Data; > + } > + > + ZwClose(key); > + } > + > + return disabled; > +}

1 0

Re: [PATCH v3 03/20] ntoskrnl.exe: Implement loading plug and play devices
by Thomas Faber Sept. 6, 2016

Sept. 6, 2016

> @@ -2791,10 +2794,188 @@ done: > } > > > +static NTSTATUS WINAPI internal_complete( DEVICE_OBJECT *device, IRP *irp, void *context ) > +{ > + SetEvent( irp->UserEvent ); > + return STATUS_MORE_PROCESSING_REQUIRED; > +} > + > + > +static NTSTATUS send_device_irp( DEVICE_OBJECT *device, IRP *irp, ULONG_PTR *info ) > +{ > + NTSTATUS status; > + IO_STACK_LOCATION *irpsp; > + HANDLE event = CreateEventA( NULL, FALSE, FALSE, NULL ); > + > + irp->UserEvent = event; > + irpsp = IoGetNextIrpStackLocation( irp ); > + irpsp->CompletionRoutine = internal_complete; > + irpsp->Control = SL_INVOKE_ON_SUCCESS | SL_INVOKE_ON_ERROR; You probably want a call on cancel too? > + IoCallDriver( device, irp ); > + > + if (irp->IoStatus.u.Status == STATUS_PENDING) > + WaitForSingleObject( event, INFINITE ); STATUS_PENDING is returned from the dispatch routine, it doesn't make sense in irp->IoStatus. (Between your call to IoCallDriver and your completion routine firing, somebody else owns the IRP, meaning you're not allowed to touch it.) > + status = irp->IoStatus.u.Status; > + if (info) > + *info = irp->IoStatus.Information; > + IoCompleteRequest( irp, IO_NO_INCREMENT ); > + CloseHandle( event ); > + return status; > +} > + > + > +static NTSTATUS get_device_id( DEVICE_OBJECT *device, BUS_QUERY_ID_TYPE type, WCHAR **id ) > +{ > + IO_STACK_LOCATION *irpsp; > + IO_STATUS_BLOCK irp_status; > + IRP *irp; > + > + if (!(irp = IoBuildSynchronousFsdRequest( IRP_MJ_PNP, device, NULL, 0, NULL, NULL, &irp_status ))) > + return STATUS_NO_MEMORY; > + > + irpsp = IoGetNextIrpStackLocation( irp ); > + irpsp->MinorFunction = IRP_MN_QUERY_ID; > + irpsp->Parameters.QueryId.IdType = type; You should initialize irp->IoStatus.Status to STATUS_NOT_SUPPORTED, drivers depend on this (they're supposed to leave it untouched for requests they don't support). > + return send_device_irp( device, irp, (ULONG_PTR *)id ); > +} > + > + > +static BOOL get_driver_for_id( const WCHAR *id, WCHAR *driver ) > +{ > + static const WCHAR serviceW[] = {'S','e','r','v','i','c','e',0}; > + static const UNICODE_STRING service_str = { sizeof(serviceW) - sizeof(WCHAR), sizeof(serviceW), (WCHAR *)serviceW }; > + static const WCHAR critical_fmtW[] = > + {'\\','R','e','g','i','s','t','r','y', > + '\\','M','a','c','h','i','n','e', > + '\\','S','y','s','t','e','m', > + '\\','C','u','r','r','e','n','t','C','o','n','t','r','o','l','S','e','t', > + '\\','C','o','n','t','r','o','l', > + '\\','C','r','i','t','i','c','a','l','D','e','v','i','c','e','D','a','t','a','b','a','s','e', > + '\\','%','s',0}; > + WCHAR buffer[sizeof(KEY_VALUE_PARTIAL_INFORMATION) + MAX_SERVICE_NAME * sizeof(WCHAR)]; > + KEY_VALUE_PARTIAL_INFORMATION *info = (KEY_VALUE_PARTIAL_INFORMATION *)buffer; > + OBJECT_ATTRIBUTES attr; > + UNICODE_STRING key; > + NTSTATUS status; > + HANDLE hkey; > + WCHAR *keyW; > + DWORD len; > + > + if (!(keyW = RtlAllocateHeap( GetProcessHeap(), 0, sizeof(critical_fmtW) + strlenW(id) * sizeof(WCHAR) ))) > + return STATUS_NO_MEMORY; > + > + sprintfW( keyW, critical_fmtW, id ); > + RtlInitUnicodeString( &key, keyW ); > + InitializeObjectAttributes( &attr, &key, OBJ_CASE_INSENSITIVE, NULL, NULL ); This should have OBJ_KERNEL_HANDLE in native, not sure if that is, or ever will be, relevant for Wine. > + status = ZwOpenKey( &hkey, KEY_ALL_ACCESS, &attr ); > + RtlFreeUnicodeString( &key ); > + if (status != STATUS_SUCCESS) > + { > + WARN_(plugplay)( "failed to open critical device database\n" ); > + return FALSE; > + } > + > + status = ZwQueryValueKey( hkey, &service_str, KeyValuePartialInformation, > + info, sizeof(buffer) - sizeof(WCHAR), &len ); > + ZwClose( hkey ); > + if (status != STATUS_SUCCESS || info->Type != REG_SZ) > + { > + TRACE_(plugplay)( "no driver found for %s\n", debugstr_w(id) ); > + return FALSE; > + } > + > + memcpy( driver, info->Data, info->DataLength ); > + driver[ info->DataLength / sizeof(WCHAR) ] = 0; > + TRACE_(plugplay)( "found driver %s for %s\n", debugstr_w(driver), debugstr_w(id) ); > + return TRUE; > +} > + > + > +static void handle_bus_relations( DEVICE_OBJECT *device ) > +{ > + static const WCHAR driverW[] = {'\\','D','r','i','v','e','r','\\',0}; > + WCHAR buffer[MAX_SERVICE_NAME + sizeof(servicesW)/sizeof(WCHAR)]; > + WCHAR driver[MAX_SERVICE_NAME] = {0}; > + DRIVER_OBJECT *driver_obj; > + UNICODE_STRING string; > + WCHAR *ids, *ptr; > + NTSTATUS status; > + > + TRACE_(plugplay)( "(%p)\n", device ); > + > + /* We could (should?) do a full IRP_MN_QUERY_DEVICE_RELATIONS query, > + * but we dont have to, We have the DEVICE_OBJECT of the new device > + * so we can simply handle the process here */ > + > + status = get_device_id( device, BusQueryCompatibleIDs, &ids ); PNP (and power, and generally most) IRPs should be sent to the top of the device stack. That generally means you want to throw an IoGetAttachedDeviceReference call in before sending an IRP. > + if (status != ERROR_SUCCESS || !ids) STATUS_SUCCESS? > + { > + ERR_(plugplay)( "Failed to get device IDs\n" ); > + return; > + } > + > + for (ptr = ids; *ptr; ptr += strlenW(ptr) + 1) > + { > + if (get_driver_for_id( ptr, driver )) > + break; > + } > + RtlFreeHeap( GetProcessHeap(), 0, ids ); > + > + if (!driver[0]) > + { > + ERR_(plugplay)( "No matching driver found for device\n" ); > + return; > + } > + > + strcpyW( buffer, servicesW ); > + strcatW( buffer, driver ); > + RtlInitUnicodeString( &string, buffer ); > + if (ZwLoadDriver( &string ) != STATUS_SUCCESS) > + { > + ERR_(plugplay)( "Failed to load driver %s\n", debugstr_w(driver) ); > + return; > + } > + > + strcpyW( buffer, driverW ); > + strcatW( buffer, driver ); > + RtlInitUnicodeString( &string, buffer ); > + if (ObReferenceObjectByName( &string, OBJ_CASE_INSENSITIVE, NULL, > + 0, NULL, KernelMode, NULL, (void **)&driver_obj ) != STATUS_SUCCESS) > + { > + ERR_(plugplay)( "Failed to locate loaded driver %s\n", debugstr_w(driver) ); > + return; > + } > + > + if (driver_obj->DriverExtension->AddDevice) > + status = driver_obj->DriverExtension->AddDevice( driver_obj, device ); > + else > + status = STATUS_NOT_IMPLEMENTED; > + > + ObDereferenceObject( driver_obj ); > + > + if (status != STATUS_SUCCESS) > + ERR_(plugplay)( "AddDevice failed for driver %s\n", debugstr_w(driver) ); > +} > + > + > /*********************************************************************** > * IoInvalidateDeviceRelations (NTOSKRNL.EXE.@) > */ > void WINAPI IoInvalidateDeviceRelations( DEVICE_OBJECT *device_object, DEVICE_RELATION_TYPE type ) > { > - FIXME( "(%p, %i): stub\n", device_object, type ); > + TRACE( "(%p, %i)\n", device_object, type ); > + > + switch (type) > + { > + case BusRelations: > + handle_bus_relations( device_object ); > + break; > + default: > + FIXME( "unhandled relation %i\n", type ); > + break; > + } > } > >

2 4

Re: taskschd: Added ITaskDefinition::get_Triggers implementation.
by Marvin Sept. 6, 2016

Sept. 6, 2016

Hi, While running your changed tests on Windows, 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=25577 Your paranoid android. === w1064 (32 bit scheduler) === scheduler.c:159: Test failed: expected ERROR_INVALID_NAME, got 0x80070002 scheduler.c:204: Test failed: expected ERROR_PATH_NOT_FOUND, got 0x80070002 scheduler.c:354: Test failed: expected ERROR_FILE_NOT_FOUND, got 0 scheduler.c:767: Test failed: expected ERROR_PATH_NOT_FOUND, got 0x80070002 scheduler.c:773: Test failed: expected ERROR_PATH_NOT_FOUND, got 0x80070002 scheduler.c:941: Test failed: expected ERROR_FILE_NOT_FOUND, got 0 === w1064 (64 bit scheduler) === scheduler.c:159: Test failed: expected ERROR_INVALID_NAME, got 0x80070002 scheduler.c:204: Test failed: expected ERROR_PATH_NOT_FOUND, got 0x80070002 scheduler.c:354: Test failed: expected ERROR_FILE_NOT_FOUND, got 0 scheduler.c:767: Test failed: expected ERROR_PATH_NOT_FOUND, got 0x80070002 scheduler.c:773: Test failed: expected ERROR_PATH_NOT_FOUND, got 0x80070002 scheduler.c:941: Test failed: expected ERROR_FILE_NOT_FOUND, got 0

1 0

Re: [PATCH v2] d3d11/tests: Extend GetDC() tests
by Henri Verbeet Sept. 6, 2016

Sept. 6, 2016

On 6 September 2016 at 11:42, Nikolay Sivov <nsivov(a)codeweavers.com> wrote: > dlls/d3d11/tests/Makefile.in | 2 +- > dlls/d3d11/tests/d3d11.c | 159 ++++++++++++++++++++++++++++++++++++++----- > 2 files changed, 142 insertions(+), 19 deletions(-) > Unfortunately this fails here on Windows: d3d11.c:8502: Test failed: Got unexpected size 24 for format B8G8R8A8_UNORM. d3d11.c:8516: Test failed: Got unexpected bits 00000000 for format B8G8R8A8_UNORM. d3d11.c:8518: Test failed: Got unexpected size 2686104 for format B8G8R8A8_UNORM. d3d11.c:8520: Test failed: Got unexpected width 9502720 for format B8G8R8A8_UNORM. d3d11.c:8522: Test failed: Got unexpected height 9502720 for format B8G8R8A8_UNORM. d3d11.c:8524: Test failed: Got unexpected plane count 0 for format B8G8R8A8_UNORM. d3d11.c:8526: Test failed: Got unexpected bit count 0 for format B8G8R8A8_UNORM. d3d11.c:8528: Test failed: Got unexpected compression 0x28fd48 for format B8G8R8A8_UNORM. d3d11.c:8530: Test failed: Got unexpected image size 2686324 for format B8G8R8A8_UNORM. d3d11.c:8532: Test failed: Got unexpected horizontal resolution 4316655 for format B8G8R8A8_UNORM. d3d11.c:8534: Test failed: Got unexpected vertical resolution 1 for format B8G8R8A8_UNORM. d3d11.c:8536: Test failed: Got unexpected used colour count 4330808 for format B8G8R8A8_UNORM. d3d11.c:8538: Test failed: Got unexpected important colour count 2686152 for format B8G8R8A8_UNORM. d3d11.c:8540: Test failed: Got unexpected colour masks 0x0028fcb8 0x735a25c5 0x00000000 for format B8G8R8A8_UNORM. d3d11.c:8543: Test failed: Got unexpected section 000020A4 for format B8G8R8A8_UNORM. d3d11.c:8544: Test failed: Got unexpected offset 4228276 for format B8G8R8A8_UNORM. d3d11.c:8502: Test failed: Got unexpected size 24 for format B8G8R8A8_TYPELESS. d3d11.c:8516: Test failed: Got unexpected bits 00000000 for format B8G8R8A8_TYPELESS. d3d11.c:8518: Test failed: Got unexpected size 2686104 for format B8G8R8A8_TYPELESS. d3d11.c:8520: Test failed: Got unexpected width 9502720 for format B8G8R8A8_TYPELESS. d3d11.c:8522: Test failed: Got unexpected height 9502720 for format B8G8R8A8_TYPELESS. d3d11.c:8524: Test failed: Got unexpected plane count 0 for format B8G8R8A8_TYPELESS. d3d11.c:8526: Test failed: Got unexpected bit count 0 for format B8G8R8A8_TYPELESS. d3d11.c:8528: Test failed: Got unexpected compression 0x28fd48 for format B8G8R8A8_TYPELESS. d3d11.c:8530: Test failed: Got unexpected image size 2686324 for format B8G8R8A8_TYPELESS. d3d11.c:8532: Test failed: Got unexpected horizontal resolution 4316655 for format B8G8R8A8_TYPELESS. d3d11.c:8534: Test failed: Got unexpected vertical resolution 1 for format B8G8R8A8_TYPELESS. d3d11.c:8536: Test failed: Got unexpected used colour count 4330808 for format B8G8R8A8_TYPELESS. d3d11.c:8538: Test failed: Got unexpected important colour count 2686152 for format B8G8R8A8_TYPELESS. d3d11.c:8540: Test failed: Got unexpected colour masks 0x0028fcb8 0x735a25c5 0x00000000 for format B8G8R8A8_TYPELESS. d3d11.c:8543: Test failed: Got unexpected section 000020A4 for format B8G8R8A8_TYPELESS. d3d11.c:8544: Test failed: Got unexpected offset 4228276 for format B8G8R8A8_TYPELESS. d3d11.c:8502: Test failed: Got unexpected size 24 for format B8G8R8A8_UNORM_SRGB. d3d11.c:8516: Test failed: Got unexpected bits 00000000 for format B8G8R8A8_UNORM_SRGB. d3d11.c:8518: Test failed: Got unexpected size 2686104 for format B8G8R8A8_UNORM_SRGB. d3d11.c:8520: Test failed: Got unexpected width 9502720 for format B8G8R8A8_UNORM_SRGB. d3d11.c:8522: Test failed: Got unexpected height 9502720 for format B8G8R8A8_UNORM_SRGB. d3d11.c:8524: Test failed: Got unexpected plane count 0 for format B8G8R8A8_UNORM_SRGB. d3d11.c:8526: Test failed: Got unexpected bit count 0 for format B8G8R8A8_UNORM_SRGB. d3d11.c:8528: Test failed: Got unexpected compression 0x28fd48 for format B8G8R8A8_UNORM_SRGB. d3d11.c:8530: Test failed: Got unexpected image size 2686324 for format B8G8R8A8_UNORM_SRGB. d3d11.c:8532: Test failed: Got unexpected horizontal resolution 4316655 for format B8G8R8A8_UNORM_SRGB. d3d11.c:8534: Test failed: Got unexpected vertical resolution 1 for format B8G8R8A8_UNORM_SRGB. d3d11.c:8536: Test failed: Got unexpected used colour count 4330808 for format B8G8R8A8_UNORM_SRGB. d3d11.c:8538: Test failed: Got unexpected important colour count 2686152 for format B8G8R8A8_UNORM_SRGB. d3d11.c:8540: Test failed: Got unexpected colour masks 0x0028fcb8 0x735a25c5 0x00000000 for format B8G8R8A8_UNORM_SRGB. d3d11.c:8543: Test failed: Got unexpected section 000020A4 for format B8G8R8A8_UNORM_SRGB. d3d11.c:8544: Test failed: Got unexpected offset 4228276 for format B8G8R8A8_UNORM_SRGB.

1 0

Re: [PATCH] d3d11/tests: Extend GetDC() tests
by Henri Verbeet Sept. 6, 2016

Sept. 6, 2016

On 5 September 2016 at 21:45, Nikolay Sivov <nsivov(a)codeweavers.com> wrote: > + static const struct > + { > + const char *name; > + DXGI_FORMAT format; > + unsigned int bit_count; > + DWORD mask_r, mask_g, mask_b; > + BOOL getdc_supported; > + } > + testdata[] = > + { > + {"B8G8R8A8_UNORM", DXGI_FORMAT_B8G8R8A8_UNORM, 32, 0x00000000, 0x00000000, 0x00000000, TRUE }, > + {"B8G8R8A8_TYPELESS", DXGI_FORMAT_B8G8R8A8_TYPELESS, 32, 0x00000000, 0x00000000, 0x00000000, TRUE }, > + {"B8G8R8A8_UNORM_SRGB", DXGI_FORMAT_B8G8R8A8_UNORM, 32, 0x00000000, 0x00000000, 0x00000000, TRUE }, > + }; I guess this was inherited from the d3d9 test, but not all of these fields are very useful in the current form of the test. Do you intend to add more formats later?

2 2

Re: [PATCH v11 3/3] shell32: Implement FolderItems_get_Count.
by Marvin Sept. 6, 2016

Sept. 6, 2016

Hi, While running your changed tests on Windows, 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=25567 Your paranoid android. === build (build) === Make failed

1 0

Re: [PATCH v11 2/3] shell32: Implement FolderItems_Item.
by Marvin Sept. 6, 2016

Sept. 6, 2016

Hi, While running your changed tests on Windows, 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=25566 Your paranoid android. === build (build) === Make failed

1 0