BREAKING Linux AGAIN! Chinese Zhaoxin CPU handling in GlibC broke other, older X86 CPUs for years!

#AMD #Ryzen #Zen4 #speculative #execution #bug #Ad: You can support my work at: https://patreon.com/renerebe https://github.com/sponsors/rxrbln/ https://buymeacoffee.com/renerebe http://onlyfans.com/renerebe PCs & more @Amazon: https://services.exactcode.de/amzn.cgi?index=books&keywords=rust
https://exactcode.com https://t2sde.org https://rene.rebe.de

0:00 Introduction
5:00 Demo
7:00 Q & A

source by Bits inside by René Rebe

linux foundation

10 thoughts on “BREAKING Linux AGAIN! Chinese Zhaoxin CPU handling in GlibC broke other, older X86 CPUs for years!

  • Just got a book8088 "laptop". The BIOS is a GPL ripoff (why? They could have used the GPL BIOS legally; I guess, Chinese are so used to remove all names from code that they applied their usual process here). VGA doesn't work because of bugs in the LCD panel firmware. All DOS/Windows software they put on it doesn't work (USB driver c375, windows 3.0 starts in CGA mode). They are simply not testing the stuff they put out in the wild. Minix 2 hangs on an IRQ/timer issue because something is not quite ok on the PCB. DMA is broken and needs a soldering job. Otherwise, the product would really be nice, if it had some quality.

  • Fun fact, china "design" for arm cpus. Does the same loop for debugging fun.

  • I don't think you need an Oxford Dictionary, you effectively communicated

  • interesting. more streams would be dope. stream whatever

  • omfg the while(1) really got me shouting at the screen !

  • Hey rene, thanks as always for the great work and content, i was wondering how can i get mainline to revert a specific commit that breaks a vintage sound card i have, the driver is the snd_emu10k1 and basically a developer decided to completely alter the way all professional e-mu edsp sound cards are initialized and as a result my older e-mu 8810 sound card can't be initialized.

  • It would be really helpful if you could engage more with community, by at least opening a bug o glibc. Recent I fixed a AMD performance issue that took almost 3 months to AMD to acknowledge the fix, so it takes times and it would more profitable than constantly ranting against open source projects… but just saying

  • The debate about upstreaming reminds me of that parody page "Gentoo users are ricers" which recommended the GCC flags -funroll-loops and -fuck-upstream.

  • any time, any second spent in GlibC or kernel is a blessing and this means the livestream will be in my background while I'm also doing some debugging. This is therapy !!! lol pls keep up this awesome work

Comments are closed.