Re: 80186: [PATCH 2/2] dinput: Restore checks for DirectX 3 buffer size.

[Dan Kegel]

Hi Vitaliy,
could you check on win64?  I had a test failure there with your patch.
- Dan

On Thu, Oct 20, 2011 at 7:49 AM,   wrote:
> This is an experimental automated build and test service.
> Please feel free to ignore this email while we work the kinks out.
>
> For more info about this message, see http://wiki.winehq.org/BuildBot
>
> The Buildbot has detected a failed build on builder runtests-default-x86_64 while building Wine.
> Full details are available at: http://buildbot.kegel.com/builders/runtests-default-x86_64/builds/769 (though maybe not for long, as I'm still reinstalling the buildbot periodically while experimenting)
> BUILD FAILED: failed shell_3
>
> Errors:
> alarum: failed command was ../../../wine dinput_test.exe.so device.c
> device.c:117: Test failed: GetDeviceData() should have failed: 80040207
> device.c:120: Test failed: GetDeviceData() should have failed: 80040207
> device.c:117: Test failed: GetDeviceData() should have failed: 80040207
> device.c:120: Test failed: GetDeviceData() should have failed: 80040207
> device.c:117: Test failed: GetDeviceData() should have failed: 80040207
> device.c:120: Test failed: GetDeviceData() should have failed: 80040207
> device.c:117: Test failed: GetDeviceData() should have failed: 80040207
> device.c:120: Test failed: GetDeviceData() should have failed: 80040207
> make: *** [device.ok] Error 8

80186: [PATCH 2/2] dinput: Restore checks for DirectX 3 buffer size.

[Vitaliy Margolen]

Could be because of definition problems. Here's what I get compiling wine64:
../../../../wine.git/dlls/dinput/tests/device.c:117:5: warning: comparison 
is always false due to limited range of data type
../../../../wine.git/dlls/dinput/tests/device.c:120:5: warning: comparison 
is always false due to limited range of data type

HRESULT defined as "typedef LONG HRESULT;". But error code defined as:
#define DIERR_NOTBUFFERED 0x80040207L

Not sure what to do here. Test passed on all win64 bit platforms according 
to test bot. So unless we want to redefine all LONG constants in PSDK (and 
remove that last L) I'd say ignore this.

Vitaliy.

On 10/20/2011 09:46 AM, Dan Kegel wrote:
> Hi Vitaliy,
> could you check on win64?  I had a test failure there with your patch.
> - Dan
>
> On Thu, Oct 20, 2011 at 7:49 AM,  wrote:
> > This is an experimental automated build and test service.
> > Please feel free to ignore this email while we work the kinks out.
> >
> > For more info about this message, see http://wiki.winehq.org/BuildBot
> >
> > The Buildbot has detected a failed build on builder runtests-default-x86_64 while building Wine.
> > Full details are available at: http://buildbot.kegel.com/builders/runtests-default-x86_64/builds/769 (though maybe not for long, as I'm still reinstalling the buildbot periodically while experimenting)
> > BUILD FAILED: failed shell_3
> >
> > Errors:
> > alarum: failed command was ../../../wine dinput_test.exe.so device.c
> > device.c:117: Test failed: GetDeviceData() should have failed: 80040207
> > device.c:120: Test failed: GetDeviceData() should have failed: 80040207
> > device.c:117: Test failed: GetDeviceData() should have failed: 80040207
> > device.c:120: Test failed: GetDeviceData() should have failed: 80040207
> > device.c:117: Test failed: GetDeviceData() should have failed: 80040207
> > device.c:120: Test failed: GetDeviceData() should have failed: 80040207
> > device.c:117: Test failed: GetDeviceData() should have failed: 80040207
> > device.c:120: Test failed: GetDeviceData() should have failed: 80040207
> > make: *** [device.ok] Error 8

80186: [PATCH 2/2] dinput: Restore checks for DirectX 3 buffer size.

[Dan Kegel]

On Thu, Oct 20, 2011 at 5:15 PM, Vitaliy Margolen
 wrote:
> > > device.c:117: Test failed: GetDeviceData() should have failed: 80040207
> > > device.c:120: Test failed: GetDeviceData() should have failed: 80040207
> > > make: *** [device.ok] Error 8
>
> Could be because of definition problems. Here's what I get compiling wine64:
> ../../../../wine.git/dlls/dinput/tests/device.c:117:5: warning: comparison
> is always false due to limited range of data type
> ../../../../wine.git/dlls/dinput/tests/device.c:120:5: warning: comparison
> is always false due to limited range of data type
>
> HRESULT defined as "typedef LONG HRESULT;". But error code defined as:
> #define DIERR_NOTBUFFERED 0x80040207L
>
> Not sure what to do here. Test passed on all win64 bit platforms according
> to test bot. So unless we want to redefine all LONG constants in PSDK (and
> remove that last L) I'd say ignore this.

How can we ignore it?  Once we check it in, the 64 bit build will
start failing its tests :-(
- Dan

80186: [PATCH 2/2] dinput: Restore checks for DirectX 3 buffer size.

[Vitaliy Margolen]

On 10/20/2011 06:18 PM, Dan Kegel wrote:
> On Thu, Oct 20, 2011 at 5:15 PM, Vitaliy Margolen
>   wrote:
> > > > device.c:117: Test failed: GetDeviceData() should have failed: 80040207
> > > > device.c:120: Test failed: GetDeviceData() should have failed: 80040207
> > > > make: *** [device.ok] Error 8
> >
> > Could be because of definition problems. Here's what I get compiling wine64:
> > ../../../../wine.git/dlls/dinput/tests/device.c:117:5: warning: comparison
> > is always false due to limited range of data type
> > ../../../../wine.git/dlls/dinput/tests/device.c:120:5: warning: comparison
> > is always false due to limited range of data type
> >
> > HRESULT defined as "typedef LONG HRESULT;". But error code defined as:
> > #define DIERR_NOTBUFFERED 0x80040207L
> >
> > Not sure what to do here. Test passed on all win64 bit platforms according
> > to test bot. So unless we want to redefine all LONG constants in PSDK (and
> > remove that last L) I'd say ignore this.
>
> How can we ignore it?  Once we check it in, the 64 bit build will
> start failing its tests :-(

So you saying we should modify all long error constants in PSDK? I'm not 
sure what else can be done.

Vitaliy.

80186: [PATCH 2/2] dinput: Restore checks for DirectX 3 buffer size.

[Vitaliy Margolen]

On 10/20/2011 06:26 PM, Vitaliy Margolen wrote:
> On 10/20/2011 06:18 PM, Dan Kegel wrote:
> > On Thu, Oct 20, 2011 at 5:15 PM, Vitaliy Margolen
> >  wrote:
> > > > > device.c:117: Test failed: GetDeviceData() should have failed: 80040207
> > > > > device.c:120: Test failed: GetDeviceData() should have failed: 80040207
> > > > > make: *** [device.ok] Error 8
> > >
> > > Could be because of definition problems. Here's what I get compiling wine64:
> > > ../../../../wine.git/dlls/dinput/tests/device.c:117:5: warning: comparison
> > > is always false due to limited range of data type
> > > ../../../../wine.git/dlls/dinput/tests/device.c:120:5: warning: comparison
> > > is always false due to limited range of data type
> > >
> > > HRESULT defined as "typedef LONG HRESULT;". But error code defined as:
> > > #define DIERR_NOTBUFFERED 0x80040207L
> > >
> > > Not sure what to do here. Test passed on all win64 bit platforms according
> > > to test bot. So unless we want to redefine all LONG constants in PSDK (and
> > > remove that last L) I'd say ignore this.
> >
> > How can we ignore it? Once we check it in, the 64 bit build will
> > start failing its tests :-(
>
> So you saying we should modify all long error constants in PSDK? I'm not
> sure what else can be done.

Ok, casting error code to HRESULT seemed to work.

Vitaliy.

80186: [PATCH 2/2] dinput: Restore checks for DirectX 3 buffer size.

[Michael Stefaniuc]

Vitaliy Margolen wrote:
> On 10/20/2011 06:18 PM, Dan Kegel wrote:
> > On Thu, Oct 20, 2011 at 5:15 PM, Vitaliy Margolen
> >   wrote:
> > > > > device.c:117: Test failed: GetDeviceData() should have failed:
> > > > > 80040207
> > > > > device.c:120: Test failed: GetDeviceData() should have failed:
> > > > > 80040207
> > > > > make: *** [device.ok] Error 8
> > >
> > > Could be because of definition problems. Here's what I get compiling
> > > wine64:
> > > ../../../../wine.git/dlls/dinput/tests/device.c:117:5: warning:
> > > comparison
> > > is always false due to limited range of data type
> > > ../../../../wine.git/dlls/dinput/tests/device.c:120:5: warning:
> > > comparison
> > > is always false due to limited range of data type
> > >
> > > HRESULT defined as "typedef LONG HRESULT;". But error code defined as:
> > > #define DIERR_NOTBUFFERED 0x80040207L
> > >
> > > Not sure what to do here. Test passed on all win64 bit platforms
> > > according
> > > to test bot. So unless we want to redefine all LONG constants in PSDK
> > > (and
> > > remove that last L) I'd say ignore this.
> >
> > How can we ignore it?  Once we check it in, the 64 bit build will
> > start failing its tests :-(
>
> So you saying we should modify all long error constants in PSDK? I'm not
> sure what else can be done.

Yes, that's the right way. I got that as a janitorial task from
Alexandre during the last WineConf. One commit is in already
http://source.winehq.org/git/wine.git/?a=commit;h=0324fc57469f402102b677c4e3...

While the search and replace is trivial it uncovers interesting
challenges and opportunities in the Wine code... That's why I'm not done
 yet with this task.

bye
    michael