December 2015 - Wine-Devel - list.winehq.org

Re: [PATCH 1/2] ntdll/tests: Add SIMD exception test for floating point invalid operation fault (try 3).

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=18719 Your paranoid android. === wvistau64 (32 bit exception) === exception.c:485: Test failed: 42: Wrong exception address 003D0000/003D0001 exception: unhandled exception 80000004 at 003D0000 exception.c:485: Test failed: 42: Wrong exception address 77BAED97/003D0001 === w7u (32 bit exception) === exception.c:485: Test failed: 42: Wrong exception address 002F0000/002F0001 exception: unhandled exception 80000004 at 002F0000 exception.c:485: Test failed: 42: Wrong exception address 77CDE34F/002F0001 === w1064 (32 bit exception) === Failure running script in VM: network read timed out (Connect:AgentVersion.h:0/9)

Dec. 2, 2015

Re: loader/wine.inf.in: add 'HKLM\Software\Classes\Applications\iexplore.exe\shell\open\command' key

by Jacek Caban

Hi Austin, On 12/02/15 07:31, Austin English wrote: > > From a8309f37269bc374359e6ee473e66465551b867f Mon Sep 17 00:00:00 2001 > From: Austin English <austinenglish(a)gmail.com> > Date: Wed, 2 Dec 2015 01:27:15 -0500 > Subject: [PATCH] loader/wine.inf.in: add > 'HKLM\Software\Classes\Applications\iexplore.exe\shell\open\command' key > > Signed-off-by: Austin English <austinenglish(a)gmail.com> > --- > loader/wine.inf.in | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/loader/wine.inf.in b/loader/wine.inf.in I believe this should go to programs/iexplore/iexplore.inf. > index 58a8d1d..02b92f4 100644 > --- a/loader/wine.inf.in > +++ b/loader/wine.inf.in > @@ -245,6 +245,7 @@ HKCR,https\shell\open\ddeexec,"NoActivateHandler",2,"" > HKCR,https\shell\open\ddeexec\Application,,2,"IExplore" > HKCR,https\shell\open\ddeexec\Topic,,2,"WWW_OpenURL" > HKCR,mailto\shell\open\command,,2,"%11%\winebrowser %1" > +HKLM,Software\Classes\Applications\iexplore.exe\shell\open\command,,2,"""%16422%\Internet Explorer\iexplore.exe"" %1" Why don't you use HKCR here? Thanks, Jacek

Dec. 2, 2015

Re: [PATCH] ntdll: Fix valgrind heap realloc notifications.

by Alexandre Julliard

Thomas Faller <tfaller1(a)gmx.de> writes: > Valgrind can't handle a realloc, if the new size is zero. > Fix for bug https://bugs.winehq.org/show_bug.cgi?id=37646 I'd suggest fixing valgrind, or if not possible, using a size of 1. Your change seems like it would make notifications less useful in several cases. -- Alexandre Julliard julliard(a)winehq.org

Dec. 2, 2015

Re: Simplified local IP reordering patch for default routes

by Bruno Jesus

On Wed, Dec 2, 2015 at 5:56 AM, Jamie Taylor <Jamie.Taylor(a)pobox.com> wrote: > On Thu, Nov 26, 2015 at 11:00:01PM -0500, Jamie Taylor wrote: >> On Thu, Nov 26, 2015 at 08:11:20PM +0800, Bruno Jesus wrote: >> > Hi, Jamie and Joachim and list. This is a short version of a long >> > patch I'm working on, it orders the IP list with the default routes on >> > top and it supports more than one default route. It fixes the problem >> > on Linux and FreeBSD, could anyone also test it on OSX, please? >> >> I am away from my test setup until next week due to the Thanksgiving >> holiday in the USA, but I will do a test on OSX when I get back to it, > > Works for me on OSX. Thanks for the test, patch will be submitted after code freeze.

Dec. 2, 2015

Re: shlwapi/tests: Test NULL handle duplication in SHMapHandle()

by Olivier F. R. Dierick

Le mardi 01 décembre 2015 à 19:51 +0800, Bruno Jesus a écrit : > Related to https://bugs.winehq.org/show_bug.cgi?id=36838 > > Not sure when Olivier Dierick will be back to this so I'm sending the > test I made. > > Signed-off-by: Bruno Jesus <00cpxxx(a)gmail.com> Thank you. The shlwapi functions are undocumented in MSDN. This is were I found documentation for the SHMapHandle() function: http://www.geoffchappell.com/studies/windows/shell/shlwapi/api/shared/mapha… -- Olivier F. R. Dierick o.dierick(a)piezo-forte.be

Dec. 2, 2015

Re: Simplified local IP reordering patch for default routes

by Jamie Taylor

On Thu, Nov 26, 2015 at 11:00:01PM -0500, Jamie Taylor wrote: > On Thu, Nov 26, 2015 at 08:11:20PM +0800, Bruno Jesus wrote: > > Hi, Jamie and Joachim and list. This is a short version of a long > > patch I'm working on, it orders the IP list with the default routes on > > top and it supports more than one default route. It fixes the problem > > on Linux and FreeBSD, could anyone also test it on OSX, please? > > I am away from my test setup until next week due to the Thanksgiving > holiday in the USA, but I will do a test on OSX when I get back to it, Works for me on OSX. Jamie

Dec. 1, 2015

Re: Shell32: Check null handles before freeing

by Nikolay Sivov

On 01.12.2015 22:02, Olivier F. R. Dierick wrote: > Le mercredi 25 novembre 2015 à 11:11 +0300, Nikolay Sivov a écrit : >> On 25.11.2015 10:02, Olivier F. R. Dierick wrote: >>> Fix one of the error leak source that prevents the installer to succeed >>> (bug 36838). >>> >>> The original code did not check if the handles were null before freeing >>> them. For some reason they happen to be null and last error is set to >>> ERROR_INVALID_HANDLE. The game installer chokes at some point when the >>> last error is anything but zero and fails to complete the installation. >>> Checking if the handles are non-null before freeing them avoids this. >>> According to MSDN, it is not an error to have null values for those >>> handles. >>> >>> Signed-off-by: Olivier F. R. Dierick <o.dierick(a)piezo-forte.be> >>> --- >>> dlls/shell32/changenotify.c | 6 +++--- >>> 1 file changed, 3 insertions(+), 3 deletions(-) >>> >> >> We most likely have many more places that do same thing, it will need >> tests for SHFree() and CoTaskMemFree(). If it turns out error is not >> changed on NULLs, it's easier to return early from SHFree or >> CoTaskMemFree than fix all places where those are used. >> >> >> > > Hello, > > I sent a test for CoTaskMemFree() to the Wine test bot. The test passed > with 0 failures. Can someone confirm that the test is valid? Here is the > link to the test job: > https://testbot.winehq.org/JobDetails.pl?Key=18703 Test is valid, but todo_wine is not - it doesn't fail for me in Wine, without todo. > > Also, I put the test in a new file (dlls/ole32/tests/ifs.c). This file > contains only one test. Is it worth submission? In general I think testing last errors from OLE and COM methods/functions is not very useful, because it's usually designed to return specific error code directly, and set last error doesn't indicate failure in this case, but instead side effect from underlying code. For you specific case, like I said, I don't see CoTaskMemFree setting error on NULL input, so SHFree won't set it. If your patch still helps without SHFree input pointer checks, the problem is in shlwapi.SHFreeShared instead.

Dec. 1, 2015

Random err:winediag:nulldrv_CreateWindow errors related to windows and explorer

by Bruno Jesus

I believe that since code freeze started (or maybe 1.7.55) a random behavior was introduced related to these 2 messages: err:winediag:nulldrv_CreateWindow Application tried to create a window, but no driver could be loaded. err:winediag:nulldrv_CreateWindow The explorer process failed to start. Sometimes together there is also: err:ole:apartment_createwindowifneeded CreateWindow failed with error 0 I'm not seeing this alone. at November 23 there was a mention to it in bug https://bugs.winehq.org/show_bug.cgi?id=39384 and also in December 1 there was a mention at bug https://bugs.winehq.org/show_bug.cgi?id=39057 The issue seems to manifest more when the computer is heavily loaded (cpu/ram). I believe a regression test to be very hard since the issue is random, sorry. Is anyone else affected by this? I'm still affected with today's git. Best wishes, Bruno

Dec. 1, 2015

Re: Shell32: Check null handles before freeing

by Olivier F. R. Dierick

Le mercredi 25 novembre 2015 à 11:11 +0300, Nikolay Sivov a écrit : > On 25.11.2015 10:02, Olivier F. R. Dierick wrote: > > Fix one of the error leak source that prevents the installer to succeed > > (bug 36838). > > > > The original code did not check if the handles were null before freeing > > them. For some reason they happen to be null and last error is set to > > ERROR_INVALID_HANDLE. The game installer chokes at some point when the > > last error is anything but zero and fails to complete the installation. > > Checking if the handles are non-null before freeing them avoids this. > > According to MSDN, it is not an error to have null values for those > > handles. > > > > Signed-off-by: Olivier F. R. Dierick <o.dierick(a)piezo-forte.be> > > --- > > dlls/shell32/changenotify.c | 6 +++--- > > 1 file changed, 3 insertions(+), 3 deletions(-) > > > > We most likely have many more places that do same thing, it will need > tests for SHFree() and CoTaskMemFree(). If it turns out error is not > changed on NULLs, it's easier to return early from SHFree or > CoTaskMemFree than fix all places where those are used. > > > Hello, I sent a test for CoTaskMemFree() to the Wine test bot. The test passed with 0 failures. Can someone confirm that the test is valid? Here is the link to the test job: https://testbot.winehq.org/JobDetails.pl?Key=18703 Also, I put the test in a new file (dlls/ole32/tests/ifs.c). This file contains only one test. Is it worth submission? -- Olivier F. R. Dierick o.dierick(a)piezo-forte.be

Dec. 1, 2015

Re: dinput: Implement device property DIPROP_USERNAME.

by Sebastian Lackner

On 01.12.2015 14:54, Bernhard Übelacker wrote: > https://bugs.winehq.org/show_bug.cgi?id=39667 > > Probably same issue as in https://bugs.winehq.org/show_bug.cgi?id=12432 . > (Attached backtrace seems equal.) > > Steps to reproduce: > - start launcher > - "Configure Controller" > - leave dialog with "Cancel" > - crash > > MotoGP 3 demo launcher uses ConfigureDevices for the key mapping. > The crash seems to happen because the result of a > GetProperty(DIPROP_USERNAME) is used without checking. > > This affects just the launcher. The game starts without problems with > a default key mapping. > > Signed-off-by: Bernhard Übelacker <bernhardu(a)vr-web.de> > --- > dlls/dinput/device.c | 44 ++++++++++++++++++++++++++++++++++++++++++++ > dlls/dinput/device_private.h | 1 + > dlls/dinput8/tests/device.c | 29 +++++++++++++++++++++++++++-- > 3 files changed, 72 insertions(+), 2 deletions(-) > > diff --git a/dlls/dinput/device.c b/dlls/dinput/device.c > index e525f01..1095c4c 100644 > --- a/dlls/dinput/device.c > +++ b/dlls/dinput/device.c > @@ -783,6 +783,7 @@ HRESULT _set_action_map(LPDIRECTINPUTDEVICE8W iface, LPDIACTIONFORMATW lpdiaf, L > DIOBJECTDATAFORMAT *obj_df = NULL; > DIPROPDWORD dp; > DIPROPRANGE dpr; > + DIPROPSTRING dps; > WCHAR username[MAX_PATH]; > DWORD username_size = MAX_PATH; > int i, action = 0, num_actions = 0; > @@ -863,6 +864,19 @@ HRESULT _set_action_map(LPDIRECTINPUTDEVICE8W iface, LPDIACTIONFORMATW lpdiaf, L > else > lstrcpynW(username, lpszUserName, MAX_PATH); > > + dps.diph.dwSize = sizeof(dps); > + dps.diph.dwHeaderSize = sizeof(DIPROPHEADER); > + dps.diph.dwObj = 0; > + dps.diph.dwHow = DIPH_DEVICE; > + lstrcpynW(dps.wsz, username, sizeof(dps.wsz)); > + IDirectInputDevice8_SetProperty(iface, DIPROP_USERNAME, &dps.diph); > + > + if (dwFlags & DIDSAM_NOUSER) > + { > + HeapFree(GetProcessHeap(), 0, This->username); > + This->username = NULL; > + } When I understand it correctly, if DISAM_NOUSER is set, you set the username just to reset it manually again afterwards? Why not pass an empty string to IDirectInputDevice8_SetProperty? > + > /* Save the settings to disk */ > save_mapping_settings(iface, lpdiaf, username); > > @@ -1100,6 +1114,9 @@ ULONG WINAPI IDirectInputDevice2WImpl_Release(LPDIRECTINPUTDEVICE8W iface) > /* Free action mapping */ > HeapFree(GetProcessHeap(), 0, This->action_map); > > + /* Free username */ > + HeapFree(GetProcessHeap(), 0, This->username); > + > EnterCriticalSection( &This->dinput->crit ); > list_remove( &This->entry ); > LeaveCriticalSection( &This->dinput->crit ); > @@ -1251,6 +1268,17 @@ HRESULT WINAPI IDirectInputDevice2WImpl_GetProperty(LPDIRECTINPUTDEVICE8W iface, > TRACE("buffersize = %d\n", pd->dwData); > break; > } > + case (DWORD_PTR) DIPROP_USERNAME: > + { > + LPDIPROPSTRING ps = (LPDIPROPSTRING)pdiph; > + > + if (pdiph->dwSize != sizeof(DIPROPSTRING)) return DIERR_INVALIDPARAM; > + > + ps->wsz[0] = 0; > + if (This->username) > + lstrcpynW(ps->wsz, This->username, sizeof(ps->wsz)); The last argument is wrong, it should be a WCHAR count. > + break; > + } > case (DWORD_PTR) DIPROP_VIDPID: > FIXME("DIPROP_VIDPID not implemented\n"); > return DIERR_UNSUPPORTED; > @@ -1324,6 +1352,22 @@ HRESULT WINAPI IDirectInputDevice2WImpl_SetProperty( > LeaveCriticalSection(&This->crit); > break; > } > + case (DWORD_PTR) DIPROP_USERNAME: > + { > + LPCDIPROPSTRING ps = (LPCDIPROPSTRING)pdiph; > + > + if (pdiph->dwSize != sizeof(DIPROPSTRING)) return DIERR_INVALIDPARAM; > + > + if (!This->username) > + This->username = HeapAlloc(GetProcessHeap(), 0, sizeof(ps->wsz)); > + if (!This->username) > + return DIERR_OUTOFMEMORY; > + > + This->username[0] = 0; > + if (ps->wsz) > + lstrcpynW(This->username, ps->wsz, sizeof(ps->wsz)); Same here. Also, it could make sense to look at the actual string length, instead of allocating a fixed size buffer. > + break; > + } > default: > WARN("Unknown property %s\n", debugstr_guid(rguid)); > return DIERR_UNSUPPORTED; > diff --git a/dlls/dinput/device_private.h b/dlls/dinput/device_private.h > index 52bbec4..26ec7c2 100644 > --- a/dlls/dinput/device_private.h > +++ b/dlls/dinput/device_private.h > @@ -81,6 +81,7 @@ struct IDirectInputDeviceImpl > /* Action mapping */ > int num_actions; /* number of actions mapped */ > ActionMap *action_map; /* array of mappings */ > + WCHAR *username; /* set in SetActionMap */ > }; > > extern BOOL get_app_key(HKEY*, HKEY*) DECLSPEC_HIDDEN; > diff --git a/dlls/dinput8/tests/device.c b/dlls/dinput8/tests/device.c > index 6495559..af8667d 100644 > --- a/dlls/dinput8/tests/device.c > +++ b/dlls/dinput8/tests/device.c > @@ -223,8 +223,8 @@ static BOOL CALLBACK enumeration_callback(const DIDEVICEINSTANCEA *lpddi, IDirec > dps.wsz[0] = '\0'; > > hr = IDirectInputDevice_GetProperty(lpdid, DIPROP_USERNAME, &dps.diph); > - todo_wine ok (SUCCEEDED(hr), "GetProperty failed hr=%08x\n", hr); > - todo_wine ok (!lstrcmpW(usernameW, dps.wsz), "Username not set correctly expected=%s, got=%s\n", wine_dbgstr_wn(usernameW, -1), wine_dbgstr_wn(dps.wsz, -1)); > + ok (SUCCEEDED(hr), "GetProperty failed hr=%08x\n", hr); > + ok (!lstrcmpW(usernameW, dps.wsz), "Username not set correctly expected=%s, got=%s\n", wine_dbgstr_wn(usernameW, -1), wine_dbgstr_wn(dps.wsz, -1)); > > /* Test buffer size */ > memset(&dp, 0, sizeof(dp)); > @@ -275,6 +275,7 @@ static void test_action_mapping(void) > HINSTANCE hinst = GetModuleHandleA(NULL); > IDirectInput8A *pDI = NULL; > DIACTIONFORMATA af; > + DIPROPSTRING dps; > struct enum_data data = {pDI, &af, NULL, NULL, NULL, 0}; > HWND hwnd; > > @@ -342,6 +343,30 @@ static void test_action_mapping(void) > > af.dwDataSize = 4 * sizeof(actionMapping) / sizeof(actionMapping[0]); > af.dwNumActions = sizeof(actionMapping) / sizeof(actionMapping[0]); > + > + /* test DIDSAM_NOUSER */ > + dps.diph.dwSize = sizeof(dps); > + dps.diph.dwHeaderSize = sizeof(DIPROPHEADER); > + dps.diph.dwObj = 0; > + dps.diph.dwHow = DIPH_DEVICE; > + dps.wsz[0] = '\0'; > + > + hr = IDirectInputDevice_GetProperty(data.keyboard, DIPROP_USERNAME, &dps.diph); > + ok (SUCCEEDED(hr), "GetProperty failed hr=%08x\n", hr); > + ok (dps.wsz[0] != 0, "Expected any username, got=%s\n", wine_dbgstr_wn(dps.wsz, -1)); > + > + hr = IDirectInputDevice8_SetActionMap(data.keyboard, data.lpdiaf, NULL, DIDSAM_NOUSER); > + ok (SUCCEEDED(hr), "SetActionMap failed hr=%08x\n", hr); > + > + dps.diph.dwSize = sizeof(dps); > + dps.diph.dwHeaderSize = sizeof(DIPROPHEADER); > + dps.diph.dwObj = 0; > + dps.diph.dwHow = DIPH_DEVICE; > + dps.wsz[0] = '\0'; > + > + hr = IDirectInputDevice_GetProperty(data.keyboard, DIPROP_USERNAME, &dps.diph); > + ok (SUCCEEDED(hr), "GetProperty failed hr=%08x\n", hr); > + ok (dps.wsz[0] == 0, "Expected empty username, got=%s\n", wine_dbgstr_wn(dps.wsz, -1)); > } > > if (data.mouse != NULL) >

Dec. 1, 2015