[Bug rtl-optimization/43305] New: ICE: in emit_unop_insn, at optabs.c:3838 with -Os -ffast-math and ilogbl()

[Bug rtl-optimization/43305] New: ICE: in emit_unop_insn, at optabs.c:3838 with -Os -ffast-math and ilogbl()

[zsojka at seznam dot cz]

Command line:
gcc -Os -ffast-math testcase.c
or
gcc -Os -fno-math-errno -funsafe-math-optimizations testcase.c
crashes with -m32 as well

-------- testcase.c --------
extern int ilogbl(long double);
int foo(long double x)
{
  return ilogbl(x);
}
----------------------------
(reduced from testsuite/gcc.dg/builtins-38.c)

Tested revisions:
trunk r157161 - crash
trunk r153685 - crash
4.4 r157120 - crash (checking)
4.4 r155365 - crash (release checking)

Output:
$ /mnt/svn/gcc-trunk/binary-157161/bin/gcc -Os -ffast-math testcase.c
testcase.c: In function 'foo':
testcase.c:4:3: internal compiler error: in emit_unop_insn, at optabs.c:3838


-- 
           Summary: ICE: in emit_unop_insn, at optabs.c:3838 with -Os -
                    ffast-math and ilogbl()
           Product: gcc
           Version: 4.5.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: rtl-optimization
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: zsojka at seznam dot cz
  GCC host triplet: x86_64-pc-linux-gnu
GCC target triplet: x86(_64)-pc-linux-gnu


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43305

[Bug target/43305] ICE: in emit_unop_insn, at optabs.c:3838 with -Os -ffast-math and ilogbl()

[jakub at gcc dot gnu dot org]

------- Comment #1 from jakub at gcc dot gnu dot org  2010-03-09 15:01 -------
emit_unop_insn doesn't allow the expansion to fail, but ilogbxf2 has FAIL if
-Os.


-- 

jakub at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43305

[Bug target/43305] [4.4/4.5 Regression] ICE: in emit_unop_insn, at optabs.c:3838 with -Os -ffast-math and ilogbl()

[zsojka at seznam dot cz]

------- Comment #2 from zsojka at seznam dot cz  2010-03-09 15:19 -------
I don't have 4.3 (or older) builds with enabled checking available. But without
checking, the code below prints "0.000000" for 4.4/4.5 and "6.000000" for 4.3
and older (4.2.4, 4.1.2, 3.3.6, 3.4.6).

Command line:
gcc -Os -ffast-math main.c -lm && ./a.out
------- main.c ------
extern int ilogbl(long double);
extern int printf(const char *format, ...);

__attribute__((noinline, noclone))
int foo(long double x) { return ilogbl(x); }
int main() { printf("%f\n", (float)foo(100)); }
---------------------

In 4.4/4.5 without checking, foo() is reduced to:
        xorl    %eax, %eax
        ret
while in 4.3/4.2/4.1, it is:
        jmp     ilogbl


-- 

zsojka at seznam dot cz changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |ice-checking, ice-on-valid-
                   |                            |code, wrong-code
            Summary|ICE: in emit_unop_insn, at  |[4.4/4.5 Regression] ICE: in
                   |optabs.c:3838 with -Os -    |emit_unop_insn, at
                   |ffast-math and ilogbl()     |optabs.c:3838 with -Os -
                   |                            |ffast-math and ilogbl()


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43305

[Bug target/43305] [4.4/4.5 Regression] ICE: in emit_unop_insn, at optabs.c:3838 with -Os -ffast-math and ilogbl()

[zsojka at seznam dot cz]

------- Comment #3 from zsojka at seznam dot cz  2010-03-09 15:23 -------
Created an attachment (id=20058)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=20058&action=view)
more readable testcase for comment #2


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43305

[Bug target/43305] [4.4/4.5 Regression] ICE: in emit_unop_insn, at optabs.c:3838 with -Os -ffast-math and ilogbl()

[rguenth at gcc dot gnu dot org]

------- Comment #4 from rguenth at gcc dot gnu dot org  2010-03-09 15:49 -------
Confirmed.  Introduced by Honza with the optimize_insn_for_size_p () changes.


-- 

rguenth at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |hubicka at gcc dot gnu dot
                   |                            |org
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
   Last reconfirmed|0000-00-00 00:00:00         |2010-03-09 15:49:38
               date|                            |
   Target Milestone|---                         |4.4.4


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43305

[Bug target/43305] [4.4/4.5 Regression] ICE: in emit_unop_insn, at optabs.c:3838 with -Os -ffast-math and ilogbl()

[hjl dot tools at gmail dot com]

------- Comment #5 from hjl dot tools at gmail dot com  2010-03-09 18:59 -------
(In reply to comment #4)
> Confirmed.  Introduced by Honza with the optimize_insn_for_size_p () changes.
> 

That is revision 138835:

http://gcc.gnu.org/ml/gcc-cvs/2008-08/msg00394.html


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43305

[Bug target/43305] [4.4/4.5 Regression] ICE: in emit_unop_insn, at optabs.c:3838 with -Os -ffast-math and ilogbl()

[rguenth at gcc dot gnu dot org]

-- 

rguenth at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43305

[Bug target/43305] [4.4/4.5 Regression] ICE: in emit_unop_insn, at optabs.c:3838 with -Os -ffast-math and ilogbl()

[matz at gcc dot gnu dot org]

------- Comment #6 from matz at gcc dot gnu dot org  2010-03-18 16:47 -------
Mine.


-- 

matz at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |matz at gcc dot gnu dot org
                   |dot org                     |
             Status|NEW                         |ASSIGNED
   Last reconfirmed|2010-03-09 15:49:38         |2010-03-18 16:47:09
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43305

[Bug target/43305] [4.4/4.5 Regression] ICE: in emit_unop_insn, at optabs.c:3838 with -Os -ffast-math and ilogbl()

[matz at gcc dot gnu dot org]

------- Comment #7 from matz at gcc dot gnu dot org  2010-03-19 12:37 -------
Subject: Bug 43305

Author: matz
Date: Fri Mar 19 12:37:28 2010
New Revision: 157567

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=157567
Log:
        PR 43305
        * builtins.c (expand_builtin_interclass_mathfn,
        expand_builtin_signbit): Use maybe_emit_unop_insn, emit libcalls
        if that fails.

testsuite/
        * gcc.dg/pr43305.c: New testcase.

Added:
    trunk/gcc/testsuite/gcc.dg/pr43305.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/builtins.c
    trunk/gcc/testsuite/ChangeLog


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43305

[Bug target/43305] [4.4/4.5 Regression] ICE: in emit_unop_insn, at optabs.c:3838 with -Os -ffast-math and ilogbl()

[matz at gcc dot gnu dot org]

------- Comment #8 from matz at gcc dot gnu dot org  2010-03-19 12:38 -------
Fixed.


-- 

matz at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43305

[Bug target/43305] [4.4 Regression] ICE: in emit_unop_insn, at optabs.c:3838 with -Os -ffast-math and ilogbl()

[jakub at gcc dot gnu dot org]

------- Comment #9 from jakub at gcc dot gnu dot org  2010-03-19 15:46 -------
Well, just fixed on the trunk.  This fix looks simple enough that it should be
backportable.


-- 

jakub at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
      Known to fail|                            |4.4.3
      Known to work|                            |4.5.0 4.3.0
         Resolution|FIXED                       |
            Summary|[4.4/4.5 Regression] ICE: in|[4.4 Regression] ICE: in
                   |emit_unop_insn, at          |emit_unop_insn, at
                   |optabs.c:3838 with -Os -    |optabs.c:3838 with -Os -
                   |ffast-math and ilogbl()     |ffast-math and ilogbl()


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43305

[Bug target/43305] [4.4 Regression] ICE: in emit_unop_insn, at optabs.c:3838 with -Os -ffast-math and ilogbl()

[zsojka at seznam dot cz]

------- Comment #10 from zsojka at seznam dot cz  2010-03-19 16:08 -------
How does the test work, would it fail on non-patched trunk?

- extern int printf(const char *format, ...); 

- printf("%d\n", foo(100));
+ if (foo(100) != 6)
+   __builtin_abort ();

I *think* this would do the job, but I really don't know (and I can't test it
now).


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43305

[Bug target/43305] [4.4 Regression] ICE: in emit_unop_insn, at optabs.c:3838 with -Os -ffast-math and ilogbl()

[matz at gcc dot gnu dot org]

------- Comment #11 from matz at gcc dot gnu dot org  2010-03-19 16:23 -------
I'm not sure what you're asking.  When the unpatched compiler the testcase
(with the printf or the x!=6->abort) will ICE with a checking compiler,
and produce wrong output ("0" or an abort) with a nonchecking compiler.

With a patched compiler the printf testcase will print "6", and the abort
testcase will not abort.


-- 

matz at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|matz at gcc dot gnu dot org |unassigned at gcc dot gnu
                   |                            |dot org
             Status|REOPENED                    |NEW


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43305

[Bug target/43305] [4.4 Regression] ICE: in emit_unop_insn, at optabs.c:3838 with -Os -ffast-math and ilogbl()

[zsojka at seznam dot cz]

------- Comment #12 from zsojka at seznam dot cz  2010-03-19 16:34 -------
Thank you for the answer. I didn't realise check is usually done only with
compilers built with checking. (and thank you for fixing the issue, of course)


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43305

[Bug target/43305] [4.4 Regression] ICE: in emit_unop_insn, at optabs.c:3838 with -Os -ffast-math and ilogbl()

[jakub at gcc dot gnu dot org]

------- Comment #13 from jakub at gcc dot gnu dot org  2010-03-22 15:00 -------
Subject: Bug 43305

Author: jakub
Date: Mon Mar 22 15:00:20 2010
New Revision: 157632

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=157632
Log:
        Backport from mainline:
        2010-03-19  Michael Matz  <matz@suse.de>

        PR target/43305
        * builtins.c (expand_builtin_interclass_mathfn,
        expand_builtin_signbit): Use maybe_emit_unop_insn, emit libcalls
        if that fails.

        * gcc.dg/pr43305.c: New testcase.

Added:
    branches/gcc-4_4-branch/gcc/testsuite/gcc.dg/pr43305.c
Modified:
    branches/gcc-4_4-branch/gcc/ChangeLog
    branches/gcc-4_4-branch/gcc/builtins.c
    branches/gcc-4_4-branch/gcc/testsuite/ChangeLog


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43305

[Bug target/43305] [4.4 Regression] ICE: in emit_unop_insn, at optabs.c:3838 with -Os -ffast-math and ilogbl()

[jakub at gcc dot gnu dot org]

------- Comment #14 from jakub at gcc dot gnu dot org  2010-03-22 15:29 -------
Fixed.


-- 

jakub at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
      Known to work|4.5.0 4.3.0                 |4.5.0 4.3.0 4.4.4
         Resolution|                            |FIXED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43305