www.digitalmars.com         C & C++   DMDScript  

digitalmars.D.bugs - [Issue 5809] New: feqrel fails assert in std.math on x86_64 systems

reply d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=5809

           Summary: feqrel fails assert in std.math on x86_64 systems
           Product: D
           Version: D1 & D2
          Platform: x86_64
        OS/Version: Linux
            Status: NEW
          Severity: normal
          Priority: P2
         Component: DMD
        AssignedTo: nobody puremagic.com
        ReportedBy: moritzwarning web.de


--- Comment #0 from Moritz Warning <moritzwarning web.de> 2011-04-03 16:28:16
PDT ---
On x86_64 systems this assert fails because feqrel returns to 0.

assert(feqrel(real.min_normal/8,real.min_normal/17)==3);

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
Apr 03 2011
next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=5809



--- Comment #1 from Moritz Warning <moritzwarning web.de> 2011-04-03 16:28:44
PDT ---
Here is the link into Phobos:
https://github.com/D-Programming-Language/phobos/blob/master/std/math.d#L3634

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
Apr 03 2011
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=5809



--- Comment #2 from Moritz Warning <moritzwarning web.de> 2011-04-03 16:30:07
PDT ---
It works when a dummy function call is inserted before "if (pd[F.EXPPOS_SHORT]
== 0)" inside feqrel.

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
Apr 03 2011
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=5809


Don <clugdbug yahoo.com.au> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |wrong-code
                 CC|                            |clugdbug yahoo.com.au


--- Comment #3 from Don <clugdbug yahoo.com.au> 2011-12-21 13:02:13 PST ---
This sounds like a compiler bug. Does it still fail on 2.057?

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
Dec 21 2011
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=5809


Walter Bright <bugzilla digitalmars.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |bugzilla digitalmars.com


--- Comment #4 from Walter Bright <bugzilla digitalmars.com> 2012-04-23
17:34:36 PDT ---
(In reply to comment #3)

 Does it still fail on 2.057?


Works on 32 bit Linux, fails on 64 bit Linux. Test program:


import std.math;

void main() {
  assert(feqrel(real.min_normal/8,real.min_normal/17)==3);
}

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
Apr 23 2012
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=5809


Don <clugdbug yahoo.com.au> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|feqrel fails assert in      |[64 bit] wrong code for
                   |std.math on x86_64 systems  |*p==0, when widening
                   |                            |conversion occurs
           Severity|normal                      |blocker


--- Comment #5 from Don <clugdbug yahoo.com.au> 2012-05-08 22:08:29 PDT ---
Original title: feqrel fails assert in std.math on x86_64 systems

Reduced test case shows it is a backend bug.
The common subexpression *px gets extended to int when calculating b. Inside
the assert it is treated as *cast(int *)px, so that it picks up x[1] instead of
just x[0]. It seems to only apply to comparisons with 0.
Subtle and disastrous, raising severity to blocker.

void main()
{
    ushort[2] x = void;
    x[0] = 0;
    x[1] = 0x1234;

    ushort *px =  &x[0];

    uint b = px[0];  // this causes the next line to fail

    assert(px[0] == 0);
 }

It generates
   cmp    dword ptr -010h[RBP],0
should be word ptr, which doesn't exist in 64bit.

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
May 08 2012
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=5809


Don <clugdbug yahoo.com.au> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |pull


--- Comment #6 from Don <clugdbug yahoo.com.au> 2012-05-10 00:53:24 PDT ---
https://github.com/D-Programming-Language/dmd/pull/937

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
May 10 2012
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=5809



--- Comment #7 from Don <clugdbug yahoo.com.au> 2012-05-10 07:02:10 PDT ---
My pull request is wrong. The correct fix is in cgcod.c, line 1795.

In moving to 64 bit, commit 3cdb9f520e0bac17f6da00ef9de86c81e23429aa
made this change:
                            c = genc(NULL,0x81 ^ byte,modregrm(2,7,BPRM),
                                         FLcs,i, FLconst,(targ_uns) 0);
-                            if (REGSIZE == 4 && sz == 2)
+                            if (I32 && sz == 2)
                                 c->Iflags |= CFopsize;
                         }

But this is wrong, it should be 
+                        if ((I32 || I64) && sz == 2) 

This code is now in cod3.c, testcse(), line 1795.

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
May 10 2012
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=5809



--- Comment #8 from Don <clugdbug yahoo.com.au> 2012-05-10 07:22:16 PDT ---
Fixed pull request.

https://github.com/D-Programming-Language/dmd/pull/938

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
May 10 2012
prev sibling next sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=5809



--- Comment #9 from github-bugzilla puremagic.com 2012-05-10 10:05:33 PDT ---
Commit pushed to master at https://github.com/D-Programming-Language/dmd

https://github.com/D-Programming-Language/dmd/commit/7c2ea37c476a2d7a289f54cbf1ee0e692d7a69a0
Fix issue 5809: [64 bit] wrong code for *p==0, when widening conversion occurs

Bug was introduced in commit 3cdb9f52, but it never worked on 64bits, so it is
not a regression.

-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
May 10 2012
prev sibling parent d-bugmail puremagic.com writes:
http://d.puremagic.com/issues/show_bug.cgi?id=5809


Walter Bright <bugzilla digitalmars.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED


-- 
Configure issuemail: http://d.puremagic.com/issues/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
May 10 2012