Jerky_san

Things are looking good so far for me.

o-o Thank you @limetech installing now

Turns out it is CPU limiting.. The processes called "unraidDD" with the number I assume of the disk behind it are all spiking to 100% all the time. It also appears there are more b2sum processes running than originally meets the eye. I'm unsure though why the unraidDD process is requiring so much processing power but assume it has something to do with the lookup of files on that particular disk.

5x 160-190MB/s = 800-950MB/s Also during things like parity checks it(with the extra 2 drives for parity) it will go an average of 110 megabytes a second average check speed(which is a limitation of the back-plane. Since the back plane only has a single SAS connection 6GBx4 + overhead.

So having an issue or two.. First Full Disclosure Processor 2990wx - 24 physical cores 48 threads available for use Ram 128GB Using BLAKE2 for hashing I was doing checks to see how many disks I could check at once.. Started with All and started killing checks starting with disk 1. I noticed however that disk 1's job never stopped for some reason. It has kept running. That is my first issue. It appears repeatable as well. When I kill disk 1 it appears to kill another job instead. Problem two is I can seemingly only do 5 disks at a time if I want any kind of speed. After that the MB checked per second drops substantially. If I do all drives at once it will drop to about 10mb a second(18 Drives). So looking at top I noticed two processes seem to run at once per drive. So I cut it down to 12 drives thinking 12x2 = 24 and no one core was maxing so should be great. But it only raised it to about 30 megabytes a second. A 3x increase but still was hoping with as many physical cores as I have it would be able to do more. Is this expected or perhaps I have something configured wrong?

Thanks I can't wait. Really need the new temp monitoring

Welcome to the party.. I'm glad lime technology is growing like it is.

We need a lot more information.. What errors are you getting? Diagnostics file?

Basically there was a person on the VM part of this forum with a 3970 having performance issues. It is related to the chiplet design and how the different CPU cache levels are presented. They were getting less performance than expected due to (I'm assuming) cache hit misses. The 5.x kernel is coming in 6.9(whenever that maybe) though I've actually made a special request in the "General Support" board about potentially getting an RC1 with JUST the kernel. Anyways Ryzen in general requires more tuning then Intel so just keep that in mind. My 2990wx had a lot of tuning and tweaks in the beginning that have been slowly fixed by patches on of QEMU in unraid releases. I assume the next release will have QEMU updates for the 3k series to fix the cache as well. If I had a 3k series I'd probably know more about fixing the issue but honestly don't know if I'm going to upgrade again for a while. Been waiting to see what 4k threadripper series brings.

TO my knowledge it will with some limitations.. Limitations include QEMU isn't fully supporting the 3k series yet and without the 5.x kernel temp monitoring will be fubar..

I was wondering if we could get a timeline of when it will come out or maybe just a 6.9RC1 release with the new linux kernel 5.x series? I assume 6.9 will have a lot of new features but the temp monitoring and other really nice implementations are very nice and I've been seriously considering trying to downgrade to a 6.8RC release with the kernel though I'd prefer to stay on the latest. Would be nice if RC1 just contained the kernel update for people on ryzen series that need it. P.S. Know it's not really a support thing but didn't honestly know where to stick it. Thank you for reading.

Following up on this.. any detailed release plan coming up? Really really want the improved kernel stuff. Honestly debating downgrading at this point and applying the vulnerability patches. Btw Not directed at you testdasi =0

Go to your bios.. look for power supply idle. You will want to set it to "normal idle power". See what happens

Yeah I'm excited for the improved monitoring for temps and stuff.

Thank you very much for the thorough explanation.

It did really help with my ftp transfers. Program no longer locks up. Do you believe it will eventually be able to run with hard links enabled or is that kind of up in the air at this time? Also sorry for not completely reading the thread before responding.

There are certain versions you will need to roll back to and there is actually a specialized kernel as well. But hopefully limetech will release 6.9rc1 and it will also do this..

I see nothing on the "OP"s post or profile that says any of this.

So everything is set to stock? No DOCP no memory set higher than 2133 or whatever "base" your board is any of that? (Just confirming)

We found yesterday there are some issues with the 3rd gen ryzen and caching. There isn't a way to fix it 100%. Technically your cache should be fixed with the below but doubt it will.. If it works you should see a boost in performance but there is a fix needed for virtio/qemu <cpu mode='host-passthrough' check='none'> <topology sockets='1' cores='8' threads='2'/> <cache mode='passthrough'/> <feature policy='require' name='topoext'/> <feature policy='disable' name='monitor'/> <feature policy='require' name='hypervisor'/> <feature policy='disable' name='svm'/> <feature policy='disable' name='x2apic'/> </cpu>

almost sounds like you have a memory problem. If it was caused by the idle stuff you'd have it freeze on you instead. Are you overclocked on anything CPU/Memory?

I experience this as well. When I'm doing SFTPs to these folders it can cause my whole vm to lag in an attempt to list the folders. It can take a few seconds to bring up all the folders but honestly I have the folders as condensed as I can without plex hating me.

Hmm that is very interesting.. Your the first person to have a 3960 here I believe. At least that I've seen talk about it. As to your question I would say yes but I got a feeling your going to get slapped with performance hitching. My test game for this is "Dying Light" as it has a very good multicore implementation. The Good news though is you won't have to wait long as AMD is basically the performance crown in almost all types I figure they will start really ramping up fixes/patches in QEMU/VirtIO. I'll say though it took nearly a year for the 2990wx to start "running well" compared to baremetal sadly. It would seem they are rolling fixes to GA faster via QEMU/VirtIO so hopefully they will be picked up by limetech as quick as well.(generally they are as limetech seems to always deliver 😃 ) Any ways if you have dying light try playing it and see if you notice any hitching or frame drops. Don't play with gsync on btw. At least last time that was discussed it caused a lot of fun issues itself. You might also want to check what your DCP latency is along with what kind of memory performance your getting inside your VM with AIDA. If your latency is very high on your L1-L3 you maybe in for a ride. I get very close to bare metal on all performance metrics that I test but it was a long time of tuning. You can also check out reddit's virtio subreddit.(guess you have been but sometimes the threadripper convos get buried sadly) They don't have a lot of threadripper convos but there have been some from time to time that are very helpful. Baremetal 2990wx AIDA64 VM 2990wx AIDA64

Does appear that because you have the newer arch it isn't doing the caching properly. On the 2990wx I have(and my old 1700) it was basically a requirement. What you could "technically" do is try the old way we fixed it before they fixed it on QEMU. Adjust cores below to match whatever your doing. It will be half of whatever you have assigned.. This will pass it as an "EPYC" processor. See if CPUZ sees your cache the same as your baremetal with this. If not you may have to wait till they resolve the issue. <cpu mode='custom' match='exact' check='full'> <model fallback='forbid'>EPYC</model> <topology sockets='1' cores='22' threads='2'/> <feature policy='require' name='topoext'/> <feature policy='disable' name='monitor'/> <feature policy='require' name='hypervisor'/> <feature policy='disable' name='svm'/> <feature policy='disable' name='x2apic'/> </cpu> Cache matching with settings. If your cache doesn't match you'll get hitching and stuff due to cache hit misses. It is ESPECIALLY important for L1 and L2 but also important for L3 given how much cache the 3960 has. 2990wx Baremetal CPUZ 2990wx VM CPUZ

Could be that his cores are mismatched. It happened in earlier versions of the 2990wx bios where the cores shown were not the matching physical/hyper threaded. When your making an XML file your basically telling the machine what the host looks like. I remember some people had to manually assign even the cores in the right order. It's just a thought.