v1.12.9 high cpu usage

[Notation]

Hello, after upgrading from v1.12.8 to v1.12.9, the cpu usage of core-geth continues to be high. How can I solve this problem? HDD io is OK

System information

• core-geth info: CoreGeth Version: 1.12.9-stable Git Commit: c17c5ae18e393971d3f7910330305526d902e422 Architecture: amd64 Go Version: go1.18.10 Operating System: linux GOPATH= GOROOT=/opt/hostedtoolcache/go/1.18.10/x64
• system info: Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Byte Order: Little Endian Address sizes: 46 bits physical, 48 bits virtual CPU(s): 8 On-line CPU(s) list: 0-7 Thread(s) per core: 2 Core(s) per socket: 4 Socket(s): 1 NUMA node(s): 1 Vendor ID: GenuineIntel CPU family: 6 Model: 85 Model name: Intel(R) Xeon(R) Platinum 8269CY CPU @ 2.50GHz Stepping: 7 CPU MHz: 2500.014 BogoMIPS: 5000.02 Hypervisor vendor: KVM Virtualization type: full L1d cache: 128 KiB L1i cache: 128 KiB L2 cache: 4 MiB L3 cache: 35.8 MiB NUMA node0 CPU(s): 0-7 Vulnerability Itlb multihit: KVM: Vulnerable Vulnerability L1tf: Mitigation; PTE Inversion Vulnerability Mds: Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown Vulnerability Meltdown: Mitigation; PTI Vulnerability Spec store bypass: Vulnerable Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization Vulnerability Spectre v2: Mitigation; Full generic retpoline, IBPB conditional, IBRS_FW, STIBP conditional, RSB filling Vulnerability Srbds: Not affected Vulnerability Tsx async abort: Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 s s ht syscall nx pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid tsc_known_freq pni p clmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16 c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single pti ibrs ibpb stibp fsgsbase tsc_adjus t bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx avx512f avx512dq rdseed adx smap avx512cd avx512bw a vx512vl xsaveopt xsavec xgetbv1 arat

Expected behaviour

less cpu usage

Actual behaviour

v1.12.8

v1.12.9

Steps to reproduce the behaviour

upgrade from v1.12.8 to v1.12.9

Backtrace

[meowsbits]

Thanks for the report. I'm just in the middle of looking at this now.

txSenderCacher seems suspect.

ia@ianuc:~/Downloads$ go tool pprof profile\(2\)                                                                                                                                                                
File: geth                                                                                                                                                                                                      Build ID: 2a4f22ee8855c29f44a2a89684537cf6e595fa37                                                                                                                                                              
Type: cpu                                                                                                                                                                                                       
Time: Feb 6, 2023 at 11:07am (PST)                                                                                                                                                                              
Duration: 5.18s, Total samples = 57.12s (1102.14%)                                                                                                                                                              Entering interactive mode (type "help" for commands, "o" for options)                                                                                                                                           
(pprof) top                                                                                                                                                                                                     
Showing nodes accounting for 57.04s, 99.86% of 57.12s total                                                                                                                                                     Dropped 34 nodes (cum <= 0.29s)                                                                         
      flat  flat%   sum%        cum   cum%                                                                                                                                                                      
    21.40s 37.46% 37.46%     57.05s 99.88%  github.com/ethereum/go-ethereum/core.(*txSenderCacher).cache                                                                                                        
    15.99s 27.99% 65.46%     21.36s 37.39%  runtime.chanrecv                                                                                                                                                    
    14.28s 25.00% 90.46%     35.64s 62.39%  runtime.selectnbrecv                                                                                                                                                
     5.37s  9.40% 99.86%      5.37s  9.40%  runtime.empty (inline)