cross-posted from: https://lemmy.world/post/2852886
For those out of the loop, some AMD users have been suffering from stuttering issues caused by the AMD fTPM random number generator. A firmware/BIOS update appears to fix the issue for some users, but not others, leading to more bug reports being sent in. Last week, Linus Torvalds said “let’s just disable the stupid fTPM hwrnd thing”, and, as of today the Linux kernel has gone ahead and blanket disabled RNG use for all current AMD fTPMs.
This is the best summary I could come up with:
As a follow-up to the first-on-Phoronix article last month that highlighted Linus Torvalds’ frustrated views on the AMD fTPM random number generator continuing to cause problems for users even with updated firmware/BIOS, as of today the Linux kernel has gone ahead and blanket disabled RNG use for all current AMD fTPMs.
Mario summed up in that commit: tpm: Disable RNG for all AMD fTPMs
The TPM RNG functionality was previously disabled on a subset of AMD fTPM series, but reports continue to show problems on some systems causing stutter root caused to TPM RNG functionality.
Expand disabling TPM RNG use for all AMD fTPMs whether they have versions that claim to have fixed or not.
Thus over the next few days this change in behavior for modern AMD Ryzen systems will be rolling out in the next set of stable kernel point releases.
Hopefully this will lay to rest the various AMD stutter issues that continue to be reported by Linux users on recent kernel versions.
I’m a bot and I’m open source!
Does this affect older Ryzens as well? I’ve noticed some periodic hitching in games/videos cropping up on occasion.
Hard to say. It’s a firmware problem so presumably it could be affecting all of them. The CPUs aren’t involved. My guess given the details mentioned by Linus is that it could affect some boards and not others, depending on what else is going on in the boards’ firmware at the time when the rand call to is made. In other words, there might be other bugs present in the firmware of say ASUS that trigger this, which aren’t present on MSI. And then there might be bugs present in this or that particular model’s firmware. It’s a whackamole which is why they decided to disable it altogether.
Just to add a perspective from the other side of the fence, I have a gaming laptop running Windows 11 (yes I know) where this (or a very similar) issue has been plaguing Ryzen users for at least a year and a half. The issue is that TPM per se is not causing issues if turned on, but if BitLocker encryption is on it will cause occasional audio stutters and intermittent complete system halts. The only thing that reliably helps is completely turning off Bitlocker, the TPM chip can stay on and is of course needed for W11. OEMs and AMD have been digging their heads in the sand like ostritches and they have released the odd fix that does nothing to fix the underlying issue. I can’t see MS doing anything to reverse course on requirements and am getting a bit fed up with their BS lately, browsing what distro might suit me best and might pull the trigger and finally switch…
Have a look at Tumbleweed with KDE.
Thanks for the heads up, I’m distro hopping these days and looking for options on where to settle
Sweet. I’m switching to Ryzen next week and was hoping to see this resolved just in case.
You’ll need to use a bleeding edge kernel to get this patch unless it’s backported to older kernels by your distro’s maintainers. I doubt this will happen for say Debian or Ubuntu. Instead, you’d have to wait for a new HWE that has this new kernel or whatever the equivalent in Debian is.
You can also solve this problem by disabling the TPM in the BIOS settings, assuming your motherboard has such a setting. No TPM, no problem.
