Closed obaudys closed 4 years ago
cyring
commented
4 years ago Thank you for letting me know.
Can you please try this hot-fix:
git clone --branch hotfix_LookupProcessor --single-branch https://github.com/cyring/CoreFreq.gitAbout Turbo:
LookupProcessor() is to provide such values during the driver startup.
obaudys
commented
4 years ago Unfortunately, I still get the following with the hotfix:
CoreFreq(3:9): Processor [ 8F_71] Architecture [Zen/Whitehaven] SMT [12/12]
cyring
commented
4 years ago Can you show me the processor brand string resulting from corefreq-cli -s and the version corefreq-cli -v
Please make sure not to mix source code version
obaudys
commented
4 years ago Here are the strings you wanted, both of these are from hotfix_LookupProcessor:
Processor [AMD Ryzen 5 3600 6-Core Processor ]
|- Architecture [Zen/Whitehaven]
|- Vendor ID [AuthenticAMD]
|- Microcode [ 141561875]
|- Signature [ 8F_71]
|- Stepping [ 0]
|- Online CPU [ 12/ 12]./corefreq-cli -v
1.75.2Hello, Can you try the new fix ?
cyring
commented
4 years ago One more fix pushed. Sorry for the mess.
obaudys
commented
4 years ago The new fix works for me! Thanks for your work!
CoreFreq(5:11): Processor [ 8F_71] Architecture [Zen2/Matisse] SMT [12/12]All the speeds are correct too:
Processor [AMD Ryzen 5 3600 6-Core Processor ]
|- Architecture [Zen2/Matisse]
|- Vendor ID [AuthenticAMD]
|- Microcode [ 141561875]
|- Signature [ 8F_71]
|- Stepping [ 0]
|- Online CPU [ 12/ 12]
|- Base Clock [100.001]
|- Frequency (MHz) Ratio
Min 400.00 < 4 >
Max 3600.03 < 36 >
|- Factory [100.000]
3600 [ 36 ]
|- Performance
|- OSPM
TGT 2200.02 < 22 >
|- Turbo Boost [ UNLOCK]
XFR 4300.04 [ 43 ]
CPB 4200.04 [ 42 ]
1C 2800.02 < 28 >
2C 2200.02 < 22 > The new fix works for me! Thanks for your work!
CoreFreq(5:11): Processor [ 8F_71] Architecture [Zen2/Matisse] SMT [12/12]All the speeds are correct too:
Processor [AMD Ryzen 5 3600 6-Core Processor ] |- Architecture [Zen2/Matisse] |- Vendor ID [AuthenticAMD] |- Microcode [ 141561875] |- Signature [ 8F_71] |- Stepping [ 0] |- Online CPU [ 12/ 12] |- Base Clock [100.001] |- Frequency (MHz) Ratio Min 400.00 < 4 > Max 3600.03 < 36 > |- Factory [100.000] 3600 [ 36 ] |- Performance |- OSPM TGT 2200.02 < 22 > |- Turbo Boost [ UNLOCK] XFR 4300.04 [ 43 ] CPB 4200.04 [ 42 ] 1C 2800.02 < 28 > 2C 2200.02 < 22 >
Glad to see a Base Clock estimation close to the factory one at 100 MHz.
Usually it falls around 99.8 MHz
What is your motherboard ? Can you post a screenshot of the CoreFreq UI ?
obaudys
commented
4 years ago My motherboard is an MSI Gaming Pro Carbon AC (B450 chipset)
cyring
commented
4 years ago When using a plain Xterm, the UI will show better the dark gray color which is used for disabled features and other decorations. You may have a color profile option within your terminal, such as Linux theme.
obaudys
commented
4 years ago Are you asking for a screenshot with dark gray background? Well, here's one:
cyring
commented
4 years ago No, I mean you are missing separators, unit symbols and labels, such as, in the footer, between SMT and BOOST... Try with Xterm or with the console, you should see the difference.
I'm also noticing a bug that Min values Watt and Joules, per Core, are not shown or computed
obaudys
commented
4 years ago Ah, I see it now, it looks even better! To get it working with gnome-terminal, I enabled the "Show bold text in bright colors" preference. I'm not sure how to get it working through MacOS terminal via ssh.
I had assumed the power values and certain C-states etc were not implemented or exported by the Ryzen CPUs. If you need any help tracking down bugs, please let me know.
cyring
commented
4 years ago Ah, I see it now, it looks even better! To get it working with gnome-terminal, I enabled the "Show bold text in bright colors" preference. I'm not sure how to get it working through MacOS terminal via ssh.
I did make tests with OS/X terminals and it was ok through ssh but I have to set a standard color palette.
I had assumed the power values and certain C-states etc were not implemented or exported by the Ryzen CPUs. If you need any help tracking down bugs, please let me know.
Indeed AMD has not so far document the counters registers to measure the C-States. I'm understanding that the ACPI space can provide them. I believe most other tools are going through this route. But, I'm not sure if the ACPI provides the same accuracy that the full 64 bits MSR ? Can the ACPI be polled as low as 100 ms interval ? Does it provide all levels C3, C6, Cn per Core, Cluster CCX, CCD, Package ? I feel confident that if the TSC and Instructions counters are published then it is just a matter of time for the manufacturer to unveil the remaining.
please AMD, release the full BKDG
obaudys
commented
4 years ago I'm satisfied that the issue is resolved, since the architecture and frequencies are correctly detected; you can close this if you like.
cyring
commented
4 years ago Using the develop branch, could you test again for non regression.
I have indeed refactor the Zen tables to split the AMD families per CPUID signature ; which means that any future Zen processor should fall through into its inherent class.
Please post the full output of corefreq-cli -s
Thanks for your help.
obaudys
commented
4 years ago The develop branch does not show the correct architecture or frequencies; here's the output of corefreq-cli -s:
Processor [AMD Ryzen 5 3600 6-Core Processor ]
|- Architecture [AMD Zen]
|- Vendor ID [AuthenticAMD]
|- Microcode [ 141561875]
|- Signature [ 8F_71]
|- Stepping [ 0]
|- Online CPU [ 12/ 12]
|- Base Clock [100.000]
|- Frequency (MHz) Ratio
Min 400.00 [ 4 ]
Max 3599.99 [ 36 ]
|- Factory [100.000]
3600 [ 36 ]
|- Performance
|- OSPM
TGT 2199.99 < 22 >
|- Turbo Boost [ LOCK]
XFR 3599.99 [ 36 ]
CPB 3599.99 [ 36 ]
1C 2799.99 [ 28 ]
2C 2199.99 [ 22 ]
|- Uncore [ LOCK]
Instruction Set Extensions
|- 3DNow!/Ext [N/N] ADX [Y] AES [Y] AVX/AVX2 [Y/Y]
|- AVX-512 [N] BMI1/BMI2 [Y/Y] CLFLUSH [Y] CMOV [Y]
|- CMPXCHG8B [Y] CMPXCHG16B [Y] F16C [Y] FPU [Y]
|- FXSR [Y] LAHF/SAHF [Y] MMX/Ext [Y/Y] MONITOR/X[Y/Y]
|- MOVBE [Y] MPX [N] PCLMULQDQ [Y] POPCNT [Y]
|- RDRAND [Y] RDSEED [Y] RDTSCP [Y] SEP [Y]
|- SGX [N] SSE [Y] SSE2 [Y] SSE3 [Y]
|- SSSE3 [Y] SSE4.1/4A [Y/Y] SSE4.2 [Y] SYSCALL [Y]
Features
|- 1 GB Pages Support 1GB-PAGES [Capable]
|- 100 MHz multiplier Control 100MHzSteps [Missing]
|- Advanced Configuration & Power Interface ACPI [Capable]
|- Advanced Programmable Interrupt Controller APIC [Capable]
|- Core Multi-Processing CMP Legacy [Capable]
|- L1 Data Cache Context ID CNXT-ID [Missing]
|- Direct Cache Access DCA [Missing]
|- Debugging Extension DE [Capable]
|- Debug Store & Precise Event Based Sampling DS, PEBS [Missing]
|- CPL Qualified Debug Store DS-CPL [Missing]
|- 64-Bit Debug Store DTES64 [Missing]
|- Fast-String Operation Fast-Strings [Missing]
|- Fused Multiply Add FMA | FMA4 [Capable]
|- Hardware Lock Elision HLE [Missing]
|- Long Mode 64 bits IA64 | LM [Capable]
|- LightWeight Profiling LWP [Missing]
|- Machine-Check Architecture MCA [Capable]
|- Model Specific Registers MSR [Capable]
|- Memory Type Range Registers MTRR [Capable]
|- No-Execute Page Protection NX [Capable]
|- OS-Enabled Ext. State Management OSXSAVE [Capable]
|- Physical Address Extension PAE [Capable]
|- Page Attribute Table PAT [Capable]
|- Pending Break Enable PBE [Missing]
|- Process Context Identifiers PCID [Missing]
|- Perfmon and Debug Capability PDCM [Missing]
|- Page Global Enable PGE [Capable]
|- Page Size Extension PSE [Capable]
|- 36-bit Page Size Extension PSE36 [Capable]
|- Processor Serial Number PSN [Missing]
|- Restricted Transactional Memory RTM [Missing]
|- Safer Mode Extensions SMX [Missing]
|- Self-Snoop SS [Missing]
|- Supervisor-Mode Execution Prevention SMEP [Capable]
|- Time Stamp Counter TSC [Invariant]
|- Time Stamp Counter Deadline TSC-DEADLINE [Missing]
|- Virtual Mode Extension VME [Capable]
|- Virtual Machine Extensions VMX [Missing]
|- Extended xAPIC Support x2APIC [Missing]
|- XSAVE/XSTOR States XSAVE [Capable]
|- xTPR Update Control xTPR [Missing]
Technologies
|- System Management Mode SMM-Lock [ ON]
|- Simultaneous Multithreading SMT [ ON]
|- PowerNow! CnQ [OFF]
|- Core Performance Boost CPB < ON>
|- Virtualization SVM [ ON]
|- I/O MMU AMD-V [OFF]
|- Hypervisor [OFF]
Performance Monitoring
|- Version PM [ 0]
|- Counters: General Fixed
| 6 x 64 bits 3 x 64 bits
|- Enhanced Halt State C1E < ON>
|- Core C6 State CC6 < ON>
|- Package C6 State PC6 < ON>
|- Frequency ID control FID [ ON]
|- Voltage ID control VID [ ON]
|- P-State Hardware Coordination Feedback MPERF/APERF [ ON]
|- Hardware-Controlled Performance States HWP <OFF>
|- Capabilities (MHz) Ratio
Lowest N/A [ 0 ]
Efficient N/A [ 0 ]
Guaranteed N/A [ 0 ]
Highest N/A [ 0 ]
|- Hardware Duty Cycling HDC [OFF]
|- Package C-State
|- Configuration Control CONFIG [ LOCK]
|- Lowest C-State LIMIT [ 0]
|- I/O MWAIT Redirection IOMWAIT [Disable]
|- Max C-State Inclusion RANGE [ 0]
|- MONITOR/MWAIT
|- State index: #0 #1 #2 #3 #4 #5 #6 #7
|- Sub C-State: 1 1 0 0 0 0 0 0
|- Core Cycles [Capable]
|- Instructions Retired [Capable]
|- Reference Cycles [Capable]
|- Last Level Cache References [Missing]
|- Last Level Cache Misses [Missing]
|- Branch Instructions Retired [Missing]
|- Branch Mispredicts Retired [Missing]
Power & Thermal
|- Clock Modulation ODCM [Disable]
|- DutyCycle [ 0.00%]
|- Power Management PWR MGMT [ LOCK]
|- Energy Policy Bias Hint [ 0]
|- Energy Policy HWP EPP [ 0]
|- Junction Temperature TjMax [ 0: 0]
|- Digital Thermal Sensor DTS [Capable]
|- Power Limit Notification PLN [Missing]
|- Package Thermal Management PTM [Missing]
|- Thermal Monitor 1 TTP [Capable]
|- Thermal Monitor 2 HTC [Capable]
|- Thermal Design Power TDP [Missing]
|- Minimum Power Min [Missing]
|- Maximum Power Max [Missing]
|- Units
|- Power watt [ 0.125000000]
|- Energy joule [ 0.000015259]
|- Window second [ 0.000976562]Can you pull last source code in develop and test again.
Be sure to fully rebuild with a make clean all
Thank you
obaudys
commented
4 years ago Done. It seems to be good. Here's a dump of corefreq-cli -s:
Processor [AMD Ryzen 5 3600 6-Core Processor ]
|- Architecture [Zen2/Matisse]
|- Vendor ID [AuthenticAMD]
|- Microcode [ 141561875]
|- Signature [ 8F_71]
|- Stepping [ 0]
|- Online CPU [ 12/ 12]
|- Base Clock [100.000]
|- Frequency (MHz) Ratio
Min 400.00 < 4 >
Max 3600.00 < 36 >
|- Factory [100.000]
3600 [ 36 ]
|- Performance
|- OSPM
TGT 2200.00 < 22 >
|- Turbo Boost [ UNLOCK]
XFR 4300.00 [ 43 ]
CPB 4200.00 [ 42 ]
1C 2800.00 < 28 >
2C 2200.00 < 22 >
|- Uncore [ LOCK]
Instruction Set Extensions
|- 3DNow!/Ext [N/N] ADX [Y] AES [Y] AVX/AVX2 [Y/Y]
|- AVX-512 [N] BMI1/BMI2 [Y/Y] CLFLUSH [Y] CMOV [Y]
|- CMPXCHG8B [Y] CMPXCHG16B [Y] F16C [Y] FPU [Y]
|- FXSR [Y] LAHF/SAHF [Y] MMX/Ext [Y/Y] MONITOR/X[Y/Y]
|- MOVBE [Y] MPX [N] PCLMULQDQ [Y] POPCNT [Y]
|- RDRAND [Y] RDSEED [Y] RDTSCP [Y] SEP [Y]
|- SGX [N] SSE [Y] SSE2 [Y] SSE3 [Y]
|- SSSE3 [Y] SSE4.1/4A [Y/Y] SSE4.2 [Y] SYSCALL [Y]
Features
|- 1 GB Pages Support 1GB-PAGES [Capable]
|- 100 MHz multiplier Control 100MHzSteps [Missing]
|- Advanced Configuration & Power Interface ACPI [Capable]
|- Advanced Programmable Interrupt Controller APIC [Capable]
|- Core Multi-Processing CMP Legacy [Capable]
|- L1 Data Cache Context ID CNXT-ID [Missing]
|- Direct Cache Access DCA [Missing]
|- Debugging Extension DE [Capable]
|- Debug Store & Precise Event Based Sampling DS, PEBS [Missing]
|- CPL Qualified Debug Store DS-CPL [Missing]
|- 64-Bit Debug Store DTES64 [Missing]
|- Fast-String Operation Fast-Strings [Missing]
|- Fused Multiply Add FMA | FMA4 [Capable]
|- Hardware Lock Elision HLE [Missing]
|- Long Mode 64 bits IA64 | LM [Capable]
|- LightWeight Profiling LWP [Missing]
|- Machine-Check Architecture MCA [Capable]
|- Model Specific Registers MSR [Capable]
|- Memory Type Range Registers MTRR [Capable]
|- No-Execute Page Protection NX [Capable]
|- OS-Enabled Ext. State Management OSXSAVE [Capable]
|- Physical Address Extension PAE [Capable]
|- Page Attribute Table PAT [Capable]
|- Pending Break Enable PBE [Missing]
|- Process Context Identifiers PCID [Missing]
|- Perfmon and Debug Capability PDCM [Missing]
|- Page Global Enable PGE [Capable]
|- Page Size Extension PSE [Capable]
|- 36-bit Page Size Extension PSE36 [Capable]
|- Processor Serial Number PSN [Missing]
|- Restricted Transactional Memory RTM [Missing]
|- Safer Mode Extensions SMX [Missing]
|- Self-Snoop SS [Missing]
|- Supervisor-Mode Execution Prevention SMEP [Capable]
|- Time Stamp Counter TSC [Invariant]
|- Time Stamp Counter Deadline TSC-DEADLINE [Missing]
|- Virtual Mode Extension VME [Capable]
|- Virtual Machine Extensions VMX [Missing]
|- Extended xAPIC Support x2APIC [Missing]
|- XSAVE/XSTOR States XSAVE [Capable]
|- xTPR Update Control xTPR [Missing]
Technologies
|- System Management Mode SMM-Lock [ ON]
|- Simultaneous Multithreading SMT [OFF]
|- PowerNow! CnQ [OFF]
|- Core Performance Boost CPB < ON>
|- Virtualization SVM [ ON]
|- I/O MMU AMD-V [OFF]
|- Hypervisor [OFF]
Performance Monitoring
|- Version PM [ 0]
|- Counters: General Fixed
| 6 x 64 bits 3 x 64 bits
|- Enhanced Halt State C1E < ON>
|- Core C6 State CC6 < ON>
|- Package C6 State PC6 < ON>
|- Frequency ID control FID [ ON]
|- Voltage ID control VID [ ON]
|- P-State Hardware Coordination Feedback MPERF/APERF [ ON]
|- Hardware-Controlled Performance States HWP <OFF>
|- Capabilities (MHz) Ratio
Lowest N/A [ 0 ]
Efficient N/A [ 0 ]
Guaranteed N/A [ 0 ]
Highest N/A [ 0 ]
|- Hardware Duty Cycling HDC [OFF]
|- Package C-State
|- Configuration Control CONFIG [ LOCK]
|- Lowest C-State LIMIT [ 0]
|- I/O MWAIT Redirection IOMWAIT [Disable]
|- Max C-State Inclusion RANGE [ 0]
|- MONITOR/MWAIT
|- State index: #0 #1 #2 #3 #4 #5 #6 #7
|- Sub C-State: 1 1 0 0 0 0 0 0
|- Core Cycles [Capable]
|- Instructions Retired [Capable]
|- Reference Cycles [Capable]
|- Last Level Cache References [Missing]
|- Last Level Cache Misses [Missing]
|- Branch Instructions Retired [Missing]
|- Branch Mispredicts Retired [Missing]
Power & Thermal
|- Clock Modulation ODCM [Disable]
|- DutyCycle [ 0.00%]
|- Power Management PWR MGMT [ LOCK]
|- Energy Policy Bias Hint [ 0]
|- Energy Policy HWP EPP [ 0]
|- Junction Temperature TjMax [ 0: 0]
|- Digital Thermal Sensor DTS [Capable]
|- Power Limit Notification PLN [Missing]
|- Package Thermal Management PTM [Missing]
|- Thermal Monitor 1 TTP [Capable]
|- Thermal Monitor 2 HTC [Capable]
|- Thermal Design Power TDP [Missing]
|- Minimum Power Min [Missing]
|- Maximum Power Max [Missing]
|- Units
|- Power watt [ 0.125000000]
|- Energy joule [ 0.000015259]
|- Window second [ 0.000976562]Another bug which happens only AMD:
Processor [AMD Ryzen 5 3600 6-Core Processor ]|- Hardware-Controlled Performance States HWP <OFF>Hello,
Can you pull develop and test the Minimum Energy and Power fix.
To un-correlate the Joule and Watt values, please change the Interval to 500 ms (for example) before you screenshot the Power view
EDIT: it is also released the fix to the processor brand name cleanup that you can check easier in the output of corefreq-cli -s
obaudys
commented
4 years ago I have set my settings to match those above in the following screenshot... ie:
Thermal scope SMP
Voltage scope Pkg
Power scope None
In addition, the processor brand name still contains the trailing space:
Processor [AMD Ryzen 5 3600 6-Core Processor ]
cyring
commented
4 years ago Can you leave all scopes as default, untouched ( b/c your AMD has power counters per Core, whereas my Intel has them only per Package )
About the brand string, can you post the output of:
echo "[$(cat /proc/cpuinfo | grep -i 'model name' | tail -n 1 | cut -f 2 -d ':')]"Here's with the scopes untouched:
I'd say the package power is correct; in Windows Ryzen Master shows 80W when running with all cores at 100% (I'm running folding@home right now)
Here's the output you requested:
$ echo "[$(cat /proc/cpuinfo | grep -i 'model name' | tail -n 1 | cut -f 2 -d ':')]"
[ AMD Ryzen 5 3600 6-Core Processor]Thanks for your returns but I don't understand why the ACCU, Power and Energy are now all showing zero. The fix I made was only the Minimum columns. So do you confirm that from scratch
developthe values are zero ?
obaudys
commented
4 years ago Yes I can confirm all of the following:
developAll values are still zero except package power
cyring
commented
4 years ago Switching back to the master repo, do you get values ?
(be sure to unload current driver before switching)
obaudys
commented
4 years ago Yes, switching to master I do get values:
cyring
commented
4 years ago Can you download and test last develop
It will be version 1.76.1
obaudys
commented
4 years ago 1.76.1 still shows all 0 values
cyring
commented
4 years ago 1.76.1 still shows all 0 values
Looking at the above develop screenshots, I can see that some values are garbage compared to master
For example: L3=64 vs L3=16384 ; also SMT is disabled in develop
Usually when values are inconsistent, it's a sign of a major change in the API carried by the memory shared between the driver and the daemon. As a footprint, the version is computed and stored in the shared memory and checked by Processes to be sure they are working on the same API
Don't you have any Oops in the kernel log ? see dmesg
cyring
commented
4 years ago Do you have installed CoreFreq from the AUR or enter make install ?
obaudys
commented
4 years ago There are no oopses in the kernel log.
Apr 11 20:08:14 orion kernel: [15033.524536] CoreFreq: Unload
Apr 11 20:09:52 orion kernel: [15131.722332] CoreFreq(0:-1): Processor [ 8F_71] Architecture [Zen2/Matisse] CPU [12/12]I'm running CoreFreq from my local git repo, after pulling in changes and switching branches is required:
rmmod corefreqk; make clean all; sudo insmod ./corefreqk.ko; sudo ./corefreqd & sleep 1; ./corefreq-cliThere are no oopses in the kernel log.
Apr 11 20:08:14 orion kernel: [15033.524536] CoreFreq: Unload Apr 11 20:09:52 orion kernel: [15131.722332] CoreFreq(0:-1): Processor [ 8F_71] Architecture [Zen2/Matisse] CPU [12/12]
The log is strange:
-1 confirmes no thread CPU detected thus SMT is disabled ; but we see the Zen2/Matisse CPUID family ID 8F_71
I'm running CoreFreq from my local git repo, after pulling in changes and switching branches is required:
rmmod corefreqk; make clean all; sudo insmod ./corefreqk.ko; sudo ./corefreqd & sleep 1; ./corefreq-cli
Is it virtualized ?
obaudys
commented
4 years ago No, not virtualised, this is a bare metal PC. Running Ubuntu 19.10 with a 5.4.0 kernel.
Here's a dump of my kernel logs since boot, grepping for CoreFreq. SMT was only detected in master, not develop
pr 11 15:59:07 orion kernel: [ 86.312860] CoreFreq(11:-1): Processor [ 8F_71] Architecture [Zen2/Matisse] CPU [12/12]
Apr 11 16:03:10 orion kernel: [ 329.855418] CoreFreq: Unload
Apr 11 16:03:30 orion kernel: [ 349.389771] CoreFreq(7:-1): Processor [ 8F_71] Architecture [Zen2/Matisse] CPU [12/12]
Apr 11 16:47:28 orion kernel: [ 2988.025105] CoreFreq: Unload
Apr 11 16:47:55 orion kernel: [ 3014.840460] CoreFreq(2:8): Processor [ 8F_71] Architecture [Zen2/Matisse] SMT [12/12]
Apr 11 20:08:14 orion kernel: [15033.524536] CoreFreq: Unload
Apr 11 20:09:52 orion kernel: [15131.722332] CoreFreq(0:-1): Processor [ 8F_71] Architecture [Zen2/Matisse] CPU [12/12]No, not virtualised, this is a bare metal PC. Running Ubuntu 19.10 with a 5.4.0 kernel.
OK, thanks for your answer. Doing a code review for regression but still don't know why SMT has for instance disappear.
cyring
commented
4 years ago I think it comes from this commit f7c5dbecf14588fa3e68a8ec1db1ce411508b2f0
EDIT I have reverted the commit, you can just pull and try
obaudys
commented
4 years ago I have pulled the latest develop, and built, but the issues persist: No SMT, wrong cache sizes, and per-core power shows 0.
cyring
commented
4 years ago Can you please post the dump with corefreq-cli -u
CPU #0 will be enough
obaudys
commented
4 years ago CPU #0 function EAX EBX ECX EDX
|- 00000000:00000000 00000010 68747541 444d4163 69746e65
|- Largest Standard Function=00000010
|- 80000000:00000000 80000020 68747541 444d4163 69746e65
|- Largest Extended Function=80000020
|- 00000001:00000000 00870f10 000c0800 7ed8320b 178bfbff
|- 00000002:00000000 00000000 00000000 00000000 00000000
|- 00000003:00000000 00000000 00000000 00000000 00000000
|- 00000004:00000000 00000000 00000000 00000000 00000000
|- 00000004:00000001 00000000 00000000 00000000 00000000
|- 00000004:00000002 00000000 00000000 00000000 00000000
|- 00000004:00000003 00000000 00000000 00000000 00000000
|- 00000005:00000000 00000040 00000040 00000003 00000011
|- 00000006:00000000 00000004 00000000 00000001 00000000
|- 00000007:00000000 00000000 219c91a9 00400004 00000000
|- 00000007:00000001 00000000 00000000 00000000 00000000
|- 00000009:00000000 00000000 00000000 00000000 00000000
|- 0000000a:00000000 00000000 00000000 00000000 00000000
|- 0000000b:00000000 00000001 00000002 00000100 00000000
|- 0000000d:00000000 00000207 00000340 00000380 00000000
|- 0000000d:00000001 0000000f 00000340 00000000 00000000
|- 0000000d:00000002 00000100 00000240 00000000 00000000
|- 0000000d:00000003 00000000 00000000 00000000 00000000
|- 0000000d:00000004 00000000 00000000 00000000 00000000
|- 0000000d:0000003e 00000000 00000000 00000000 00000000
|- 0000000f:00000000 00000000 000000ff 00000000 00000002
|- 0000000f:00000001 00000000 00000040 000000ff 00000007
|- 00000010:00000000 00000000 00000002 00000000 00000000
|- 00000010:00000001 0000000f 00000000 00000004 0000000f
|- 00000010:00000002 00000000 00000000 00000000 00000000
|- 00000010:00000003 00000000 00000000 00000000 00000000
|- 00000012:00000000 00000000 00000000 00000000 00000000
|- 00000012:00000001 00000000 00000000 00000000 00000000
|- 00000012:00000002 00000000 00000000 00000000 00000000
|- 00000014:00000000 00000000 00000000 00000000 00000000
|- 00000014:00000001 00000000 00000000 00000000 00000000
|- 00000015:00000000 00000000 00000000 00000000 00000000
|- 00000016:00000000 00000000 00000000 00000000 00000000
|- 00000017:00000000 00000000 00000000 00000000 00000000
|- 00000017:00000001 00000000 00000000 00000000 00000000
|- 00000017:00000002 00000000 00000000 00000000 00000000
|- 00000017:00000003 00000000 00000000 00000000 00000000
|- 00000018:00000000 00000000 00000000 00000000 00000000
|- 00000018:00000001 00000000 00000000 00000000 00000000
|- 0000001a:00000000 00000000 00000000 00000000 00000000
|- 0000001b:00000000 00000000 00000000 00000000 00000000
|- 0000001f:00000000 00000000 00000000 00000000 00000000
|- 80000001:00000000 00870f10 20000000 75c237ff 2fd3fbff
|- 80000002:00000000 20444d41 657a7952 2035206e 30303633
|- 80000003:00000000 432d3620 2065726f 636f7250 6f737365
|- 80000004:00000000 20202072 20202020 20202020 00202020
|- 80000005:00000000 ff40ff40 ff40ff40 20080140 20080140
|- 80000006:00000000 48006400 68006400 02006140 01009140
|- 80000007:00000000 00000000 0000001b 00000000 00006799
|- 80000008:00000000 00003030 010cb657 0000700b 00010000
|- 8000000a:00000000 00000001 00008000 00000000 0013bcff
|- 80000019:00000000 f040f040 00000000 00000000 00000000
|- 8000001a:00000000 00000006 00000000 00000000 00000000
|- 8000001b:00000000 000003ff 00000000 00000000 00000000
|- 8000001c:00000000 00000000 00000000 00000000 00000000
|- 8000001d:00000000 00004121 01c0003f 0000003f 00000000
|- 8000001d:00000001 00004122 01c0003f 0000003f 00000000
|- 8000001d:00000002 00004143 01c0003f 000003ff 00000002
|- 8000001d:00000003 00014163 03c0003f 00003fff 00000001
|- 8000001e:00000000 00000000 00000100 00000000 00000000
|- 40000000:00000000 00000000 00000000 00000000 00000000
|- 40000001:00000000 00000000 00000000 00000000 00000000
|- 40000002:00000000 00000000 00000000 00000000 00000000
|- 40000003:00000000 00000000 00000000 00000000 00000000
|- 40000004:00000000 00000000 00000000 00000000 00000000
|- 40000005:00000000 00000000 00000000 00000000 00000000
|- 40000006:00000000 00000000 00000000 00000000 00000000
Edit: I have taken a diff between the dump from master and develop and it seems develop has a few more entries:
diff -c cpu0.master.dump cpu0.develop.dump
*** cpu0.master.dump 2020-04-12 10:22:50.744997569 +0930
--- cpu0.develop.dump 2020-04-12 10:22:20.193091420 +0930
***************
*** 13,18 ****
--- 13,19 ----
|- 00000005:00000000 00000040 00000040 00000003 00000011
|- 00000006:00000000 00000004 00000000 00000001 00000000
|- 00000007:00000000 00000000 219c91a9 00400004 00000000
+ |- 00000007:00000001 00000000 00000000 00000000 00000000
|- 00000009:00000000 00000000 00000000 00000000 00000000
|- 0000000a:00000000 00000000 00000000 00000000 00000000
|- 0000000b:00000000 00000001 00000002 00000100 00000000
***************
*** 39,44 ****
--- 40,50 ----
|- 00000017:00000001 00000000 00000000 00000000 00000000
|- 00000017:00000002 00000000 00000000 00000000 00000000
|- 00000017:00000003 00000000 00000000 00000000 00000000
+ |- 00000018:00000000 00000000 00000000 00000000 00000000
+ |- 00000018:00000001 00000000 00000000 00000000 00000000
+ |- 0000001a:00000000 00000000 00000000 00000000 00000000
+ |- 0000001b:00000000 00000000 00000000 00000000 00000000
+ |- 0000001f:00000000 00000000 00000000 00000000 00000000
|- 80000001:00000000 00870f10 20000000 75c237ff 2fd3fbff
|- 80000002:00000000 20444d41 657a7952 2035206e 30303633
|- 80000003:00000000 432d3620 2065726f 636f7250 6f737365 Yes those come from these commits 5322e6809102143db22e460b29c42a7645dd5097 ec3bb72be314034fa321f2ca769dff5dd1efb653
cyring
commented
4 years ago Hello,
Can you please try the latest develop ?
obaudys
commented
4 years ago Things look good now!
obaudys
commented
4 years ago I spoke too soon, I see a glitch in the power summary:
When i resize the window:
cyring
commented
4 years ago When i resize the window:
This a bug in consequence of the previous glitch
cyring
commented
4 years ago I spoke too soon, I see a glitch in the power summary:
Looks like a bug due to negative values converted or printed as unsigned
cyring
commented
4 years ago Do you get the same bug with corefreq-cli -W ?
obaudys
commented
4 years ago Yes, I do:
CPU Freq(MHz) Accumulator Min Energy(J) Max Min Power(W) Max
000 3924.37 000000000000267281 0.02 4.08 281474976645124.06 0.02 4.08 281474976645124.06At this line: https://github.com/cyring/CoreFreq/blob/2b5775c82888fc9aecf95cab2b3ddbd560de1964/corefreqk.c#L9032 Can you replace with the following statement
/* Read the Physical Core RAPL counter. */
if (Core->T.ThreadID == 0)
{
RDCOUNTER(Core->Counter[1].Power.ACCU,MSR_AMD_PP0_ENERGY_STATUS);
Core->Counter[1].Power.ACCU &= 0xffffffff;
if (Core->Counter[1].Power.ACCU > Core->Counter[0].Power.ACCU)
{
Core->Delta.Power.ACCU = Core->Counter[1].Power.ACCU
- Core->Counter[0].Power.ACCU;
} else {
Core->Delta.Power.ACCU = 0;
}
Core->Counter[0].Power.ACCU = Core->Counter[1].Power.ACCU;
}then, rebuild, reload and test, please.
obaudys
commented
4 years ago Everything looks good now:
Up until commit 298af3fe5a8664d8b8094385277b533a44281a7a my CPU was detected correctly as
Zen2/Matisse. eg:CoreFreq(4:10): Processor [ 8F_71] Architecture [Zen2/Matisse] SMT [12/12]From that commit onwards, it is detected as
Zen/Summit Ridgeand in the latest commits asZen/Whitehaven, eg:CoreFreq(3:9): Processor [ 8F_71] Architecture [Zen/Summit Ridge] SMT [12/12]The turbo frequency is also incorrect, I assume that is determined by the architecture.
BTW, this project is awesome!