March 2017 - Wine-devel - list.winehq.org

Re: [PATCH 2/2] comctl32/propsheet: Implement PSPCB_ADDREF/PSPCB_RELEASE notifications
by Dmitry Timoshkov 02 Mar '17

02 Mar '17

Nikolay Sivov <nsivov(a)codeweavers.com> wrote: > Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com> > --- > dlls/comctl32/propsheet.c | 41 +++++++++++++++++++++++++++++++++++------ > dlls/comctl32/tests/propsheet.c | 4 ---- > 2 files changed, 35 insertions(+), 10 deletions(-) And of course it's a pure coincidence that you immediately started to work on PSPCB_ADDREF/PSPCB_RELEASE once a patch for it has been added to wine-staging. I'd suggest to stop such ill-behaved practices, or at least give a credit to the source if you decided to "borrow" an idea. Or is this an indication that you just ran out of your own ideas to work on? :) -- Dmitry.

3 4

Re: [PATCH 1/2] mciavi32: Implement MCI_BREAK
by Andrew Eikum 02 Mar '17

02 Mar '17

On Wed, Mar 01, 2017 at 04:10:43AM -0300, Bruno Jesus wrote: > + wma->mci_break.flags = dwFlags; > + wma->mci_break.parms = *lpParms; The rest of this looks OK to me, but please guard access to wma's members with wma->cs. Andrew

1 0

Re: [v2 2/4] dinput: Handle username in EnumDevicesBySemantics.
by Andrew Eikum 02 Mar '17

02 Mar '17

Thanks for fixing these up, they look good to me. However this patch introduces a test failure: dinput.c:573: Test succeeded inside todo block: No device should be available after action mapping available=0 dinput.c:596: Test succeeded inside todo block: This user should own no devices owned=0 This is obviously fixed by your 4/4 patch, but we like to keep the tests clean in every commit. It's fine to squash the test changes into the code changes to create a 2-patch sequence with no test failures. Then I think it should be good to go in. Thanks, Andrew On Thu, Mar 02, 2017 at 12:11:24AM +0200, Jetro Jormalainen wrote: > EnumDevicesBySemantics enums only devices with given username when > DIEDBSFL_THISUSER is set and only unowned devices when > DIEDBSFL_AVAILABLEDEVICES is set. > > v2: Fixed problems noted by Andrew Eikum. > > Tested on Arch Linux and Rally Trophy Demo. > > Signed-off-by: Jetro Jormalainen <jje-wine(a)jv.jetro.fi> > --- > dlls/dinput/dinput_main.c | 111 +++++++++++++++++++++++++++++++++++++++------- > 1 file changed, 96 insertions(+), 15 deletions(-) > > diff --git a/dlls/dinput/dinput_main.c b/dlls/dinput/dinput_main.c > index 2bf9f38ba4..a59217472c 100644 > --- a/dlls/dinput/dinput_main.c > +++ b/dlls/dinput/dinput_main.c > @@ -923,6 +923,47 @@ static HRESULT WINAPI IDirectInput8WImpl_FindDevice(LPDIRECTINPUT8W iface, REFGU > return IDirectInput2WImpl_FindDevice( &This->IDirectInput7W_iface, rguid, pszName, pguidInstance ); > } > > +static BOOL should_enumerate_device(const WCHAR *username, DWORD dwFlags, > + struct list *device_players, REFGUID guid) > +{ > + BOOL should_enumerate = TRUE; > + struct DevicePlayer *device_player; > + > + /* Check if user owns this device */ > + if (dwFlags & DIEDBSFL_THISUSER && username && *username) > + { > + should_enumerate = FALSE; > + LIST_FOR_EACH_ENTRY(device_player, device_players, struct DevicePlayer, entry) > + { > + if (IsEqualGUID(&device_player->instance_guid, guid)) > + { > + if (*device_player->username && !lstrcmpW(username, device_player->username)) > + return TRUE; /* Device username matches */ > + break; > + } > + } > + } > + > + /* Check if this device is not owned by anyone */ > + if (dwFlags & DIEDBSFL_AVAILABLEDEVICES) { > + BOOL found = FALSE; > + should_enumerate = FALSE; > + LIST_FOR_EACH_ENTRY(device_player, device_players, struct DevicePlayer, entry) > + { > + if (IsEqualGUID(&device_player->instance_guid, guid)) > + { > + if (*device_player->username) > + found = TRUE; > + break; > + } > + } > + if (!found) > + return TRUE; /* Device does not have a username */ > + } > + > + return should_enumerate; > +} > + > static HRESULT WINAPI IDirectInput8AImpl_EnumDevicesBySemantics( > LPDIRECTINPUT8A iface, LPCSTR ptszUserName, LPDIACTIONFORMATA lpdiActionFormat, > LPDIENUMDEVICESBYSEMANTICSCBA lpCallback, > @@ -939,6 +980,7 @@ static HRESULT WINAPI IDirectInput8AImpl_EnumDevicesBySemantics( > int device_count = 0; > int remain; > DIDEVICEINSTANCEA *didevis = 0; > + WCHAR *username_w = 0; > > FIXME("(this=%p,%s,%p,%p,%p,%04x): semi-stub\n", This, debugstr_a(ptszUserName), lpdiActionFormat, > lpCallback, pvRef, dwFlags); > @@ -955,6 +997,14 @@ static HRESULT WINAPI IDirectInput8AImpl_EnumDevicesBySemantics( > > didevi.dwSize = sizeof(didevi); > > + if (ptszUserName) > + { > + int len = MultiByteToWideChar(CP_ACP, 0, ptszUserName, -1, 0, 0); > + > + username_w = HeapAlloc(GetProcessHeap(), 0, sizeof(WCHAR)*len); > + MultiByteToWideChar(CP_ACP, 0, ptszUserName, -1, username_w, len); > + } > + > /* Enumerate all the joysticks */ > for (i = 0; i < NB_DINPUT_DEVICES; i++) > { > @@ -968,7 +1018,8 @@ static HRESULT WINAPI IDirectInput8AImpl_EnumDevicesBySemantics( > > /* Default behavior is to enumerate attached game controllers */ > enumSuccess = dinput_devices[i]->enum_deviceA(DI8DEVCLASS_GAMECTRL, DIEDFL_ATTACHEDONLY | dwFlags, &didevi, This->dwVersion, j); > - if (enumSuccess == S_OK) > + if (enumSuccess == S_OK && > + should_enumerate_device(username_w, dwFlags, &This->device_players, &didevi.guidInstance)) > { > if (device_count++) > didevis = HeapReAlloc(GetProcessHeap(), 0, didevis, sizeof(DIDEVICEINSTANCEA)*device_count); > @@ -980,8 +1031,15 @@ static HRESULT WINAPI IDirectInput8AImpl_EnumDevicesBySemantics( > } > > remain = device_count; > + /* Add keyboard and mouse to remaining device count */ > if (!(dwFlags & DIEDBSFL_FORCEFEEDBACK)) > - remain += sizeof(guids)/sizeof(guids[0]); > + { > + for (i = 0; i < sizeof(guids) / sizeof(guids[0]); i++) > + { > + if (should_enumerate_device(username_w, dwFlags, &This->device_players, guids[i])) > + remain++; > + } > + } > > for (i = 0; i < device_count; i++) > { > @@ -991,26 +1049,38 @@ static HRESULT WINAPI IDirectInput8AImpl_EnumDevicesBySemantics( > if (lpCallback(&didevis[i], lpdid, callbackFlags, --remain, pvRef) == DIENUM_STOP) > { > HeapFree(GetProcessHeap(), 0, didevis); > + HeapFree(GetProcessHeap(), 0, username_w); > return DI_OK; > } > } > > HeapFree(GetProcessHeap(), 0, didevis); > > - if (dwFlags & DIEDBSFL_FORCEFEEDBACK) return DI_OK; > + if (dwFlags & DIEDBSFL_FORCEFEEDBACK) > + { > + HeapFree(GetProcessHeap(), 0, username_w); > + return DI_OK; > + } > > /* Enumerate keyboard and mouse */ > for(i=0; i < sizeof(guids)/sizeof(guids[0]); i++) > { > - callbackFlags = diactionformat_priorityA(lpdiActionFormat, actionMasks[i]); > + if (should_enumerate_device(username_w, dwFlags, &This->device_players, guids[i])) > + { > + callbackFlags = diactionformat_priorityA(lpdiActionFormat, actionMasks[i]); > > - IDirectInput_CreateDevice(iface, guids[i], &lpdid, NULL); > - IDirectInputDevice_GetDeviceInfo(lpdid, &didevi); > + IDirectInput_CreateDevice(iface, guids[i], &lpdid, NULL); > + IDirectInputDevice_GetDeviceInfo(lpdid, &didevi); > > - if (lpCallback(&didevi, lpdid, callbackFlags, sizeof(guids)/sizeof(guids[0]) - (i+1), pvRef) == DIENUM_STOP) > - return DI_OK; > + if (lpCallback(&didevi, lpdid, callbackFlags, --remain, pvRef) == DIENUM_STOP) > + { > + HeapFree(GetProcessHeap(), 0, username_w); > + return DI_OK; > + } > + } > } > > + HeapFree(GetProcessHeap(), 0, username_w); > return DI_OK; > } > > @@ -1049,7 +1119,8 @@ static HRESULT WINAPI IDirectInput8WImpl_EnumDevicesBySemantics( > > /* Default behavior is to enumerate attached game controllers */ > enumSuccess = dinput_devices[i]->enum_deviceW(DI8DEVCLASS_GAMECTRL, DIEDFL_ATTACHEDONLY | dwFlags, &didevi, This->dwVersion, j); > - if (enumSuccess == S_OK) > + if (enumSuccess == S_OK && > + should_enumerate_device(ptszUserName, dwFlags, &This->device_players, &didevi.guidInstance)) > { > if (device_count++) > didevis = HeapReAlloc(GetProcessHeap(), 0, didevis, sizeof(DIDEVICEINSTANCEW)*device_count); > @@ -1061,8 +1132,15 @@ static HRESULT WINAPI IDirectInput8WImpl_EnumDevicesBySemantics( > } > > remain = device_count; > + /* Add keyboard and mouse to remaining device count */ > if (!(dwFlags & DIEDBSFL_FORCEFEEDBACK)) > - remain += sizeof(guids)/sizeof(guids[0]); > + { > + for (i = 0; i < sizeof(guids) / sizeof(guids[0]); i++) > + { > + if (should_enumerate_device(ptszUserName, dwFlags, &This->device_players, guids[i])) > + remain++; > + } > + } > > for (i = 0; i < device_count; i++) > { > @@ -1083,13 +1161,16 @@ static HRESULT WINAPI IDirectInput8WImpl_EnumDevicesBySemantics( > /* Enumerate keyboard and mouse */ > for(i=0; i < sizeof(guids)/sizeof(guids[0]); i++) > { > - callbackFlags = diactionformat_priorityW(lpdiActionFormat, actionMasks[i]); > + if (should_enumerate_device(ptszUserName, dwFlags, &This->device_players, guids[i])) > + { > + callbackFlags = diactionformat_priorityW(lpdiActionFormat, actionMasks[i]); > > - IDirectInput_CreateDevice(iface, guids[i], &lpdid, NULL); > - IDirectInputDevice_GetDeviceInfo(lpdid, &didevi); > + IDirectInput_CreateDevice(iface, guids[i], &lpdid, NULL); > + IDirectInputDevice_GetDeviceInfo(lpdid, &didevi); > > - if (lpCallback(&didevi, lpdid, callbackFlags, sizeof(guids)/sizeof(guids[0]) - (i+1), pvRef) == DIENUM_STOP) > - return DI_OK; > + if (lpCallback(&didevi, lpdid, callbackFlags, --remain, pvRef) == DIENUM_STOP) > + return DI_OK; > + } > } > > return DI_OK; > -- > 2.11.0 > > >

1 0

Re: [PATCH v3 1/5] winebus.sys: Handle linux input event device via udev
by Sebastian Lackner 01 Mar '17

01 Mar '17

I haven't done a full review yet, but here are some first comments. On 01.03.2017 17:16, Aric Stewart wrote: > + subsystem = udev_device_get_subsystem(dev); > usbdev = udev_device_get_parent_with_subsystem_devtype(dev, "usb", "usb_device"); > if (usbdev) > { > + #ifdef HAS_PROPER_INPUT_HEADER > + const platform_vtbl *other_vtbl = NULL; > + DEVICE_OBJECT *dup = NULL; > + if (strcmp(subsystem, "hidraw") == 0) > + other_vtbl = &lnxev_vtbl; > + else if (strcmp(subsystem, "input") == 0) > + other_vtbl = &hidraw_vtbl; > + > + if (other_vtbl) > + dup = bus_enumerate_hid_devices(other_vtbl, check_same_device, dev); Shouldn't we have some preference of either hidraw or linux event devices? Otherwise this could lead to different results in each run. > } > > +DEVICE_OBJECT* bus_enumerate_hid_devices(const platform_vtbl *vtbl, enum_func function, void* context) > +{ > + struct pnp_device *dev; > + DEVICE_OBJECT *ret = NULL; > + > + TRACE("(%p)\n", vtbl); > + > + EnterCriticalSection(&device_list_cs); > + LIST_FOR_EACH_ENTRY(dev, &pnp_devset, struct pnp_device, entry) > + { > + struct device_extension *ext = (struct device_extension *)dev->device->DeviceExtension; > + if (ext->vtbl != vtbl) continue; > + if (function(dev->device, context) == 0) > + ret = dev->device; > + break; This is wrong and will always abort after the first function call. > + } > + LeaveCriticalSection(&device_list_cs); > + return ret; > +} > +

2 1

Re: [PATCH 3/6] wined3d: Handle raw thread group shared memory declarations.
by Sebastian Lackner 01 Mar '17

01 Mar '17

On 28.02.2017 14:42, Józef Kucia wrote: > + reg_maps->tgsm_count = reg_idx + 1; Wouldn't it be better to set it to max(reg_maps->tgsm_count, ...) to allow decreasing reg indices (even if not seen in practice yet)? Alternatively you could also get rid of this variable and update patch 6 to check the index against the capacity. Best regards, Sebastian

2 1

Re: [PATCH 2/2] winmm: Dispatch MCI_BREAK to the driver
by Sebastian Lackner 01 Mar '17

01 Mar '17

On 01.03.2017 08:10, Bruno Jesus wrote: > + TRACE("(%08x, %08X, vkey %04X\n", wDevID, dwFlags, lpParms->nVirtKey); Did you forget to add a ")" ?

2 1

Re: [PATCH 1/2] server: Added server side named pipe read and write implementation and use it for message mode pipes.
by Sebastian Lackner 01 Mar '17

01 Mar '17

On 28.02.2017 14:14, Jacek Caban wrote: > + LIST_FOR_EACH_ENTRY( message, &pipe_end->message_queue, struct pipe_message, entry ) > + avail += message->iosb->in_size - message->read_pos; > + reply_size = min( reply_size, avail ); > + > + if (avail) > + { > + message = LIST_ENTRY( list_head(&pipe_end->message_queue), struct pipe_message, entry ); > + reply_size = min( reply_size, message->iosb->in_size ); > + } Isn't there a problem with this size check? Lets assume we have two messages with 10 byte, for the first one the read_pos = 5. Then avail == 15, and reply_size is clamped to 10 - but it will attempt to copy 10 bytes starting position 5? > + > + if (!(buffer = set_reply_data_size(FIELD_OFFSET( FILE_PIPE_PEEK_BUFFER, Data[reply_size] )))) return; > + buffer->NamedPipeState = 0; /* FIXME */ > + buffer->ReadDataAvailable = avail; > + buffer->NumberOfMessages = 0; /* FIXME */ > + buffer->MessageLength = 0; /* FIXME */ > + if (avail) memcpy( buffer->Data, (const char *)message->iosb->in_data + message->read_pos, reply_size ); > +} > +

2 1

Re: [PATCH v3] oledb32: Correct length calculation in DataConvert for DBTYPE_STR
by Huw Davies 01 Mar '17

01 Mar '17

On Tue, Feb 28, 2017 at 11:17:29PM +0000, Alistair Leslie-Hughes wrote: > --- a/dlls/oledb32/convert.c > +++ b/dlls/oledb32/convert.c > @@ -816,23 +816,23 @@ static HRESULT WINAPI convert_DataConvert(IDataConvert* iface, > case DBTYPE_STR: > { > BSTR b; > - DBLENGTH bstr_len; > + DBLENGTH length; > INT bytes_to_copy; > - hr = IDataConvert_DataConvert(iface, src_type, DBTYPE_BSTR, src_len, &bstr_len, > + hr = IDataConvert_DataConvert(iface, src_type, DBTYPE_BSTR, src_len, &length, > src, &b, sizeof(BSTR), src_status, dst_status, > precision, scale, flags); > if(hr != S_OK) return hr; > - bstr_len = SysStringLen(b); > - *dst_len = bstr_len * sizeof(char); /* Doesn't include size for '\0' */ > + *dst_len = SysStringLen(b); /* Doesn't include size for '\0' */ You've gone back to confusing the length of a BSTR with the length of the char * str. We seem to be going round in circles here, so I've just sent in the correct fix. Please take a look at it to figure out where you were going wrong. Huw.

1 0

Re: [PATCH] d3dx9_36/tests: Fix a memory leak in test_createtext.
by Matteo Bruni 01 Mar '17

01 Mar '17

2017-02-27 7:52 GMT+01:00 Alex Henrie <alexhenrie24(a)gmail.com>: > diff --git a/dlls/d3dx9_36/tests/mesh.c b/dlls/d3dx9_36/tests/mesh.c > index 95dd06e466..64b7b8823c 100644 > --- a/dlls/d3dx9_36/tests/mesh.c > +++ b/dlls/d3dx9_36/tests/mesh.c > @@ -4340,8 +4340,7 @@ static void test_createtext(IDirect3DDevice9 *device, HDC hdc, const char *text, > if (!compute_text_mesh(&mesh, text, deviation, extrusion, otm.otmEMSquare, glyphs)) > { > skip("Couldn't create mesh\n"); > - d3dxmesh->lpVtbl->Release(d3dxmesh); > - return; > + goto error; > } > mesh.fvf = D3DFVF_XYZ | D3DFVF_NORMAL; I would go in the other direction and strip all the error checking branches and gotos from those tests. compute_text_mesh() returns FALSE only on allocation failure and that's not supposed to happen in the controlled conditions of the test. It's probably better to just fail and crash in those cases.

2 1