HyperZ: Errata & The Catalyst Command Stream
Following yesterday's article about Radeon Gallium3D HyperZ support defeating open-source developers, Jerome Glisse has clarified the situation after trying to make this code work properly for more than a half-year.
Jerome Glisse's blog post from this morning is entitled Hyperz, it was a long time ago. To condense several paragraphs into something more elegant: the open-source HyperZ problems are likely due to a R600+ GPU hardware bug. However, there's nothing cited as an errata within the public documentation and its unlikely any internal AMD Catalyst driver engineers would remember the details around the issue since the hardware in question is about five years old.
Due to hardware errata, the only sure way to tell how to control the hardware is by looking at the closed-source driver's command stream to see how the binary blob is programming the hardware. However, for the open-source Gallium3D driver to mirror the Catalyst driver's command stream as it pertains to HyperZ would mean re-writing much of the Gallium3D driver.
Jerome Glisse's blog post from this morning is entitled Hyperz, it was a long time ago. To condense several paragraphs into something more elegant: the open-source HyperZ problems are likely due to a R600+ GPU hardware bug. However, there's nothing cited as an errata within the public documentation and its unlikely any internal AMD Catalyst driver engineers would remember the details around the issue since the hardware in question is about five years old.
Due to hardware errata, the only sure way to tell how to control the hardware is by looking at the closed-source driver's command stream to see how the binary blob is programming the hardware. However, for the open-source Gallium3D driver to mirror the Catalyst driver's command stream as it pertains to HyperZ would mean re-writing much of the Gallium3D driver.
But no matter how much i followed the AMD documentation advice, i still run into issue. I went back to look at what fglrx was doing and of course i found several things that i believe was no were documented, such as never reset htile preloading if resetting same surface, or first depth clear can't be a fast clear because you need to initialize the htile surface. Maybe i just missread or missunderstand documentation i was provided and i apologize if so.Simply put, end-users and open-source Radeon driver developers are screwed right now if hoping to reliably use the performance-boosting HyperZ feature on modern Radeon graphics card. Though if you haven't upgraded from an old ATI Radeon graphics card, at least HyperZ works on the old GPUs.
In the end, from a register value point of view in each use case my patch now pretty much exactly match the register value fglrx uses. Yet on some specific use case i am still hitting lockup. So i am left with little option here, either i am missing a single bit somewhere (despite my automatic command stream comparison i might still miss thing). Or the order in which you do thing matter much more that what we believe ie you need to program some register in some specific order to avoid issues. I believe this is the issue i am left with, but trying to match fglrx order means huge overhaul of how r600g build its command stream.
So the fact is, in the end the closed source driver is the reference implementation that got all the informations in it. So looking at closed source driver command stream is always the saffest way to be sure to have all the informations. That's at least my opinion.
52 Comments