Search results

@TrevorH - the kernel is afaik OK. I do have: IRQ_REMAP [=y] Depends on: IOMMU_SUPPORT [=y] && X86_64 [=y] && X86_IO_APIC [=y] && PCI_MSI [=y] && ACPI [=y] As it is a pre-requisite for: X86_X2APIC [=y] Depends on: X86_LOCAL_APIC [=y] && X86_64 [=y] && (IRQ_REMAP [=y] || HYPERVISOR_GUEST [=n])...

Well, the BIOS setting does say Extended APIC, not x2APIC like I googled now and saw on some screenshots. Yet the EMR should have an x2APIC - so I assume that is dependant on the BIOS support then? X2 does exist from Intel Core era ... so.. why its not there on (my) ES EMR? (or is that a...

Hi friends. I was updating today my Gentoo on this MS03-CE0 and ran into a problem with X2APIC after I enabled it in the kernel. ACPI: Core revision 20251212 clocksource: hpet: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 76450417870 ns APIC: Switch to symmetric I/O mode setup x2apic...

To add to the last topic: Would the accelerator appear without the on-demand license? Or would the shown accelerators that appear on PCI bus refuse opeation without such a license? (any tip for a command to query the status of that?)

Had anybody tried to run own FW build (eg gentoo linux) on an SRX000 series (eg SRX320) ? Is it visible from the serial console, whether there is any sort of secure boot going on, or its a classic uboot process at the start? (afaik on a similar product - an USG, also with Octeon, its possible)

So finally replying to the original poster ... I got the same issues on Linux, but managed to add an entry for firmware that is available as of now: https://github.com/RemixVSL/vsl_downloads/issues/3

As I was informed - we have not seen a drop in the GFLops rating when the HT was disabled. So even 1 thread on 1 core can fully utilize the vector unit(s) present there.

I've got a pair of 8272CL / SRF89, but they do not work in X11DAi-N (tried CPU1 only, with either of the two). Same result - board powers on the PSU, but the power draw is ~25W and it does never jump to 50W+ as with a other working CPU (4208) on the same board. Do I still need to reprogram the...

Which systems use EC4 DDR5 UDIMMs? I have seen SODIMMs in that configuration but I am unble to figure out what systems are they meant for. Any tip appreciated! And also - would EC4 work as non-ECC in a system that was not designed for ECC ? like the majority of DDR4 ECC UDIMM work in a non-ECC...

Nice, I was not aware of EL1000, probably the easiest path would be to get me one as well. I have a collection of various Moonshot CPU modules, not all are listed as compatible though with el1000. Previously I have managed to power up one, and there was a bios on serial port (pin header on...

Has anybody started reversing the pinout of the Moonshot edge connector, or is it described somewhere? I would like to make a "micro-moon" single node appliance - for router / NAS / home server use, and them something like a smaller cluster of 3-4 nodes, targetting HA setups.

Hello guys, I have bought an Akamai Epyc mobo (with case and psu), this one (runs with 2nd gen cpu, 1st gen has 0x00 post code and wont boot, which is exactly the same behaviour as on my normal KNPA-U16 flashed for Rome): https://www.ebay.com/itm/125882817902 The state is basically the same -...

Yes, the trace which the resistors connect to - has the same voltage as the TPM Pin 9 (+3V is a typo, it is actually a normal +3.3V rail), which measures on my board +3.456 V. That is different from the +3.3VSB (pin TPM 15), which is powered also in power off (standby) state. I have to fix my...

Sure, here are the pictures. I have modified a C2750D4I rev 1.02, but also own a C2550D4I rev 1.03. That one had one of the resistors already there (first in group of three) and it measures 131.6 ohms (in-circuit), so my mods using 120ohm probably wont break anything. That signal is the TPM...

I have manged to bring my C2750D4I back to life by soldering the 5+2 resistors - mine missed two and your pictures shows soldering on those two places as well - one of them is in fact a TPM signal (IRQ), but they preferred to populate the unpopulated location within the fix, the remaining...