cyring
Incorrect temperature on 9950x
❯ corefreq-cli -C 1
CPU Freq(MHz) VID Vcore TMP(C) Accumulator Energy(J) Power(W)
000 45.69 144 0.9000 34 000000000000017680 0.269775391 0.269775391
001 72.80 144 0.9000 34 000000000000014518 0.221527100 0.221527100
002 55.16 144 0.9000 34 000000000000009318 0.142181396 0.142181396
003 38.04 144 0.9000 34 000000000000012885 0.196609497 0.196609497
004 33.44 144 0.9000 34 000000000000005478 0.083587646 0.083587646
005 111.49 144 0.9000 34 000000000000022974 0.350555420 0.350555420
006 11.37 144 0.9000 34 000000000000002650 0.040435791 0.040435791
007 6.79 144 0.9000 34 000000000000002502 0.038177490 0.038177490
008 1.45 144 0.9000 35 000000000000000955 0.014572144 0.014572144
009 1.72 144 0.9000 35 000000000000000926 0.014129639 0.014129639
010 0.82 144 0.9000 35 000000000000006141 0.093704224 0.093704224
011 1.25 144 0.9000 35 000000000000005767 0.087997437 0.087997437
012 1.00 144 0.9000 35 000000000000001133 0.017288208 0.017288208
013 2.71 144 0.9000 35 000000000000000746 0.011383057 0.011383057
014 1.80 144 0.9000 35 000000000000001279 0.019515991 0.019515991
015 1.16 144 0.9000 35 000000000000000735 0.011215210 0.011215210
016 13.98 144 0.9000 34 000000000000000000 0.000000000 0.000000000
017 18.64 144 0.9000 34 000000000000000000 0.000000000 0.000000000
018 6.12 144 0.9000 34 000000000000000000 0.000000000 0.000000000
019 6.35 144 0.9000 34 000000000000000000 0.000000000 0.000000000
020 4.55 144 0.9000 34 000000000000000000 0.000000000 0.000000000
021 100.82 144 0.9000 34 000000000000000000 0.000000000 0.000000000
022 11.95 144 0.9000 34 000000000000000000 0.000000000 0.000000000
023 5.20 144 0.9000 34 000000000000000000 0.000000000 0.000000000
024 1.03 144 0.9000 35 000000000000000000 0.000000000 0.000000000
025 0.97 144 0.9000 35 000000000000000000 0.000000000 0.000000000
026 84.94 144 0.9000 35 000000000000000000 0.000000000 0.000000000
027 35.21 144 0.9000 35 000000000000000000 0.000000000 0.000000000
028 3.04 144 0.9000 35 000000000000000000 0.000000000 0.000000000
029 2.03 144 0.9000 35 000000000000000000 0.000000000 0.000000000
030 2.54 144 0.9000 35 000000000000000000 0.000000000 0.000000000
031 1.04 144 0.9000 35 000000000000000000 0.000000000 0.000000000
Package[0] Cores Uncore Memory Platform
Energy(J): 40.838531494 1.612655640 0.000000000 0.000000000 0.000000000
Power(W) : 40.838531494 1.612655640 0.000000000 0.000000000 0.000000000
❯ sensors k10temp-pci-00c3
k10temp-pci-00c3
Adapter: PCI adapter
Tctl: +52.1°C
The temperature reported by k10temp is correct.
Additionally, corefreq-cli -C 1 sometimes fails to respond, and I need to press Ctrl+C and rerun it multiple times to obtain the output.
System Information
❯ corefreq-cli -s
Processor [AMD Ryzen 9 9950X 16-Core Processor]
|- Architecture [Zen5/Granite Ridge]
|- Vendor ID [AuthenticAMD]
|- Microcode [0x0b404032]
|- Signature [ BF_44]
|- Stepping [ 0]
|- Online CPU [ 32/ 32]
|- Base Clock [100.000]
|- Frequency (MHz) Ratio
Min 3000.00 < 30 >
Max 4299.99 < 43 >
|- Factory [100.000]
4300 [ 43 ]
|- Performance
TGT 4299.99 < 43 >
|- CPPC
Min 3699.99 < 37 >
Max 400.00 < 4 >
TGT AUTO < 0 >
|- Boost [ UNLOCK]
XFR 5699.99 [ 57 ]
CPB 5699.99 [ 57 ]
|- P-State
P1 3000.00 < 30 >
|- Uncore [ LOCK]
CLK 1600.00 [ 16 ]
MEM 3199.99 [ 32 ]
Instruction Set Extensions
|- 3DNow!/Ext [N/N] ADX [Y] AES [Y] AVX/AVX2 [Y/Y]
|- AVX512-F [Y] AVX512-DQ [Y] AVX512-IFMA [Y] AVX512-PF [N]
|- AVX512-ER [N] AVX512-CD [Y] AVX512-BW [Y] AVX512-VL [Y]
|- AVX512-VBMI [Y] AVX512-VBMI2 [Y] AVX512-VNNI [Y] AVX512-ALG [Y]
|- AVX512-VPOP [Y] AVX512-VNNIW [N] AVX512-FMAPS [N] AVX512-VP2I [Y]
|- AVX512-BF16 [Y] AVX-VNNI-VEX [Y] AVX-FP128 [N] AVX-FP256 [N]
|- BMI1/BMI2 [Y/Y] CLWB [Y] CLFLUSH [Y] CLFLUSH-OPT [Y]
|- CLAC-STAC [Y] CMOV [Y] CMPXCHG8B [Y] CMPXCHG16B [Y]
|- F16C [Y] FPU [Y] FXSR [Y] LAHF-SAHF [Y]
|- MMX/Ext [Y/Y] MON/MWAITX [Y/Y] MOVBE [Y] PCLMULQDQ [Y]
|- POPCNT [Y] RDRAND [Y] RDSEED [Y] RDTSCP [Y]
|- SEP [Y] SHA [Y] SSE [Y] SSE2 [Y]
|- SSE3 [Y] SSSE3 [Y] SSE4.1/4A [Y/Y] SSE4.2 [Y]
|- SERIALIZE [N] SYSCALL [Y] RDPID [Y] UMIP [Y]
|- VAES [Y] VPCLMULQDQ [Y] PREFETCH/W [Y] LZCNT [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]
|- Advanced Virtual Interrupt Controller AVIC [Capable]
|- APIC Timer Invariance ARAT [Capable]
|- LOCK prefix to read CR8 AltMov [Capable]
|- Clear Zero Instruction CLZERO [Capable]
|- Core Multi-Processing CMP Legacy [Capable]
|- L1 Data Cache Context ID CNXT-ID [Missing]
|- Collaborative Processor Performance Control CPPC [Capable]
|- 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 Short REP CMPSB|SCASB FSRC [Capable]
|- Fast Short REP MOVSB FSRM [Capable]
|- Fast Short REP STOSB FSRS [Capable]
|- Fast-String Operation ERMS [Capable]
|- Fused Multiply Add FMA4 [Missing]
|- Fused Multiply Add FMA [Capable]
|- Hardware Lock Elision HLE [Missing]
|- Hyper-Threading Technology HTT [Capable]
|- Hardware P-state control HwP [Capable]
|- Instruction Based Sampling IBS [Capable]
|- Instruction INVLPGB INVLPGB [Missing]
|- Instruction INVPCID INVPCID [Capable]
|- Long Mode 64 bits IA64 | LM [Capable]
|- LightWeight Profiling LWP [Missing]
|- Memory Bandwidth Enforcement MBE [Capable]
|- Machine-Check Architecture MCA [Capable]
|- Instruction MCOMMIT MCOMMIT [Missing]
|- Model Specific Registers MSR [Capable]
|- Memory Type Range Registers MTRR [Capable]
|- No-Execute Page Protection NX [Capable]
|- OS-Enabled Ext. State Management OSXSAVE [Capable]
|- OS Visible Work-around OSVW [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]
|- PREFETCHIT0/1 Instructions PREFETCHI [Capable]
|- Resource Director Technology/PQE RDT-A [Capable]
|- Resource Director Technology/PQM RDT-M [Capable]
|- Read Processor Register at User level RDPRU [Capable]
|- Restricted Transactional Memory RTM [Missing]
|- Safer Mode Extensions SMX [Missing]
|- Self-Snoop SS [Missing]
|- Supervisor-Mode Access Prevention SMAP [Capable]
|- Supervisor-Mode Execution Prevention SMEP [Capable]
|- Trailing Bit Manipulation TBM [Missing]
|- Translation Cache Extension TCE [Capable]
|- Time Stamp Counter TSC [Invariant]
|- Time Stamp Counter Deadline TSC-DEADLINE [Missing]
|- TSX Force Abort MSR Register TSX-ABORT [Missing]
|- TSX Suspend Load Address Tracking TSX-LDTRK [Missing]
|- User-Mode Instruction Prevention UMIP [Capable]
|- Virtual Mode Extension VME [Capable]
|- Virtual Machine Extensions VMX [Missing]
|- Write Back & Do Not Invalidate Cache WBNOINVD [Capable]
|- Extended xAPIC Support x2APIC [ xAPIC]
|- AVIC controller for x2APIC x2AVIC [Capable]
|- XSAVE/XSTOR States XSAVE [Capable]
|- xTPR Update Control xTPR [Missing]
|- Extended Operation Support XOP [Missing]
Mitigation mechanisms
|- Indirect Branch Restricted Speculation IBRS [Capable]
|- IBRS Always-On preferred by processor [ Unable]
|- IBRS preferred over software solution [Capable]
|- IBRS provides same speculation limits [Capable]
|- Indirect Branch Prediction Barrier IBPB [Capable]
|- Selective Branch Predictor Barrier SBPB [Capable]
|- Single Thread Indirect Branch Predictor STIBP [ Enable]
|- Speculative Store Bypass Disable SSBD [Capable]
|- SSBD use VIRT_SPEC_CTRL register [ Unable]
|- SSBD not needed on this processor [ Unable]
|- No Speculative Return Stack Overflow SRSO_NO [ Unable]
|- No SRSO at the User-Kernel boundary [Capable]
|- No Branch Type Confusion BTC_NO [Capable]
|- BTC on Non-Branch instruction BTC-NOBR [ Unable]
|- Limited Early Redirect Window AGENPICK [ Unable]
|- Arch - No Fast Predictive Store Forwarding PSFD [Capable]
|- Arch - Enhanced Predictive Store Forwarding EPSF [Capable]
|- Arch - Cross Processor Information Leak XPROC_LEAK [ Unable]
Security Features
|- CET Shadow Stack features CET-SS [Capable]
|- Secure Init and Jump with Attestation SKINIT [Capable]
|- Secure Encrypted Virtualization SEV [Missing]
|- SEV - Encrypted State SEV-ES [Missing]
|- SEV - Secure Nested Paging SEV-SNP [Missing]
|- Guest Mode Execute Trap GMET [Capable]
|- Supervisor Shadow Stack SSS [Capable]
|- VM Permission Levels VMPL [Missing]
|- VMPL Supervisor Shadow Stack VMPL-SSS [Missing]
|- Secure Memory Encryption SME [Capable]
|- Transparent SME TSME [ Enable]
|- Secure Multi-Key Memory Encryption SME-MK [Missing]
|- DRAM Data Scrambling Scrambler [ Enable]
Technologies
|- Instruction Cache Unit
|- L1 IP Prefetcher L1 HW IP < ON>
|- Data Cache Unit
|- L1 Prefetcher L1 HW < ON>
|- Cache Prefetchers
|- L2 Prefetcher L2 HW < ON>
|- L1 Stride Prefetcher L1 Stride < ON>
|- L1 Region Prefetcher L1 Region < ON>
|- L1 Burst Prefetch Mode L1 Burst < ON>
|- L2 Stream HW Prefetcher L2 Stream < ON>
|- L2 Up/Down Prefetcher L2 Up/Down < ON>
|- System Management Mode SMM-Lock [ ON]
|- Simultaneous Multithreading SMT [ ON]
|- PowerNow! CnQ [ ON]
|- Core C-States CCx [ ON]
|- Core Performance Boost CPB < ON>
|- Watchdog Timer WDT < ON>
|- Virtualization SVM [ ON]
|- I/O MMU AMD-V [OFF]
|- Version [ N/A]
|- Hypervisor [OFF]
|- Vendor ID [ N/A]
Performance Monitoring
|- Version PM [ 2]
|- Counters: General Fixed
| { 6, 6, 16 } x 48 bits 3 x 64 bits
|- Enhanced Halt State C1E <OFF>
|- C2 UnDemotion C2U < ON>
|- C3 UnDemotion C3U < ON>
|- Core C6 State CC6 < ON>
|- Package C6 State PC6 < ON>
|- Legacy Frequency ID control FID [OFF]
|- Legacy Voltage ID control VID [OFF]
|- P-State Hardware Coordination Feedback MPERF/APERF [ ON]
|- Core C-States
|- C-States Base Address BAR [ 0x413 ]
|- ACPI Processor C-States _CST [ 3]
|- MONITOR/MWAIT
|- State index: #0 #1 #2 #3 #4 #5 #6 #7
|- Sub C-State: 1 2 0 0 0 0 0 0
|- Monitor-Mwait Extensions EMX [Capable]
|- Interrupt Break-Event IBE [Capable]
|- Core Cycles [Capable]
|- Instructions Retired [Capable]
|- Reference Cycles [Capable]
|- Last Level Cache References [Capable]
|- Global Time Stamp Counter [Missing]
|- Data Fabric Performance Counter [Capable]
|- Core Performance Counter [Capable]
|- Processor Performance Control _PCT [ Enable]
|- Performance Supported States _PSS [ 2]
|- Performance Present Capabilities _PPC [ 0]
|- Continuous Performance Control _CPC [Missing]
Power, Current & Thermal
|- Temperature Offset:Junction TjMax [ 49: 95 C]
|- CPPC Energy Preference EPP < 128>
|- Digital Thermal Sensor DTS [Capable]
|- Power Limit Notification PLN [Missing]
|- Package Thermal Management PTM [Missing]
|- Thermal Monitor 1 TTP [ Enable]
|- Thermal Monitor 2 HTC [ Enable]
|- Thermal Design Power TDP [Missing]
|- Minimum Power Min [Missing]
|- Maximum Power Max [Missing]
|- Thermal Design Power Package [Disable]
|- Power Limit PL1 [ 0 W]
|- Time Window TW1 [ 0 ns]
|- Power Limit PL2 [ 0 W]
|- Time Window TW2 [ 0 ns]
|- Thermal Design Power Core [Disable]
|- Power Limit PL1 [ 0 W]
|- Time Window TW1 [ 0 ns]
|- Thermal Design Power Uncore [Disable]
|- Power Limit PL1 [ 0 W]
|- Time Window TW1 [ 0 ns]
|- Thermal Design Power DRAM [Disable]
|- Power Limit PL1 [ 0 W]
|- Time Window TW1 [ 0 ns]
|- Thermal Design Power Platform [Disable]
|- Power Limit PL1 [ 0 W]
|- Time Window TW1 [ 0 ns]
|- Power Limit PL2 [ 0 W]
|- Time Window TW2 [ 0 ns]
|- Package Power Tracking PPT [Missing]
|- Electrical Design Current EDC [Missing]
|- Thermal Design Current TDC [Missing]
|- Core Thermal Point
|- Package Thermal Point
|- Thermal Monitor Trip Limit [ 115 C]
|- HTC Temperature Limit Limit [ 127 C]
|- HTC Temperature Hysteresis Threshold [ 2 C]
|- Units
|- Power watt [ Missing]
|- Energy joule [ 0.000015259]
|- Window second [ 0.000976562]
Hi the other 9950X did not encounter this issue.
What values are you reading when Cores are fully stressed ? Are those still different than k10temp ?
Hi the other 9950X did not encounter this issue.
What values are you reading when Cores are fully stressed ? Are those still different than k10temp ?
Yes, but they don't vary that much:
~ ▓▒░·································································································
❯ corefreq-cli -C 1
CPU Freq(MHz) VID Vcore TMP(C) Accumulator Energy(J) Power(W)
000 5426.59 192 1.2000 80 000000000000678095 10.346908569 10.346908569
001 5426.64 192 1.2000 80 000000000000685677 10.462600708 10.462600708
002 5426.64 192 1.2000 80 000000000000675926 10.313812256 10.313812256
003 5426.59 192 1.2000 80 000000000000695846 10.617767334 10.617767334
004 5426.64 192 1.2000 80 000000000000687470 10.489959717 10.489959717
005 5426.64 192 1.2000 80 000000000000695677 10.615188599 10.615188599
006 5426.64 190 1.1875 80 000000000000684540 10.445251465 10.445251465
007 5426.64 190 1.1875 80 000000000000693621 10.583816528 10.583816528
008 5376.40 190 1.1875 81 000000000000693943 10.588729858 10.588729858
009 5376.35 190 1.1875 81 000000000000680012 10.376159668 10.376159668
010 5376.30 190 1.1875 81 000000000000697079 10.636581421 10.636581421
011 5376.35 190 1.1875 81 000000000000685131 10.454269409 10.454269409
012 5376.30 190 1.1875 81 000000000000695940 10.619201660 10.619201660
013 5376.30 190 1.1875 81 000000000000691096 10.545288086 10.545288086
014 5376.35 190 1.1875 81 000000000000701210 10.699615479 10.699615479
015 5376.35 190 1.1875 81 000000000000679342 10.365936279 10.365936279
016 5426.64 190 1.1875 80 000000000000000000 0.000000000 0.000000000
017 5426.64 190 1.1875 80 000000000000000000 0.000000000 0.000000000
018 5426.65 190 1.1875 80 000000000000000000 0.000000000 0.000000000
019 5426.65 190 1.1875 80 000000000000000000 0.000000000 0.000000000
020 5426.65 190 1.1875 80 000000000000000000 0.000000000 0.000000000
021 5426.65 190 1.1875 80 000000000000000000 0.000000000 0.000000000
022 5426.65 190 1.1875 80 000000000000000000 0.000000000 0.000000000
023 5426.65 190 1.1875 80 000000000000000000 0.000000000 0.000000000
024 5376.40 190 1.1875 81 000000000000000000 0.000000000 0.000000000
025 5376.30 190 1.1875 81 000000000000000000 0.000000000 0.000000000
026 5376.30 190 1.1875 81 000000000000000000 0.000000000 0.000000000
027 5376.34 190 1.1875 81 000000000000000000 0.000000000 0.000000000
028 5376.40 190 1.1875 81 000000000000000000 0.000000000 0.000000000
029 5376.35 190 1.1875 81 000000000000000000 0.000000000 0.000000000
030 5376.30 190 1.1875 81 000000000000000000 0.000000000 0.000000000
031 5376.35 190 1.1875 81 000000000000000000 0.000000000 0.000000000
Package[0] Cores Uncore Memory Platform
Energy(J): 220.729217529 167.909622192 0.000000000 0.000000000 0.000000000
Power(W) : 220.729217529 167.909622192 0.000000000 0.000000000 0.000000000
~ ▓▒░·································································································
❯ sensors k10temp-pci-00c3
k10temp-pci-00c3
Adapter: PCI adapter
Tctl: +83.1°C
Additionally,
corefreq-cli -C 1sometimes fails to respond, and I need to press Ctrl+C and rerun it multiple times to obtain the output.
This might be because I have another instance of corefreq running. corefreq-cli -C 1 returns promptly once I close the other one.
Hi the other 9950X did not encounter this issue. What values are you reading when Cores are fully stressed ? Are those still different than k10temp ?
Yes, but they don't vary that much:
~ ▓▒░································································································· ❯ corefreq-cli -C 1 CPU Freq(MHz) VID Vcore TMP(C) Accumulator Energy(J) Power(W) 000 5426.59 192 1.2000 80 000000000000678095 10.346908569 10.346908569 001 5426.64 192 1.2000 80 000000000000685677 10.462600708 10.462600708 002 5426.64 192 1.2000 80 000000000000675926 10.313812256 10.313812256 003 5426.59 192 1.2000 80 000000000000695846 10.617767334 10.617767334 004 5426.64 192 1.2000 80 000000000000687470 10.489959717 10.489959717 005 5426.64 192 1.2000 80 000000000000695677 10.615188599 10.615188599 006 5426.64 190 1.1875 80 000000000000684540 10.445251465 10.445251465 007 5426.64 190 1.1875 80 000000000000693621 10.583816528 10.583816528 008 5376.40 190 1.1875 81 000000000000693943 10.588729858 10.588729858 009 5376.35 190 1.1875 81 000000000000680012 10.376159668 10.376159668 010 5376.30 190 1.1875 81 000000000000697079 10.636581421 10.636581421 011 5376.35 190 1.1875 81 000000000000685131 10.454269409 10.454269409 012 5376.30 190 1.1875 81 000000000000695940 10.619201660 10.619201660 013 5376.30 190 1.1875 81 000000000000691096 10.545288086 10.545288086 014 5376.35 190 1.1875 81 000000000000701210 10.699615479 10.699615479 015 5376.35 190 1.1875 81 000000000000679342 10.365936279 10.365936279 016 5426.64 190 1.1875 80 000000000000000000 0.000000000 0.000000000 017 5426.64 190 1.1875 80 000000000000000000 0.000000000 0.000000000 018 5426.65 190 1.1875 80 000000000000000000 0.000000000 0.000000000 019 5426.65 190 1.1875 80 000000000000000000 0.000000000 0.000000000 020 5426.65 190 1.1875 80 000000000000000000 0.000000000 0.000000000 021 5426.65 190 1.1875 80 000000000000000000 0.000000000 0.000000000 022 5426.65 190 1.1875 80 000000000000000000 0.000000000 0.000000000 023 5426.65 190 1.1875 80 000000000000000000 0.000000000 0.000000000 024 5376.40 190 1.1875 81 000000000000000000 0.000000000 0.000000000 025 5376.30 190 1.1875 81 000000000000000000 0.000000000 0.000000000 026 5376.30 190 1.1875 81 000000000000000000 0.000000000 0.000000000 027 5376.34 190 1.1875 81 000000000000000000 0.000000000 0.000000000 028 5376.40 190 1.1875 81 000000000000000000 0.000000000 0.000000000 029 5376.35 190 1.1875 81 000000000000000000 0.000000000 0.000000000 030 5376.30 190 1.1875 81 000000000000000000 0.000000000 0.000000000 031 5376.35 190 1.1875 81 000000000000000000 0.000000000 0.000000000 Package[0] Cores Uncore Memory Platform Energy(J): 220.729217529 167.909622192 0.000000000 0.000000000 0.000000000 Power(W) : 220.729217529 167.909622192 0.000000000 0.000000000 0.000000000 ~ ▓▒░································································································· ❯ sensors k10temp-pci-00c3 k10temp-pci-00c3 Adapter: PCI adapter Tctl: +83.1°C
High temperature are closed but not idle !
Additionally,
corefreq-cli -C 1sometimes fails to respond, and I need to press Ctrl+C and rerun it multiple times to obtain the output.This might be because I have another instance of corefreq running.
corefreq-cli -C 1returns promptly once I close the other one.
But you must not do that. I will insist on this in the Readme
Meanwhile, Power(W) also seems strange, and the readings for CCD2 are missing
Meanwhile,
Power(W)also seems strange, and the readings for CCD2 are missing
Are max Power 220 W Package and 167 W total Cores what you expect ?
Can you report about the specifications values ?
Where do you find CCD2 ? Don't know which register it is referring to ?
Meanwhile,
Power(W)also seems strange, and the readings for CCD2 are missingAre max Power
220 WPackage and167 Wtotal Cores what you expect ? Can you report about the specifications values ?
Guessing so, but why is the per-core power shown as ~10W?
Where do you find CCD2 ? Don't know which register it is referring to ?
oops, maybe CPU 0-15 are physical cores while 16-31 are hyperthreading logical cores so their power is shown as zero? I mistakenly thought 16-31 were CCD2...
Meanwhile,
Power(W)also seems strange, and the readings for CCD2 are missingAre max Power
220 WPackage and167 Wtotal Cores what you expect ? Can you report about the specifications values ?Guessing so, but why is the per-core power shown as ~10W?
Where do you find CCD2 ? Don't know which register it is referring to ?
oops, maybe CPU 0-15 are physical cores while 16-31 are hyperthreading logical cores so their power is shown as zero? I mistakenly thought 16-31 were CCD2...
Indeed you won't find RAPL registers in a logical Core In short, CoreFreq is doing the sum of all Physical Cores. Package watt is coming from SMU and may add/include the consumption of other IPs
Back to the temperature issue: I have no clue why idle value makes the gape between CoreFreq and k10temp. Don't you have other source of software to show ? Even temperature reporting from tools for Windows ?
Back to the temperature issue: I have no clue why idle value makes the gape between CoreFreq and k10temp. Don't you have other source of software to show ? Even temperature reporting from tools for Windows ?
hwinfo and aida64 idle temperatures are consistent with k10temp, so CoreFreq must be problematic.
@vient @madoverlord40 @huajian628
Hello,
Can you tell if latest version of CoreFreq reports wrong temperature with your 9950X ? Especially low temperature during the idle states.
Started with idle, then run stress 0-3 for 5s, stress 8-15 for 5s, stress 0-15 for 10s.
https://github.com/user-attachments/assets/1a07d821-730a-4b7d-b918-8131d3ab8b34
Kinda looks like k10temp has an exponential smoothing, with larger alpha when real temperature is bigger than current approximation. Overall looks pretty similar.
Don't know what k10temp is showing, Zen 5 support was added in 6.16 (so not even released yet) while I'm running 6.11.
Don't know what k10temp is showing, Zen 5 support was added in 6.16 (so not even released yet) while I'm running 6.11.
Awesome, thank you.
@moetayuko What is your Linux kernel version ?
You can post here the output of corefreq-cli -k
Don't know what k10temp is showing, Zen 5 support was added in 6.16 (so not even released yet) while I'm running 6.11.
Awesome, thank you.
@moetayuko What is your Linux kernel version ? You can post here the output of
corefreq-cli -k
❯ corefreq-cli -k
Linux:
|- Release [6.15.3-zen1-1-zen]
|- Version [#1 ZEN SMP PREEMPT_DYNAMIC Thu, 19 Jun 2025 14:41:01 +0000]
|- Machine [x86_64]
Memory:
|- Total RAM 98399876 KB
|- Shared RAM 308060 KB
|- Free RAM 78658588 KB
|- Buffer RAM 4556 KB
|- Total High 0 KB
|- Free High 0 KB
Clock Source < tsc>
CPU-Freq driver [ amd-pstate-epp]
Governor [ Missing]
CPU-Idle driver [ acpi_idle]
|- Idle Limit [ C3]
|- State POLL C1 C2 C3
|- CPUIDLE ACPI FF ACPI IO ACPI IO
|- Power -1 0 0 0
|- Latency 0 1 18 350
|- Residency 0 2 36 700
Just in case you really want to dig in, 6.11 has some code about 1Ah models which was removed in 6.12. ccd_offset is missing so I guess my k10temp only reports Tctl while CoreFreq reports Tccd1 and Tccd2 (?)
Don't know what k10temp is showing, Zen 5 support was added in 6.16 (so not even released yet) while I'm running 6.11.
The commit description mentions that it's adding support for retrieving CCD temperatures, ig CCD temperature/package temperature/per-core temperature are different concepts?
Right, README mentions that Tctl is some "synthetic" value used for fan control (guess that means that smoothing is performed by chip itself), while Tdie may provide real value and TccxN may also be available.
I also like the note in driver's code
The information [about converting raw sensor values to temperature] is not confirmed from chip datasheets, but experiments suggest that it provides reasonable temperature values.
@moetayuko but before I go into code investigation, can you provide the idle and stressed temperature values from a third party software. Preferably a Windows temperature software provided by your mainboard manufacturer ?
@moetayuko but before I go into code investigation, can you provide the idle and stressed temperature values from a third party software. Preferably a Windows temperature software provided by your mainboard manufacturer ?
idle temperature in windows is around 52℃, close to k10temp
stressed temperature is another story.
I use cinebench r23 in windows, and the temperature reaches the 95℃ threshold as set in bios, meanwhile package power reaches 270w.
in linux, the stress command is used, the temperature reaches only 81-83℃ as shown in both k10temp and corefreq, and the package power given by corefreq is only 220w.
I can't tell if the stressed temperature reported by k10temp is wrong, or stress failed to provide enough stress as cinebench did.
Should be right, stress is extremely simple. You can try stress-ng instead, something like
docker run --rm ghcr.io/colinianking/stress-ng --tz --rapl -t 60 --vnni -1
idle temperature in windows is around 52℃
Thank you. Too me that's the main issue.
@moetayuko
Based on latest version, can you edit source code for your architecture [AMD_Zen5_Eldora] at this line:
https://github.com/cyring/CoreFreq/blob/deb3da7f0049f41794ba27854591f84bac6cc9e2/x86_64/corefreqk.h#L13381
you replace with that line:
.Query = Query_AMD_F17h_PerCluster,
then you rebuild:
make clean; make
restart CoreFreq and please post the idle temperature
Don't know what k10temp is showing, Zen 5 support was added in 6.16 (so not even released yet) while I'm running 6.11.
I have checked: CoreFreq is already employing the same register address offset https://github.com/cyring/CoreFreq/blob/deb3da7f0049f41794ba27854591f84bac6cc9e2/x86_64/amd_reg.h#L212
Based on latest version, can you edit source code for your architecture
[AMD_Zen5_Eldora]at this line:Line 13381 in deb3da7 .Query = Query_AMD_F19h_61h_PerCluster,
you replace with that line:
.Query = Query_AMD_F17h_PerCluster,
then you rebuild:
make clean; make
restart CoreFreq and please post the idle temperature
all temperatures become zero after applying this to 2.0.7
all temperatures become zero after applying this to 2.0.7
Thank you for trying.
You can now reset the file to its original content
cd CoreFreq
git checkout x86_64/corefreqk.h
We will try to apply a second negative offset which will be added in the temperature formula
Temp = ((Sensor * 5 / 40) - Param.Offset[1]) - Param.Offset[2]
Specific to 9950X, edit file at this line: https://github.com/cyring/CoreFreq/blob/deb3da7f0049f41794ba27854591f84bac6cc9e2/x86_64/corefreqk.h#L9288
and change offset parameter as bellow
.Param.Offset = {95, 0, -18},
Rebuild and restart CoreFreq to test the temperature
By the way, I will also need the topology to check the CCD/CCX clusters.
Please provide the output of corefreq-cli -m
Thank you
Just in case you really want to dig in, 6.11 has some code about 1Ah models which was removed in 6.12.
ccd_offsetis missing so I guess my k10temp only reports Tctl while CoreFreq reports Tccd1 and Tccd2 (?)
Indeed in k10temp_read_temp() function k10temp can switch to /* Tctl */ or /* Tdie */ or /* Tccd{1-12} */ based on its channel numbering
On his side, CoreFreq sticks with its CCD temperature function.
So what would be the best temperature to report per Core ? Tctl or Tccd ?
Other thing.
k10temp appears to unconditionally subtract 49C to Tccd
*val = (regval & ZEN_CCD_TEMP_MASK) * 125 - 49000;
whereas CoreFreq monitors bit 11 (CurTempRangeSel) to conditionally subtract the 49 offset
https://github.com/cyring/CoreFreq/blob/deb3da7f0049f41794ba27854591f84bac6cc9e2/x86_64/corefreqk.c#L16516
Specific to 9950X, edit file at this line:
Line 9288 in deb3da7 .Param.Offset = {95, 0, 0},
and change offset parameter as bellow
.Param.Offset = {95, 0, -18},
Rebuild and restart CoreFreq to test the temperature
something wrong with this change
❯ corefreq-cli -C 1
*** buffer overflow detected ***: terminated
[1] 7045 IOT instruction (core dumped) corefreq-cli -C 1
By the way, I will also need the topology to check the CCD/CCX clusters. Please provide the output of
corefreq-cli -m
❯ corefreq-cli -m
CPU Pkg Apic Core/Thread Caches (w)rite-Back (i)nclusive
# ID ID CCD CCX ID/ID L1-Inst Way L1-Data Way L2 Way L3 Way
000:BSP 0 0 0 0 0 32 8 48 12 1024 16 i 65536 16w
001: 0 2 0 0 1 0 32 8 48 12 1024 16 i 65536 16w
002: 0 4 0 0 2 0 32 8 48 12 1024 16 i 65536 16w
003: 0 6 0 0 3 0 32 8 48 12 1024 16 i 65536 16w
004: 0 8 0 1 4 0 32 8 48 12 1024 16 i 65536 16w
005: 0 10 0 1 5 0 32 8 48 12 1024 16 i 65536 16w
006: 0 12 0 1 6 0 32 8 48 12 1024 16 i 65536 16w
007: 0 14 0 1 7 0 32 8 48 12 1024 16 i 65536 16w
008: 0 16 1 2 8 0 32 8 48 12 1024 16 i 65536 16w
009: 0 18 1 2 9 0 32 8 48 12 1024 16 i 65536 16w
010: 0 20 1 2 10 0 32 8 48 12 1024 16 i 65536 16w
011: 0 22 1 2 11 0 32 8 48 12 1024 16 i 65536 16w
012: 0 24 1 3 12 0 32 8 48 12 1024 16 i 65536 16w
013: 0 26 1 3 13 0 32 8 48 12 1024 16 i 65536 16w
014: 0 28 1 3 14 0 32 8 48 12 1024 16 i 65536 16w
015: 0 30 1 3 15 0 32 8 48 12 1024 16 i 65536 16w
016: 0 1 0 0 0 1 32 8 48 12 1024 16 i 65536 16w
017: 0 3 0 0 1 1 32 8 48 12 1024 16 i 65536 16w
018: 0 5 0 0 2 1 32 8 48 12 1024 16 i 65536 16w
019: 0 7 0 0 3 1 32 8 48 12 1024 16 i 65536 16w
020: 0 9 0 1 4 1 32 8 48 12 1024 16 i 65536 16w
021: 0 11 0 1 5 1 32 8 48 12 1024 16 i 65536 16w
022: 0 13 0 1 6 1 32 8 48 12 1024 16 i 65536 16w
023: 0 15 0 1 7 1 32 8 48 12 1024 16 i 65536 16w
024: 0 17 1 2 8 1 32 8 48 12 1024 16 i 65536 16w
025: 0 19 1 2 9 1 32 8 48 12 1024 16 i 65536 16w
026: 0 21 1 2 10 1 32 8 48 12 1024 16 i 65536 16w
027: 0 23 1 2 11 1 32 8 48 12 1024 16 i 65536 16w
028: 0 25 1 3 12 1 32 8 48 12 1024 16 i 65536 16w
029: 0 27 1 3 13 1 32 8 48 12 1024 16 i 65536 16w
030: 0 29 1 3 14 1 32 8 48 12 1024 16 i 65536 16w
031: 0 31 1 3 15 1 32 8 48 12 1024 16 i 65536 16w
something wrong with this change
Can you stop completely CoreFreq Delete its project directory and then clone it again from latest version.
First build and run straight to check if it is working back.
Next attempt to make the offset change Param.Offset = {95, 0, -18}
But make sure to fully recompile and reload everything
make clean
make
rmmod corefreqk
insmod build/corefreqk.ko
something wrong with this change
Can you stop completely CoreFreq Delete its project directory and then clone it again from latest version.
First build and run straight to check if it is working back.
Next attempt to make the offset change
Param.Offset = {95, 0, -18}But make sure to fully recompile and reload everything
make clean make rmmod corefreqk insmod build/corefreqk.ko
same. I clean built everything and restarted to ensure kernel module is reloaded. below are the backtraces:
#0 __GI_abort () at abort.c:91
#1 0x00007fc11697d5c9 in __libc_message_impl (fmt=fmt@entry=0x7fc116b09216 "*** %s ***: terminated\n") at ../sysdeps/posix/libc_fatal.c:134
#2 0x00007fc116a7ba09 in __GI___fortify_fail (msg=msg@entry=0x7fc116b091fd "buffer overflow detected") at fortify_fail.c:24
#3 0x00007fc116a7b374 in __GI___chk_fail () at chk_fail.c:28
#4 0x000055e5947a2e8d in memcpy (__dest=0x55e59820ce7d, __src=0x55e59820cea0, __len=89) at /usr/include/bits/string_fortified.h:29
#5 Sensors (iter=<optimized out>) at x86_64/corefreq-cli.c:7285
#6 0x000055e5947879ef in main (argc=3, argv=<optimized out>) at x86_64/corefreq-cli.c:22721
(gdb) bt full
#0 __GI_abort () at abort.c:91
act = {__sigaction_handler = {sa_handler = 0x0, sa_sigaction = 0x0}, sa_mask = {__val = {18446744073709551615, 0 <repeats 15 times>}}, sa_flags = 0, sa_restorer = 0x0}
#1 0x00007fc11697d5c9 in __libc_message_impl (fmt=fmt@entry=0x7fc116b09216 "*** %s ***: terminated\n") at ../sysdeps/posix/libc_fatal.c:134
ap = {{gp_offset = 16, fp_offset = 0, overflow_arg_area = 0x7fffe4840c10, reg_save_area = 0x7fffe4840ba0}}
fd = 2
iov = {{iov_base = 0x7fc116b09216, iov_len = 4}, {iov_base = 0x7fc116b091fd, iov_len = 24}, {iov_base = 0x7fc116b0921c, iov_len = 17}, {iov_base = 0x7fffe4840b60, iov_len = 9537642091124368896}, {iov_base = 0x7fffe4840b70, iov_len = 9537642091124368896}, {
iov_base = 0x7fffe4840a50, iov_len = 140467284565916}, {iov_base = 0x1f, iov_len = 152}}
iovcnt = <optimized out>
total = <optimized out>
cp = <optimized out>
#2 0x00007fc116a7ba09 in __GI___fortify_fail (msg=msg@entry=0x7fc116b091fd "buffer overflow detected") at fortify_fail.c:24
No locals.
#3 0x00007fc116a7b374 in __GI___chk_fail () at chk_fail.c:28
No locals.
#4 0x000055e5947a2e8d in memcpy (__dest=0x55e59820ce7d, __src=0x55e59820cea0, __len=89) at /usr/include/bits/string_fortified.h:29
No locals.
#5 Sensors (iter=<optimized out>) at x86_64/corefreq-cli.c:7285
idx = <optimized out>
sdx = 77
pw = <optimized out>
cpu = <optimized out>
rdx = <optimized out>
ldx = 89
out = 0x55e59820c350 "CPU Freq(MHz) VID Vcore TMP(C) Accumulator Energy(J) Power(W)\n000 45.16 152 0.9500 4294901808 ", '0' <repeats 14 times>, "2947 0.044967651 0.044967651\n001 99.80 152 0.9500 429490180"...
row = 0x55e59820cea0 "\n", ' ' <repeats 13 times>, "Package[0] Cores Uncore Memory Platform\nEnergy(J):"
#6 0x000055e5947879ef in main (argc=3, argv=<optimized out>) at x86_64/corefreq-cli.c:22721
theme = <optimized out>
_dest = <optimized out>
tty = <optimized out>
tty = <optimized out>
tty = <optimized out>
tty = <optimized out>
iter = <optimized out>
iter = <optimized out>
iter = <optimized out>
iter = <optimized out>
iter = <optimized out>
iter = <optimized out>
usrIdx = <optimized out>
roSize = 290816
rwSize = <optimized out>
stat_st = {ro = {st_dev = 24, st_ino = 2, st_nlink = 1, st_mode = 33060, st_uid = 0, st_gid = 0, __pad0 = 0, st_rdev = 0, st_size = 290816, st_blksize = 4096, st_blocks = 568, st_atim = {tv_sec = 1751096343, tv_nsec = 55553474}, st_mtim = {tv_sec = 1751096343,
tv_nsec = 55991986}, st_ctim = {tv_sec = 1751096343, tv_nsec = 55991986}, __glibc_reserved = {0, 0, 0}}, rw = {st_dev = 24, st_ino = 3, st_nlink = 1, st_mode = 33206, st_uid = 0, st_gid = 0, __pad0 = 0, st_rdev = 0, st_size = 4096, st_blksize = 4096, st_blocks = 8,
st_atim = {tv_sec = 1751096343, tv_nsec = 55553474}, st_mtim = {tv_sec = 1751096343, tv_nsec = 55935175}, st_ctim = {tv_sec = 1751096343, tv_nsec = 55935175}, __glibc_reserved = {0, 0, 0}}}
fd = {ro = 3, rw = 4}
idx = 2
program = 0x55e59820b2a0 "corefreq-cli"
appName = 0x55e59820b2a0 "corefreq-cli"
option = 67 'C'
trailing = 0 '\000'
reason = <optimized out>