Linux 5.14 Supports Some Exciting Features With RISC-V

[uid313]

Which is good. Both "established standards" are pile of cr*p.
New approach would freshen things a bit.

A shit standard is still better than no standard at all. At least with x86 you can boot them using a generic kernel that works on all x86 systems with a standard boot loader like GRUB.

Market always like cheaper, unlimited and unencumbered products.
RISC-V has no historic baggage, patent ownerships nor licence red tapes.

Yes, which mans RISC-V will be used as an internal component of some bigger system, such as power management on a graphics card or microcontroller on a SSD, in order to reduce the costs to bring more profit to rich.

End-users like you and me won't get any benefit from RISC-V such as having more competition or having an open system without binary blobs and proprietary firmware.

Pure CPU is foundation of everything else. Which is coming along.

Things don't come from nowhere, and when someone wants to create a device they need more than just a CPU. With just a CPU it is useless for anything other than being a small part of a bigger system and used to reduce costs to make rich even richer.

The only interest any company would have in RISC-V is to avoid paying licenses so they can get lower their production costs to increase their profit margin.

[coder]

The only interest any company would have in RISC-V is to avoid paying licenses so they can get lower their production costs to increase their profit margin.

Can we please just focus on technical details and not take entirely unnecessary detours into macroeconomics, corporate governance, and wealth distribution?

Whether the benefits of a royalty-free ISA flow primarily to shareholders or consumers, it does open up the market to more competitors and allow for fully libre implementations. It's hard to argue those are bad things (at least, not for anyone other than the entrenched players).

[uid313]

Whether the benefits of a royalty-free ISA flow primarily to shareholders or consumers, it does open up the market to more competitors and allow for fully libre implementations. It's hard to argue those are bad things (at least, not for anyone other than the entrenched players).

Well if a royalty-free ISA only benefits shareholders then I don't care about it, then I might as well continue using x86 or ARM, they're established have good toolchains and support. In that case RISC-V would be a bad thing for me, because it would just be fragmentation and less mature toolchains with a smaller and less supported ecosystem.

A free, royalty-free, open ISA is only of interest to me, if it benefits the end-users.

[coder]

A free, royalty-free, open ISA is only of interest to me, if it benefits the end-users.

That's difficult both to quantify and predict. I'm guessing most of us think it will.

[Vistaus]

RISC-V is a pipe dream that will never be.

Tell that to the growing amount of embedded systems using RISC-V.

[…] and there are no manufacturers interested in bringing forth any devices, hardware or products with it.

This week in China a new RISC-V test board landed that will compete with ARM next year, so clearly there *are* manufacturers working on it, also because it's already used in a growing amount of embedded systems. Those boards don't produce themselves (yet).

Also, Samsung is currently using RISC-V: https://www.anandtech.com/show/15228...se-riscv-cores
Surely Samsung would qualify as a big manufacturer, no?

[uid313]

Tell that to the growing amount of embedded systems using RISC-V.

There are no growing amount of embedded systems using RISC-V as their primary processor, it's just used in some on-board component to reduce costs.

This week in China a new RISC-V test board landed that will compete with ARM next year, so clearly there *are* manufacturers working on it, also because it's already used in a growing amount of embedded systems. Those boards don't produce themselves (yet).

I don't know what these unnamed board of yours is. Compete with ARM? In what segment?
Not against the Cortex-A or Cortex-X cores, perhaps to replace some Cortex-M0 core as a means to reduce costs in order to bring a larger profit margin to the board of directors.

Also, Samsung is currently using RISC-V: https://www.anandtech.com/show/15228...se-riscv-cores
Surely Samsung would qualify as a big manufacturer, no?

Samsung have absolutely no plans to ever sell any laptop, phone or smartwatch with a RISC-V as its main CPU. Of course Samsung is looking at ways to further reduce costs, so they might have some little circuit here and there that they will replace with a RISC-V counterpart in order to save on license costs to create put more money in the pockets for the board of directors.

[Alexmitter]

RISC-V is a pipe dream that will never be.

Hmmm, then I wonder what "be" means to you as RISC-V already happened. Its a instruction set and its founders want to kill ARM for all kinds of usage where ARM does not bring any benefits over being an ISA, not liberate your workstation usage. It is not a replacement for your x86 CPU.

All boards and booted and initialized in their own quirky unique way, there is no standard for booting such as UEFI. There is no power management such as ACPI or sleep states.

That is wrong, RISC-V chips all use something called OpenSBI. OpenSBI is basically like Coreboot and like Coreboot supporting multiple payloads like SeaBIOS, Tianocore UEFI and uboot, OpenSBI does too. This is about as standardized as it gets.

There are no consumer demand for it, and there are no manufacturers interested in bringing forth any devices, hardware or products with it.

Someone who buys a Android Phone or a Chromebook does not care, those are not the customers for such a platform.

Any adoption of it would lead to fragmentation and incompatibility due to proprietary extensions.

You want to run a general purpose operating system, like GNU/Linux. Then for that, you do not want a micro controller, you want a CPU with a certain set of instructions and there is a class for that, its called RVGC, what it supports beyond that does not matter.

A pure CPU has little appeal these days values being placed on integrated platforms where there is expectations of GPU, sound, network, 5G and AI accelerators.

Those things are called IP blocks, and like in the ARM world, chip makers can license those and put them on their RISC-V Silicon. Beside the popular Mali GPUs, non of those common IP blocks are made by ARM.

[Alexmitter]

There are no growing amount of embedded systems using RISC-V as their primary processor, it's just used in some on-board component to reduce costs.

The embedded space usually is 10 years behind the consumer electronic space. Focusing on reliability rather then new features, also with Intel CPUs.
A still popular CPU for example is the AMD Geode that is a classic 90s Intel 486 clone.
NXP, a major ARM CPU producer in the embedded space prepares to release a RISC-V CPU based on its its popular i.MX ARM CPU range.

I don't know what these unnamed board of yours is. Compete with ARM? In what segment?
Not against the Cortex-A or Cortex-X cores, perhaps to replace some Cortex-M0 core as a means to reduce costs in order to bring a larger profit margin to the board of directors.

SiFive for example has various RISC-V IP Block cores that compete with Cortex-A. Similarly StarFive and T-Head, both making cores that well compete with modern Cortex class chips.

Samsung have absolutely no plans to ever sell any laptop, phone or smartwatch with a RISC-V as its main CPU. Of course Samsung is looking at ways to further reduce costs, so they might have some little circuit here and there that they will replace with a RISC-V counterpart in order to save on license costs to create put more money in the pockets for the board of directors.

You seem to have some serious industry insider knowledge.

[andresdju]

RISC-V is a pipe dream that will never be. All boards and booted and initialized in their own quirky unique way, there is no standard for booting such as UEFI.

Booting and initialization aren't part of a CPU microarchitecture. Also, neither have x86 motherboards a standard booting and initialization process: you need to port UEFI, Coreboot or whatever to your specific x86-64 SOC, motherboard chipset and other motherboard's specifics. That's why you can't just install Coreboot on any x86-64 before porting it to the board. For RISC-V it's the same, as with ARM: you have UEFI, Coreboot or U-Boot (which also implements the booting part of the EFI standard), you just need to port it to your specific board.

There is no power management such as ACPI or sleep states.

AFAIK, ACPI is just data on the firmware, nothing to do with the CPU architecture.

Any adoption of it would lead to fragmentation and incompatibility due to proprietary extensions.

i don't know of any CPU architecture more plagued with extensions than x86-64... and all of them proprietary. Intel and AMD just cross-license them. On RISC-V all standard extensions are open, and in the worst case proprietary extensions will be specific of one company (instead of specific of two US companies). I don't expect popular free (libre) distributions to require proprietary RISC-V extensions, so I'm not very worried about that.

A pure CPU has little appeal these days values being placed on integrated platforms where there is expectations of GPU, sound, network, 5G and AI accelerators.

Most x86-64 SOC's don't have GPU, sound, network, 5G and AI accelerator (most of them are at best in the chipsets, most often on specific chips). Many of the AMD SOC'S don't have _any_ of them. Furthermore, future RISC-V SOC's for portable devices will have what you are requesting, for example Allwinner D1 has display controller, video engine, camera inputs, sound, USB, ethernet... in addition to other typical embedded interfaces like UARTs, SDIO, PWM, GPIO, ADCs etc.

[coder]

There are no growing amount of embedded systems using RISC-V as their primary processor,

That's a bold statement. I'm not sure how you can say that, since a lot of chips don't advertise what cores they're using. RISC-V is already getting big in the microcontroller segment.

Compete with ARM? In what segment?
Not against the Cortex-A or Cortex-X cores, perhaps to replace some Cortex-M0 core

Do you have any idea how long it took ARM to get to the point where it could start to challenge x86? Do you think a new ISA comes out and just dominates overnight?