[Bug target/18331] New: [4.0 regression] mmix-knuth-mmixware testsuite failure: execute/ieee/fp-cmp-8.c

[Bug target/18331] New: [4.0 regression] mmix-knuth-mmixware testsuite failure: execute/ieee/fp-cmp-8.c

[hp at gcc dot gnu dot org]

With LAST_UPDATED: "Sat Nov  6 19:25:03 GMT 2004" I get:
FAIL: gcc.c-torture/execute/ieee/fp-cmp-8.c compilation,  -O1
FAIL: gcc.c-torture/execute/ieee/fp-cmp-8.c compilation,  -O2
FAIL: gcc.c-torture/execute/ieee/fp-cmp-8.c compilation,  -O3
-fomit-frame-pointer
FAIL: gcc.c-torture/execute/ieee/fp-cmp-8.c compilation,  -O3
-fomit-frame-pointer -funroll-loops
FAIL: gcc.c-torture/execute/ieee/fp-cmp-8.c compilation,  -O3
-fomit-frame-pointer -funroll-all-loops -finline-functions
FAIL: gcc.c-torture/execute/ieee/fp-cmp-8.c compilation,  -O3 -g
FAIL: gcc.c-torture/execute/ieee/fp-cmp-8.c compilation,  -Os
FAIL: gcc.c-torture/execute/ieee/fp-cmp-8l.c compilation,  -O1
FAIL: gcc.c-torture/execute/ieee/fp-cmp-8l.c compilation,  -O2
FAIL: gcc.c-torture/execute/ieee/fp-cmp-8l.c compilation,  -O3
-fomit-frame-pointer
FAIL: gcc.c-torture/execute/ieee/fp-cmp-8l.c compilation,  -O3 -g
FAIL: gcc.c-torture/execute/ieee/fp-cmp-8l.c compilation,  -Os
(the 8 and 8l are the same: "long" == "long double")

With the message in the .log being:
/gcc/testsuite/gcc.c-torture/execute/ieee/fp-cmp-8.c: In function
'test_islessequal':
/gcc/testsuite/gcc.c-torture/execute/ieee/fp-cmp-8.c:35: error: MMIX Internal:
Trying to output invalidly reversed condition:
(gt (reg:CC_FP 0 $0 [278])
    (const_int 0 [0x0]))
/gcc/testsuite/gcc.c-torture/execute/ieee/fp-cmp-8.c:35: internal compiler
error: in mmix_output_condition, at config/mmix/mmix.c:2768

Last known to work on: "Fri Jul 9 02:34:14 GMT 2004".
Known to fail on: "Sat Sep 11 14:54:38 GMT 2004".

-- 
           Summary: [4.0 regression] mmix-knuth-mmixware testsuite failure:
                    execute/ieee/fp-cmp-8.c
           Product: gcc
           Version: 4.0.0
            Status: UNCONFIRMED
          Keywords: ice-on-valid-code
          Severity: normal
          Priority: P2
         Component: target
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: hp at gcc dot gnu dot org
                CC: gcc-bugs at gcc dot gnu dot org
  GCC host triplet: i686-pc-linux-gnu
GCC target triplet: mmix-knuth-mmixware


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18331

[Bug target/18331] [4.0 regression] mmix-knuth-mmixware testsuite failure: execute/ieee/fp-cmp-8.c

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-11-07 01:43 -------
I think we generate different trees now this which is where the problem comes from.  When expanding 
from trees to RTL the back-end is not expecting things like this.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.0.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18331

[Bug rtl-optimization/18331] [4.0 regression] mmix-knuth-mmixware testsuite failure: execute/ieee/fp-cmp-8.c

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-11-07 02:13 -------
Confirmed, this is a bug in the rtl optimizers.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
          Component|target                      |rtl-optimization
     Ever Confirmed|                            |1
   Last reconfirmed|0000-00-00 00:00:00         |2004-11-07 02:13:30
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18331

[Bug rtl-optimization/18331] [4.0 regression] mmix-knuth-mmixware testsuite failure: execute/ieee/fp-cmp-8.c

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-11-07 02:16 -------
Here is the reduced testcase:
double
test_islessequal(double x, double y, double a, double b)
{
  return __builtin_islessequal(x, y)?a:b;
}



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18331

[Bug rtl-optimization/18331] [4.0 regression] mmix-knuth-mmixware testsuite failure: execute/ieee/fp-cmp-8.c

[hp at gcc dot gnu dot org]

------- Additional Comments From hp at gcc dot gnu dot org  2004-11-07 03:28 -------
Another failure,
FAIL: gcc.dg/20040625-1.c (test for excess errors)
also has the corresponding ICE message.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18331

[Bug rtl-optimization/18331] [4.0 regression] mmix-knuth-mmixware testsuite failure: execute/ieee/fp-cmp-8.c

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-11-07 03:30 -------
The line for gcc.dg/20040625-1.c which is most likely causing the problem is:
  x = x < 70 ? x : 70;


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18331

[Bug rtl-optimization/18331] [4.0 regression] mmix-knuth-mmixware testsuite failure: execute/ieee/fp-cmp-8.c gcc.dg/20040625-1.c

[mmitchel at gcc dot gnu dot org]

------- Additional Comments From mmitchel at gcc dot gnu dot org  2004-11-19 22:26 -------
MMIX bugs are low-priority.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P2                          |P3


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18331

[Bug rtl-optimization/18331] [4.0 regression] mmix-knuth-mmixware testsuite failure: execute/ieee/fp-cmp-8.c gcc.dg/20040625-1.c

[hp at gcc dot gnu dot org]

------- Additional Comments From hp at gcc dot gnu dot org  2004-11-25 01:56 -------
I'm on it.  Looks like a target bug.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
   Last reconfirmed|2004-11-07 02:13:30         |2004-11-25 01:56:17
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18331

[Bug SWING/18331] [4.0 regression] mmix-knuth-mmixware testsuite failure: execute/ieee/fp-cmp-8.c gcc.dg/20040625-1.c

[hp at gcc dot gnu dot org]

------- Additional Comments From hp at gcc dot gnu dot org  2004-11-28 09:35 -------
Recat.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|rtl-optimization            |SWING


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18331

[Bug SWING/18331] [4.0 regression] mmix-knuth-mmixware testsuite failure: execute/ieee/fp-cmp-8.c gcc.dg/20040625-1.c

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2004-11-28 09:48 -------
Subject: Bug 18331

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	hp@gcc.gnu.org	2004-11-28 09:48:28

Modified files:
	gcc            : ChangeLog 
	gcc/config/mmix: mmix.md 

Log message:
	PR target/18331
	* config/mmix/mmix.md ("*movdicc_real_reversible"): Rename from
	"*movdicc_real".  Only match reversible comparisons.
	("*movdicc_real_nonreversible"): New pattern, split from
	"*movdicc_real".  Match only nonreversible comparisons.  Exclude
	reversed alternatives.
	("*movdfcc_real_reversible", "*movdfcc_real_nonreversible"):
	Similar.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/ChangeLog.diff?cvsroot=gcc&r1=2.6596&r2=2.6597
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/config/mmix/mmix.md.diff?cvsroot=gcc&r1=1.24&r2=1.25



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18331

[Bug target/18331] [4.0 regression] mmix-knuth-mmixware testsuite failure: execute/ieee/fp-cmp-8.c gcc.dg/20040625-1.c

[hp at gcc dot gnu dot org]

------- Additional Comments From hp at gcc dot gnu dot org  2004-11-28 10:11 -------
No URL; ml lag.  See title "Committed: Fix PR target/18331, MMIX" at
<URL:http://gcc.gnu.org/ml/gcc-patches/2004-11/> whenever it shows up.
This bug exists on the 3.4 branch too and may be uncovered as a
regression there.  When that happens, I'll re-open this PR.

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=18331