[Issue 4443] Optimizer produces wrong code for || or && with struct arrays

[via Digitalmars-d-bugs]

https://issues.dlang.org/show_bug.cgi?id=4443

Andrei Alexandrescu  changed:

   What|Removed |Added

Version|D1 & D2 |D2

--

[Issue 4443] Optimizer produces wrong code for || or && with struct arrays

[d-bugmail]

http://d.puremagic.com/issues/show_bug.cgi?id=4443



--- Comment #6 from Don  2010-08-06 01:24:53 PDT ---
*** Issue 3761 has been marked as a duplicate of this issue. ***

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
--- You are receiving this mail because: ---

[Issue 4443] Optimizer produces wrong code for || or && with struct arrays

[d-bugmail]

http://d.puremagic.com/issues/show_bug.cgi?id=4443


Walter Bright  changed:

   What|Removed |Added

 Status|NEW |RESOLVED
 CC||bugzi...@digitalmars.com
 Resolution||FIXED


--- Comment #5 from Walter Bright  2010-08-05 
14:22:53 PDT ---
http://www.dsource.org/projects/dmd/changeset/601

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
--- You are receiving this mail because: ---

[Issue 4443] Optimizer produces wrong code for || or && with struct arrays

[d-bugmail]

http://d.puremagic.com/issues/show_bug.cgi?id=4443



--- Comment #4 from Don  2010-08-04 00:07:00 PDT ---
(In reply to comment #3)
> if (!sregs)
> sregs = ALLREGS & ~rretregs;
> c3 = allocreg(&sregs,®,ty);
> }
> + // BUG: We should ensure there is only register in retregs
> reg1 = findreg(retregs);

That should be:

if (!sregs)
sregs = ALLREGS & ~rretregs;
c3 = allocreg(&sregs,®,ty);
}
+assert( (retregs & (retregs-1)) == 0); // Must be only one
register
reg1 = findreg(retregs);

There are probably contrived cases where this bug could occur in C++ code, but
I don't think it would ever occur in practice.

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
--- You are receiving this mail because: ---

[Issue 4443] Optimizer produces wrong code for || or && with struct arrays

[d-bugmail]

http://d.puremagic.com/issues/show_bug.cgi?id=4443


Don  changed:

   What|Removed |Added

   Keywords||patch


--- Comment #3 from Don  2010-08-03 21:56:32 PDT ---
This is a really subtle code generation bug. The buggy bit of code is below.
In this section of code, the expression size is one register. 
In the bit of code at L13, it checks to see if the addition factor is already
in a register.
In this case, it is, BUT it's actually in a double register: the {ptr, length}
pair is a ulong, and here it's been cast to uint to get the length.
So isregvar() returns BOTH registers in regm.
Then, at the end of this section of code, the register to use is selected with
a call to findreg(). But it just returns "the first register" which is correct
only if there is only one register.

In this patch, I just use the LSW returned from isregvar. It would also be
possible to change isregvar() to only return the LSW in the case where a cast
to smaller type has occured, but I don't know what other code expects from
isregvar. 

Not sure if the tysize() check is necessary. Maybe it could just be retregs = 1
<< reg1;

PATCH: cod2.c, cdorth(), line 362. (Does 1Ety)]<= REGSIZE)
+   retregs = 1 << reg1; // Only want the LSW
+else
retregs = regm;
c1 = NULL;
freenode(e11);
}
else
c1 = codelem(e11,&retregs,FALSE);
}
rretregs = ALLREGS & ~retregs;
c2 = scodelem(ebase,&rretregs,retregs,TRUE);
{
regm_t sregs = *pretregs & ~rretregs;
if (!sregs)
sregs = ALLREGS & ~rretregs;
c3 = allocreg(&sregs,®,ty);
}
+ // BUG: We should ensure there is only register in retregs
reg1 = findreg(retregs);


Also noticed that in cod4.c, cdmsw() line 2838, there's an incorrect comment.
retregs &= mMSW;// want LSW only
This should of course be // want MSW only

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
--- You are receiving this mail because: ---

[Issue 4443] Optimizer produces wrong code for || or && with struct arrays

[d-bugmail]

http://d.puremagic.com/issues/show_bug.cgi?id=4443


Don  changed:

   What|Removed |Added

Summary|Optimizer produces wrong|Optimizer produces wrong
   |code for statements after   |code for || or && with
   |loop|struct arrays
   Severity|normal  |critical


--- Comment #2 from Don  2010-07-12 13:08:53 PDT ---
Further reduction shows that it is unrelated to for loops. Seems to require
either || or && in the if statement.
This bug existed in prehistoric times (tested on DMD0.140).
--
struct Struct4443
{
int x;
char[5] unused;
}

void foo4443(Struct4443 *dest, Struct4443[] arr)
{
int junk = arr[$-1].x;
if (dest || arr[$-1].x) {
*dest = arr[$-1];
}
}

void main()
{
Struct4443[1] a;
Struct4443 info;
foo4443(&info, a);
}

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
--- You are receiving this mail because: ---