@linux4kix

Now, we wait.

 https://discord.gg/yFnRAWF
Joined September 2011
  • Tweets3K
  • Following116
  • Followers441
237Media
I am contemplating the possibility of having a completely fan-less home office now....well except for my scope, and 3D printer but they aren't on all the time.
1
4
148
The first #Industrial #PC in its category with 10GbE capabilities. Bedrock V3000 is set for harsh environments for various applications: Energy, robotics & manufacturing control. Watch our expert explain the advantages of AMD #Ryzen based fanless computer: youtu.be/pQgAn2MVoIA
1
3
5
673
Discover how high-performance #Industrial #PCs with 10 GbE ports are revolutionizing data-intensive applications at the network edge. Stay ahead of the competition with real-time analytics, seamless connectivity, and reliable communication. Read now: solid-run.com/blog/articles/…
1
3
540
We're excited to share our latest updates: A release of a lineup of cutting-edge technology solutions, recent participation in industry events and more! #SolidRun brings you valuable innovations to enhance your overall experience. Read our newsletter: mailchi.mp/solid-run/updates…
2
2
467
It was great participating in the #AI Tech Day event in #Tokyo! We were thrilled to showcase our latest products and advancement. Thank you, @RenesasGlobal, for this fantastic opportunity to share and expand our knowledge and insights. Visit our website: solid-run.com/
1
3
376
Always a privilege to meet our partners. Great products come from solid partnerships. We thank @RenesasGlobal for hosting us at their 2022 FAE event in #Vienna. We would never pass on an opportunity that allows us to provide better products and support to our customers. #Solidrun
1
5
.@solid_run has unveiled one of the first systems based on @AMD Ryzen Embedded V3000 SoC. Designed for #industrial applications, it supports up to 64GB DDR5, 3x NVMe SSDs, 10GbE and 2.5GbE interfaces, and more. #ryzen #factory #networking #storage cnx-software.com/2022/11/22/…

Bedrock V3000 Basic fanless industrial computer is powered by an AMD Ryzen Embedded V3000 processor...

SolidRun Bedrock V3000 Basic is an industrial fanless computer based on the new AMD Ryzen Embedded V3000 Zen3 processor family designed for storage and

cnx-software.com
9
19
I echo this sentiment. Amazing a single person could screw something up so badly in such a short amount of time. Truly a genius
This quoted post is unavailable.
Try viewing anyway · View on X
6
Our #SolidSense platform combines the connectivity and low power of the #iMX8M nano with @NordicTweets #nRF52833 open CPU wireless module. An expandable, flexible gateway solution for @Wirepas or #Bluetooth mesh deployments. Read more: nordicsemi.com/News/2022/11/…
6
8
#Linux 6.2 AMDGPU Driver Gets Newer @Radeon GPUs Working On #Arm 64-bit (AArch64), SR-IOV Fixes For #RDNA3/GFX11 phoronix.com/news/Linux-6.2-…

Linux 6.2 AMDGPU Driver Gets Newer GPUs Working On Arm, SR-IOV Fixes For RDNA3/GFX11

AMD on Friday submitted a new round of patches for AMDGPU kernel graphics driver updates they have ready for the upcoming Linux 6.2 cycle.

phoronix.com
12
2
51
glibc moving to memcpy_simd for aarch64 by default with the comment, this is the fastest on all cores. patchwork.sourceware.org/pro… .... not by my benchmarks, luckily I have libmemcpy.so github.com/jnettlet/cortex_a…
6
With the increasing demand for data centers to secure their infrastructure and scale performance, SolidNET Software-Defined #DPU offers isolation and acceleration and avoids potential hardware lock based on open standards. Read more: solid-run.com/blog/the-first…
2
3
4
Uggh....Toddlers and Dogs just don't understand Daylight Savings Time
1
4
#Aarch64 Architects, I was debugging a performance inconsistency today and figured I would pick twitter's brain. I found I could replicate the issue using stress-ng's matrix math benchmark.
2
1
1
The vector math benchmark does not exhibit the same degradation.
1
and running 1 vector math stressor and 1 matrix math stressor on each cluster shows no slowdown.
if I spread the workload as 1 core in 8 clusters I get 1.02 insn / cycle, but if I spread the workload as 2 cores in 4 clusters this drops down to .7 insn / cycle. I don't see how the L2 cache being shared could cause this much of a drop off, and only for floating point.
1
What a fantastic week at the #OCP summit! It was fun meeting customers, partners and friends in person. Many great conversations and lots of activity on SolidNET Software-Defined DPU and DC-SCM. Thanks to the team for their hard work in making the show a success! #sustainability
2
5
Well look at that. Even an Ampere Altra with a Navi card has the same Xorg corruption has the corruption that the amdgpu developers blamed on "broken" hardware. #HoneyComb_LX2 Of course my workarounds fix it. #notcrazy
3
1
13
Interesting, github.com/Tencent/TencentOS…

ampere/arm64: Work around Ampere Altra erratum #82288 PCIE_65 · Tencent/TencentOS-kernel@f454797

Altra's PCIe controller may generate incorrect addresses when receiving writes from the CPU with a discontiguous set of byte enables. Attempt to work around this by handing out Device-nGnRE...

github.com
1
and @never_released you will find this commit that is required by that errata fix interesting, github.com/Tencent/TencentOS…

ampere/arm64: Add a fixup handler for alignment faults in aarch64 code · Tencent/TencentOS-kernel...

A later patch will hand out Device memory in some cases to code which expects a Normal memory type, as an errata workaround. Unaligned accesses to Device memory will fault though, so here we add a ...

github.com
1
Load more