Top all-time · #11953
BLOGS.FAU.DE
Personal blog · blogs.fau.de/hagersystemshardware
BLOGS.FAU.DE is a personal blog at blogs.fau.de about systems and hardware. Since 2014, 6 different Hacker News readers have linked it 8 times, most often “Intel’s port 7 AGU blunder (2019)”. It is still writing: the last post was on 2026-10-16, and you can follow it by feed.
Most-linked posts?
- Intel’s port 7 AGU blunder (2019)2 people · 2 mentions · first linked 2020-01-16
What Hacker News says?
SSE allows you to do non-temporal loads & stores, within certain constraints. "A case for the non-temporal store" https://blogs.fau.de/hager/archives/2103
Comment linking /hager/archives/2103
on “Cache eviction: when are randomized algorithms better than LRU? (2014)”Submitted Intel’s port 7 AGU blunder (2019) as a story · 81 points
HN discussionAgreed. To my surprise, he kinda argues against this in one of his items ( https://blogs.fau.de/hager/archives/6558 ): Always compare the fully parallel accelerator code with a sequential (single-core) CPU version. You can even argue in favor of this, saying that “serial CPU code constitutes a well-defined baseline.” Be sure to rush ahead to the next slide before anyone dares to ask why you don’t use a single GPGPU…
Comment linking /hager/archives/6558
on “Fooling the masses with performance results on parallel computers (2010)”Submitted Fooling the masses with performance results on parallel computers (2010) as a story · 53 points
HN discussionSubmitted A case for the non-temporal store (2008) as a story · 9 points
HN discussionSubmitted Intel’s port 7 AGU blunder as a story · 2 points
HN discussionSubmitted Fooling the masses with performance results on parallel computers (2010) as a story · 2 points
HN discussionSubmitted CloverLeaf on Intel Multi-Core CPUs: A Case Study in Write-Allocate Evasion as a story · 1 points
HN discussion
