[Bug testsuite/40130] using RUNTESTFLAGS="--target_board 'foo{-mxyz,-mjkl,}'" screws up ieee.exp (and possibly others?)

[Bug testsuite/40130] using RUNTESTFLAGS="--target_board 'foo{-mxyz,-mjkl,}'" screws up ieee.exp (and possibly others?)

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=40130

--- Comment #10 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Uroš Bizjak from comment #9)
> BTW: Since you are looking in this area, can you add
> __attribute__((noinline)) to the tests? Some of them are too easy to
> optimize for recent optimizing passes ;)

-fno-inline was added by r0-95831-g9e2ceea07aaab0 .









I tried to reproduce the original reported bug on x86_64 with:
make -k check-gcc RUNTESTFLAGS="--target_board 'unix{,-m32}{-march=k8,}'
ieee.exp=20010114-2.c"

But it is looking like it is working:
Executing on host: /bajas/pinskia/src/upstream-gcc-match/gcc/objdir/gcc/xgcc
-B/bajas/pinskia/src/upstream-gcc-match/gcc/objdir/gcc/ 
/home/apinski/src/upstream-gcc-match/gcc/gcc/testsuite/gcc.c-torture/execute/ieee/20010114-2.c
 -m32 -march=k8   -fdiagnostics-plain-output  -w  -O0  -fno-inline
-ffloat-store  -lm  -o
/bajas/pinskia/src/upstream-gcc-match/gcc/objdir/gcc/testsuite/gcc/20010114-2.x0
   (timeout = 300)


...
Executing on host: /bajas/pinskia/src/upstream-gcc-match/gcc/objdir/gcc/xgcc
-B/bajas/pinskia/src/upstream-gcc-match/gcc/objdir/gcc/ 
/home/apinski/src/upstream-gcc-match/gcc/gcc/testsuite/gcc.c-torture/execute/ieee/20010114-2.c
 -m32   -fdiagnostics-plain-output  -w  -O0  -fno-inline -ffloat-store  -lm  -o
/bajas/pinskia/src/upstream-gcc-match/gcc/objdir/gcc/testsuite/gcc/20010114-2.x0
   (timeout = 300)

[Bug testsuite/40130] using RUNTESTFLAGS="--target_board 'foo{-mxyz,-mjkl,}'" screws up ieee.exp (and possibly others?)

[ubizjak at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=40130

Uroš Bizjak <ubizjak at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|UNCONFIRMED                 |RESOLVED
   Target Milestone|---                         |4.5.0

--- Comment #11 from Uroš Bizjak <ubizjak at gmail dot com> ---
Fixed long time ago.

[Bug testsuite/40130] using RUNTESTFLAGS="--target_board 'foo{-mxyz,-mjkl,}'" screws up ieee.exp (and possibly others?)

[ubizjak at gmail dot com]

------- Comment #9 from ubizjak at gmail dot com  2009-05-18 16:37 -------
(In reply to comment #8)

> No, it's just i686-pc-linux-gnu.  Without --target_board "etc." it
> works; with I got this spurious failure.
> 
> But it's not a wrong code, so I think we can close as invalid.

BTW: Since you are looking in this area, can you add __attribute__((noinline))
to the tests? Some of them are too easy to optimize for recent optimizing
passes ;)


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40130

[Bug testsuite/40130] using RUNTESTFLAGS="--target_board 'foo{-mxyz,-mjkl,}'" screws up ieee.exp (and possibly others?)

[bonzini at gnu dot org]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 677 bytes --]



------- Comment #8 from bonzini at gnu dot org  2009-05-14 22:01 -------
Subject: Re:  using RUNTESTFLAGS="--target_board 
        'foo{-mxyz,-mjkl,}'" screws up ieee.exp (and possibly others?)

> Paolo, is it possible that you're using a target triplet that doesn't match the
> checks in ieee.exp for using -ffloat-store?  Those are
>
>    [istarget "i\[34567\]86-*-*"]
>    { [istarget "x86_64-*-*"] && [check_effective_target_ilp32] }

No, it's just i686-pc-linux-gnu.  Without --target_board "etc." it
works; with I got this spurious failure.

But it's not a wrong code, so I think we can close as invalid.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40130

[Bug testsuite/40130] using RUNTESTFLAGS="--target_board 'foo{-mxyz,-mjkl,}'" screws up ieee.exp (and possibly others?)

[janis at gcc dot gnu dot org]

------- Comment #7 from janis at gcc dot gnu dot org  2009-05-14 21:23 -------
I can't reproduce this on my x86 laptop.  When I use the "make check" line
shown in comment #6, each compile for the ieee.exp tests uses -ffloat-store.

Paolo, is it possible that you're using a target triplet that doesn't match the
checks in ieee.exp for using -ffloat-store?  Those are

    [istarget "i\[34567\]86-*-*"]
    { [istarget "x86_64-*-*"] && [check_effective_target_ilp32] }


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40130

[Bug testsuite/40130] using RUNTESTFLAGS="--target_board 'foo{-mxyz,-mjkl,}" screws up ieee.exp (and possibly others?)

[bonzini at gnu dot org]

------- Comment #6 from bonzini at gnu dot org  2009-05-14 14:54 -------
Yes, indeed ieee.exp adds -ffloat-store so I was wrong yesterday in my first
report.  However I can reproduce the problem with

  make -k check RUNTESTFLAGS="--target_board
'unix{-march=i486,-march=pentium2,}'"

where testsuite/gcc/gcc.log does not show -ffloat-store anymore.  So it's
probably a bug either in the harness or in dejagnu itself.

Changing component and CCing Janis.


-- 

bonzini at gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |janis at gcc dot gnu dot org
          Component|target                      |testsuite
           Keywords|wrong-code                  |
            Summary|rint from gcc.c-            |using RUNTESTFLAGS="--
                   |torture/execute miscompiled |target_board 'foo{-mxyz,-
                   |with -march=i486            |mjkl,}" screws up ieee.exp
                   |                            |(and possibly others?)


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40130