Re: Loading vmm

From: John F Carr <jfc_at_mit.edu>
Date: Sun, 14 Apr 2024 12:17:22 UTC
See bug https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=277559.  It was suspected to be specific to the heterogenous Rockchip RK3399 but perhaps it is more widespread.

When I first hit the bug there was a 30-50% chance that the first kldload vmm would hang the system so hard I couldn't even invoke the debugger with break on a serial console.  Later, with no relevant changes to source code, it would take about 2,000 load/unload pairs to hang the system.  If that 1-in-2000 chance is typical this bug could affect all systems.  Give it a try on your 128 core system:

# while kldload vmm ; do echo -n . ; kldunload vmm ; done

If anybody can tell me how to debug at the hardware level using JTAG, I can give it a try on my RockPro64 and see where the kernel is stuck.  I have used hardware level debugging tools, but they were always black boxes my employer bought with instructions included.

> On Apr 14, 2024, at 04:50, tuexen@freebsd.org wrote:
> 
> Dear all,
> 
> using the sources for kernel/world of yesterday running
> kldload vmm
> on a 128 core Ampere system runs just fine:
> 
> CPU  0: ARM Neoverse-N1 r3p1 affinity: 18  0  0
>                  Cache Type = <64 byte D-cacheline,64 byte I-cacheline,PIPT ICache,64 byte ERG,64 byte CWG,IDC>
> Instruction Set Attributes 0 = <DP,RDM,Atomic,CRC32,SHA2,SHA1,AES+PMULL>
> Instruction Set Attributes 1 = <RCPC-8.3,DCPoP>
> Instruction Set Attributes 2 = <>
>        Processor Features 0 = <CSV3,CSV2,RAS,GIC,AdvSIMD+HP,FP+HP,EL3,EL2,EL1,EL0 32>
>        Processor Features 1 = <PSTATE.SSBS MSR>
>     Memory Model Features 0 = <TGran4,TGran64,TGran16,SNSMem,BigEnd,16bit ASID,256TB PA>
>     Memory Model Features 1 = <XNX,PAN+ATS1E1,LO,HPD+TTPBHA,VH,16bit VMID,HAF+DS>
>     Memory Model Features 2 = <EVT-8.2,32bit CCIDX,48bit VA,UAO,CnP>
>            Debug Features 0 = <DoubleLock,SPE,2 CTX BKPTs,4 Watchpoints,6 Breakpoints,PMUv3p1,Debugv8p2>
>            Debug Features 1 = <>
>        Auxiliary Features 0 = <>
>        Auxiliary Features 1 = <>
> AArch32 Instruction Set Attributes 5 = <RDM,CRC32,SHA2,SHA1,AES+VMULL,SEVL>
> AArch32 Media and VFP Features 0 = <FPRound,FPSqrt,FPDivide,DP VFPv3+v4,SP VFPv3+v4,AdvSIMD>
> AArch32 Media and VFP Features 1 = <SIMDFMAC,FPHP Arith,SIMDHP Arith,SIMDSP,SIMDInt,SIMDLS,FPDNaN,FPFtZ>
> 
> However, doing the same on a 32 core Ampere system results in
> the system becoming unresponsive. No reaction on the console,
> no message there, no response over the network.
> 
> CPU  0: APM eMAG 8180 r3p2 affinity:  0  0
>                Cache Type = <64 byte D-cacheline,64 byte I-cacheline,PIPT ICache,64 byte ERG,64 byte CWG>
> Instruction Set Attributes 0 = <CRC32,SHA2,SHA1,AES+PMULL>
> Instruction Set Attributes 1 = <>
> Instruction Set Attributes 2 = <>
>      Processor Features 0 = <GIC,AdvSIMD,FP,EL3,EL2,EL1 32,EL0 32>
>      Processor Features 1 = <>
>   Memory Model Features 0 = <TGran4,TGran64,TGran16,SNSMem,BigEnd,16bit ASID,4TB PA>
>   Memory Model Features 1 = <8bit VMID>
> Trying to mount root from ufs:/dev/ada0p2 [rw]...
>   Memory Model Features 2 = <32bit CCIDX,48bit VA>
>          Debug Features 0 = <DoubleLock,2 CTX BKPTs,4 Watchpoints,6 Breakpoints,PMUv3,Debugv8>
>          Debug Features 1 = <>
>      Auxiliary Features 0 = <>
>      Auxiliary Features 1 = <>
> AArch32 Instruction Set Attributes 5 = <CRC32,SHA2,SHA1,AES+VMULL,SEVL>
> AArch32 Media and VFP Features 0 = <FPRound,FPSqrt,FPDivide,DP VFPv3+v4,SP VFPv3+v4,AdvSIMD>
> AArch32 Media and VFP Features 1 = <SIMDFMAC,FPHP DP Conv,SIMDHP SP Conv,SIMDSP,SIMDInt,SIMDLS,FPDNaN,FPFtZ>
> 
> Any idea, what is going wrong?
> 
> Best regards
> Michael