A Brazilian electronics engineer and assembly programmer named Jessé says he has tested two AMD processors extensively and found that the chips' hardware random number generator, accessed via instructions like rdrand and rdseed, has not produced a single zero value in over nine days of continuous testing. He has reported the issue to AMD, which has escalated it internally, though no technical explanation has been provided yet. Jessé also updated his testing application to benchmark RNG speed and support processors lacking the rdseed instruction, noting large performance differences across CPU generations.
board.flatassembler.net
· 2026-09-22
A technical analysis compares Microsoft's Prism binary translator, which lets x86-64 Windows software run on Arm-based aarch64 processors, against native Geekbench 7 performance. Using a pro key from Primate Labs founder John Poole to profile individual workloads, the tester measures how much performance is lost when translating x86-64 instructions instead of running aarch64 code directly.
chipsandcheese.com
· 2026-09-10
A technical blog post by Martin Uecker examines how GCC compiles nested functions using trampolines placed on the stack, which normally require the stack to be executable. He shows that by extracting the code address and static chain values embedded in the trampoline and passing them to the __builtin_call_with_static_chain built-in, developers can indirectly call nested functions on older GCC versions without needing an executable stack.
uecker.codeberg.page
· 2026-08-29
GoKawiil