Linux 5.15 Is A Very Exciting Kernel For AMD

[skeevy420]

Use ZFS for raid?

That's what I do.

My "multimedia" mirror started as a single 2TB disk back in 2015 two computers ago (Intel q6600)...split 75% ZFS and 25% NTFS...no shit... Went to 100% ZFS around 2016/2017, also changed motherboards (HP to an ASUS). Upgraded to a new (to me) PC about a year after that (Dell T5500, dual x5687) and moved my disk into that.

It then became a mirror on that PC in 2020/end of 2019, I attached a 4TB disk to my 2TB pool. Was such a tease having 2TB I couldn't access. A few months later I finally replaced the original disk when I built my current computer (custom R5 4650G build) and attached a 2nd 4TB disk to it.

I feel like all of that would have been a severe pain in the ass to do with any other file system or setup. I love being able to buy disks as I can afford them or as shit happens and it becomes necessary. I've gone through multiple motherboards, CPUs, operating systems, ZFS versions and features, and I'm not even using the same disk as when I started, but all of my important data and games have come with me.

If that's not a ZFS success story I don't know what is.

[mdedetrich]

I get that you never used an Intel Bay Trail or Cherry Trail device on Linux then? Those were really rough in the first years:

• 1000+ comment kernel bugzilla report that cstates causes Bay Trail systems to crash
• Intel not implementing major/essential features on Bay Trail and Cherry Trail platforms, leaving them to the community (the effort is still ongoing...)

It's not just some forgotten old Atom CPUs either. Intel's most recently launched Rocket Lake CPU's iGPU didn't work out of the box with the then current kernel because Intel did not include the necessary PCI IDs.

I am the first one to admit that AMD sucks at timely integrating new hardware support into the kernel, but Intel has got their fair share of duds too.

I don't think anyone is claiming that Intel hasn't had issues but at this point you are just cherry picking specific problems (which you can always do since you will always have bugs).

The difference between Intel and AMD has historically been that AMD generally has been problematic and not some specific SKU's. For example their sensor code i nthe Kernel has generally been terrible for all of the fairly recent lines of AMD CPU's all the way up until Zen3 (apparently).

I mean this is not surprising, Intel is much larger and hence has a much bigger engineering team specifically for Linux support. As stated before AMD is only solving these issues recently.

[wallcarpet40]

I hope to see Freesync HDMI work on my 5700XT.

[chithanh]

their day one CPU support has been lackluster, while Intel has been on the ball.

I get that you never used an Intel Bay Trail or Cherry Trail device on Linux then? Those were really rough in the first years:

• 1000+ comment kernel bugzilla report that cstates causes Bay Trail systems to crash
• Intel not implementing major/essential features on Bay Trail and Cherry Trail platforms, leaving them to the community (the effort is still ongoing...)

It's not just some forgotten old Atom CPUs either. Intel's most recently launched Rocket Lake CPU's iGPU didn't work out of the box with the then current kernel because Intel did not include the necessary PCI IDs.

I am the first one to admit that AMD sucks at timely integrating new hardware support into the kernel, but Intel has got their fair share of duds too.

[brent]

Intel for sure doesn't work 100% either. Intel's reputation is better but there's no good reason for that.

There have been various fatal hardware issues in the last decade that required Intel to replace hardware (both CPUs and chipsets). Intel had to deactivate various features after the fact due to brokenness (e.g. TSX) as well. AMD arguably had less hardware issues, even though they had well-known problems with Phenom and reliability issues with some early Ryzen CPUs.

I personally experienced and to some degree still experience the following problems:

- Intel WiFi driver/firmware issues with AX200 (firmware crashes, I have to reboot to fix it)
- Bluetooth issues with AX200 as well (random HCI errors and device disconnects)
- Frequency scaling and power usage regressions on older hardware
- RC6 power management issues plagued the GPU drivers for a LONG time (a year or so?)
- Panel self-refresh (PSR) is semi-broken on my XPS 13 (and I've reported these and assisted to fix... TWICE!)

[MadCatX]

I haven't had any Intel gear that 100% worked either. I'm not impressed with my experience with Intel hardware either. I'm thoroughly disgusted with how buggy practically everything is nowadays.

Exactly. When you buy a new rig nowadays, you don't ask yourself "Will I have any issues with it? but "How troublesome will it be this time?" (Still waiting for Lenovo to fix the battery drain issue on AMD T14 Gen1)

[grigi]

This kind of shell game is beloved of marketing pukes, but misleading and tiresome to the people whose money makes the company run: the customer. Hopefully it won't be as bad as the 200 series, which included:

Terascale
GCN 1.0
GCN 1.1
GCN 1.2

Buyer beware.

Yet the 7000 series right before it only had GCN1.0/1.1
The 200 series was a right mess.

[mdedetrich]

The thing I most want AMD to get their act together on is RAID. I've inherited a system which due to what it is used for is stuffed full of GPUs and has no room for a dedicated RAID controller card... while the AMD motherboard/software/fake/whatever-you-want-to-call-it works fine in Windows with the appropriate driver, Linux won't even see it. I tried mdadm RAID but the array falls apart on every reboot and I won't risk not being able to put it together again.

I'm not actually enormously fussed about per-core temperature monitoring. It was nice when I was into overclocking in a big way, but now I'm more interested in systems being stable for months at a time. My 3900X system is fairly well behaved, but the odd kernel regression does hit it occasionally - one landed yesterday; my wireless mouse now won't work through the KVM switch USB port unless I replug it after I'm at a desktop. Works fine with previous kernel version.

If you are talking about fake RAID (i.e. the RAID that is typically enabled in BIOS) then just don't use this RAID. It doesn't matter if its AMD or Intel, fake RAID historically has been trash for various reasons. Either get a HBA or in your case (since you don't have room) then just use software raid, mdadm or solutions like BTRFS/ZFS.

[PSSGCSim]

The thing I most want AMD to get their act together on is RAID. I've inherited a system which due to what it is used for is stuffed full of GPUs and has no room for a dedicated RAID controller card... while the AMD motherboard/software/fake/whatever-you-want-to-call-it works fine in Windows with the appropriate driver, Linux won't even see it. I tried mdadm RAID but the array falls apart on every reboot and I won't risk not being able to put it together again.

I'm not actually enormously fussed about per-core temperature monitoring. It was nice when I was into overclocking in a big way, but now I'm more interested in systems being stable for months at a time. My 3900X system is fairly well behaved, but the odd kernel regression does hit it occasionally - one landed yesterday; my wireless mouse now won't work through the KVM switch USB port unless I replug it after I'm at a desktop. Works fine with previous kernel version.

In my experience mdadm RAID is way more solid than any motherboard RAID and even some RAID cards. "mdadm RAID but the array falls apart on every reboot" this sounds more like you have not set it up properly or the on board SATA controller is faulty. You can also look at ZFS as a really solid alternative but it requires some additional installation under Linux.

[peterdk]

Use ZFS for raid?