Closed notgood closed 2 years ago
cyring
commented
2 years ago Hello,
What I have noticed in BIOS is that those limits have to be changed from [AUTO] to any value, lower than their max. Then I able to increase/decrease the limits, from CoreFreq
Can you give a try in that way ?
cyring
commented
2 years ago EDIT:
10W.142W, plus 10W
notgood
commented
2 years ago Went to BIOS, lowered TDC and EDC to 100A, PPT to 100W, temp to 80C.
Corefreq still incorrectly displays stock and non-adjustable values for 5950X (TDC 95A, EDC 140A, PPT 142W)
As I mentioned, real and correct PPT is displayed in Corefreq "Power Limit (0 sec) PL1" field. I can adjust in any steps as well (until I hit some other temperature or current limits) Here is a some screenshots of burn test: adjusting PL1 to 80W, 100W and 120W results same consumption in Package power label.
cyring
commented
2 years ago 
cyring
commented
2 years ago PL1 and PL2 somehow reflected by dialing values into BIOSEDC TDCare not taken into account
0x59800) but couldn't find this BIOS value among bits.notgood
commented
2 years ago OK, tried installing https://gitlab.com/leogx9r/ryzen_smu/ driver and https://github.com/hattedsquirrel/ryzen_monitor frontend.
ryzen_smu is able to correctly retrieve EDC/TDC/PPT/TempLimit, both current and limits. It also can adjust all four variables.
So perhaps required registers can be found in ryzen_smu code.
cyring
commented
2 years ago OK, tried installing https://gitlab.com/leogx9r/ryzen_smu/ driver and https://github.com/hattedsquirrel/ryzen_monitor frontend.
ryzen_smu is able to correctly retrieve EDC/TDC/PPT/TempLimit, both current and limits. It also can adjust all four variables.
So perhaps required registers can be found in ryzen_smu code.
Those are undocumented, unspecified registers from Manufacturer.
I will rather implement the "safe" i2c/RSMI, RBI protocols.
notgood
commented
2 years ago From my limited testing, both corefreqk and ryzen_smu kernel drivers can coexist. Frontends also seem to run OK at the same time.
I was able to run the following things simultaneously:
Fits my needs I guess.
cyring
commented
2 years ago From my limited testing, both corefreqk and ryzen_smu kernel drivers can coexist. Frontends also seem to run OK at the same time.
I was able to run the following things simultaneously:
- burn test in corefreqd
- watch all the related data in corefreq-cli and ryzen_monitor
- adjust max allowed temperature using /sys/kernel/ryzen_smu_drv/rsmu_cmd
Fits my needs I guess.
Hello,
Can you give a try to the develop branch for those cases:
Feel free to add screenshots; thank you.
notgood
commented
2 years ago Installed git AUR packages: corefreq-client-git 1.88.4.r19.g39eeddd-1 corefreq-dkms-git 1.88.4.r19.g39eeddd-1 corefreq-server-git 1.88.4.r19.g39eeddd-1
Memory controller corefreq vs zentimings:
notgood
commented
2 years ago Corefreq sensors
notgood
commented
2 years ago Limits corefreq vs AMDMaster, TDC/EDC/PPT still stock instead of actual ones.
cyring
commented
2 years ago @notgood Thank you for your screenshots.
Apparently looks like based on develop branch : do you confirm ?
diocletiann
commented
2 years ago TDC/EDC/PPT, they are indeed still stock. I can't beat Ryzen-Master because only AMD knows those registers. Perhaps one day.
CTR/Hydra can do it on Windows, FYI.
notgood
commented
2 years ago AMD Ryzen TDC/EDC/PPT can be reliably read and changed using open source software on both Linux (ryzen_smu) and Windows (SMUDebugTool), I've tested it myself. SMU commands, addresses and arguments required to to so are all available in relevant gitlab/github repositories. To my understanding, this information was acquired by snooping on commands sent to CPU by Ryzen Master.
But I also understand and respect @cyring wish to only use officialy available and documented commands.
cyring
commented
2 years ago AMD Ryzen TDC/EDC/PPT can be reliably read and changed using open source software on both Linux (ryzen_smu) and Windows (SMUDebugTool), I've tested it myself. SMU commands, addresses and arguments required to to so are all available in relevant gitlab/github repositories. To my understanding, this information was acquired by snooping on commands sent to CPU by Ryzen Master.
But I also understand and respect @cyring wish to only use officialy available and documented commands.
Issue postponed until I get better TDC , EDC registers specs. PPT of 270W appears ok with RM
Running CoreFreq 1.88.4 on Ryzen 5950X, here is a sysinfo output:
corefreq-cli -s
Processor [AMD Ryzen 9 5950X 16-Core Processor] |- Architecture [Zen3/Vermeer] |- Vendor ID [AuthenticAMD] |- Firmware [ 56.53.0-2] |- Microcode [0x0a201016] |- Signature [ AF_21] |- Stepping [ 0] |- Online CPU [ 32/ 32] |- Base Clock [100.005] |- Frequency (MHz) Ratio Min 2200.15 < 22 > Max 3400.23 < 34 > |- Factory [100.000] 3400 [ 34 ] |- Performance |- P-State TGT 2200.15 < 22 > |- Turbo Boost [ UNLOCK] XFR 5100.35 [ 51 ] CPB 5000.34 [ 50 ] 1C 2800.19 < 28 > 2C 2200.15 < 22 > |- Uncore [ LOCK] Min 1800.12 [ 18 ] Max 1800.12 [ 18 ] |- TDP Level [ 0:0 ] |- Programmable [ UNLOCK] Instruction Set Extensions |- 3DNow!/Ext [N/N] ADX [Y] AES [Y] AVX/AVX2 [Y/Y] |- AVX512-F [N] AVX512-DQ [N] AVX512-IFMA [N] AVX512-PF [N] |- AVX512-ER [N] AVX512-CD [N] AVX512-BW [N] AVX512-VL [N] |- AVX512-VBMI [N] AVX512-VBMI2 [N] AVX512-VNMI [N] AVX512-ALG [N] |- AVX512-VPOP [N] AVX512-VNNIW [N] AVX512-FMAPS [N] AVX512-VP2I [N] |- AVX512-BF16 [N] BMI1/BMI2 [Y/Y] CLWB [Y] CLFLUSH/O [Y/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 [N] UMIP [N] 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] |- Collaborative Processor Performance Control CPPC [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 [Capable] |- Fused Multiply Add FMA | FMA4 [Capable] |- Hardware Lock Elision HLE [Missing] |- Instruction Based Sampling IBS [Capable] |- Instruction INVLPGB INVLPGB [Missing] |- 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] |- Memory Protection Extensions MPX [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] |- 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] |- 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] |- 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 [Missing] |- Virtual Mode Extension VME [Capable] |- Virtual Machine Extensions VMX [Missing] |- Extended xAPIC Support x2APIC [ xAPIC] |- XSAVE/XSTOR States XSAVE [Capable] |- xTPR Update Control xTPR [Missing] Mitigation mechanisms |- Indirect Branch Restricted Speculation IBRS [Capable] |- Indirect Branch Prediction Barrier IBPB [Capable] |- Single Thread Indirect Branch Predictor STIBP [Capable] |- Speculative Store Bypass Disable SSBD [Capable] |- Architectural - Predictive Store Forwarding PSFD [Capable] Technologies |- Data Cache Unit |- L1 Prefetcher L1 HW < ON> |- L2 Prefetcher L2 HW < ON> |- System Management Mode SMM-Lock [ ON] |- Simultaneous Multithreading SMT [ ON] |- PowerNow! CnQ [OFF] |- Core C-States CCx [ ON] |- Core Performance Boost CPB < ON> |- Watchdog Timer WDT < ON> |- Virtualization SVM [ ON] |- I/O MMU AMD-V [ ON] |- Version [ 0.1] |- Hypervisor [OFF] |- Vendor ID [ N/A] Performance Monitoring |- Version PM [N/A] |- Counters: General Fixed | 6 x 64 bits 3 x 64 bits |- Enhanced Halt State C1E
|- C2 UnDemotion C2U
|- 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]
|- Hardware-Controlled Performance States HWP [ ON]
|- Core C-States
|- C-States Base Address BAR [ 0x413 ]
|- 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 [Capable]
|- Performance Time Stamp Counter [Missing]
|- Data Fabric Performance Counter [Capable]
|- Core Performance Counter [Capable]
Power, Current & Thermal
|- Junction Temperature TjMax [49: 90C]
|- 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 [ 105 W]
|- Minimum Power Min [ 105 W]
|- Maximum Power Max [ 105 W]
|- Thermal Design Power Package < Enable>
|- Power Limit (0 sec) PL1 < 270 W>
|- Power Limit (0 sec) PL2 < 1200 W>
|- Thermal Design Power Core
|- Power Limit PL1 [Missing]
|- Thermal Design Power Uncore
|- Power Limit PL1 [Missing]
|- Thermal Design Power DRAM
|- Power Limit PL1 [Missing]
|- Thermal Design Power Platform
|- Power Limit PL1 [Missing]
|- Power Limit PL2 [Missing]
|- Package Power Tracking PPT [ 142 W]
|- Electrical Design Current EDC [ 140 A]
|- Thermal Design Current TDC < 95 A>
|- Units
|- Power watt [ 0.125000000]
|- Energy joule [ 0.000015259]
|- Window second [ 0.000976562]
What I'm really interested in is ability to display and control power/thermal limits: TDC, EDC, PPT and max temperature.
CoreFreq currently displays stock PPT, EDC, and TDC values for 5950X, instead of actual live ones. TDC field is editable, but adjusting it doesn't work. At the same time actual PPT value is displayed in "Power Limit (0 sec) PL1", and adjusting it does work.
User adjustable max temperature ("Platform Thermal Throttle Limit" in AMD BIOS, allows to cap max temperature below TjMax) seems no be unavailable in CoreFreq.
Here are my actual values, set in UEFI, seen in Win10 tools: PPT 270W, TDC 160A, EDC 180A, Temp Limit 80C.