Non-server CPU compatibility - Ryzen and Intel
-
@olivierlambert - my friend with the Ryzen 7900 isn't experiencing the same issues that @BlueBadger and myself (with the 7900X) are having - his system is working fine (both of us are running xcp-ng-8.3.testing-2023.02.15-12.19-install.iso).
-
With the same motherboard and the same BIOS settings/version?
-
His is closer to your board (I believe), and to one that @BlueBadger tried (ASUS Prime B650M-A II-CSM)
My friend's board is: ASUS Prime B650M-A AX:
https://www.asus.com/us/motherboards-components/motherboards/prime/prime-b650m-a-ax/He's using the Ryzen 9 7900 and isn't experiencing any problems with Linux VMs in XCP-ng.
-
An idea investigation would be to swap the X and non-X CPU and see if there's a diff.
I'm under the impression it's more a motherboard issue (BIOS, or version) than anything else however
-
I talked to my friend about doing a processor swap test, but he's happy with the way his system is running and doesn't want to take a chance of messing something up. Sorry about that
-
Maybe there's others people in the community that could bring that info
-
I ordered a Ryzen 7900 last night and got it this morning. (Thanks Amazon).
I just replaced the 7950x with the 7900 and things seem to work better.I can run now a AlmaLinux 8 VM without the nopv flag.
I can now run Xen Orchestra without the nopv flag.I will do some more testing.
-
@BlueBadger - Thank you! Looking forward to see what else you learn.
-
I was having issues (download stalls) with the onboard 2.5Gb NIC (RTL8125) on the ASUS Prime B650M-A II-CSM motherboard even after I switched from the Ryzen 7950x to the 7900.
My setup also includes a X540 10Gb NIC which seemed to be working well.I swapped the motherboard back to the ASRock B650M PG Riptide and was still having issues with the onboard 2.5Gb NIC.
I disabled the onboard NIC and installed a second X540 and have not have any network issues so far.
I'm guessing there might be an issue with the r8125 driver.
Excluding the onboard 2.5Gb NIC, XCP-ng seems to run well on both motherboards.
The BIOS errors in dmesg don't seem to be causing any issues.
(The ASRock B650M PG Riptide seems like a nicer motherboard.) -
That'sā¦ interesting
So the "X" series seems to have some issues in the end? It's weird since it should be very different than it's non-X counterpart. -
So, we've had reports on xen-devel which look a little like this.
@BlueBadger are you able to switch back to your 7950x and try booting Xen with
x2apic_phys=true? It appears that the -X processors are missing a feature in their IOMMU and Xen was getting confused when setting up interrupt handling.https://xenbits.xen.org/gitweb/?p=xen.git;a=commitdiff;h=0d2686f6b66b4b1b3c72c3525083b0ce02830054 is at least part of the fix, but so far feedback on the mailing lists suggests it's not a complete fix.
-
@andyhhp Thanks for the info.
I plan to leave my current machine (Ryzen 7900) as is since it seems to be running well.
I plan to build a new machine with the extra 7950x. The motherboard is on back order.
I will try the new setting once it is built. -
@andyhhp I built a new machine with my Ryzen 7950x.
Booting Xen with x2apic_phys=true did not seem to fix any issues.
-
Interesting, thanks for the feedback. @andyhhp should we provide a Xen version with the initial fix and see if it's better? (maybe combined to the x2apic param)
-
I'm testing this combo:
- AMD RYZEN 9 7900X
- ASUS PRIME X670-P WIFI bios 1406
- 2x32GB KINGSTON 5600 CL40 (max QLV)
- Boot drive NVME 250gb (chipset) and SN850X 4.0 1TB on CPU.
With setting: Local APIC Mode = X2APIC and UEFI set to Other OS. Installed 8.3 alpha and updated, got errors. Test installing XOA took too much time and booting was painfully slow using only 1 SSD on chipset NVME.
Tried disabling IOMMU, but the same issue.
-
It's possible we got a fix, but we need sometime to generate the right build/RPM for you to test
-
@olivierlambert said in Non-server CPU compatibility - Ryzen and Intel:
It's possible we got a fix, but we need sometime to generate the right build/RPM for you to test
I have to identical setups for testing. With all the storage being pcie and interrupts. I also got a instlall with proxmox, but kind of same issues with IOMMU, trying to passtrough the 4.0 nvme. The disk was very slow, CPU wise performance on W11 machine was great.
-
I'm almost sure the future patch will fix the slow CPU behavior
Hang on! -
@olivierlambert said in Non-server CPU compatibility - Ryzen and Intel:
I'm almost sure the future patch will fix the slow CPU behavior
Hang on!Any update on when this fix is available. I got a x series cpu and am having these same problems.
-
We need to take a closer look, it's in our backlog but we also have other stuff pending (as you can imagine). Let me ping @stormi about this.
