Tagged articles

kernel performance

3 articles · Page 1 of 1

Sep 22, 2026 · Operations

Linux 7.3 Preview: Btrfs 5x Speedup, Chinese Engineers Drive Key Optimizations

Linux 7.3, expected October 18, delivers major performance gains: Btrfs and EXT4 speedups, a KSM fix reducing latency from 705ms to 1.44ms, ByteDance's scheduler restructuring cutting worst-case latency 90%, and extensive security hardening including NTFS3 fixes and BPF verifier enhancements, with new hardware support for Apple Silicon USB4 and AMD Zen 6.

Apple SiliconBPFBtrfs

0 likes · 12 min read

Linux 7.3 Preview: Btrfs 5x Speedup, Chinese Engineers Drive Key Optimizations

Linux Kernel Journey

Sep 30, 2025 · Fundamentals

How Merging cfs_rq and sched_entity Boosts Linux Scheduler Cache Hits by 31%

Google engineer Zecheng Li’s patch series merges the CFS scheduler’s cfs_rq and sched_entity structures and switches to a per‑CPU allocator, cutting LLC miss rates by up to 31% and raising kernel IPC by up to 25% on AMD and Intel platforms, with negligible regression.

CFSLinuxcache optimization

0 likes · 8 min read

How Merging cfs_rq and sched_entity Boosts Linux Scheduler Cache Hits by 31%

Liangxu Linux

Sep 2, 2020 · Fundamentals

Why System Calls Are So Expensive: From INT 0x80 to vDSO

This article explains how Linux system calls work, compares their overhead to regular function calls, and details the three implementation methods—software interrupts, fast syscall instructions, and the virtual dynamic shared object (vDSO)—including their performance trade‑offs and practical usage.

INT 0x80Linuxkernel performance

0 likes · 16 min read

Why System Calls Are So Expensive: From INT 0x80 to vDSO

kernel performance

Linux 7.3 Preview: Btrfs 5x Speedup, Chinese Engineers Drive Key Optimizations

How Merging cfs_rq and sched_entity Boosts Linux Scheduler Cache Hits by 31%

Why System Calls Are So Expensive: From INT 0x80 to vDSO

Why System Calls Are So Expensive: From INT 0x80 to vDSO