Re: 10.0-BETA1 i386 on VirtualBox

On Mon, Nov 04, 2013 at 12:30:09PM +0100, Maciej Milewski wrote:
> 
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> On 03.11.2013 15:41, Konstantin Belousov wrote:
> > On Sun, Nov 03, 2013 at 05:46:01PM +0400, Gleb Smirnoff wrote:
> >>   Maciej, Boris,
> >>
> >> On Sun, Nov 03, 2013 at 04:29:11PM +0400, Boris Bobrov wrote:
> >> B> > I traced this down to r248521:
> >> B> > svn log -r248521
> >> B> >
> -----------------------------------------------------------------------
> >> B> > - r248521 | kib | 2013-03-19 16:08:15 +0100 (Tue, 19 Mar 2013) | 5
> >> B> > lines
> >> B> >
> >> B> > UFS support of the unmapped i/o for the user data buffers.
> >> B> >
> >> B> > Sponsored by:   The FreeBSD Foundation
> >> B> > Tested by:      pho, scottl, jhb, bf
> >> B> >
> >> B> >
> -----------------------------------------------------------------------
> >> B> > -
> >> B> >
> >> B> > The last working revision is 248520.
> >> B>
> >> B> Yes, I confirm that.
> >>
> >> Thanks for you help!
> >>
> >> Can you please now try unpatched vanilla 10.0-BETA2 iso file and when it
> >> boots set in loader (before kernel):
> >>
> >> OK set vfs.unmapped_buf_allowed=0
> >> OK boot
> >
> > Show what disk driver is used,  also show the output of hw.ncpu, both
> > from the guest.
> Konstantin,
> do you need any other/more info from guest host?
> Other than hw.ncpu=1
> Disk is
> ada0 at ahcich0 bus 0 scbus2 target 0 lun 0
> and second test instance is
> ada0 at ata0 bus 0 scbus0 target 0 lun 0
So both ahci and ata attached disks show the bug ?

Also please show me the CPU features banner from the boot in the VB,
like this:
CPU: Intel(R) Core(TM) i7-4770S CPU _at_ 3.10GHz (2993.13-MHz K8-class CPU)
  Origin = "GenuineIntel"  Id = 0x306c3  Family = 0x6  Model = 0x3c  Stepping = 3
  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=0x7ffafbff<SSE3,PCLMULQDQ,DTES64,MON,DS_CPL,VMX,SMX,EST,TM2,SSSE3,<b11>,FMA,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,x2APIC,MOVBE,POPCNT,TSCDLT,AESNI,XSAVE,OSXSAVE,AVX,F16C,RDRAND>
  AMD Features=0x2c100800<SYSCALL,NX,Page1GB,RDTSCP,LM>
  AMD Features2=0x21<LAHF,ABM>
  Standard Extended Features=0x2fbb<GSFSBASE,TSCADJ,BMI1,HLE,AVX2,SMEP,BMI2,ENHMOVSB,INVPCID,RTM>
  TSC: P-state invariant, performance statistics
real memory  = 34359738368 (32768 MB)


> 
> Is Gleb's loader hint the final right solution for this problem?

Probably not.  This smells like a bug in the vbox.  It is especially
suspicious that only some i/o transfers cause corruption.