4 ms·
Apple chips famously have Imagination GPUs that have zero open-source support, a far ways off even from Nvidia. They are the absolute worst. That's before you
by stefan_ 8y ago
Apple chips famously have Imagination GPUs that have zero open-source support, a far ways off even from Nvidia. They are the absolute worst.
That's before you consider that all the stuff Apple did to that poor ARM probably means you can't even boot Linux into a terminal on it today.
- mepian 8y ago>Apple chips famously have Imagination GPUs Not anymore, Apple switched to in-house GPUs recently.
- stefan_ 8y agoThat makes it worse? The Imagination crap is shipped on Android phones, too, so there are at least vendor kernel drops to work from.
- craftyguy 8y agoThe vendor kernels are worthless, tainted, piles of crap. Vendors rarely, if ever, maintain them once they are posted publicly. The GPU still requires proprietary blobs to be even remotely useful.
- stefan_ 8y agoOf course. But they contain non-zero amounts of information on the chip so you can do a proper implementation or, much less ambitious, update it for newer kernel versions.
- craftyguy 8y agoNot in my experience. The code in the kernel is usually just a shim for loading the blobs that do the real work.
- stefan_ 8y agoAll the reverse-engineered GPU drivers we have now (freedreno, etnaviv, nouveau, lima) started out with intercepting kernel ioctls, being able to observe the blobs at work. Many of them at one point ran with the vendor kernel interface but open-source userspace. In many ways the kernel part of a GPU is arguably the least important part and it's the userland where the shader compiler sits and produces the magic, but if you start from nothing it's very useful to have a working kernel interface that initializes the hardware and accepts command streams to run.
- evilduck 8y agoThe odds of Apple releasing a development ARM board is probably roughly equal to that of them releasing complete documentation and linux kernel driver support to go along with it.
- rbirkby 8y agoApple's GPU is designed just down the road from Imaginations HQ. Guess where the designers previously worked...
- valarauca1 8y agoThis is marketing hype. Apple’s commits to the LLVM (where the shader gets compiled into GPU executable code) states the A11 GPU target is “PowerVR Rogue” ISA. Which is identical to how it compiles for the A10. So it’s still using foreign IP for the instruction set. Maybe Apple moved things around/optimized internals? Get better path optimization, or improve IPC? But it’s weird they didn’t add optimizations specifically for that A11. So an A10/A11 GPU just run the same opcodes.
- bronxbomber92 8y agoCitation for those commits?
- nsxwolf 8y agoIt's a completely locked down system - does it matter if the GPU isn't open source?
- stefan_ 8y agoMaybe I misunderstood "portable MAME cabinets, video transcoders, scene displays" and the comparison to the RPi but surely the intention is to run your own code on it from the bootloader up?
- nsxwolf 8y agoOk, but Apple devotes considerable resources to make all of that impossible. Why pick on the GPU specifically?