News for hackers. Stuff that matters.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Monday September 04, 2023 @01:15AM (#37377325)
I'm bullish on RISC-V, but sort of bearish on SiFive. The last several systems I've used have been StarFive systems (which is distinct from SiFive.) They (SiFive) seem to be more of a custom IP shop, where you go when you want to spin some custom silicon but not pay ARM. But now that we've broken with China re: chip production, is there anyone on this side of the pacific I can get catalog parts from?

So... their cores are great, but if I'm only going to see them if I write a very large check to SiFive, they're sort of irrelevant to my world. I would love to have SiFive IP that doesn't use Rocket/Scala/Chisel and does produce catalog parts.

That could be just me. I enjoy getting my hands dirty.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by photonbeam on Monday September 04, 2023 @10:48AM (#37380988)
You normally only see ARM when you cut a big check too, they sell IP that others get fab’d
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Monday September 04, 2023 @01:19PM (#37382600)
Well... our ARM cores use more standard tooling. I sort of detest System Verilog, but I detest it a lot less than Rocket/Chisel. And as annoying as it is to write a big check to ARM, at least you can hire people who know how to use your IP.

It probably wasn't obvious from my earlier comment, but I was comparing SiFive to ARM and then I was comparing SiFive with other RISC-V IP vendors (and open source projects.)

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by fidotron on Monday September 04, 2023 @11:11AM (#37381226)
I agree with this, and suspect that if SiFive did get big enough they would start creating patented extensions as a defensive strategy.

Personally I'm more curious about RISC-V in tiny microcontrollers, and were I guiding semiconductor strategy I'd be way more concerned about ability to build embedded systems than GPU farms. The ecosystem to do the former can evolve into the latter, but skipping to the end is just a route to burn absolutely billions.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Monday September 04, 2023 @11:25AM (#37381388)
FWIW, after listening to Chuck Peddle's CHM Oral History interview, I sort of realized I didn't need the ARM ecosystem and just designed my own 6-bit microcontroller. Granted, I'm only doing some very simple control applications and fabbing at 800nm, so it's not going to be competing with an i7 or an A72 anytime soon.

IIRC, the RISC-V spec includes an "embedded profile" that specifies a CPU with 16 32-bit registers, so that should make RV32 even smaller than it would be otherwise. SiFive designed a dog-simple 2-stage pipeline 32 bit core that could even run without I-cache. I mean, it's not a speed demon, but it was implementable in many fewer gates than their earlier E cores.

I worked at SiFive for a year and couldn't make sense of their product strategy. But maybe that's a me problem and not a SiFive problem.

I really do sort of wonder what they're doing with 175m in Series F funding. Where do you go from there?

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by fidotron on Monday September 04, 2023 @11:33AM (#37381474)
I don’t think it’s you.

The only true thing that could be done is to have some top secret insight into CPU design optimization and apply it to RISC-V guiding the actual open spec along with you as necessary. Which is close to what Apple appear to have done with ARM64, but then Apple did hire/acquire an absolutely incredible team to get them there, and have the money to reward them.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Tuesday September 05, 2023 @02:44PM (#37395985)
Well... all the tech committees in the foundation report up through one of the SiFive principals, so if you want to change the spec, it at least appears SiFive has veto power.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by photonbeam on Monday September 04, 2023 @11:33AM (#37381471)
Presumably lots of runway for salaries until next funding round or exit event
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Tuesday September 05, 2023 @03:05PM (#37396266)
Intel publicly walked away from a $2B offer for SiFive, which means that at least one large chip manufacturer does not believe their IP is worth more than $2B. I'm not sure if the Round F post-money valuation was made public, but if it's more than $2B, SiFive will need to have a solid story for why a potential buyer should pay more than what Intel didn't want to pay. And if they're going to IPO, it would be helpful to have a reference customer for newer designs which seem to be on par with ARM offerings from four years ago.

It's certainly not impossible, but it seems a tough row to hoe.

But the cool part of this is that the ISA and many supporting packages (rocket, boom, etc.) are "open" and "free as in beer." Even if SiFive did fail, an IP investment in RISC-V would not necessarily have been a bad move. Which is why I said I'm fairly bullish on RISC-V (cause even if SiFive fails, customers existing investment in the ISA is not lost) but bearish on SiFive (because I fear their existing valuation may be too inflated.)

I'm not trying to tell anyone their SiFive baby is ugly. I'm not saying designing a RISC-V solution is a bad move. I'm just saying from my perspective, with incomplete information about what's going on inside SiFive, it looks like they have a difficult road ahead. I would POSITIVELY WELCOME any information regarding reference customers or catalog parts which might imply there's a large customer who's signed on the dotted line to pay SiFive a royalty for using their designs.

Full disclosure: I own no SIFIVE stock (or options) nor do I own stock in any company that competes in the RISC-V design space (Esperanto, Andes, Sandisk/WD, etc.) I still have stock in TXN though, but I haven't heard that they're deep in RISC-V. I am a former employee of SiFive, but was not employed in a management position; don't take my comments as being based on "inside information."

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by brucehoult on Tuesday September 05, 2023 @09:19PM (#37400213)
Intel walked away? It was my impression the Intel offer was unsolicited and it was SiFive who said "nah, not interested".
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by imtringued on Monday September 04, 2023 @01:01PM (#37382374)
> SiFive designed a dog-simple 2-stage pipeline 32 bit core that could even run without I-cache. I mean, it's not a speed demon, but it was implementable in many fewer gates than their earlier E cores.

How about SERV... That core can be configured to not use a register file (everything stored in BRAM).

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by brucehoult on Monday September 04, 2023 @07:55PM (#37386343)
SiFive's 2-stage pipeline E20 (and Arm's Cortex M0+) still execute 1 instruction per clock cycle on straight line code -- they just take an extra cycle on any taken branch .. doesn't matter whether conditional, unconditional, call, return.

SERV takes 32 clock cycles for most instructions, 64 cycles for some such as shifts and branches.

Yes, there are some things SERV is fast enough for, but E20 is not all that much slower than 5-stage pipe plus branch prediction E31 -- E20 does 1.2 DMIPS/MHz compared to E31 1.6 (and Arm Cortex M0+ 0.95, M3/M4 1.25)

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by Pet_Ant on Tuesday September 05, 2023 @10:03AM (#37391786)
> BRAM

Never heard of this despite being casually interested in hardware, it is apparently this?:

https://nandland.com/lesson-15-what-is-a-block-ram-bram/

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Tuesday September 05, 2023 @03:10PM (#37396337)
Ha! Yes! The serial RV32 implementation! That was Olof Kindgren's design, right? It's another great design at one corner of the design space. I mentioned the SiFive core since that was the one I was most familiar with. I've heard Olaf speak at conferences. Smart guy.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by taway32r41 on Monday September 04, 2023 @04:19PM (#37384574)
Here's a class at MIT that takes undergraduates who don't even know verilog and at then end of the class they have a mostly passable RV32I. Admittedly pushing that through VLSI CAD will not be push-button easy but many universities also have an undergraduate tape-out class, including MIT...

https://ocw.mit.edu/courses/6-004-computation-structures-spr...

Via MPW and assembly service one can easily have that die manufactured, packaged and mounted on custom PCBs for under $50K, a.k.a. much less than you'll pay in salary to have it designed, fabricated and tested. Even at non-American labor rates.

In 2023 producing a computing core, albeit not a state-of-the-art one, is just not the moat it once was...

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Tuesday September 05, 2023 @02:42PM (#37395970)
Yes. I saw that too. It turns out that "building a chip" is different than "building a chip that meets specfic requirements we got from the customer."

I'm not saying "building a RV32x is impossible," I'm saying there are people with existing tool-chains that favour Verilog or SystemVerilog (or even VHDL, though I don't know of them personally.) And telling them "no, you should use Scala / Chisel / Firrtl to model the features you want to add to the system in order to meet customer requirements" is kind of a hard sell for many customers.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by taway32r41 on Wednesday September 06, 2023 @02:18PM (#37408934)
I definitely agree with the primary point, "building a chip that meets specfic requirements we got from the customer" is not easy and is what matters.

However, RISCV cores abound. In pretty much any HDL known to man with varying design trade-offs and capabilities. It's extremely difficult to differentiate at the RTL level at this time. Not impossible, but it would be a significant investment, which is I guess SiFive's business model. Sell IP at prices cheaper than that.

Here is a high quality, well documented, SystemVerilog version intended for embedded applications that I know has been included in multiple ASIC and FPGA designs successfully.

https://github.com/lowRISC/ibex

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by brucehoult on Monday September 04, 2023 @07:46PM (#37386292)
> couldn't make sense of their product strategy. But maybe that's a me problem

Well, you drove a Bambino on a 60 miles each way commute over a huge hill, so there's something questionable going on there :p

I can't see much wrong with the "start with microcontrollers, work your way up to single-issue Linux-capable, then dual-issue in-order, then small then bigger and bigger OoO" strategy.

What else can you do? Try to take on Xeon day #1?

On the other hand, sure, you can ask a lot of questions about how you take that to the market, and who the customers will be.

SiFive early on obviously thought there were a lot of people out there who really wanted to make little custom chips for this and that but didn't have the skills, and SiFive (via buying OpenSilicon) could provide a one stop shop. And even provide a lot of automation around that, in the SiFive Chip Designer -- there used to be a web page (similar to SiFive Core Designer) but it's long gone, but see e.g. this Yunsup's presentation https://www.youtube.com/watch?v=pxd93jb1OAk

It seems no one really wanted to do that -- or at least SiFive didn't find them. So they got rid of OpenSilicon again (at a HUGE profit) and seem to be just selling CPU core IP to people who were already making chips, had all the people and knowledge to do that, but just wanted to change away from Arm or Tensilica or 8051 or whatever.

> what they're doing with 175m in Series F funding. Where do you go from there?

That's still absolute chickenfeed compared to what Intel or AMD or probably Arm spend on designing a single core.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Tuesday September 05, 2023 @02:38PM (#37395922)
Hey Bruce!

I'm talking more about how they thought they were going to interact with customers w/ a design language no one seemed to want to use. I think we agree on that; NO ONE seemed to be asking for Rocket / Scala / Chisel / Firrtl and it seemed odd that every time a potential customer asked for Verilog, they kept pushing Chisel.

But now that they're not pushing the core designer, how are they justifying their multiple? They're no longer a SaaS play, they're an IP design play which makes it a lot more difficult to hit the rule of 40.

But Intel famously passed on the $2B offer, which means they have to do something to convince someone out there they ARE worth whatever their post-money, round F valuation is lest the round C investors lose money and the round A investors not make as much money as they had hoped.

And yes $175m is less than ARM spends to design a core, but that's not what the $175m round F was for. Or at least I hope it's not what it's for, 'cause if they did sink it into designing a core that's just now on par with ARM's a72 (or was the recent goodness supposed to be on par with the a77?) from 4 years ago, what's the motivation for using a RISC-V core instead of an ARM core?

"We're as good as ARM was 4 years ago" isn't a great marketing slogan.

Your information about the details of the OpenSilicon spin-off must be better than mine. "Huge" profit is not what I heard. But neither of us were in San Mateo when that happened.

I would LOVE to be able to buy a part that lets me test any of the recent cores (I mean for less than $100,000.) Or I would love to hear about any customers using the new cores. It shouldn't be that hard to find a reference customer.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by panick21_ on Monday September 04, 2023 @06:11PM (#37385618)
RISC-V will not do X and then Y. RISC-V is going in all directions at the same time. People are pushing to the smallest micro controlers, to data centers and to niche applications like space all at the same time.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by codedokode on Monday September 04, 2023 @08:29PM (#37386535)
I see no value in targeting microcontrollers. They usually do not require high speed so can be designed by literally anyone with decent result.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by bmitc on Monday September 04, 2023 @11:05PM (#37387384)
Power usage and efficiency are important there.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by hajile on Tuesday September 05, 2023 @02:52PM (#37396082)
The ability to run your code many places instead of learning many proprietary ISAs is its own benefit.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Tuesday September 05, 2023 @03:13PM (#37396375)
Which is why I designed my own controller instead of using an RV32E.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by rwmj on Monday September 04, 2023 @11:15AM (#37381265)
They seem to be positioning themselves as the "RISC-V Arm" equivalent. Of course there are significant differences, like they don't completely control the architecture (as Arm does). But Arm have made a decent business out of it, so SiFive should be able to do the same.

By the way StarFive were SiFive China until SiFive divested them (the split was a bit acrimonious).

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by qwytw on Monday September 04, 2023 @11:30AM (#37381434)
> But Arm have made a decent business

They are making peanuts compared pretty much all companies that actually make chips. Their IP seems to be worth much more than the business.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by rwmj on Monday September 04, 2023 @12:57PM (#37382329)
Just because Arm isn't huge, it's still not nothing. They're about to go into a floatation to try to raise $50-55 billion, which is not too shabby! (Even if it's down from the original target of over $60b).
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by qwytw on Monday September 04, 2023 @02:15PM (#37383270)
True (assuming the stock price doesen't crash immediately after the IPO) but still it's not really justifiable based on their current financial figures or future growth prospects.

I'm not even sure what the valuation this valuation might even be based on (which admittedly applies to a bunch of other companies which actually end up doing quite well). They might conceivably be worth 100 P/E as an acquisition but an IPO would only make it less likely?

$50-55 billion would mean that ARM is more overvalued than Nvidia and Nvidia actually has a realistic chance to grow into that valuation.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by brucehoult on Monday September 04, 2023 @08:17PM (#37386467)
"Try" is doing a lot there.

I can't see how Arm is even worth the $32 billion SoftBank paid for them now. RISC-V didn't exist outside of Berkeley in 2016 when SoftBank bought Arm, but now it's becoming a serious rival and especially a serious limitation on how much Arm can raise prices.

Arm have around 2.7 billion in annual sales with $500m or $600m of profit.

That makes $32b a P/E ratio of 50 to 60, and $50b a P/E ratio of 80 to 100.

In simple terms, if you pay $50 billion to buy Arm, and they make the same profit every year as they are making now, it will take 80 to 100 years for you to make your investment back.

It's like putting your money into a bank paying 1% interest, but a lot riskier.

Such P/E ratios can only be justified by the company and its profits growing rapidly.

Arm's revenue actually fell 1% from 2022 to 2023. The last four years have been roughly 2.0, 2.7, 2.75, 2.7.

Is that a peak, and it's only downhill from now on, because of RISC-V (or otherwise). Or will they start to grow again?

Either way, it sure doesn't look like the very high annual growth rate needed to justify a price of $30b let alone $50b.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Tuesday September 05, 2023 @03:39PM (#37396705)
> In simple terms, if you pay $50 billion to buy Arm, and they make the same profit every year as they are making now, it will take 80 to 100 years for you to make your investment back.

But that's not how you make money by purchasing companies. You make money by buying the company low and then selling the company high. So if you bought ARM for $50B and sold it for $70B in 5 years, you would make $20B on an investment of $50B

But I'm certainly not in the position to guess how much ARM would appreciate over the next 5 years. I pulled that example out of a hat (or some other, less polite to mention part of my anatomy.)

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by brucehoult on Tuesday September 05, 2023 @09:29PM (#37400302)
Sure, that's the principle, but the question is: where do you find the Bigger Fool?

Fundamentals count for something, no?

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by fidotron on Monday September 04, 2023 @02:27PM (#37383375)
It is worth considering why you expect anyone else could be more successful.

On the one hand you have ARM investors wanting them to extract more revenue, and the ARM customers do not want to pay more or have more restrictive licenses. They are in a bit of a nasty equilibrium but one that will hit all IP companies eventually.

An old colleague of mine has got into the idea businesses are now going to put the big squeeze on all suppliers that count as part of gross margins, which would include ARM, while essentially ignoring operational costs. The implication would be companies looking to use RISC-V would be more inclined to pay chip design houses on a contract basis more upfront than have anything including royalties. I dislike the idea but think he is right, and that the future of electronics is about to go through the same sea change that software did with open source and the emergence of SaaS.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by rbanffy on Tuesday September 05, 2023 @07:24AM (#37390286)
> ARM is making peanuts

ARM is a fabless pure play IP house. The big bucks are not in IP design but in manufacture.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Tuesday September 05, 2023 @03:40PM (#37396724)
Yeah, but it's also where the big costs are.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by rbanffy on Tuesday September 05, 2023 @06:11PM (#37398476)
It's big bucks coming in and even bigger bucks coming out.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by vintermann on Monday September 04, 2023 @11:38AM (#37381525)
> the split was a bit acrimonious

That's another thing they have in common with Arm.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by Fordec on Monday September 04, 2023 @12:35PM (#37382095)
If RISC-V is just their platform to provide a different company to license IP from, I'm just sticking to Arm.

RISC-V is only interesting as a platform for a different business model.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by rwmj on Monday September 04, 2023 @12:50PM (#37382233)
For sure, but I guess you're not in SiFive's target market. You're not a customer of Arm either, unless you are licensing chip designs from Arm. Someone who merely buys Arm chips is a customer of the company that licenses from Arm, not of Arm.

Nevertheless it seems plausible there are companies that would like to license a RISC-V design (rather than implement their own or use one of the slower free cores) and might therefore be a SiFive customer. I'm not saying it's going to be a huge market, probably a much smaller one even than Arm, but it's a possible one.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Tuesday September 05, 2023 @03:31PM (#37396588)
Very true. I'm not in SiFive's target market. Though I do use M0 and M4 cores fairly frequently. But I'm not a direct customer of ARM, though. I think that's what you were implying.

I think part of what I'm getting at is it's TOUGH to get people to adopt new IP. And it's TOUGH to be all things to all people so you can compete in the low end against M0s and MSP430s, in the mid range against A77s and in the high end against i9's.

That being said, if SiFive fabbed a catalog part with one or two E20s at a reasonable cost, I have an immediate need of 200 of them. I only wish there were another 500 of me with a similar need so there was a chance we could get it fabbed at a decent price.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by brucehoult on Tuesday September 05, 2023 @09:28PM (#37400284)
> if SiFive fabbed a catalog part with one or two E20s at a reasonable cost

Why would they do that? That's not the business SiFive is in. Arm doesn't do that -- they leave that to people such as the Raspberry Pi Foundation with the RP2040.

Perhaps (if you insist on a SiFive core) you could look at something like the BL602? $0.99 here.

https://pine64.com/product/pineseed-bl602-wifi-ble5-soc/

If you're willing to use non-SiFive cores (and why not), the CH32V003 is pretty awesome for $0.10 (8 pin) to $0.14 (20 pin), assuming 2K RAM and 16K flash is enough for your purposes.

But I don't know what is "a decent price" for you.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by crote on Monday September 04, 2023 @12:57PM (#37382321)
The biggest benefit of RISC-V is that you are not married to a single IP vendor.

If ARM decides to change their license terms, you are screwed. If SiFive decides to change their license terms, you could just switch to a competing IP vendor without any changes to your software ecosystem.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Monday September 04, 2023 @01:15PM (#37382551)
ARM seems to be half-way moving in this direction. They allowed "royalty free" licenses to educational institutions for some of their low end M0/3/4 cores. But yeah, I think you're right. The cool thing about RISC-V is you can just pick up the ISA and run with it. I'm pretty sure the Rocket core generator is released under a BSD license. SiFive sort of makes sense if you want to integrate proprietary IP with open IP, and some people do actually want to do that. I don't know if there's enough demand for that, though.

But yeah. Even though we have compilers, it's not like you can spin up support for a new ISA in the drop of a hat. If you launched a project with a StarFive part and then decided you wanted to use SiFive IP, you should be able to use the same tooling. The impact of that shouldn't be under-estimated.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by RetroTechie on Monday September 04, 2023 @06:10PM (#37385616)
That's an important advantage: go Arm, and ultimately you'll be dealing with Arm (the IP licensing company), or with a 3rd party which deals with them.

Go x86, and you'll end up dealing with Intel, or AMD (okay, there's Via & maybe a few makers of legacy x86 parts).

Go RISC-V, and roll your own or pick from a growing list of vendors, are both valid options. And switch supplier while keeping your toolchain ecosystem.

I'd expect at least that this will spur innovation in the field. Likely increase consumer choice.

More suppliers of [insert product type here] is almost always a good thing.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by pjmlp on Tuesday September 05, 2023 @02:42AM (#37388605)
RISC-V is only a basic CPU specification, everything else is vendor specific.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by snvzz on Tuesday September 05, 2023 @07:16AM (#37390242)
RISC-V, as of current ratified specifications, is anything but basic: It has feature parity with x86 and ARM.

And the specs extend further to non-ISA, such as standardizing boot process and platform.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by pjmlp on Tuesday September 05, 2023 @08:13AM (#37390631)
If we forget about everything else needed to put that paper specification on a real motherboard.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by adgjlsfhk1 on Tuesday September 05, 2023 @12:07PM (#37393717)
The important part is from a software perspective, none of that stuff matters much. The benefit of RiscV is you can be sure that your low level software (e.g. an OS or HPC thing that requires hand coded assembly) won't be tied to a specific company's hardware.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Tuesday September 05, 2023 @03:24PM (#37396520)
I read your comment to imply that the ISA is the only thing that matters since that's what defines the software interface and that software is the only thing that matters.

If that is what you're saying, I think you're wrong.

As to the low level hand coded assembly. There's a fair amount of stuff in the Intel architecture that's not officially published. There are tools (from Intel) like VTune and the Intel C compilers that use non-published interfaces to optimize C or assembly, sometimes significantly. It is true you can write software without them, but many people purchase these Intel tools to make sure their systems run quickly and efficiently.

I can't remember if the SiFive U-X4 core equivalents were defined in the open RISC-V specs or as part of SiFive specific interfaces. If the former, then your point is even stronger. If the latter, your point is still valid, though modulo the impact of having relied on a SiFive specific interface.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by pjmlp on Tuesday September 05, 2023 @06:36PM (#37398756)
Ignoring extensions and motherboard features used from userspace.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Tuesday September 05, 2023 @03:15PM (#37396407)
Or... like my team did... you can design your own IP that specifically fits your needs.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by pjmlp on Tuesday September 05, 2023 @02:35AM (#37388563)
Depends on the extensions being used, and board designs.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by imtringued on Monday September 04, 2023 @01:04PM (#37382411)
Jim Keller's Tenstorrent worked with SiFive instead, because ARM didn't want to customize their cores for machine learning workloads and now they are developing a RISC-V with a ridiculous performance profile.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Monday September 04, 2023 @01:08PM (#37382472)
Well. They effectively control the architecture. All the technical committees in the foundation report up through one of the three principals at SiFive. I can't imagine a technical spec would be published without the consent of SiFive.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by ChuckMcM on Monday September 04, 2023 @03:51PM (#37384266)
The ugly truth though is that "catalog parts" carry with it a whole boatload of overhead in the form of sockets/memory infrastructure/voltage regulators/cooling standards etc. The current "big money" is in being the core that gets integrated into a phone or tablet. It makes me sad too as this is a pretty spiffy micro-architecture to play around with.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Tuesday September 05, 2023 @04:09PM (#37397060)
Oh hey Chuck!

Yes. That's where the big money is. It seems their current top-of-the-line core is on par with an A72 or A77. I'm not sure how it stacks up against an Apple M1.

My take on SiFive is it's at the point that it COULD compete, but they have to get more mind-share. If you're Samsung or Apple (or MediaTek or Allwinner or Rockchip or ...), you probably have experience with the ARM ecosystem and tools. Presumably you have a history with a business model and have factored in whatever royalties you're paying ARM into your cost of operations. It's a risk to set that aside and start using a different architecture, so you would want to have some quantifiable benefit before you ditched ARM for RISC-V.

It seems to me that clearly elucidating that benefit to potential customers is an important part of this equation. SiFive's earlier designs seemed to have smaller PPA numbers, so that's nice. But they were for in-order execution w/o macro fusion or speculative branching, so they were a little closer to A53s in performance per watt. If you had a RV64 design with the performance per watt of an A77, but with a power/area budget 20% lower, that might have a REAL impact on yield and cost.

That would be a thing to crow about. But that's not the type of crowing I see happening on the SiFive web page.

I keep thinking a catalog part with maybe a pair of U-54s, a pair of 270s and a pair of 670s plus typical SoC peripherals would be a great way to introduce developers to what SiFive designs could do. They could give dev boards to Samsung, Zero, Google, LTE engineers who could tweak the existing android RISC-V android images to get an idea for performance and power use.

Maybe they could sell Apple on the idea of acquihiring SiFive. Apple's never been skiddish about changing processor architectures or suppliers.

So... to recap... I'm thinking catalog parts mostly as a vehicle to prove to potential customers your IP can be fabbed to meet their power / performance / yield / cost targets. And if I can buy a couple hundred of these boards for a different project, then I'm even happier.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by brucehoult on Tuesday September 05, 2023 @09:46PM (#37400447)
> It seems their current top-of-the-line core is on par with an A72 or A77

Their current core, the P870 (the subject of the post we are commenting on) is on par with Cortex X2, announced just over 2 years ago in May 2021.

A76 class was the P550, announced in June 2021, 3 years after the A76's May 2018 announcement.

A72 class was their U84, back in October 2019, 4 1/2 years after the A72's February 2015 announcement.

The time gap has been coming steadily down.

All of these cores, both Arm and SiFive, take about four years to get into cheap mass produced catalog part chips and boards e.g. the A76 RK3588 just started appearing in SBCs around May 2022. A72 to Pi 4 was 4 1/2 years.

I don't know how much you can pick up a JH7110 with four 1.5 GHz U74s (plus that S7) for, but Milk-V are selling a compute module board with one for $34

https://arace.tech/products/milk-v-mars-cm?variant=423421722...

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by ChuckMcM on Wednesday September 06, 2023 @09:01PM (#37413478)
> So... to recap... I'm thinking catalog parts mostly as a vehicle to prove to potential customers your IP can be fabbed to meet their power / performance / yield / cost targets. And if I can buy a couple hundred of these boards for a different project, then I'm even happier.

Oh I don't disagree, lowering the cost to field a system out of your micro-architecture increases the likely hood a design engineer will design it into their own design. This effect singlehandedly sustained the x86 product line until it didn't.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by pama on Monday September 04, 2023 @05:56PM (#37385494)
Is there any good benchmark/comparison of the current risk-v boards for use in hobby projects?
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by brucehoult on Monday September 04, 2023 @08:23PM (#37386500)
If by "good" you mean "something that's probably about as good as Dhrystone or Coremark but a lot simpler and harder to game" then I've got a benchmark and quite a few results collected on various things over the last seven years here:

https://hoult.org/primes.txt

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by pama on Tuesday September 05, 2023 @10:10AM (#37391867)
Thanks. So some of these cpu, say VisionFive 2 U74, start to look interesting.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by pama on Wednesday September 06, 2023 @10:22PM (#37413972)
Just noting for my own reference that the TH1520 is well within the interesting range.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Tuesday September 05, 2023 @04:16PM (#37397138)
Though... RISC-V support in gnu c and clang have evolved significantly over the last 5 years. Did you re-run any of the tests with more recent compilers? When we were doing benchmarking 5 years ago, it seemed we got slightly better performance every month as Palmer integrated improvements into the gnu toolchain. There were at least a couple months there where we REALLY didn't want to publish benchmarks because we were expecting the numbers to be noticeably better in n days.
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by brucehoult on Tuesday September 05, 2023 @09:53PM (#37400494)
This is very straightforward code (which I wrote in 2016 before I'd even heard of RISC-V) that maps very cleanly and obviously to instructions and any gcc from the last 20 years will make essentially the same code. A large part of my goal when I wrote it was to make it far more impervious to compiler "improvements" and gaming that Dhrystone and Coremark -- and I think I succeeded very well.

The RISC-V B extension's `sh2add` and friends make a small difference -- that's basically the 2.7% gap between the VisionFive 2 and the HiFive Unmatched. If I disable Zba on the VF2 then the times are just about identical.

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by eigenform on Monday September 04, 2023 @06:59PM (#37385986)
I'm curious what your objections to Chisel are! (I don't really care for Scala either .. unless you mean that like, this somehow affects the way that IP is delivered to you as a customer? I thought you just get a big blob of verilog in the end, and I'm not sure how that'd affect you?)
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by OhMeadhbh on Tuesday September 05, 2023 @12:53PM (#37394547)
I'm not the worlds biggest fan of Verilog, but Verilog's semantics are relatively fixed. The interpretation of verilog has remained fairly stable for a decade at least. Chisel is in some ways still a work in progress. So if you want to use it, you have to keep track not only of Chisel productions, but also the version of the boom or rocket generator you're using with it. Chisel was never at the center of our design flow, and none of the people I currently work with went to UCB. When we used the Rocket generator we were often "nervous" about whether the version we were using would interpret Chisel productions in the way in which it had previously.

But yes... this led us to a working style where we tried very hard to avoid having to use rocket or boom generators. I didn't want to be in the business of maintaining RISC-V Verilog tools. At around the same time we realized RISC-V was overkill and built a very poorly designed Lisp DSL which output Verilog for a dog-simple 6-bit controller. Not an ideal solution to be sure, but it worked better for us than spending a couple years learning Rocket/Scala/Chisel.

Nick Tredennick (who I believe was the main designer of the 68k) has a quote in his book "Logic Design: The Flowchart Method":

  The problem with many texts is that we lie about
  details.  We are sloppy in areas that are not our primary
  concern.  Academics are method fanatics.  Practitioners
  are solution fanatics.  In school, we glorify the methods
  and lie about the sophisticated problems we solved.  In
  industry, we glorify the problems and lie about the
  sophisticated methods we used.  Each side loses
  credibility the minute one side reads the other's
  literature.  The academic knows that the practitioner's
  "method" is (ugh!) arbitrary, just as the practitioner
  knows the academic's "solution is (ugh!) not applicable.
  Because we oversell method and solution, it takes too
  long to figure out what really works.
I don't want to say Rocket / Chisel / Firrtl is bad, but it always seemed to be more in support of a "process focused" methodology than the "solution oriented" focus I'm used to. If it works for you, I'm absolutely not going to tell you it's a bad idea to use it. It's just too much overhead for the small projects I work on.

That being said... SiFive was working on CoC (theorum prover) extensions to validate designs using formal methods. THAT sounds fascinating, but I'm not sure how they would monetize it.

[Edit: To recap, I don't have a problem with Rocket / Chisel / Firrtl per se, but they operate at a level I'm uncomfortable with and seem to be less stable than I would prefer. I completely understand this is not the case for many other people.]

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by eigenform on Tuesday September 05, 2023 @01:31PM (#37395066)
I can totally understand it being too "heavy-handed" a solution for tons of cases, definitely! Thanks for sharing :)
Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by taway32r41 on Thursday September 07, 2023 @02:05AM (#37415288)
> SiFive was working on CoC (theorum prover) extensions to validate designs using formal methods. THAT sounds fascinating, but I'm not sure how they would monetize it.

Synopsys seems to think it's a decent business to be in. :)

https://www.synopsys.com/verification/static-and-formal-veri...

Re:Hot Chips 2023: SiFive’s P870 Takes RISC-V Further by cinquemb on Tuesday September 05, 2023 @01:05AM (#37388086)
Makes me think that asia is gonna have this cheap compute available alot faster at the consumer level than anywhere else for at least the next 5-10 years.