Daemon Segmentation Fault on X570 + Ryzen 3

[olejon]

I updated CoreFreq a couple of days ago using my automatic script and it worked just fine.

Now I did it again, and loading the module works but the daemon does not load and the UI shows nothing.

---

# corefreqd -d

CoreFreq Daemon 1.78.1  Copyright (C) 2015-2020 CYRIL INGENIERIE

Processor [AMD Ryzen 5 3600X 6-Core Processor]

Architecture [Zen2/Matisse] 12/12 CPU Online.

SleepInterval(1000), SysGate(2000), 2326 tasks

    Thread [7fa23dc86700] Init CHILD 000
    Thread [7fa23d485700] Init CHILD 001
    Thread [7fa23cc84700] Init CHILD 002
    Thread [7fa237fff700] Init CHILD 003
    Thread [7fa22ffff700] Init CHILD 004
    Thread [7fa235ffb700] Init CHILD 008
    Thread [7fa2367fc700] Init CHILD 007
    Thread [7fa236ffd700] Init CHILD 006
    Thread [7fa234ff9700] Init CHILD 010
    Thread [7fa22f7fe700] Init CHILD 011
    Thread [7fa2377fe700] Init CHILD 005
    Thread [7fa2357fa700] Init CHILD 009

Segmentation fault

---

Setup is same as before when you asked about my entire setup. I see you did some changes recently.

[cyring]

More than 195W is scarry and with 82°C, I wonder when thermal throatling is going to engage ? I'm gonna review the driver to see if I had set a thermal sticky bit into code, which translates into HOT being highlighted.

[olejon]

• First of all it's an official Enabled/Disabled setting in UEFI and the Help Text simply says something like "Can boost performance at the cost of power saving"
• But when Idle, Temperature and Watts = normal
• Also my 3600X with the stock cooler often got above 75°C, which AMD considers "High" for all these CPUs, and I've never had problems with the temperature being about what Intel (85°C) or AMD consider "High"
• A MacBook Air, many other laptops and iMacs too with Intel reach close to 100°C ("Hysterical") quickly with just some moderate multitasking, as easily seen as they start to throttle. Then CoreFreq correctly for my MacBook Air with Haswell reports 3 indicators for HOT, but there's no way to avoid that and they're built that way, and BIOS set to throttle instead of shut down for CPU protection
• When my workstation ran Skylake i5 I set all cores to Turbo and overclocked all so the CPU reached 95°C under stress without problems, just used a sensor tool to simply reboot the PC if I wasn't using it, and for some reason say some process went crazy and it reached 96°C (never happened but tested and worked fine)
• I have only read that the "High" value is a temperature you generally want to stay below but it definitely can go higher in periods, just prolonged close to "Hysterical" can damage it over time, but reaching "Hysterical" causes the PC to shut off (or throttle if that's set in BIOS)
• If any Intel laptop goes 85°C+ constantly, it can't be a problem for an AMD desktop CPU to go 82°C for a few minutes? Only benchmarks cause this
• And regarding benchmarks, Blender Benchmark did NOT cause an even higher temperature BUT 203 Watts. Temperature actually floated at a slightly lower one than Conic IIRC
• Now, Watts is another thing. Blender Benchmark showing 203 Watts, wow. My PSU is just 500 Watts so kind of surprised, so:
• Must check Ryzen Master if CoreFreq's reading is right
• Also must check if Ryzen Master gives any warning when running Blender Benchmark on Windows
• BTW with this I got exactly 1 second less on both your results from Blender Benchmark :smiley:
• Will read a bit more about this option before stressing the system again, but it's nowhere close to this at regular use, it is now at 14 Watts and 35°C as ambient is 29°C

[cyring]

Thanks for your temperature calibration effort. Can't wait to see your results.

Meanwhile I've checked code and driver has no ThermalTrip query in families 17h & 18h. Thus no HOT will appear with Zen processors. Perhaps kernel is logging it. I will search for such thermal status register which may be queried onto SMU

[olejon]

Many bullet points. Please read them all.

• You made me curious although I haven't slept yet :stuck_out_tongue:
• First, correction, according to k10temp "HIGH" is 70°C, BUT forget that because:
• Have to look at Ryzen Master, which only lists 1 temperature and that's called Limit 95°C, as said before. Goes for all the Gen 3 CPUs
• I ran a full Blender Benchmark classroom and the system was totally stable
• Ryzen Master did not even reach Orange color for temperature or any other
• I am almost 100% sure I saw the Orange color on my previous 3600X when stressing that
• In fact I manually overclocked my 3600X to Ratio 41 (4100 MHz), just 1 above standard 40, and when stressed it didn't like that, temperature rose and the system simply shut off, and I could without problems boot into UEFI and revert that setting to default. Same with my Skylake i5
• Temperature towards the end was at 85°C, but still Yellow color and NO warning
• As said, on Linux that temperature is lower during the same Blender Benchmark run
• Ryzen Master now shows 395W, and 50% of that is 197.5W as in Screenshot 2
• Ryzen Master Screenshots when Idle (almost, I did some things) and During Blender Benchmark (also see points below these):

Almost Idle

During Blender Benchmark

---

• Absolutely no warnings from the kernel when pushed up above 80°C. I checked with dmesg, journalctl, sensors (which shows an alarm/alert for certain HW) and used grep on ALL files in /var/log and there was nothing matching any relevant keyword
• The last message was simply that CoreFreq was loaded, others were standard stuff from server services and desktop environment
• Blender Benchmark classroom finished faster on Windows as well, but result isn't valid as I had Ryzen Master open + I took a screenshot during the run
• Blender Benchmark is very sensitive to any other stuff going on (system should be totally idle), and it stresses both Windows and Linux so much for me that the desktop lags a bit and is a bit slower to open say the file manager
• The below says "... expense of idle power savings", which may be true for the RAM but I don't see the CPU being more active when idle (and there's no integrated graphics as you know so that doesn't count)
• Screenshot of the Name and Description of the Setting in UEFI. I see nothing that screams Danger! Danger! Do you?

Both Red Rectangles say the same

[cyring]

Very interesting investigation you did, although I will go very slowly on those settings t'il we have a clear understanding on what is going on, and their side effects. I'm still not sure if hardware is well protected against over-voltage, over-consumption and so on. Don't fry your brand new Ryzen !

• This morning I went through the last PPR, this document revision applies to model 20h which concerns APU to my understanding. What we found is a list of registers not unveiled in previous families. Most of them can be queried through wire protocols: SM-Bus, i2C, GPIO, etc I bet AMD has some MSR to ease the readings but those are not public yet.

We find wanted registers a bit every-where, sometimes through :

• ProcHot, ThermalTrip : on SMIstatus, SmiControl5 & Power Management (PM) Registers

• C1e : PMx07E [CStateEn]

• IdleCount (aka "idle count from the latest monitored period") at MISCx060 [IdleCntrl] (FCH::MISC::IdleCntrl). Is this what RM called Sleep from ?

• Have a rest, you deserve it !

[olejon]

Don't fry your brand new Ryzen !

• How can that happen when:
• I ran a FULL Blender Benchmark with ALL the scenes, more stress regular usage will ever do
• SSH + CoreFreq to monitor Temperature and Watts
• Temperature: 78-85°C
• Watts: 173-203
• The scene victor took very long to load ("Warm up"), it seemed like it tried several times so I was on alert, but CoreFreq showed something was going on at least, and it did load and logical as it was the definitely longest one, but Watts during dropped to the lowest during that at times, to 173 Watts
• In fact 203 Watts only seen during the first 2 ones: bmw27 and classroom, but never above 85°C during those
• If going to 86-87°C it only took 1-2 seconds to drop back down to 85°C
• During pavillion_barcelona - I think, if not koro - it managed to push it to 87-88°C, but not above. Watts was well under 200 then
• BUT: If running those scenes again alone, it stays at 85°C and around 200 Watts, so I suspect it was because they load quickly and maybe didn't allow the CPU to cool down a little during meanwhile, maybe
• So maybe I have to repeat this to see if it happens again during 1 of those 2...
• Again my PSU is only 500 Watts, as I don't have an insane GPU, and don't plan to, as they freakin' now require a whopping typical minimum of 750 Watts PSU as 2 PCIe Power Cables are required (8 + 8 pins = 16 pins). Why didn't they fix this with PCIe4? Enough power in the slot please
• Or GPU manufacturers, please look at what Apple fits into some square centimeters if not millimeters on iPad Pros with NO fans...
• Really goes for all cabling, 1 standard please. I hope they look at USB-C + Thunderbolt, and use something like that for every cable (I like the design of SATA cables so e.g. they can keep that design). I've used the most fancy enterprise docks and they support tons of monitors, audio and I/O of all types, from 1 cable from a laptop

[olejon]

• Read the above first
• I can only find ASRock documenting "Uncore OC Mode" and it seems to have to do with RAM and Infinity Fabric only
• I've not touched any of these manual settings, just turned off Gear Down which should be totally safe, and another Power Saving setting for RAM which is a sub-menu under that part again
• Still system performs better. So is it simply the fact that it now actually uses my RAM to the max as D.C.O.P. = Enabled or what?
• All people discussing this setting also do manual overclocking (entering values manually) but I don't

This does NOT seem to discuss that setting but rather manual overclocking in the AI Tweaker part of the UEFI, although maybe interesting for you? Remember to see the 2 screenshots below too.

"OC mode" is automatically engaged when P0 frequency is raised above base frequency. In OC mode, the SMU inside Ryzen CPU is disabled, which means sophisticated power management and power saving features are not functional. The presense of OC mode can be verified by calling GetCurrentOCMode() via the RyzenMaster SDK or code 0C via a Port 80 debugger card. HWiNFO has 3 voltage readings related to CPU. If it's individual core VID or motherboard Vcore that is scaling down to 0.2 Volts, then that's expected. What's important is "CPU Core Voltage (SVI2 TFN)". I think this value can tell whether the CPU has gone into "OC mode" or not. If this value is not scaling down, then it most likely means OC mode is engaged

Screenshots from 2 Manuals for ASRock MBs. Not much info:

---

• PS: I've NOT touched this voltage!

---

[cyring]

• Read the above first

I did but I don't feel comfortable with the power consumed. 173-203 are far too high to my taste See my way in next post

• I can only find ASRock documenting "Uncore OC Mode" and it seems to have to do with RAM and Infinity Fabric only

I'm not familiar with the Zen Uncore to tweak it yet. Especially its voltage limit.

• I've not touched any of these manual settings, just turned off Gear Down which should be totally safe, and another Power Saving setting for RAM which is a sub-menu under that part again

I've also noticed this one and I don't know what's behind the hood. For now I have stabilized my DRAM at its OC factory mode and it will be easier to understand those extra settings when I will have a clear understanding of how the Ryzen Memory controller manages the [sub]-timings . So far, I have no specs how to query the configured DDR speed, geometry, latency, and so on

• Still system performs better. So is it simply the fact that it now actually uses my RAM to the max as D.C.O.P. = Enabled or what?
• All people discussing this setting also do manual overclocking (entering values manually) but I don't

I have chosen D.C.O.P(XMP) to auto configure the G.Skill kit.

"OC mode" is automatically engaged when P0 frequency is raised above base frequency. In OC mode, the SMU inside Ryzen CPU is disabled, which means sophisticated power management and power saving features are not functional. The presense of OC mode can be verified by calling GetCurrentOCMode() via the RyzenMaster SDK or code 0C via a Port 80 debugger card. HWiNFO has 3 voltage readings related to CPU. If it's individual core VID or motherboard Vcore that is scaling down to 0.2 Volts, then that's expected. What's important is "CPU Core Voltage (SVI2 TFN)". I think this value can tell whether the CPU has gone into "OC mode" or not. If this value is not scaling down, then it most likely means OC mode is engaged

Screenshots from 2 Manuals for ASRock MBs. Not much info:

• PS: I've NOT touched this voltage!

Those Windows tools are so lucky to get so many specs details from Manufacturers I'm trying to stick to Linux and to offer an Open Source software. Road is long.

[olejon]

• Correction: It is the scene fishy_cat that causes the highest temperature. See if you can replicate that, no matter settings
• I ran the full Blender Benchmark with Ryzen Master open and saw the exact same behavior as on Linux with CoreFreq open, all of Temperature / Watts / MHz
• Sat there during the whole run on Windows ready to close Blender Benchmark in case Ryzen Master showed 89°C, BUT it never did, as in Linux, reached 87-88°C during fishy_cat, other than that at 80-85°C
• Speed gain: From a little above 3800 MHz to a little above 4000 MHz, not massive but clearly better results
• Totally stable on both Windows and Linux and no alarms/alerts
• I also have enabled D.C.O.P. and not touched any of the values it sets (UEFI shows what you have changed when you choose "Save & Reset")
• I found my "Manual OC"!
• I've always wondered why there are 2 settings for Precision Boost Override (PBO)
• Ryzen Master only activates the one under AMD Overclocking
• The 1 in AI Tweaker > Precision Boost Override is what makes the difference
• Using the Drop-Downs to change both from Auto to Enabled + 100 MHz, the same as in the other, and BTW Ryzen Master still shows PBO is 100 despite setting it both places
• Again, I have not entered any value manually, just Drop-Downs
• Take a look at the differences in PPT, TDC and EDC, from 142 W to 395 W, NOT that it uses that, it uses 49-50% of that:

With both set to Auto

---

Set to Enabled and 100MHz

---

This is UEFI setting alone causes that big difference in RM's Max Values, not that they are reached at all

---

• In fact, if both set to Auto as in Screenshot 1, a Blender Benchmark causes PPT, TDC and EDC to go to Orange Color, almost 100%
• Again, not that it does anything, Temperature is the SAME, system is stable and it does not show any sign of wanting to shut down, no alarm/alert, nothing
• I think I will keep this "Overclock" on, as it's virtually impossible to reach above 70-75°C during regular use. Only benchmarks cause this... :man_shrugging:
• If it goes wrong for some reason, I will put back in my 3600X and let you know :smile:
• As Ryzen Master does NOT drop the MHz lower than when using CoreFreq, I guess a comparison is fine. Linux beats Windows, no surprise, especially in victor as it is the longest running one, but also the one with Lowest Temperature, actually:

Linux

---

Windows

[cyring]

Power tuning

BIOS settings: all [AUTO] beside a few

• D.C.O.P standard with XMP but Memory Frequency set to 3600MHz

• Applying a voltage offset decrease to use the minimum juice when CPU is idle but 4.7GHz single Core still capable

• As automatically set by XMP, DRAM voltage is kept at 1.35V

• Using Conic Compute... > Hyperboloid of two sheets which is the most computing algorithm, I'm getting ~ 145W Package with a max Temp of 67°C at 1.3V

• BIOS: PBO disabled, XFR disabled, CPB enabled

• SMT available

• DDR timings: Beside Manufacturer's 16-16-16-36 values, all set to [AUTO]

• CPU Power Thermal Control set by first POST. I don't know what's the purpose of this setting ?

Day to day working state

• Single Core can go as high as 4.7GHz at 1.5V
• An idle session such as Programming, Web surfing, PDF readings, etc claims less than 1V

[olejon]

That came quick! Remember to see the one I just commented.

Then we only have 2 differences in UEFI:

• I have NOT touched CPU Core Voltage Offset. No manually entered value at all
• I have PBO enabled to 100 MHz both places
• Ryzen Master shows no PBO difference because of that, says Boost Override CPU = 100 and OC Mode = Auto Overclocking, which sets the UEFI AMD Overclocking setting automatically, but not the one in AI Tweaker, which for you is called Extreme Tweaker (same settings though)

---

• I see you only get 1 reading under CoreFreq > Voltage, while I get for all of them
• You know why probably?
• You get a slightly better speed for Hyperboloid of two sheets, while I get a slightly better for Single Core, and both are totally stable in MHz
• I see virtually no fluctuations during the full Blender Benchmark, all cores stay at exactly the same frequency, around 4065 MHz (think it varies a little depending on scene)
• Screenshots of the same:

Temperature stays at 50°C

Temperature eventually reaches 80°C

Idle less than 1V - 33°C with Ambient 27°C

[cyring]

How CoreFreq is reading a Vcore of 0.2V : is it a voltage you set in BIOS ?

[olejon]

You mean the one at bottom right? Saying V[0.20]. Is that bad or anything?

I'm not into Voltage tuning AT ALL.

I haven't set ANY integer or string manually, only Drop-Downs. Only the D.C.O.P. changes automatically what you already know:

1. RAM voltage to 1.35
2. RAM timings, for me 16-18-18-36
3. RAM frequency for me 3200 MHz

Yours show V[0.91]. What's the diff?

[cyring]

You mean the one at bottom right? Saying V[0.20]. Is that bad or anything?

I'm not into Voltage tuning AT ALL.

I haven't set ANY integer or string manually, only Drop-Downs. Only the D.C.O.P. changes automatically what you already know:

1. RAM voltage to 1.35
2. RAM timings, for me 16-18-18-36
3. RAM frequency for me 3200 MHz

Yours show V[0.91]. What's the diff?

OK, so it's a CoreFreq bug Adding it to the todo list

[olejon]

• Which reading? Yours or mine?
• Could it be because I have Uncore OC Mode enabled?
• A bit tired of rebooting and checking, people also use my server services during the day

[olejon]

Hmm, look at previous IDLE Screenshot:

https://github.com/cyring/CoreFreq/issues/186#issuecomment-646917760

• It says V[0.91]
• Suggesting it's because of Uncore OC Mode + AI Tweaker PBO which wasn't on when that screenshot was taken

[cyring]

Manual OC

• Let's do things differently

  BIOS: we're disabling CPB, PBO, XFR

  Vcore is restored to [AUTO] which means that Processor posts at less than 1V

• Ready to reboot

  Doing the OC manually

  Idle state

  

• BOOST register is confirmed OFF To answer to your previous question: Voltage Core scope is changed to Package in Settings menu

  Experimental mode

• Let's increase the P-State ratio to 42 for all CPUs and stress my best Core
• Let's go straight to AMD specs with a boosted ratio of 47
• But can we go above ? Let's try one bin This 4.8GHz frequency appears very shortly but fallback immediately to 4.7GHz Like something is monitoring against this P-State changed. Interesting isn't it !

[olejon]

I suppose :man_shrugging: :smile:

• You're using the CoreFreq CPU driver, no?
• You have CPB 2 places in your UEFI? Maybe I do too but haven't looked, just found it 1 place, but I know it hasn't been touched (tried disabling it once as said and everything went slower indeed)
• BTW, as said my Voltage at Idle was 0.91V as yours before enabling Uncore OC Mode
• When it now says 0.2V, which Value in Ryzen Master should I look at? Tell me the name from the Screenshots of Ryzen Master

Differences:

• You have faster RAM
• Your MB is same brand but costs twice as much so it would be weird if you don't get better results than me with the same settings (you do, I must have Uncore OC Mode Enabled to get same Blender Benchmark results)
• Newer kernel (my experience = better performance with every version for Ryzen Gen 3)
• As said you use the CoreFreq CPU driver(?)

[cyring]

I suppose

• You're using the CoreFreq CPU driver, no?

Yes all P-State tweaks done with CoreFreq. But I've not programmed a governor for AMD yet. Thus OSPM is under the control of the Linux ACPI module.

• You have CPB 2 places in your UEFI? Maybe I do too but haven't looked, just found it 1 place, but I know it hasn't been touched (tried disabling it once as said and everything went slower indeed)

Yes, the Crosshair is also showing this CPB issue, 2 places for the same name. It's confusing.

• BTW, as said my Voltage at Idle was 0.91V as yours before enabling Uncore OC Mode
• When it now says 0.2V, which Value in Ryzen Master should I look at? Tell me the name from the Screenshots of Ryzen Master

I believe it is called CPU Voltage under Voltage Control

Differences:

• You have faster RAM
• Your MB is same brand but costs twice as much so it would be weird if you don't get better results than me with the same settings (you do, I must have Uncore OC Mode Enabled to get same Blender Benchmark results)
• Newer kernel (my experience = better performance with every version for Ryzen Gen 3)
• As said you use the CoreFreq CPU driver(?)

Perhaps but the Processor P-States remain what they are, whatever the bells and whistles of the motherboard. My goal here is to control the P-States: next steps will be to set some manual frequency ratios in BIOS then dump the registers changes within CoreFreq driver to understand how BIOS is stabilizing frequencies...

[cyring]

You mean the one at bottom right? Saying V[0.20]. Is that bad or anything? I'm not into Voltage tuning AT ALL. I haven't set ANY integer or string manually, only Drop-Downs. Only the D.C.O.P. changes automatically what you already know:

1. RAM voltage to 1.35
2. RAM timings, for me 16-18-18-36
3. RAM frequency for me 3200 MHz

Yours show V[0.91]. What's the diff?

OK, so it's a CoreFreq bug Adding it to the todo list

Checking code, threshold is 0.15V and it applies only to the lowest voltage when reading falls down to zero. Also threshold is not applied to current value (middle column) So it might not a bug and the voltage has been read at 0.2 for some reasons. https://github.com/cyring/CoreFreq/blob/ab750ba6df881c4a8e959797a23689800f54ce19/corefreq.h#L492

[olejon]

• Well I'll check later what Ryzen Master says with current config totally idle
• Then I'll Disable Uncore OC Mode and the AI Tweaker > PBO option and recheck RM and CoreFreq (surely goes back to 0.91V at idle)
• Not like I need the extra speed in day to day use
• BUT a good way to:
  • Confirm that my Noctua fan holds its promise about best turbo/overclocking headroom, using only 1 of the fans
  • Check that Linux beats Windows also when overclocked
  • Fun to push HW to its limit when only danger is that it does a soft shut off (hasn't happened, won't push more as typical for Intel where you kind of have to since stock sucks)

[olejon]

• As seen in Screenshot, CoreFreq is NOT reading the wrong Voltage. Ryzen Master also goes down to 0.2V in when Idle, although, since it's Windows, it goes much more up and down, but definitely in the 0.2V area many times during 1 minute
• When reverted to Uncore OC Mode = Disabled and AI Tweaker > PBO = All Auto, CoreFreq again shows V[0.91]
• BUT: Conic Compute > Hyperboloid of two sheets doesn't even manage to bring the CPU Temperature to 70°C (stays in the 60's)
• Voltage goes to 1.07V during that stresstest
• So why lose performance? Now it's just overkill cooling and CPU & PSU can handle the OC it seems
• I have a Wattmeter I can plug in between wall outlet and PSU and see how much the PC in total uses during a Blender Benchmark. Would be interesting to see how close to the PSU's 500W it comes

[olejon]

The 0.20V screenshot was a "lucky one":

• Voltage changes like this, in order of most frequent reading: 1 0.91V 2 0.96V 3 0.20V

So it's kind of like Ryzen Master, only that RM shows way more levels between 1-3 (and many more decimals, not that that matters).

I'm back to the OC mode. One can set a maximum temperature in AI Tweaker > Precision Boost Override too say avoid it never reaches 90°C, which I guess is when the Orange Color shows, but I have never reached above 89°C, as said, and that was just briefly before down again to 87-88, and only during Blender Benchmark fishy_cat.

[olejon]

I tend to check the ASUS website if there is a BIOS update. Saw there was one from June 17 published. Probably one for you too?

• After installing it, Linux boots fine, but see these and journalctl shows the IRQ stuff in Red

juni 24 00:52:21 tux kernel: do_IRQ: 1.55 No irq handler for vector
juni 24 00:52:21 tux kernel:   #2
juni 24 00:52:21 tux kernel: do_IRQ: 2.55 No irq handler for vector
juni 24 00:52:21 tux kernel:   #3
juni 24 00:52:21 tux kernel: do_IRQ: 3.55 No irq handler for vector
juni 24 00:52:21 tux kernel:   #4
juni 24 00:52:21 tux kernel: do_IRQ: 4.55 No irq handler for vector
juni 24 00:52:21 tux kernel:   #5
juni 24 00:52:21 tux kernel: do_IRQ: 5.55 No irq handler for vector
juni 24 00:52:21 tux kernel:   #6
juni 24 00:52:21 tux kernel: do_IRQ: 6.55 No irq handler for vector
juni 24 00:52:21 tux kernel:   #7
juni 24 00:52:21 tux kernel: do_IRQ: 7.55 No irq handler for vector
juni 24 00:52:21 tux kernel:   #8
juni 24 00:52:21 tux kernel: do_IRQ: 8.55 No irq handler for vector
juni 24 00:52:21 tux kernel:   #9
juni 24 00:52:21 tux kernel: do_IRQ: 9.55 No irq handler for vector
juni 24 00:52:21 tux kernel:  #10
juni 24 00:52:21 tux kernel: do_IRQ: 10.55 No irq handler for vector

...

juni 24 00:52:21 tux kernel: ACPI BIOS Error (bug): Failure creating named object [\_SB.PCI0.GPP8._DSM], AE_ALREADY_EXISTS (20190703/dswload2-327)
juni 24 00:52:21 tux kernel: ACPI Error: AE_ALREADY_EXISTS, During name lookup/catalog (20190703/psobject-221)

• Any ideas?
• The BIOS update seems to be AMD only stuff. You know what this is?
• After install it said it was updating the "Led firmware" and nothing more before booting
• Changelog:

PRIME X570-PRO BIOS 2203
Update AMD AM4 AGESA PI 1.0.0.1

[cyring]

I tend to check the ASUS website if there is a BIOS update. Saw there was one from June 17 published. Probably one for you too?

Thanks for the tips

• After installing it, Linux boots fine, but see these and journalctl shows the IRQ stuff in Red

juni 24 00:52:21 tux kernel: do_IRQ: 1.55 No irq handler for vector
juni 24 00:52:21 tux kernel:   #2
juni 24 00:52:21 tux kernel: do_IRQ: 2.55 No irq handler for vector
juni 24 00:52:21 tux kernel:   #3
juni 24 00:52:21 tux kernel: do_IRQ: 3.55 No irq handler for vector
juni 24 00:52:21 tux kernel:   #4
juni 24 00:52:21 tux kernel: do_IRQ: 4.55 No irq handler for vector
juni 24 00:52:21 tux kernel:   #5
juni 24 00:52:21 tux kernel: do_IRQ: 5.55 No irq handler for vector
juni 24 00:52:21 tux kernel:   #6
juni 24 00:52:21 tux kernel: do_IRQ: 6.55 No irq handler for vector
juni 24 00:52:21 tux kernel:   #7
juni 24 00:52:21 tux kernel: do_IRQ: 7.55 No irq handler for vector
juni 24 00:52:21 tux kernel:   #8
juni 24 00:52:21 tux kernel: do_IRQ: 8.55 No irq handler for vector
juni 24 00:52:21 tux kernel:   #9
juni 24 00:52:21 tux kernel: do_IRQ: 9.55 No irq handler for vector
juni 24 00:52:21 tux kernel:  #10
juni 24 00:52:21 tux kernel: do_IRQ: 10.55 No irq handler for vector

...

juni 24 00:52:21 tux kernel: ACPI BIOS Error (bug): Failure creating named object [\_SB.PCI0.GPP8._DSM], AE_ALREADY_EXISTS (20190703/dswload2-327)
juni 24 00:52:21 tux kernel: ACPI Error: AE_ALREADY_EXISTS, During name lookup/catalog (20190703/psobject-221)

• Any ideas?
• The BIOS update seems to be AMD only stuff. You know what this is?
• After install it said it was updating the "Led firmware" and nothing more before booting
• Changelog:

PRIME X570-PRO BIOS 2203
Update AMD AM4 AGESA PI 1.0.0.1

Have you clear the CMOS right after the update ? If not, you have garbage values in the nvram

[olejon]

The BIOS always resets to default values after an update and I go through all the settings again and set them as before (and check if new ones), instead of loading a saved profile.

Hm. Maybe a total power off then (PSU too). If not, does a hard CMOS reset ("short circuit" pins) do more than resetting the BIOS in UEFI, really? Fixing NVRAM shit is typically a Mac thing. Think many knows Option + Command + P + R lol (and the SMC reset one).

Had something similar, don't remember when, if was switching to Ryzen or the AMD GPU, PCIe3 to 4 NVMes... That was fixed quickly by an Ubuntu update.

[cyring]

The BIOS always resets to default values after an update and I go through all the settings again and set them as before (and check if new ones), instead of loading a saved profile.

Hm. Maybe a total power off then (PSU too). If not, does a hard CMOS reset ("short circuit" pins) do more than resetting the BIOS in UEFI, really? Fixing NVRAM shit is typically a Mac thing. Think many knows Option + Command + P + R lol (and the SMC reset one).

Had something similar, don't remember when, if was switching to Ryzen or the AMD GPU, PCIe3 to 4 NVMes... That was fixed quickly by an Ubuntu update.

It still written today in the ROG manual to clear the BIOS RAM after an update. To my knowledge, nothing is automated. A jumper should be on board ?

[olejon]

• Well tried that... The jumper actually resets less than the BIOS update... Any BIOS update does the same as Load optimized defaults (F5)
• With the jumper it's immediately clear when I see my favorites in UEFI they're still there. After F5 + Save, I see that short circuiting the jumper actually keeps some AMD RAM stuff (they're not cleared by jumper just UEFI (F5)
• I always update the BIOS over the Internet in UEFI and it always says Update successful. System will Reset! (and must set all settings again, or load previous profile but I never use older profiles on new BIOS)
• I took out the RAM chips for a while, for another reason
• Still bug. It's clearly a BIOS bug, kernel says so... (Windows maybe too if digging in logs)

juni 24 04:02:08 tux kernel: ACPI BIOS Error (bug): Failure creating named object [\_SB.PCI0.GPP8._DSM], AE_ALREADY_EXISTS (20190703/dswload2-327)
juni 24 04:02:08 tux kernel: ACPI: [Firmware Bug]: BIOS _OSI(Linux) query ignored

• Meh, unless I notice any regression. Assume this will be fixed in the not too distant future

[cyring]

Some ACPI Device Specific Method _DSM linked to GPP8 in System Bus _SB

Search for GPP8 in whole /sys/devices/LNXSYSTM Mine is located at :

cat /sys/devices/LNXSYSTM:00/LNXSYBUS:00/PNP0A08:00/device:0f/path
\_SB_.PCI0.GPP8

[olejon]

Nothing...

/sys/devices/LNXSYSTM:00# find ./ -name "*GPP8*"

Suppose there's no point in reinstalling the linux-firmware amd-microcode packages. Will check with Ubuntu 20.04 Live USB if also there. Just in case.

Think my MB has BIOS Flashback to revert, but I've never used such a feature. Next time I'll google properly before updating. The April BIOS was good.

Did you understand the changelog? Have you installed it successfully?

[cyring]

Nothing...

/sys/devices/LNXSYSTM:00# find ./ -name "*GPP8*"

Suppose there's no point in reinstalling the linux-firmware amd-microcode packages. Will check with Ubuntu 20.04 Live USB if also there. Just in case.

Think my MB has BIOS Flashback to revert, but I've never used such a feature. Next time I'll google properly before updating. The April BIOS was good.

Did you understand the changelog? Have you installed it successfully?

I won't install any BIOS. All features are working fine

Since Zen has been introduced, ACPI tables have been an issue. See the kernel history....

ROG CROSSHAIR VIII HERO had been released with:

BIOS 1201 11/18/2019

but it has however a remaining log trace ...

ACPI: [Firmware Bug]: BIOS _OSI(Linux) query ignored
ACPI BIOS Error (bug): AE_AML_PACKAGE_LIMIT, Index (0x000000007) is beyond end of object (length 0x6) (20200326/exoparg2-393)
ccp 0000:0c:00.1: ccp: unable to access the device: you might be running a broken BIOS.

... which is fine for CoreFreq because I'm working on my own [P][C]-States sub-drivers for Zen

My EFI entry tries to blacklist any of them:

title   Arch Linux
linux   /EFI/Linux/vmlinuz-linux
initrd  /EFI/Linux/amd-ucode.img
initrd  /EFI/Linux/initramfs-linux.img
options root=/dev/disk/by-label/root rw quiet break=n add_efi_memmap nmi_watchdog=0 selinux=0 loglevel=3 rd.systemd.show_status=auto rd.udev.log-priority=3 consoleblank=0 vt.color=0x03 modprobe.blacklist=pcspkr,nouveau,k10temp,acpi_cpufreq idle=halt cpu0_hotplug audit=0

which results in CoreFreq:

Linux:                                                                          
|- Release                                                       [5.7.4-arch1-1]
|- Version                      [#1 SMP PREEMPT Thu, 18 Jun 2020 16:01:07 +0000]
|- Machine                                                              [x86_64]
Memory:                                                                         
|- Total RAM                                                         32856504 KB
|- Shared RAM                                                          117756 KB
|- Free RAM                                                          30792460 KB
|- Buffer RAM                                                           79832 KB
|- Total High                                                               0 KB
|- Free High                                                                0 KB
CPU-Freq driver                                               [         Missing]
Governor                                                      [         Missing]
CPU-Idle driver                                               [         Missing]

Missing being the sign that nothing is in control, letting me choose Target P-States For example, decrease from default 35 to 22 but put back CPU #12 to 35

But target ratios are still moving on their own, resulting in CPU #12 = 36 and its SMT CPU #28 jumping between 43, 36 other CPUs are also not fully in control.

I believe its part of the collaborative stuff between Processor and some software, feeding back to it the requested Target. So far studying those entry functions: acpi_cpufreq_fast_switch , acpi_idle_enter and cppc_cpufreq_set_target

[cyring]

I'm noticing this strange behavior

• with acpi_cpufreq kmod blacklist and thus no cppc enabled (to my understanding)

• CoreFreq started with AutoClock=0 argument to read a BCLK of 100MHz

• based on the latest develop of CoreFreq which has been changed to display 3 items in the Target selector:

  1. current performance ratio
  2. current performance frequency
  3. programmed target ratio

• The frequencies plan as follow:

  |- Base Clock                                                          [100.000]
  |- Frequency            (MHz)                      Ratio                        
               Min    400.00                    <   4 >                       
               Max   3500.00                    <  35 >                       
  |- Factory                                                             [100.000]
                     3500                       [  35 ]                       
  |- Performance                                                                  
  |- OSPM                                                                      
               TGT   3500.00                    <  35 >                       
  |- Turbo Boost                                                         [ UNLOCK]
               XFR   4700.00                    [  47 ]                       
               CPB   4600.00                    [  46 ]                       
                1C   2800.00                    <  28 >                       
                2C   2200.00                    <  22 >                       

  >> meaning 3 enabled Ratio P-States: P0 = 35 , P1 = 28 and P2 = 22

• Now stressing one Core, at least one physical, preferably the best one which is CPU#0

  1. set all Target to lowest ratio P-state, value P2 = 22
  2. set its SMT Core, CPU#16 , to ratio P1 = 28 >> the obtained frequency is 2.8GHz
  3. but now set CPU#16 to P0 = 35 >> it triggers Turbo Boost !

• It's a strange behavior which mimics the Turbo conditions found in the Intel Nehalem architecture

[olejon]

Luckily it was easy to revert back to BIOS 1407 (April). It did enhance performance somewhat (not reverting but upgrading to that one from the previous).

1. Load optimized defaults (F5) in UEFI and save
2. Boot Windows (could've used Linux)
3. Format USB drive to FAT32
4. Download that BIOS file and put it on the USB
5. Opening the EZ Flash tool in UEFI
6. It automatically switched to the disk with the .CAP file
7. Flash it, and it says Update successful, System will Reset!
8. Back to normal
9. Manually go through UEFI settings again (right now all stock)

• Your BIOS version is the one AMD specifically asked people to install, fixing the "ABBA bug", after installing 1005 first, to fix PBO really not doing anything on many Ryzen 3rd Gen CPUs
• Indeed the manual says that one should not update the BIOS if the system is stable, ALL manuals for any brand/arch say this of course
• But my journey has shown improvements with every update, like benchmark performance and PCIe4 improvements, maybe also allowing modules like k10temp reading the temperatures, but that I don't know as that update came more or less at the same time as a BIOS update
• Also systemd-fstrim working finally, after not working the first half year or so (and can't be trimmed on Windows as no NTFS)
• IIRC both AMD and Corsair employees recommended to install all the 2019 BIOS updates, at the time I looked. Corsair to get the best from their new PCIe4 NVMes. Then using "Corsair Toolbox" on Windows to update the firmware. Only available on Windows, but didn't cause any harm to my NVMes which only has Linux on them
• There are 2 more 2019 updates than yours, 1404 & 1405

---

Changelogs for those:

1404 (2019):

1. Update AM4 combo PI 1.0.0.4 patch B
2. Support Ryzen™ 2000-series APU
3. You will not be able to downgrade your BIOS after updating to this BIOS version

1405 (2019):

1. Improve system performance

1407 (April 2020 one I run now):

1. Improve system stability
2. Add BCLK Frequency and SB Clock Spread Spectrum items into BIOS Advanced mode.

---

• I remember now what caused some kernel errors before (without problems). Was The radeon driver (IDK after system or BIOS update), but as said that was fixed within days by Ubuntu kernel and/or grahpic stack devs
• One thing I noticed with this June BIOS 2203 update was Ryzen Master actually showing all cores sleeping when system is idle, so some CPU stuff in it
• Why the June BIOS has such a drastic jump in BIOS version IDK, up to 2203. Only thing is Update AMD AM4 AGESA PI. Staying on 1407 until I see some more interesting changelog. Not installing it if it says one cant downgrade after updating. Also waiting at least 1 month and google first (Linux + 3950X)
• Regarding you seeing the CPU "denying" to adhere, I'm pretty sure I read many places after launch + YouTube videos that the Ryzen 3rd Gen CPUs have a lot of protection built-in. If they live in the BIOS, Motherboard or the CPU itself I don't know
• Basically have to use the UEFI tweaker, like setting Core Ratios for all to say 4200 MHz when turbo (ratio 42), but there's little headroom there without very good cooling (but safety should just turn off the system if too much OC, my experience for the 3600X), as it should, people should NOT have to fiddle like one basically must on Intel i9 CPUs to achieve the specs

[cyring]

Glad you successfully fallback to previous firmware and thank you for this detailed procedure. As I told you I'm not in a hurry to upgrade BIOS. There are still so many bits I have to discover and to program into CoreFreq

Using the latest version, are you also noticing than for a standard target performance ratio of 35, processors answers with a value of 36 ?

To monitor this, press key ! to toggle to absolute frequency

Are you also observing that 36 is the magic value which triggers Turbo Boost ?

[olejon]

• And guess what? JUST about 24 hours later came a full kernel stack update + updates to some core libraries like libc not shown here, which may have a fix for the bug :man_shrugging:
• Although say Ubuntu 20.04 has a newer kernel all bug fixes and performance updates are backported. And about 3 months after 20.04.1 or 20.04.2 has been released, 18.04 will get the 20.04 HWE stack, kernel and graphics, running the same
• But won't update to the June BIOS again to check heh
• Package names don't reflect kernel version, it's 5.3.18, but these came:

linux-image-5.3.0-61-generic linux-modules-extra-5.3.0-61-generic linux-headers-5.3.0-61-generic linux-headers-5.3.0-61 linux-modules-5.3.0-61-generic

• Can NOT test with a Live USB even with kernel hotswap as a reboot is required to see the bug-entries
• But yeah, after downgrading to 1407, they're gone now. BTW only thing showing in plain text after that update was also Updating Led Firmware, but that might have been really a downgrade from the June one
• As to your question, nope, I get the exact same values as the freq BASH alias I have on all my devices, which runs this command:

watch -n 2 cat /sys/devices/system/cpu/cpu*/cpufreq/scaling_cur_freq

• Now I can't say this isn't because of me running a newer BIOS than you or not, as per changelogs having performance and stability improvements you don't have, as I've not used the ! toggle since fall 2019 and DON'T remember how it was back then
• Screenshot:

[cyring]

• Here already running one of the latest kernel version 5.7.5 but still having ACPI issue. My BIOS might be not ACPI tables compliant or the Kernel still don't know how to process some of them.

• Your P-State ratio is showing 22 because, I believe, your processor is under the control of acpi_cpufreq. In my use case, I have blacklist any kernel modules, to study P-States in a deterministic way

Without CPUFreq drivers (confirmed by Missing) I can program things these ways:

1. You program ratio to 22 then you stress CPU and its frequency ratio is capped to22 (as expected)
2. The same happens, if you program the P-State to 28
3. But as soon as you program 35, idle shows 36 and full load goes up to 47

To summarize: I don't get the 47 ratio boost with P-States 22 and 28 but only with the P0 P-State of factory value 35

[cyring]

This is what we get from the Rocket SSD

[olejon]

• CrystalDiskMark is OK, sometimes, but can be wildly inaccurate too
• Manufacturers use ATTO Disk Benchmark when setting their specs

From Tom's Hardware:

ATTO is a simple and free application that SSD vendors commonly use to assign sequential performance specifications to their products. It also gives us insight into how the device handles different file sizes

Straight from Corsair:

ATTO Disk Benchmark has long been an industry staple for measuring drive performance

• It's just a form to get it from the official website (would not trust all the freeware sites) where you can use false name and whatever test email, and newsletter is OFF by default and I've never gotten any email afterwards My download link below should work. Just installs 3 files in same folder of your liking. Start Menu icon can be deleted of course, no desktop icon is added and uninstall is as easy as any, and zero bloatware...
• Link you get is generic and should work: https://www2.atto.com/asset/1:atto-disk-benchmark
• Made an NTFS partition at the end of my /home NVMe. Don't remember if it's the CPU bound or Chipset bound, but they get the same speed in Linux so...
• I just use these options to best match the Linux Benchmark, but you can try the default full run. For me it is unnecessary, I see these results after reaching the 128 KB test

Windows - Corsair MP600 - Basically the first proper PCIe4

[cyring]

I fear that WSL won't offer more than the TSC WiP #191

[olejon]

Oh lol, I was about to try CoreFreq on the Windows Linux subsystem one day, but assumed it wouldn't work, but worth a shot since it's easy enough to try. That if I could actually install the build-essential metapackage. Otherwise I would assume not. Also noticed screen only works as root/sudo. Don't like to run e.g. apt (or any package manager) to install/update many packages without screen or similar, on any system.

---

Kind of odd Microsoft has added that option. First they kind of had their subsystem where you could run BASH directly from the Start Menu (through search only I think, was a hidden link) or Shift + Right Click in a folder in Windows Explorer. But they removed that without saying really. BASH just exited.

But was no good anyways when actually worked. Lacked so much. Ubuntu from the Microsoft Store runs the latest LTS, or you can choose a compatible LTS. Then it works, and has more or less what macOS has by default. Easier to install more, though. Not Homebrew and Apple making that harder or screwing up installed packages for every version.

It's weird to have cmd.exe which is horrible but can do much more than one might think. Many driver packages, some software and enterprise AD use it for setup still... Has tons of AD/networking commands but very undocumented. Unless you've taken a MS certification course or something, it's hard to work with.

Powershell I played with a little. Long story short, even our Powershell expert couldn't make it do what a BASH script I wrote in 30 seconds did (read webcam barcode value and open browser to X URL with Y parameters).

So now Window's got 3! They gotta keep cmd for drivers and certain software, but hide it and just use BASH... IMHO.

Pretty nicely integrated with your / Linux folder right there in Windows Explorer and everything. Little like Chromebooks but they also run GUI software just fine like GIMP etc.

Like... Have you tried just to get the MD5 value of a file on Windows without a 3rd party program? Found some official Microsoft package to add an insanely long command to cmd to do checksums but turned out to be deprecated, without saying any option... Now that's easy. And much more basics that a freakin' OS should be able to. Lots of Windows packages does "Verifying download/install" but IDK what mechanism. Apple has md5 (not called md5sum) and shasum like Linux, but doesn't use BASH by default anymore.

[cyring]

Is Hyper-V the Hypervisor of WSL-2 ? Booting the CoreFreq ISO into a Hyper-V VM behaves the same as within WSL. Thus is WSL a particular VM of Hyper-V, but without showing up into the VM manager.

Many MSR can be queried. Although they return zero, they don't crash. They are thus not pathrough, but trapped.

But APERF/MPERF counters, which are similar to MSR, crash. I feel they are trapped by Hyper-V, but something is missing. Probably, some settings are missing in my WSL configuration, such as Xen, KVM and VBOX let you define rules when a MSR is called.

Why Hyper-V makes possible a virtualized TSC, and not APERF /MPERF ? Especially when: virtualizing the TSC, the hardest part, has been achieved.

[olejon]

Well one search and...

Does WSL 2 use Hyper-V? Will it be available on Windows 10 Home? WSL 2 will be available on all SKUs where WSL is currently available, including Windows 10 Home. The newest version of WSL uses Hyper-V architecture to enable its virtualization. This architecture will be available in the 'Virtual Machine Platform' optional component. This optional component will be available on all SKUs. You can expect to see more details about this experience soon as we get closer to the WSL 2 release.

FAQ: https://docs.microsoft.com/en-us/windows/wsl/wsl2-faq

[cyring]

Definitively it's the same Hypervisor: CoreFreq is running in an ArchLinux VM and we're reading the same Vendor ID

[cyring]

It's been so hot in Europe these last days, I tuned my CPUs to cool down: No OC, especially PBO

[olejon]

Introduction

Yeah, also very hot In Norway, for weeks :sun_with_face:

• I wanted to get to the bottom of which UEFI Settings give better performance
• For this I used Windows (Idle) + Blender Benchmark classroom + Ryzen Master to monitor
• I compare 2 findings with Linux (Idle) using Blender Benchmark classroom + CoreFreq to monitor
• For this I did reset all the UEFI Settings to default, then set:
  • Logically D.C.O.P = Enabled to actually use my RAM chips' speed
  • Logically CSM = Enabled to be able to boot Windows, since legacy boot

NOTE: I divide this into 2 Categories: AMD OC found under AMD Overclocking and Manual OC found under AI Tweaker, or for you Extreme Tweaker. From what I see they have the same Settings. One (fancy manual) setting was added in a later BIOS version than yours. I've never touched it.

1. Windows Screenshots. Every screenshot has bullet points above = Which UEFI Settings were changed
2. Linux Comparison Screenshots
3. UEFI Screenshots of the Settings
4. CONCLUSION: Temperatures, Watts and Other Observations

Windows - Run 1

• Nothing changed

Windows - Run 2

• In Ryzen Master: OC Mode set to Auto Overclocking and Reboot
• This changes AMD OC only
• Change approx. from 3850 to 3920 MHz
• Speed gain approx. 70 MHz
• RESULT: 1 Second better than Run 1

Windows - Run 3

• Also RAM Gear Down Mode = Disabled
• This changes AMD OC only
• Default is Enabled
• This setting can be seen in Ryzen Master as GearDownEn = Enabled/Disabled
• RESULT: 3 Seconds better than Run 1, meaning this Simple Setting actually helps performance as said in videos, at no cost(?)

Windows - Run 4

• Also Power Down Enable = Enabled
• This changes AMD OC only
• Default is Auto, may be affected by Gear Down Mode = Disabled, I did not check, waste of time
• This setting I can not find in Ryzen Master
• RESULT: No effect. Useless!

Windows - Run 5

• This is "Manual OC"
• Reverted the useless back to Power Down Enable = Auto
• In AI Tweaker, set Precision Boost Override = Enabled
• Let the 3 others stay at Auto
• RESULT: 14 Seconds better than Run 1

---

Linux Clone of same Settings as in Run 3

• RESULT: 24 Seconds better than comparison

Linux Clone of same Settings as in Run 5

• RESULT: 22 Seconds better than comparison

---

UEFI AMD OC Built-In PBO Settings

• NOTE: Both these set by Ryzen Master
• First Screenshot sets PBO to a hypothetical 100 MHz
• Second Screenshot simply changes Overclock = Enabled while the rest stays at Auto
• RESULT: Peak MHz about 70 MHz higher on Windows

UEFI AMD OC Built-In RAM Settings

• NOTE: Gear Down Mode = Disabled gave better Blender Benchmark result

• NOTE: Power Down Enable = Enabled default is Auto, as said may be affected by Gear Down Mode

UEFI Manual OC Settings

• Have only tried this one of all the stuff in AI Tweaker, and don't plan to mess with others
• As said in another post, it is NOT necessary to enable Uncore OC Mode (my bad) which is under AMD OC Settings
• Only have to set Precision Boost Override = Enabled and leave rest at Auto
• No reason to set 100 MHz here too
• Not only does it increase Peak MHz by approx. 200 MHz on both Windows and Linux
• It also seems to make the Peak MHz more stable, in both Ryzen Master and CoreFreq
• By more stable I mean when you look at the MHz of all the Threads they fluctuate less
• Help text is pretty clear. Allows the "CPU to run (beyond its X defined limits) to the limits of the board", sounds OK for the Motherboard... Seems scary to you?
• CPUs have their own protections, and several, no?
• I think my Skylake simply shut off when too much OC. That's why I had this sensor software to reboot it if >= 96°C in case, so server services would continue to work
• My previous 3600X simply rebooted to POST and just pressed F1 to revert a test OC I did I more or less knew wouldn't work since it had the stock cooler. I may be wrong and it is the other way around. Hasn't happened with this one so IDK. Per now this Not Enabled before I read more about it, FYI

---

CONCLUSION

• The following observed when running Blender Benchmark classroom
• When Ryzen Master's OC Mode = Default then PPT = 95-97% of 142 Watts
• When Ryzen Master's OC Mode = Auto Overclocking then PPT = 100% of 142 Watts, showing Orange Color, but since 197 Watts+ is apparently no problem, it's just an indicator for that PPT
• When run in Linux, CoreFreq shows a pretty darn stable 142 Watts too
• Guess it is a "cap", meaning if PPT, TDC or EDC reach 100%, the CPU does not give more performance
• Temperature on Windows then = 65-67°C, but NOT valid, fans barely spin up at that with my current UEFI fans config
• Temperature on Linux then = 64-66°C, but NOT valid, fans barely spin up at that with my current UEFI fans config
• POINT IS: No significant difference seen in Temperature between Ryzen Master OC Mode Default and Auto Overclocking
• Setting Gear Down Mode = Disabled did not affect Temperature either. This was as said with OC Mode = Auto Overclocking so PPT = 100% of 142 Watts, still the better result
• Remember for the following, this is what I saw during the Blender Benchmark, but all in all I think I have seen Windows give a higher Peak MHz, without "winning" because of it
• The effect of OC Mode = Auto Overclocking is as said a Peak MHz of approx. 70 MHz higher, but it is higher on Windows actually
• On Windows OC Mode = Auto Overclocking causes 3850 MHz to 3920 MHz (more or less)
• On Windows OC Mode = Auto Overclocking causes 3835 MHz to 3905 MHz (more or less)
• In this mode NO stresstest gets it up to 70°C even with 30°C+ ambient, generally it Idles at 3-5°C above ambient so all good there
• Forgot to pay attention to Voltage, but Conic Compute > Hyperboloid of two sheets = 1.09 V

---

• Now we set Precision Boost Override = Enabled under Manual OC
• Still OC Mode = Auto Overclocking and Gear Down Mode = Disabled for best results
• PPT changes from 142 Watts to 395 Watts
• TDC changes from 95 A to 255 A, but it never reaches that of course, always stays at Green Color
• EDC changes from 140 A to 255 A, but it never reaches that of course, always stays at Green Color
• During run, Ryzen Master shows PPT = 50% of 395 Watts = 197.5 Watts, and totally stable at that
• CoreFreq shows a bit more fluctuation, jumps between 193-203 Watts, but not constantly, so an average approx. Windows
• I can only guess there is some "cap" around 203 W on Linux with this particular OC, while Windows doesn't want to go over 197.5 W, it doesn't go below either, while Linux does (some balancing?)
• Temperature on Windows then = 85-86°C
• Temperature on Linux then = 84-85°C
• BTW: Temperature is not the cap. I've seen 88°C in scene fishy_cat in this mode on Windows, MHz and everything totally stable, but Watts is lower with that scene
• Forgot to pay attention to Voltage, but Conic Compute > Hyperboloid of two sheets = 1.28 V

---

• DESPITE THIS: No worries when the Winner of all CPU Benchmarks run to date here is Linux. And that from a year ago, when I bought the 3600X
• Numbers are interesting but never tell the full story between 2 very different Operating Systems. Despite stable Peak MHz indicate Windows should win, it never does...

---

• EDIT: Would be cool if CoreFreq could read PPT, TDC and EDC MAX values as Ryzen Master shows them. Well at least PPT since CoreFreq shows Watts. Possible?
• I don't expect any long answer to this. Just a test I wanted to do and why not share it when we have the same CPU now :-)

[cyring]

Awesome.

PBO is indeed helping a lot in performance score. But did you mention which policy plan is in used: Windows default or AMD policy ? The last is keeping the frequency target to a high value, at the expense of temperature; whereas Windows Balanced mode is keeping frequency among the lowest P-State ratios.

Would be cool if CoreFreq could read PPT, TDC and EDC MAX values as Ryzen Master shows them. Well at least PPT since CoreFreq shows Watts. Possible?

One software on GitHub is getting those values by reversed engineering. I've not try it yet. And specs just give the basic information. Not enough to program a safe algorithm.

[olejon]

Impossible to replicate Run 3 on Windows

• After having reset all UEFI Settings I found Linux performed the same as Run 3 with only D.C.O.P.
• So I had to check Windows, and it did a little worse, so seemed like Windows wanted the Run 3 Settings
• The Ryzen Master stress test for OC Mode: Default reached 97% of PPT (142 W) as before

But first the Voltages:

• Windows:
  • Running classroom: 1.14 - 1.16 V
  • Idle = 0.32 V (up some times, but for the vast majority of time)
• Linux:
  • Running classroom: 1.06 - 1.09 V
  • Idle: 0.20 - 0.91 - 0.96 V (vast majority of time at 0.91 V, but all cores show they've been at 0.20 V too, and occasionally goes 0.96 V)

Frequencies after having run 1 minute:

• Windows: 3825 - 3850 MHz
• Linux: Approx. 15-20 MHz lower, never hitting 3850 MHz again after the very beginning

ACTUAL RESULTS:

• So I used Ryzen Master again to set OC Mode: Auto Overclocking and Reboot
• The Ryzen Master stress test after Reboot then showed 100% of PPT (142 W) as expected
• Reboot and let Idle
• Despite this, time went up with a couple of seconds
• So I rebooted again and set Gear Down Mode = Disabled too
• Despite this, time went up with a couple of seconds
• I think first Ryzen Master showed some 95% of PPT = 135 Watts
• But this went down to 93% of PPT = 132 Watts
• Results for classroom went like this: 4m23s, 4m26s, 4m29s, 4m30s, and then I facepalmed and gave up
• Booted Linux and was afraid it would perform worse there too, for some reason...
• Linux, as before, ALWAYS uses about 96% of PPT, a little above 136 Watts
• BUT nope, Linux did just as always
• Result for classrom: 4m plus/minus 1 second, every time
• And this result it gets as said in the beginning with only D.C.O.P.
• The other UEFI settings I change do not have anything to do with OC

POSSIBLE REASONS:

• It depends on Windows' mood or time of day? :man_shrugging:
• The Chassis HDD Led does NOT blink for NVMes, and I see some blinking during run on Windows, as it is on a dedicated SATA SSD, so maybe Blender Benchmark depends on I/O speed after loading the scene?
• You have both OS' on the same NVMe so you can test that
• BUT: That still does not explain why Linux beats Windows by a lot in Geekbench

TO YOUR QUESTION:

• I use the Balanced Power Plan as it's the only one really made for Windows 10. Changing to another removes a Performance Slider from the new Settings App
• It acts just like ondemand on Linux when I see the frequencies using HWMonitor (same as ! in CoreFreq)
• The Windows Built-In Power Plans can be reset to default
• 3rd Party Power Plans can not, you must reinstall the whole package that installs them...
• If you open the ADVANCED DETAILS of a Windows Power Plan you see all the settings, including after how many minutes the monitor(s) should turn off and all that
• And if you see the Windows Built-In Performance plan compared to the Ryzen Performance plan they are pretty identical
• The Ryzen Performance does nothing more than make the CPU run at Base Clock MHz all the time
• This can be achieved in Linux using sysfs too... Setting governor to performance or a Specific Frequency for all 32 in sysfs (think this was needed, changing governor alone didn't work, it still boosted)
• I made a BASH script to toggle it when I had the 3600X, it does NOT improve benchmarks
• At best, IF the system is Idle, you get a tiny little advantage when opening something that is not in RAM... Except NVMes...
• I have tried the Performance Slider in the new Settings App, without seeing any difference, IIRC it's just 3 steps, basically a replacement for the old Power Saver, Balanced and Performance. Those old ones found in the "old" Control Panel are just there for legacy reasons it seems, considering the Slider disappears if not using Balanced as said, at least it was like that, and I'm 100% sure it disappears if you choose a 3rd Party Plan
• BUT remember, all those Windows Vista-Era plans have the same settings, if you look into them... So you can set the Balanced plan to the same settings as seen in Ryzen Performance, and IDK what the Slider is for then, maybe it disappears
• And basically just 2 settings have to do with the CPU, the one called something like "Active/Passive Cooling" and the one where you set Minimum/Maximum % of operation, which by default is 5% in all plans except the Performance ones, and what is 5% exactly? They only switch Minimum to 100%, Maximum is already 100% of course. That's is, IIRC

[cyring]

It depends on Windows' mood or time of day?

There are so many Windows service processes of any kind running that it's hard to measure and reproduce benchmark. Even RM has a 100 to 300 MHz CPU overhead for its own monitoring. Either Windows is stripped of any tasks to make it bare metal, or Linux Kernel is built with most drivers and security removed, such as page randomization. Then you will set processor registers to your will and barely get a reproductible and understandable score.

• In BIOS I have enable the Data-Fabric C-States but with no specs how to implement them in CoreFreq, I can't tell the benefits.

[olejon]

My Windows 10 is as Clean as it gets. All permanent or scheduled unnecessary services Disabled.

After boot:

• Run Task Manager until The System Idle "process" is 99% constantly and no I/O (you never see 100%, that "process" has been there since Windows 95 or something)
• So wait about 2-3 minutes
• Open Ryzen Master and see that indeed 31 Cores are "Sleeping" all the time, sometimes 1 more wakes up for 1-2 seconds but it's always like that
• I have of course run Blender Benchmark without Ryzen Master open and of course EXACT same results, the tiny 150-200 MHz it uses makes no difference for a heavy tool like Blender Benchmark
• When I runch Geekbench I NEVER have anything open other than cmd.exe

I did try the Ryzen Performance Power Plan, and observed Ryzen Master for any changes:

• No difference in Voltage
• No difference in Watts
• No difference in Amount of Cores "Sleeping", still 31 Cores sleeping for the most part
• The "Peak Speed", although hard to observe, was maybe 40 MHz above, like instead of being mostly around 160 MHz it was more around 200 MHz, but I didn't bother sitting there staring for minutes
• Remember just moving the mouse a little wakes up several Cores, so that Power Plan causes less CPU use than moving the mouse, if any at all
• CONFIRMED: The only place the Ryzen Performance Power Plan is different than the standard Windows Balanced is that it sets "Minimum CPU % = 100", which does nothing as said above
• The Ryzen Balanced Power Plan is even stupider. Its only difference is that it sets "Minimum CPU % = 99". Those using that are just stupid. No difference whatsoever in the values above

BTW: Not sure if you're ever going to find that the Cores actually Sleep. I've been watching and never seen a Core below 130 MHz - once it's below 130 MHz it says "Sleeping". So maybe "Sleeping" is just indicating very low MHz...

[cyring]

BTW: Not sure if you're ever going to find that the Cores actually Sleep. I've been watching and never seen a Core below 130 MHz - once it's below 130 MHz it says "Sleeping". So maybe "Sleeping" is just indicating very low MHz...

But AMD says Processor has C-States. Those are mentioned in ASUS BIOS. The Software Manual specifies the X86 MONITOR/MWAIT instructions to enter idle states. This has to exist but we don't have so far the counters addresses to monitor them.

I'm reading you that RM may implement a CPU load minimum threshold to say: "it's sleeping" But why such dummy algorithm ? RM is an AMD software: authors have full access to the NDA specifications to make an accurate software !

• CoreFreq version 1.79 for Hyper-V & WSL is released You can download the ISO image to try under your Hyper-V manager Or, just pull the last CoreFreq source code, following WSL procedure to build with kernel headers at #191