Vue normale

Reçu aujourd’hui — 21 mai 2025Phoronix
Reçu hier — 20 mai 2025Phoronix

Linux Scheduler Patches Aim To Address Performance Regression Since Last Year

20 mai 2025 à 20:05
A set of Linux kernel scheduler patches posted today are trying to address some performance regressions observed since the Linux 6.11 kernel that was released back in September 2024. These performance-fixing patches are flying under a "request for comments" flag and some of the regressions are tricky and perhaps not completely resolved, but it looks to be a step in the right direction...

Some Minor Performance Hits Observed With New Intel Arrow Lake 0x118 CPU Microcode

20 mai 2025 à 14:00
Last week Intel released new CPU microcode for a number of processor generations due to the Training Solo vulnerability and Branch Privilege Injection. In this article today are some benchmarks looking at the performance difference from simply upgrading to the new CPU microcode on an Intel Core Ultra 9 285K "Arrow Lake" desktop system...

Adaptive Sharpness Property Still Being Worked On For Intel Lunar Lake & Newer On Linux

20 mai 2025 à 10:22
Since last summer Intel open-source driver engineers have been working to make use of Lunar Lake's new adaptive sharpening capabilities under Linux. This is being done via a proposed DRM sharpness property for communicating sharpness preferences while the hardware capabilities within Lunar Lake equate to minimal power and performance costs when in use. The DRM sharpness property support still has yet to be mainlined but the newest iteration of the patches were posted on Monday...
Reçu avant avant-hierPhoronix

Intel Adds OpenMP Multi-Threading To Its Speedy x86-simd-sort Library

19 mai 2025 à 20:05
Intel's x86-simd-sort open-source project is a C++ template library for high performance sorting routines that can leverage AVX2 and AVX-512 for crazy fast sorting. The x86-simd-sort code in turn is used by Numpy, more recently adopted by PyTorch too, and has shown off the great performance potential of AVX-512 for very fast sorting algorithms. Out today is x86-simd-sort 7.0 and it's even faster due to now supporting OpenMP parallelization...

Samsung Back To Working On Upstreaming Tesla FSD SoC Support In The Linux Kernel

19 mai 2025 à 19:09
In early 2022 Samsung engineers began working on upstreaming support for the Tesla Full Self-Driving "FSD" SoC to the mainline Linux kernel. Those early patches were mainlined in Linux 5.18 and later in 2022 turned to working on the PCIe support for the Tesla FSD SoC with the mainline kernel, but then work seemingly ceased on this upstreaming effort. More than two years later, the work was restarted today with Samsung posting the latest patches for enabling PCI Express support for the Tesla FSD SoC in the Linux kernel...
❌