Re: pfLPointState problem

New Message Reply Date view Thread view Subject view Author view

Marcin Romaszewicz (marcin++at++asmodean.engr.sgi.com)
Tue, 7 Sep 1999 10:47:59 -0700 (PDT)

You are seeing a bug which was fixed in 2.2.4. A pfLPointState with an
intensity of 0 triggers an optimization which has an error in it. Until you
upgrade, the workaround is to use a very small positive intensity.

-- Marcin

>
>
>
> Hi,
> When I set the intensity of a pfLPointState to 0.0f, I am getting a segmentation
> fault in the draw process. The problem occurs using IRIX 6.5 and Performer 2.2.1
> on an O2. The light point state was originally created by the OpenFlight loader.
>
> The same code works without any problems using IRIX 6.5 and Performer 2.2.1 on
> an Onyx2 system.
>
> The following call stack was extracted from the core dump:
> drawX_LIGHT_POINTS_CvVv() ["gspoint.C":652]
> drawX_POINTS_CvVv(<stripped>) ["gspoint.C":390]
> pfDispList::pr_caseDL_DRAW_GSET_GSTATE() ["pfGeoSet.h":613]
> pfDispList::pr_drawFlat(<stripped>) ["pfDispList.C":3858]
> pfDispList::draw() ["pfDispList.C":660]
> pfChannel::pf_drawScene() ["pfChannel.C":2137]
> pfChannel::pf_draw() ["pfChannel.C":2112]
> pfDraw() ["pfProcess.C":6348]
> ccExecutionControl::_draw(this = 0x5802e480, channel = 0x586fa8c0)
> ["ccExecutionControl.C":870]
> ccExecutionControl::_drawCallback(channel = 0x586fa8c0, userData = 0x0)
> ["ccExecutionControl.C":1230]
> pfChannel::pf_callDrawFunc() ["pfChannel.C":2194]
> doDraw() ["pfProcess.C":6200]
> mpDraw() ["pfProcess.C":6890]
> pfConfig() ["pfProcess.C":2828]
> ccExecutionControl::configure(this = 0x5802e480, defaultConfigFile = 0x102fc000
> = "avMain.cfg", argc = 1, argv = 0x7fff2e94) ["ccExecutionControl.C":213]
> main(argc = 1, argv = 0x7fff2e94) ["avMain.C":60]
> __start(<stripped>) ["crt1text.s":177]
>
> I have found a report of the same problem in the archives (21 May, 1998), but
> unfortunately there was no reply stored. Is the O2 behaviour a known problem, or
> am I doing something wrong?
>
> In the meantime, I have worked around the problem by just setting the intensity
> to a very low non-zero value.
>
> Thanks,
> Ian Hawkes
>
>
> -----------------------------------------------------------------------
> List Archives, FAQ, FTP: http://www.sgi.com/software/performer/
> Submissions: info-performer++at++sgi.com
> Admin. requests: info-performer-request++at++sgi.com
>

New Message Reply Date view Thread view Subject view Author view

This archive was generated by hypermail 2.0b2 on Tue Sep 07 1999 - 10:48:01 PDT

This message has been cleansed for anti-spam protection. Replace '++at++' in any mail addresses with the '@' symbol.