#SiFive

eicker.news ᳇ tech newstechnews@eicker.news
2026-01-18

#KrsteAsanović, co-founder of #SiFive and a key figure in the development of #RISCV, discusses the #opensource #CPU architecture’s origins and evolution. Initially created for academic research at Berkeley, RISC-V’s modular design allowed for rapid prototyping and software development. The architecture’s openness attracted interest from both academia and industry. morethanmoore.substack.com/p/a #tech #media #news

N-gated Hacker Newsngate
2026-01-11

Ah, another brave soul attempts to shoehorn into the almighty HiFive Unmatched board. 🤡 Because an archaic educational OS to a niche board is exactly what the world needed right now. 🙄 , of course, stands by to witness this monumental achievement in software archaeology. 🥳
github.com/eyengin/xv6-riscv-u

2025-11-17

We've updated our tor relay build script for #RISCV, building #openssl and #tor from source:

code.disobey.net/EmeraldOnion/

Notably, to get rid of this tor initialization warning:
> We were built to run on a 64-bit CPU, with OpenSSL 1.0.1 or later, but with a version of OpenSSL that apparently lacks accelerated support for the NIST P-224 and P-256 groups. Building openssl with such support (using the enable-ec_nistp_64_gcc_128 option when configuring it) would make ECDH much faster.

With an added openssl configure target and option:
> linux64-riscv64 enable-ec_nistp_64_gcc_128

Our #SiFive #HiFive RISC-V 256-bits ECDH performance is synthetically boosted:

From:
> 256 bits ecdh (nistp256) 0.0029s 340.7

To:
> 256 bits ecdh (nistp256) 0.0005s 2057.1

This is a 6x P-256 handshakes boost, so this should help speed up this Tor exit relay. We are only running one Tor daemon on this hardware to see how well it does.

root@sifive:~# /usr/local/bin/openssl speed ecdhp256
Doing 256 bits  ecdh ops for 10s: 3407 256-bits ECDH ops in 10.00s
version: 3.6.0
built on: Sun Oct 12 19:13:28 2025 UTC
options: bn(64,64)
compiler: gcc -fPIC -pthread -Wa,--noexecstack -Wall -O3 -DOPENSSL_USE_NODELETE -DOPENSSL_PIC -DOPENSSL_BUILDING_OPENSSL -DNDEBUG
CPUINFO: OPENSSL_riscvcap=
                              op      op/s
 256 bits ecdh (nistp256)   0.0029s    340.7root@sifive:~# /usr/local/bin/openssl speed ecdhp256
Doing 256 bits  ecdh ops for 10s: 20550 256-bits ECDH ops in 9.99s
version: 3.6.0
built on: Sun Oct 12 19:13:28 2025 UTC
options: bn(64,64)
compiler: gcc -fPIC -pthread -Wa,--noexecstack -Wall -O3 -DOPENSSL_USE_NODELETE -DOPENSSL_PIC -DOPENSSL_BUILDING_OPENSSL -DNDEBUG
CPUINFO: OPENSSL_riscvcap=
                              op      op/s
 256 bits ecdh (nistp256)   0.0005s   2057.1
2025-10-15
2025-08-08

Kickstarterová kampaň na odlehčenou verzi VisionFive 2 spuštěna
#SiFive #RISC-V
alt-f4.cz/kickstarterova-kampa

2025-08-07

StarFive VisionFive 2 Lite is a cheap(er) RISC-V single-board computer (crowdfunding)

The VisionFive 2 Lite is a credit card-sized single-board computer (SBC) that looks a lot like a Raspberry Pi. But it’s actually a smaller, cheaper, and less powerful version of the VisionFive 2 RISC-V SBC that launched a few years ago.

The new model has a slower version of the same processor and loses a few ports and connectors, but picks up optional support for onboard WiFi and Bluetooth. […]

#crowdfunding #riscV #sbc #sifive #starfive #starfiveVisionfive2Lite

Read more: liliputing.com/starfive-vision

2025-06-06

SiFive X280 RVV benchmarks: camel-cdr.github.io/rvv-bench-

Civil was so nice to run my rvv benchmark on the SiFive X280 cores on the Tenstorrent Blackhole.

#riscv #rvv #Tenstorrent #sifive

Performance observations

These are just my first impressions of the core looking at the initial benchmark results.

    the regular vector instructions take 2/4/8/16 cycles for LMUL=1/2/4/8
    The core likes bigger LMUL and vsetvli has basically no overhead (unrolling can still be advantageous, because of the in-order scalar core)
    vl based dispatch, so a LMUL=8 instruction is as fast as a LMUL=1 instruction if vl elements fit into a LMUL=1 register
    1 cycle per element `vrgather.vv` and `vcompress.vm`, but you can still get decent speed up over scalar loads from memory. (see LUT4 and LUT6)
    Very good fault-only-first load support. Page-size-aware strlen is only about 25% faster than the fault-only-first load version.
    nf=2 segmented load/stores seems very fast. (see utf16)
    nf>2 segmented- and strided load/stores perform decently until a big drop in performance between 8 and 24 KiB of processed data. (see utf32)
Benjamin Carr, Ph.D. 👨🏻‍💻🧬BenjaminHCCarr@hachyderm.io
2025-03-06

A #RISCV Progress Check: Benchmarking #P550 and #C910
The RISC-V cores fall short of Ar #CortexA73 and well short of Intel’s #Goldmont Plus. Mediatek’s In-Order #Genio1200 hyas higher clock speeds and better DRAM latency than C910 and P550. Its #CortexA55 cores are able to catch C910 and P550 without out-of-order execution.
#SiFive’s P550 and T-HEAD’s Xuantie C910 are both notable for featuring out-of-order execution on RISC-V. Both are plagued by low clock speeds.
chipsandcheese.com/p/a-risc-v-

Benjamin Carr, Ph.D. 👨🏻‍💻🧬BenjaminHCCarr@hachyderm.io
2025-03-04

Inside #SiFive’s #P550 #RISCV Microarchitecture
It doesn’t go head-on against likes of AMD’s Zen 5, Intel’s Lion Cove, or Qualcomm’s Oryon. P550’s out-of-order engine is closer in size to something like Intel’s Core 2 from over 15 yrs ago. Combine that with much lower clock speeds than even what Core 2 ran at, and P550 is really a low power core with modest performance. This core aims for “30% higher performance in less than half the area of a comparable Arm Cortex A75.”
chipsandcheese.com/p/inside-si

Gábor SEBESTYÉN 🇭🇺🇪🇺🇺🇦segabor@czinege.social
2025-02-23

@zilahu és akkor még egy friss-ropogós teszt a Si-Five USA cég HiFive Premier P550 alaplapjáról youtube.com/watch?v=1565YYsFmd #risc-v #sifive #p550 #test

Jure Repinc :linux: :kde:JRepin@mstdn.io
2025-02-04

RISC-V Mainboard for Framework Laptop 13 is now available
🔗 frame.work/si/en/blog/risc-v-m
via @frameworkcomputer

"We’re happy to share that DeepComputing’s DC-ROMA RISC-V Mainboard for Framework Laptop 13 is now in stock and shipping in the Framework Marketplace. This is very much a developer-focused board to help accelerate maturing the software ecosystem around RISC-V."

#RISCV #RISC_V #laptop #laptops #hardware #CPU #CPUs #Framework #DeepComputing #StarFive #JH7110 #SiFive #U74

Jure Repinc :linux: :kde:JRepin@mstdn.io
2025-01-27

Inside SiFive’s P550 Microarchitecture
🔗 old.chipsandcheese.com/2025/01

"The P550 is a 3-wide out-of-order core with a 13 stage pipeline. Out-of-order execution lets the core move past a stalled instruction to extract instruction level parallelism. It’s critical for achieving high performance because cache and memory latency can be significant limiters for modern CPUs."

#RISCV #RISC_V #ComputerArchitecture #CPU #CPUs #Processor #Processors #Hardware #ComputerHardware #Eswin #EC7700X #SiFive #P550

Benjamin Carr, Ph.D. 👨🏻‍💻🧬BenjaminHCCarr@hachyderm.io
2024-12-27

#SiFive #HiFive Premier #P550 #RISCV Price Lowered to $399, ready to go with #Ubuntu 24.04 LTS Support
This RISC-V developer board features the SiFive P550 CPU, 128GB eMMC storage, 16GB or 32GB of LPDDR5 memory, PCIe connectivity, M.2 storage support, USB 3, and more.
phoronix.com/news/HiFive-Premi

Client Info

Server: https://mastodon.social
Version: 2025.07
Repository: https://github.com/cyevgeniy/lmst