January 2016 - Wine-Devel - list.winehq.org

Re: [PATCH 2/2] shell32: Set the stream and storage attributes where needed.

by Marvin

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=20039 Your paranoid android. === w1064 (32 bit shlfolder) === shlfolder.c:1591: Test failed: IShellFolder_GetDisplayNameOf(NULL) failed! hr = 80070057

Jan. 28, 2016

Re: ndis.sys: add NdisAllocateSpinLock stub

by Alexandre Julliard

Austin English <austinenglish(a)gmail.com> writes: > From 5450fedec076c6d2f2c10c2e5cfa8f79578803a0 Mon Sep 17 00:00:00 2001 > From: Austin English <austinenglish(a)gmail.com> > Date: Tue, 5 Jan 2016 17:24:12 -0600 > Subject: [PATCH] ndis.sys: add NdisAllocateSpinLock stub > > Signed-off-by: Austin English <austinenglish(a)gmail.com> > --- > dlls/ndis.sys/main.c | 5 +++++ > include/ddk/ndis.h | 7 +++++++ > 2 files changed, 12 insertions(+) You forgot the spec file. -- Alexandre Julliard julliard(a)winehq.org

Jan. 28, 2016

Re: [1/3] scrrun: Implement filesys_DriveExists. (resend)

by Alexandre Julliard

Joachim Priesner <joachim.priesner(a)web.de> writes: > - return E_NOTIMPL; > + if (!pfExists) return E_POINTER; > + *pfExists = VARIANT_FALSE; > + len = SysStringLen(DriveSpec); > + if (len >= 1 && DriveSpec[0] >= 'A' && DriveSpec[0] <= 'Z' > + && (len < 2 || DriveSpec[1] == ':') > + && (len < 3 || DriveSpec[2] == '\\')) { Surely lower-case drives should work too. -- Alexandre Julliard julliard(a)winehq.org

Jan. 28, 2016

Re: [PATCH 4/7] ddraw/tests: Add tests for invalid render states.

by Henri Verbeet

On 28 January 2016 at 00:17, Józef Kucia <jkucia(a)codeweavers.com> wrote: > Signed-off-by: Józef Kucia <jkucia(a)codeweavers.com> > --- > Tested on Windows 7 and 10. > > A similar test crashes in ddraw1. > > I'm also not sure if this test is safe for ddraw7 as it seems to crash > with higher values. Yeah, it probably corrupts memory in ddraw7. What does GetRenderState() do?

Jan. 28, 2016

Re: [PATCH 6/7] wined3d: Validate stream_idx in wined3d_device_set_stream_source_freq().

by Henri Verbeet

On 28 January 2016 at 00:17, Józef Kucia <jkucia(a)codeweavers.com> wrote: > @@ -1412,6 +1412,12 @@ HRESULT CDECL wined3d_device_set_stream_source_freq(struct wined3d_device *devic > return WINED3DERR_INVALIDCALL; > } > > + if (stream_idx >= MAX_STREAMS) > + { > + FIXME("Stream index %u out of range.\n", stream_idx); > + return WINED3DERR_INVALIDCALL; > + } This would have to be a WARN instead of a FIXME, but in general we don't do this in wined3d unless we have to.

Jan. 28, 2016

Re: [PATCH 2/7] d3d9/tests: Add test for invalid render states.

by Henri Verbeet

On 28 January 2016 at 00:17, Józef Kucia <jkucia(a)codeweavers.com> wrote: > Signed-off-by: Józef Kucia <jkucia(a)codeweavers.com> > --- > A similar test crashes in d3d8. > --- > dlls/d3d9/tests/device.c | 40 ++++++++++++++++++++++++++++++++++++++++ > 1 file changed, 40 insertions(+) > This probably just corrupts memory on Windows, if e.g. SetTextureStageState() is any indication.

Jan. 28, 2016

Re: [PATCH 1/7] wined3d: Validate render state in wined3d_device_set_render_state().

by Henri Verbeet

On 28 January 2016 at 00:17, Józef Kucia <jkucia(a)codeweavers.com> wrote: > + if (state > WINEHIGHEST_RENDER_STATE) > + { > + WARN("Unhandled render state %#x.\n", state); > + return; > + } It may make sense to do this for wined3d_device_get_render_state() as well.

Jan. 28, 2016

Re: [PATCH v2] ntoskrnl.exe: Implement IoGetDeviceProperty for DevicePropertyPhysicalDeviceObjectName

by Alexandre Julliard

Aric Stewart <aric(a)codeweavers.com> writes: > + switch (device_property) > + { > + case DevicePropertyPhysicalDeviceObjectName: > + { > + OBJECT_NAME_INFORMATION *name = HeapAlloc(GetProcessHeap(), 0, buffer_length + (sizeof(OBJECT_NAME_INFORMATION))); > + > + status = NtQueryObject(device->Reserved, ObjectNameInformation, name, buffer_length + sizeof(OBJECT_NAME_INFORMATION), result_length); > + if (status == STATUS_SUCCESS) > + memcpy(property_buffer, name->Name.Buffer, name->Name.Length); > + HeapFree(GetProcessHeap(), 0, name); > + break; result_length is not handled correctly. -- Alexandre Julliard julliard(a)winehq.org

Jan. 28, 2016

Re: [PATCH] d3dx9: Signature isn't part of the DDS header

by Henri Verbeet

On 28 January 2016 at 00:20, Matteo Bruni <matteo.mystral(a)gmail.com> wrote: >> @@ -437,8 +442,8 @@ static HRESULT load_surface_from_dds(IDirect3DSurface9 *dst_surface, const PALET >> { >> UINT size; >> UINT src_pitch; >> - const struct dds_header *header = src_data; >> - const BYTE *pixels = (BYTE *)(header + 1); >> + const struct DDS *dds = src_data; >> + const BYTE *pixels = (BYTE *)(dds + 1); > > I guess you could instead make the src_data parameter BYTE * and > compute "pixels" as src_data + sizeof(struct dds) (or however you want > to call the new struct). It's mostly a matter of taste though so it's > fine to me either way. > If you define struct dds as struct dds { DWORD signature; struct dds_header header; BYTE data[1]; }; you can just change load_surface_from_dds() to take "const struct dds *dds" as parameter instead of src_data, and pass dds->data to D3DXLoadSurfaceFromMemory(). Or just pass dds->data to load_surface_from_dds() and rename that function, since it doesn't do all that much with the information in the header. It also means you can do things like "return FIELD_OFFSET(struct dds, data[data_size]);" in calculate_dds_file_size(). You may also want to consider just merging surface.c and volume.c into texture.c. On 27 January 2016 at 10:29, Alistair Leslie-Hughes <leslie_alistair(a)hotmail.com> wrote: > + memset(dds, 0, sizeof(*dds)); > + dds->signature = MAKEFOURCC('D','D','S',' '); > + dds->header.size = sizeof(struct dds_header); + dds->header.size = sizeof(dds->header);

Jan. 28, 2016

Re: [PATCH 4/7] ddraw/tests: Add tests for invalid render states.

by Józef Kucia

On Thu, Jan 28, 2016 at 12:53 AM, Marvin <testbot(a)winehq.org> wrote: > 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=20023 > > Your paranoid android. > > > === build (build) === > Missing build status line > The build timed out > > Actual test run: ddraw2 https://testbot.winehq.org/JobDetails.pl?Key=20031 ddraw4 https://testbot.winehq.org/JobDetails.pl?Key=20032 ddraw7 https://testbot.winehq.org/JobDetails.pl?Key=20033

Jan. 28, 2016