Re: [Qemu-devel] Actual TB code doesn't look like what was intended (TCG issue)?

[Max Filippov]

> > Hello guys.
> >
> > I'm running qemu on x86_64 host.
> > It's clean build from git sources dated 2011.05.19, commit 
> > 1fddfba129f5435c80eda14e8bc23fdb888c7187
> > I have the following output from "log trace,op,out_asm":
> >
> > Trace 0x4000a310 [d0026c92]
> > OP:
> >  ---- 0xd00000c0
> >  movi_i32 tmp1,$0xfffffff4
> >  add_i32 tmp0,ar9,tmp1
> >  qemu_ld32 ar1,tmp0,$0x0
> >
> >  ---- 0xd00000c3
> >  movi_i32 tmp1,$0xfffffff0
> >  add_i32 tmp0,ar9,tmp1
> >  qemu_ld32 ar0,tmp0,$0x0
> >
> > [...snip...]
> [...]
> > 0x4000a360:  xor    %esi,%esi
> > 0x4000a362:  callq  0x52edc2
> [...]
> > (gdb) x/25i 0x4000a330
> [...]
> >   0x4000a360:  mov    $0x1,%esi
> >   0x4000a365:  callq  0x52edc2 <__ldl_mmu>
> >   0x4000a36a:  mov    %eax,%ebp
> >   0x4000a36c:  sub    $0x44,%al
> > => 0x4000a36e:  lea    -0x10(%rbx),%esp
> >   0x4000a371:  mov    %ebp,0xc(%r14)
> >   0x4000a375:  mov    %r12d,%esi
> >   0x4000a378:  mov    %r12d,%edi
> >
> > Please note how the current instruction in gdb differ from what was said in 
> > OUT. This lea corrupts stack pointer and the next callq generates segfault.
> > Could please anyone familiar with TCG take a look at this, or suggest where 
> > I should look myself?
> 
> As Peter hinted, you're not looking at the code you think :-)
> Note how your original TCG code does loads:
> 
>    qemu_ld32 ar1,tmp0,$0x0
> 
> That $0x0 will end up in %RSI.  It's the mem index used to
> distinguish from user and privileged level accesses.  In your
> examples of host code, in one case it is 0 and in the other
> it is 1, so you're definitely not really looking at the same
> block in the same running conditions.

Yes, I've noticed it (however, after I sent this mail).
But (1) quoted OUT is the last OUT for this host address range in the log and 
(2) in gdb I set "b tlb_fill if retaddr == 0x4000a369" and made some steps.
You mean that I should look at previous OUTs for this address range?

Thanks.
-- Max