[llvmlinux] 16bit boot with current patchset

Jan-Simon Möller dl9pf at gmx.de
Mon Mar 24 22:10:15 UTC 2014


Hi all!

I'm building x86_64 right now and we seem to miss  a bit in the 16bit code 
path for a successful but. I just don't know yet, what it is.

The last instructions in qemu are:
IN: 
0x0000000000006812:  add    %al,(%bx,%si)
0x0000000000006814:  add    %al,(%bx,%si)
0x0000000000006816:  add    %al,(%bx,%si)
0x0000000000006818:  add    %al,(%bx,%si)
0x000000000000681a:  add    %al,(%bx,%si)
0x000000000000681c:  add    %al,(%bx,%si)
0x000000000000681e:  add    %al,(%bx,%si)
0x0000000000006820:  lods   %ds:(%si),%al
0x0000000000006821:  or     $0xf,%al
0x0000000000006823:  add    %bl,0x14(%di)
0x0000000000006826:  sldt   (%bx,%si)

----------------
IN: 
0x0000000000006826:  sldt   (%bx,%si)

Servicing hardware INT=0x08
Servicing hardware INT=0x08
Servicing hardware INT=0x08
Servicing hardware INT=0x08
[.. stuck in servicing INT=0x08]

I assume we removed compiler flag during cleanup of 16bit code in our 
patchset.

Investigating ... 


-- 

Dipl.-Ing.
Jan-Simon Möller

jansimon.moeller at gmx.de


More information about the LLVMLinux mailing list