[p_l]

I suspect it's mainly the reduced maintenance and reduction of workload needed to support, especially with more platforms coming to be supported (not so long ago there was no ARM64 nvidia support, now they are shipping their own ARM64 servers!)

What really changed the situation is that Turing architecture GPUs bring new, more powerful management CPU, which has enough capacity to essentially run the OS-agnostic parts of driver that used to be provided as blob on linux.

[knotimpressed]

Am I correct in reading that as Turing architecture cards include a small CPU on the GPU board, running parts of the driver/other code?

[p_l]

In Turing microarchitecture, nVidia replaced their old "falcon" cpu with NV-RISCV RV64 chip, running various internal tasks.

"Open Drivers" from nVidia include different firmware that utilizes the new-found performance.

[matheusmoreira]

How well isolated is this secondary computer? Do we have reason to fear the proprietary software running on it?

[p_l]

As well isolated as anything else on the bus.

So you better actually use IOMMU

[stragies]

Ah, yes, the magical IOMMU controller, that everybody just assumes to be implemented perfectly across the board. I'm expecting this to be like Hyperthreading, where we find out 20 years later, that the feature was faulty/maybe_bugdoored since inception in many/most/all implementations.

Same thing with USB3/TB-controllers, NPUs, etc that everybody just expects to be perfectly implemented to spec, with flawless firmwares.

[p_l]

It's not perfect or anything, but it's usually a step up[1], and the funniest thing is that GPUs generally had less of ... "interesting" compute facilities to jump over from, just easier to access usually. My first 64 bit laptop, my first android smartphone, first few iPhones, had more MIPS32le cores with possible DMA access to memory than the main CPU cores, and that was just counting one component of many (the wifi chip).

Also, Hyperthreading wasn't itself faulty or "bugdoored". The tricks necessary to get high performance out of CPUs were, and then there was intel deciding to drop various good precautions in name of still higher single core performance.

Fortunately, after several years, IOMMU availability becomes more common (current laptop I'm writing this on has proper separate groups for every device it seems)

[1] There's always the OpenBSD of navel gazing about writing "secure" C code, becoming slowly obsolescent thanks to being behind in performance and features, and ultimately getting pwned because your C focus and not implementing "complex" features helping mitigate access results in pwnable SMTPd running as root.

[matheusmoreira]

> you better actually use IOMMU

Is this feature commonly present on PC hardware? I've only ever read about it in the context of smartphone security. I've also read that nvidia doesn't like this sort of thing because it allows virtualizing their cards which is supposed to be an "enterprise" feature.

[brendank310]

Relatively common nowadays. It used to be delineated as a feature in Intel chips as part of their vPro line, but I think it’s baked in. Generally an IOMMU is needed for performant PCI passthrough to VMs, and Windows uses it for DeviceGuard which tries to prevent DMA attacks.