https://forum.asrock.com/ Copyright (c) 2006-2013 Web Wiz Forums - All Rights Reserved. Mon, 20 Apr 2026 17:42:41 +0000 Sun, 07 Apr 2019 04:14:37 +0000 http://blogs.law.harvard.edu/tech/rss Web Wiz Forums 12.04 360 https://forum.asrock.com/RSS_post_feed.asp?TID=11264 https://forum.asrock.com/forum_images/web_wiz_forums_black.png https://forum.asrock.com/ https://forum.asrock.com/forum_posts.asp?TID=11264&PID=62067&title=linux-x470-taichi-hangs#62067 Author: kschendel
Subject: 11264
Posted: 07 Apr 2019 at 4:14am

paologab wrote: Not here, kernel 5.0.5. I?�m also testing deeply the ram @3466 (rated @3200) with no issues so far.

5.0.5 hung just as often as any other 5.0.x for me.
]]> Sun, 07 Apr 2019 04:14:37 +0000 https://forum.asrock.com/forum_posts.asp?TID=11264&PID=62067&title=linux-x470-taichi-hangs#62067 https://forum.asrock.com/forum_posts.asp?TID=11264&PID=62059&title=linux-x470-taichi-hangs#62059 Author: Onox
Subject: 11264
Posted: 06 Apr 2019 at 9:17pm

Hi. Yes, it's a widespread issue. There are freezes with other boards too, and even Ryzen laptops freeze with Linux :/ You can find a lot of information here : https://bugzilla.kernel.org/show_bug.cgi?id=196683

I have the X370 Taichi and I had similar issues even on Windows, until I started using the "Power Supply Idle Control" > "Typical Current Idle" option. I'm not 100% sure that this entirely fixes it for me, as I had the computer freeze when resuming from sleep twice, but only twice and that was some time ago. I also had the "Ryzen segfault bug" and had to RMA my 1700X (I'm currently waiting for the new CPU to be shipped), but since you have a 2700X you should not be affected by that flaw (as far as I know, only the Ryzen 1000 series may have it).

I also read that the default SOC voltage is too low on Asrock boards; this could explain why people who overclock never experience any freezes.]]> Sat, 06 Apr 2019 21:17:39 +0000 https://forum.asrock.com/forum_posts.asp?TID=11264&PID=62059&title=linux-x470-taichi-hangs#62059 https://forum.asrock.com/forum_posts.asp?TID=11264&PID=62058&title=linux-x470-taichi-hangs#62058 Author: paologab
Subject: 11264
Posted: 06 Apr 2019 at 8:44pm

Not here, kernel 5.0.5.
I?™m also testing deeply the ram @3466 (rated @3200) with no issues so far.

]]> Sat, 06 Apr 2019 20:44:42 +0000 https://forum.asrock.com/forum_posts.asp?TID=11264&PID=62058&title=linux-x470-taichi-hangs#62058 https://forum.asrock.com/forum_posts.asp?TID=11264&PID=62049&title=linux-x470-taichi-hangs#62049 Author: xhue
Subject: 11264
Posted: 06 Apr 2019 at 5:35pm

Thanks for the info!

AFAIK even kernel 5.0 has issues with Ryzens still.

Depending on your needs you may want to consider running Linux on a VM. This way you will let the hypervisor deal with the real power management, but you still have the other benefits of Linux.]]> Sat, 06 Apr 2019 17:35:10 +0000 https://forum.asrock.com/forum_posts.asp?TID=11264&PID=62049&title=linux-x470-taichi-hangs#62049 https://forum.asrock.com/forum_posts.asp?TID=11264&PID=62048&title=linux-x470-taichi-hangs#62048 Author: kschendel
Subject: 11264
Posted: 06 Apr 2019 at 3:52pm

This is in the way of a "public service announcement", specifically for the X470 Taichi Ultimate but I suspect that similar problems might exist on other boards.

Build is a 2700X in a Taichi Ultimate, RX 550 graphics. I had been extremely stable on kernel 4.17.19 and later since last summer, until I updated my BIOS to the 2.00 version. I started getting occasional freezes, where the entire machine would lock up hard with nothing in the MCE logs or anywhere else. I was blaming this on the RX 550 (and indeed the amdgpu driver has its issues), but with kernel 5.0.x it got so that a freeze would happen within minutes.

The cure turns out to be to avoid deep C-states. The older BIOS had a "disable C6 state" option, which is apparently gone in the new version. I haven't yet figured out whether a kernel command line of "processor.max_cstates=1" suffices, or whether one needs to disable Global C-States entirely in the BIOS (OC Tweaker, AMD CBS, CPU options); the difference is some 10 watts at idle which is minimal. I can say that "processor.max_cstates=5" does NOT work to avoid the hangs.

It's only been a day so far, but that's more uptime than I've had in a couple weeks. Other input welcome.
]]> Sat, 06 Apr 2019 15:52:28 +0000 https://forum.asrock.com/forum_posts.asp?TID=11264&PID=62048&title=linux-x470-taichi-hangs#62048