[Bf-committers] Patch #3760 Regression in 2.44

[Early Ehlinger]

Hi Shaul,

Thanks for the pointer - I'll have to make a note of it for future 
bugs.  However, this is when the fix was put in - not when it was removed.

It looks like that happened in revision 6654, which has all references 
to G.vd removed:

http://projects.blender.org/plugins/scmsvn/viewcvs.php/trunk/blender/source/blender/src/renderwin.c?root=bf-blender&r1=6491&r2=6654

And when it made its return in revision 6839, it was now unprotected:

http://projects.blender.org/plugins/scmsvn/viewcvs.php/trunk/blender/source/blender/src/renderwin.c?root=bf-blender&r1=6835&r2=6839

So what's the consensus?  Does the test need to be re-added, or is there 
a better solution?

-- Early Ehlinger, President, ResPower, Inc.


Shaul Kedem wrote:
> Early,
>  Check this out:
> http://projects.blender.org/plugins/scmsvn/viewcvs.php/trunk/blender/source/blender/src/renderwin.c?root=bf-blender&r1=6219&r2=6491&pathrev=6491
>
> Shaul
>
> On 7/8/07, Ton Roosendaal <ton at blender.org> wrote:
>   
>> Hi,
>>
>> Yes that is a good test to add in Blender.
>>
>> Note that this error only happens in rendering commands within the UI
>> (not commandline), and with no 3D window present.
>>
>> (BTW; if this takes you a couple of hours, you might consider to get
>> help with gdb! NULL pointer errors always pop up as first, without any
>> effort from developer side...)
>>
>> -Ton-
>>
>> On 8 Jul, 2007, at 16:01, Early Ehlinger wrote:
>>
>>     
>>> Hi All,
>>>
>>> I just spent a couple of hours adding Blender_Trace statements (See
>>> Patch 6161) to hunt down a crash in Blender 2.44 for a customer.  It
>>> turns out that G.vd is NULL, and the check for G.vd being NULL was
>>> removed sometime between Jan '06 and Blender 2.44.
>>>
>>> My guess is this happened when the following snip was moved from
>>> BIF_do_render to do_render, but that's just a guess:
>>>
>>>             if(G.vd->lay & 0xFF000000) {
>>>                 G.scene->lay |= G.vd->lay;
>>>                 G.scene->r.scemode |= R_SINGLE_LAYER;
>>>             }
>>>             else G.scene->lay= G.vd->lay;
>>>
>>> Can anybody else confirm that this is a problem?  I'm waiting for my
>>> customer to give permission to use his .blend file, which reproduces it
>>> every time, but I suspect that Cory's script from Jan'06 will do the
>>> trick, too.
>>>
>>> In the meantime, I'll see if I can get current CVS going and submit a
>>> patch against it.
>>>
>>> -- ResPower Support
>>> _______________________________________________
>>> Bf-committers mailing list
>>> Bf-committers at blender.org
>>> http://lists.blender.org/mailman/listinfo/bf-committers
>>>
>>>
>>>       
>> ------------------------------------------------------------------------
>> --
>> Ton Roosendaal  Blender Foundation ton at blender.org
>> http://www.blender.org
>>
>> _______________________________________________
>> Bf-committers mailing list
>> Bf-committers at blender.org
>> http://lists.blender.org/mailman/listinfo/bf-committers
>>
>>     
> _______________________________________________
> Bf-committers mailing list
> Bf-committers at blender.org
> http://lists.blender.org/mailman/listinfo/bf-committers
>
>
>   

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.blender.org/pipermail/bf-committers/attachments/20070708/ffc43728/attachment.htm