On Thu, May 28, 2020 at 05:44:31PM +0300, Paul Irofti wrote: > Hi, > > Here is a new iteration of the diff which includes support for MD high > resolution clocks. Currently only implements TSC on amd64. If the > MD function is not defined, it fallsback to the syscall. > > There is the question of the skew fix, but that will be addressed in a > separate kernel diff that will not affect the current diff at all. > > I could not find a way to find on which processor the process is running > on from userland without going through a syscall. If there is one please > let me know. It would make things easier. > > In the meantime I have also gotten positive feedback from various > testers that run this on their main machine. > > Anyway, I think we can decide on the struct name and the auxiliary > vector ID and consider this done. > > Thoughts? > > Paul
This appears to work just fine here on my desktop. cpu0: AMD A10-5700 APU with Radeon(tm) HD Graphics, 3417.45 MHz, 15-10-01 cpu0: FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,MMX,FXSR,SSE,SSE2,HTT,SSE3,PCLMUL,MWAIT,SSSE3,FMA3,CX16,SSE4.1,SSE4.2,POPCNT,AES,XSAVE,AVX,F16C,NXE,MMXX,FFXSR,PAGE1GB,RDTSCP,LONG,LAHF,CMPLEG,SVM,EAPICSP,AMCR8,ABM,SSE4A,MASSE,3DNOWP,OSVW,IBS,XOP,SKINIT,WDT,FMA4,TCE,NODEID,TBM,TOPEXT,CPCTR,ITSC,BMI1 (other bits of dmesg available on demand)