Comment 78 for bug 201330

Okay, I showed bug 195051 to Alex and chatted with him a bit.

A lot of -ati's internals were written prior to having access to register documentation, so there's a lot of educated guesses in -ati that proved to be wrong once the docs were in hand, and lead to random crashes like this, and Alex believes this is the case with this bug. He did not have a workaround to suggest, and confirmed Travis' finding that the issue is not going to be limited to a specific chipset so pci id checks wouldn't be a reliable workaround. He has plans to go through and fix up all these mappings correctly, within the next 3-6 months. Unfortunately that rules out a 8.04.1 fix (and even if it could be achieved in that timeframe, the size of the rework would probably be beyond our bug-risk threshhold.)

Meanwhile, I'll make sure alex remains aware of the bug, and try to improve the reports upstream so it will be fixed for Intrepid. The work can be tracked at bug 195051 here, and at http://bugs.freedesktop.org/show_bug.cgi?id=15096 upstream (unless someone knows of a more appropriate upstream bug?) So, I'm resetting this to WONTFIX.