I noted this /after /svn updating my 11.x box to the most-current -STABLE:
A bug was diagnosed in interaction of the |pmap_activate()| function
and TLB shootdownIPI handler on amd64 systems which
have PCID features but do not implement theINVPCID instruction. On
such machines, such as SandyBridge? and IvyBridge?
microarchitectures, set the loader
tunable |vm.pmap.pcid_enabled=0| during boot:
set vm.pmap.pcid_enabled=0
boot
Add this line to |/boot/loader.conf| for the change to persist
across reboots:
To check if the system is affected, check dmesg(8)
<http://www.freebsd.org/cgi/man.cgi?query=dmesg&sektion=8&manpath=freebsd-release-ports>
for PCID listed
in the "Features2", and absence of INVPCID in the "Structured
Extended Features". If the PCID feature is not present,
or INVPCID is present, system is not affected.
Well, I'm allegedly subject to this:
Copyright (c) 1992-2016 The FreeBSD Project.
Copyright (c) 1979, 1980, 1983, 1986, 1988, 1989, 1991, 1992, 1993, 1994
The Regents of the University of California. All rights reserved.
FreeBSD is a registered trademark of The FreeBSD Foundation.
FreeBSD 11.0-STABLE #13 r307318M: Fri Oct 14 09:23:46 CDT 2016
karl at NewFS.denninger.net:/usr/obj/usr/src/sys/KSD-SMP amd64
FreeBSD clang version 3.8.0 (tags/RELEASE_380/final 262564) (based on
LLVM 3.8.0)
VT(vga): text 80x25
CPU: Intel(R) Xeon(R) CPU E5620 @ 2.40GHz (2400.13-MHz
K8-class CPU)
Origin="GenuineIntel" Id=0x206c2 Family=0x6 Model=0x2c
Stepping=2
Features=0xbfebfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CLFLUSH,DTS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE>
Features2=0x29ee3ff<SSE3,PCLMULQDQ,DTES64,MON,DS_CPL,VMX,SMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,PCID,DCA,SSE4.1,SSE4.2,POPCNT,AESNI>
AMD Features=0x2c100800<SYSCALL,NX,Page1GB,RDTSCP,LM>
AMD Features2=0x1<LAHF>
VT-x: PAT,HLT,MTF,PAUSE,EPT,UG,VPID
TSC: P-state invariant, performance statistics
And I do _*not*_ have it turned off at this moment.
vm.pmap.invpcid_works: 0
vm.pmap.pcid_enabled: 1
But.... I've also yet to see any sort of misbehavior.
So the questions are:
1. What's the misbehavior I should expect to see if this is not shut off
(and it isn't)?
2. Should I take this machine down immediately to reboot it with
vm.pmap.pcid_enabled=0 in /boot/loader.conf?
A svn log perusal didn't see anything post the errata date that appears
to be related.... which makes me wonder.
Thanks in advance!
--
Karl Denninger
karl at denninger.net <mailto:karl at denninger.net>
/The Market Ticker/
/[S/MIME encrypted email preferred]/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 2996 bytes
Desc: S/MIME Cryptographic Signature
URL:
<http://lists.freebsd.org/pipermail/freebsd-stable/attachments/20161015/4a983be9/attachment.bin>