etnaviv: mmu issue after end of address space reached?

[Wladimir J. van der Laan]

> <3>[  549.814025] etnaviv-gpu 130000.gpu: MMU fault status 0x00000002 <- happens almost immediately
> <3>[  549.819960] etnaviv-gpu 130000.gpu: MMU 0 fault addr 0xe8783040
> <3>[  549.825889] etnaviv-gpu 130000.gpu: MMU 1 fault addr 0x00000000
> <3>[  549.831817] etnaviv-gpu 130000.gpu: MMU 2 fault addr 0x00000000
> <3>[  549.837744] etnaviv-gpu 130000.gpu: MMU 3 fault addr 0x00000000

Okay I just tried to get the same while rendering in Mesa and it doesn't happen.

It reaches the end of the address space, sets last_iova back to 0, and just continues.

So the MMU fault is somehow specific to what I'm doing. Interesting.

> What I find curious, though, is that after the search presumably starts over at
> 0 it returns 0xe8783000 instead of an earlier address. For this reason
> last_iova is stuck near the end of the address space and the problem keeps
> repeating once it's been hit.

This does happen when rendering - it keeps dealing out iovas near the end of the
address space. But that seems harmless, though maybe causes some more MMU
flushes than necessary.

Wladimir