http://www.phoronix.com/scan.php?page=news_item&px=Ryzen-Compiler-Issues
AYYMD IS FINISHED & BANRKUPT
AYYMDPOORFAGS CONFIRMED ON SUICIDE WATCH
Archive, Pastebin, or fuck off.
>>60714396
Oh no, I'm on linux and I don't have any issue, oh no wat to do...
>>60714396
>
I'm going to back the unstable overclocks theory. I had been overclocking my R7 1700 to 3.95 GHz, but after an hour or so, it would crash while doing compile jobs. Now I am running at 3.9 GHz and it runs all day everyday no problem, no instability.
tl;dr gentoo nerds lost the silicon lottery, forced to dial back the OC a bit.
>>60714396
Yes, I compile daily so that concerns me very much
>>60714396
>>60714396
>problems namely when running heavy compilation tasks, like on Arch and Gentoo
>Gentoo
>>60714396
>some people are having issues with hardware while others doing the same don't
>no mention of hardware details
Lets look at that google doc
>-march=haswell
>-march=bdver4
Lmao
Works for me, might be some combination of unstable voltages or RAM timings but you'll have to do the work yourself.
>>60714900
Or it might just be GCC or some hardware combination, google "Intel segfault" and you'll end up with mostly linux issues and a few gcc ones as well.
AMD fixed the VME issue with Agesa 1.6 without telling anyone, and looking at the Gentoo thread someone people reported better results with beta 1.6 Agesa, some not at all.
Anyway, this leads me to the conclusion that this is a combination of hardware setup and microcode, clearly this requires a microcode update if it can even be reproduced, but AMD is uncharacteristically being fast at fixing stuff, first the FMA bug and then VME.