[fpc-devel] Might need some help with this one

[J. Gareth Moreton]

Okay, so I have to specify "-MObjfpc -Sh" to get it to compile due to 
what the code uses, and the bug still requires the use of -O2 to 
appear.  However... "fpc -MObjfpc -Sh -O2 -Ooregvar -alr -sr breakp.pp" 
produces a bad binary.

I removed the register comments so it's clearer:

     ...
# [11] pChar(result)[i] := BarSym[(progress >= (0.75 + i) / divs) or (i 
= int32(divs) - 1) and (progress >= 1)];
     cvtsi2ssl    %ireg16d,%mreg34ms
     movaps    %mreg34ms,%mreg35ms
     addss    _$BREAKP$_Ld1(%rip),%mreg35ms
     movl    %ireg18d,%ireg28d
     andl    $4294967295,%ireg28d
     cvtsi2ssq    %ireg28q,%mreg37ms
     movaps    %mreg35ms,%mreg38ms
     divss    %mreg37ms,%mreg38ms
     comiss    %mreg32ms,%mreg38ms
     jp    .Lj12
     jbe    .Lj10 <---- Register %ireg27q not initialised if this 
instruction branches to .Lj10
.Lj12:
     jmp    .Lj11
.Lj11:
     movl    %ireg18d,%ireg29d
     movslq    %ireg29d,%ireg30q
     subq    $1,%ireg30q
     movslq    %ireg16d,%ireg31q
     movq    %ireg31q,%ireg27q <---- Problem register: %ireg27q
     cmpq    %ireg27q,%ireg30q
     je    .Lj14
     jmp    .Lj15
.Lj14:
     comiss    _$BREAKP$_Ld2(%rip),%mreg32ms
     jp    .Lj17
     jae    .Lj16
.Lj17:
     jmp    .Lj15
.Lj16:
     jmp    .Lj10
.Lj15:
     jmp    .Lj13
.Lj10:
     movq    $1,%ireg32q
     jmp    .Lj18
.Lj13:
     movq    $0,%ireg32q
.Lj18:
     movq    (%ireg17q),%ireg33q
     cmpq    $0,%ireg33q
     jne    .Lj19
     leaq    FPC_EMPTYCHAR(%rip),%ireg33q
.Lj19:
     leaq 
TC_$P$BREAKP$_$BAR$SINGLE$LONGWORD$$ANSISTRING_$$_BARSYM(%rip),%ireg34q
     movb    (%ireg34q,%ireg32q,1),%ireg35l
     movb    %ireg35l,(%ireg33q,%ireg27q,1) <---- Problem register: %ireg27q
     cmpl    %ireg25d,%ireg16d
     jge    .Lj9
     jmp    .Lj7
.Lj9:
.Lj6:
     ...

The problem register is %ireg27q - it's not being initialised if "jbe 
.Lj10" branches.  By your explanation, it seems that the code generator 
is buggy on one of the nodes.  Thanks for the tips in isolating where it 
could be.  To help out, I've attached the node dump file for the test 
program.

Gareth aka. Kit

On 26/11/2020 19:04, Yuriy Sydorov via fpc-devel wrote:
> Hi,
>
> On 26.11.2020 17:34, J. Gareth Moreton via fpc-devel wrote:
>> Hi everyone,
>>
>> So a couple of people have reported that -O2 sometimes produces bad 
>> code under x86_64.  So far it seems isolated to that CPU.
>>
>> https://bugs.freepascal.org/view.php?id=38129
>>
>> After my own investigations with the attached code, the problem still 
>> occurs even if the peephole optimizer is disabled, and the 
>> uninitialised register is being allocated within conditional code 
>> that is not always executed, rather than before or after it.
>>
>> Anyone with any tips on where to dig next (register allocator, node 
>> converter etc.) would be most appreciated!
>
> First compile with the "-Ooregvar -alr -sr" switches. You will get the 
> assembler output with imaginary registers and notes about register 
> allocations and de-allocations. Inspect if all is correct at this 
> stage. If not then the a code generator of some node is buggy.
>
> Then compile with "-Ooregvar -alr". If the issue is present only at 
> this stage, then the bug is in the register allocator.
>
> Yuriy.
> _______________________________________________
> fpc-devel maillist  -  fpc-devel at lists.freepascal.org
> https://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-devel
>


-- 
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus
-------------- next part --------------
A non-text attachment was scrubbed...
Name: breakp-node-dump.xml
Type: text/xml
Size: 20088 bytes
Desc: not available
URL: <http://lists.freepascal.org/pipermail/fpc-devel/attachments/20201126/947e5f13/attachment-0001.xml>