Live data from Hacker News

Bug 1950764: Work Around Crash on Intel Raptor Lake CPU

phabricator.services.mozilla.com

61–62 of 62 posts

Re: Bug 1950764: Work Around Crash on Intel Raptor Lake CPU

#61

Earlier quoted context omitted.

You're implying that if you don't run the CPU at high power and high heat it won't have problems, and that undervolting or underclocking will prevent damage. This is not correct: while that is helpful, Vmin degradation occurs during idle or light activity as well Vmin will creep up, and the headroom for undervolting will degrade. It will affect the high clocks first (they demand the highest voltage), which is why dro…

Idle and light load is bad for degradation only because that's the most common scenario where the boosting algorith will actually go to the highest clocks. More loaded cores will have the CPU target lower clocks on all cores so that it actually can get the power for it and have the CPU be coolable, but if you're idle and then some task loads just a single core for a bit the CPU will boost it the highest it can. The v…

Or perhaps it is more complicated than that

"Even under idle conditions at relatively cool temperatures, sporadic elevated voltages are observed when the processor is resumed from low power states in order to service background operations before entering a low power state again."

https://www.igorslab.de/en/search-for-the-solution-to-raptor...

Re: Bug 1950764: Work Around Crash on Intel Raptor Lake CPU

#62

Also worth reading this thread on the subject: https://mas.to/@gabrielesvelto/116630047156991279 Regarding the Raptor Lake bug I received a couple of messages from confused users that had read articles on Tomshardware and Neowin. They asked about erratas and microcode updates which puzzled me, because that was part of my early investigation into the bug and we know that the failure is not caused by a known errata and…

I don't think he's right about this AI slop comment (and I am anti AI slop!). The tom's hardware article might have used AI but it doesn't say that microcode updates fixed it, just that the microcode updates mitigated it. Yes it mentions the errata, but a human could have come to that conclusion based on the bug report. And firefox.com release notes for 151.0.1 specifically claims it has a fix for this bug! https://www.firefox.com/en-US/firefox/151.0.1/releasenotes/
Post reply on HN