[Bug c/14650] New: Testsuite failures with -mabi=64 in compile/20031023-1.c, 20031023-2.c and 20031023-3.c

[Bug c/14650] New: Testsuite failures with -mabi=64 in compile/20031023-1.c, 20031023-2.c and 20031023-3.c

[ghazi at gcc dot gnu dot org]

As noted in these reports:
http://gcc.gnu.org/ml/gcc-testresults/2004-03/msg00822.html
http://gcc.gnu.org/ml/gcc-testresults/2004-03/msg00959.html

I'm seeing testsuite failures with -mabi=64 on these testcases:
FAIL: gcc.c-torture/compile/20031023-1.c  -O0  (test for excess errors)
FAIL: gcc.c-torture/compile/20031023-1.c  -O1  (test for excess errors)
FAIL: gcc.c-torture/compile/20031023-1.c  -O2  (test for excess errors)
FAIL: gcc.c-torture/compile/20031023-1.c  -O3 -fomit-frame-pointer  (test for 
excess errors)
FAIL: gcc.c-torture/compile/20031023-1.c  -O3 -g  (test for excess errors)
FAIL: gcc.c-torture/compile/20031023-1.c  -Os  (test for excess errors)
FAIL: gcc.c-torture/compile/20031023-2.c  -O0  (test for excess errors)
FAIL: gcc.c-torture/compile/20031023-2.c  -O1  (test for excess errors)
FAIL: gcc.c-torture/compile/20031023-2.c  -O2  (test for excess errors)
FAIL: gcc.c-torture/compile/20031023-2.c  -O3 -fomit-frame-pointer  (test for 
excess errors)
FAIL: gcc.c-torture/compile/20031023-2.c  -O3 -g  (test for excess errors)
FAIL: gcc.c-torture/compile/20031023-2.c  -Os  (test for excess errors)
FAIL: gcc.c-torture/compile/20031023-3.c  -O0  (test for excess errors)
FAIL: gcc.c-torture/compile/20031023-3.c  -O1  (test for excess errors)
FAIL: gcc.c-torture/compile/20031023-3.c  -O2  (test for excess errors)
FAIL: gcc.c-torture/compile/20031023-3.c  -O3 -fomit-frame-pointer  (test for 
excess errors)
FAIL: gcc.c-torture/compile/20031023-3.c  -O3 -g  (test for excess errors)
FAIL: gcc.c-torture/compile/20031023-3.c  -Os  (test for excess errors)


In the .log files I see messages like this from the native assembler:

as: Error: /var/tmp//ccfiHqDu.s, line 96: Overflow: 1099511627808
      .frame $fp,1099511627808,$31
as: Error: /var/tmp//ccfiHqDu.s, line 145: Overflow: 1099511627840
      .frame $fp,1099511627840,$31

-- 
           Summary: Testsuite failures with -mabi=64 in compile/20031023-
                    1.c, 20031023-2.c and 20031023-3.c
           Product: gcc
           Version: 3.4.0
            Status: UNCONFIRMED
          Keywords: rejects-valid
          Severity: normal
          Priority: P2
         Component: c
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: ghazi at gcc dot gnu dot org
                CC: echristo at redhat dot com,gcc-bugs at gcc dot gnu dot
                    org,rsandifo at redhat dot com
 GCC build triplet: mips-sgi-irix6.5
  GCC host triplet: mips-sgi-irix6.5
GCC target triplet: mips-sgi-irix6.5


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14650

[Bug c/14650] Testsuite failures with -mabi=64 in compile/20031023-1.c, 20031023-2.c and 20031023-3.c

[ebotcazou at gcc dot gnu dot org]

------- Additional Comments From ebotcazou at gcc dot gnu dot org  2004-03-19 15:04 -------
These testcases excercise the large frame (>2GB) support on 64-bit platforms.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14650

[Bug target/14650] Testsuite failures with -mabi=64 in compile/20031023-1.c, 20031023-2.c and 20031023-3.c

[pinskia at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|c                           |target
           Keywords|rejects-valid               |wrong-code


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14650

[Bug target/14650] Testsuite failures with -mabi=64 in compile/20031023-1.c, 20031023-2.c and 20031023-3.c

[rsandifo at gcc dot gnu dot org]

------- Additional Comments From rsandifo at gcc dot gnu dot org  2004-03-19 16:53 -------
Yeah, this is the native assembler rejecting frame sizes >=32k.
I don't think there's anything gcc can do about it, given that
the frame really is >=32k.

The tests work fine if you use gas.  For that reason, I'd be
reluctant to xfail them for mips-sgi-irix6.5.


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |WAITING


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14650

[Bug target/14650] Testsuite failures with -mabi=64 in compile/20031023-1.c, 20031023-2.c and 20031023-3.c

[rsandifo at redhat dot com]

------- Additional Comments From rsandifo at redhat dot com  2004-03-19 16:56 -------
Subject: Re:  Testsuite failures with -mabi=64 in
 compile/20031023-1.c, 20031023-2.c and 20031023-3.c

"rsandifo at gcc dot gnu dot org" <gcc-bugzilla@gcc.gnu.org> writes:
> ------- Additional Comments From rsandifo at gcc dot gnu dot org  2004-03-19 16:53 -------
> Yeah, this is the native assembler rejecting frame sizes >=32k.
> I don't think there's anything gcc can do about it, given that
> the frame really is >=32k.

Sigh.  s/32k/4G/ of course.  I've got 16-bit values on the brain.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14650

[Bug target/14650] Testsuite failures with -mabi=64 in compile/20031023-1.c, 20031023-2.c and 20031023-3.c

[ghazi at caip dot rutgers dot edu]

------- Additional Comments From ghazi at caip dot rutgers dot edu  2004-03-19 18:14 -------
Subject: Re:  Testsuite failures with -mabi=64 in compile/20031023-1.c, 20031023-2.c and 20031023-3.c

Does GCC define some macro so we can test whether we're using GAS?
There are various tests (beyond this one) that fail with native tools
and I'd like to be able to wrap that code with a GAS check to avoid
spurious testsuite failures.

If it doesn't exist, would there be any objection to adding one?

		--Kaveh
--
Kaveh R. Ghazi			ghazi@caip.rutgers.edu


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14650

[Bug target/14650] Testsuite failures with -mabi=64 in compile/20031023-1.c, 20031023-2.c and 20031023-3.c

[echristo at redhat dot com]

------- Additional Comments From echristo at redhat dot com  2004-03-19 18:20 -------
Good for me.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14650

[Bug target/14650] Testsuite failures with -mabi=64 in compile/20031023-1.c, 20031023-2.c and 20031023-3.c

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-04-12 13:00 -------
Changing back to new as this is definitely a bug.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |NEW
     Ever Confirmed|                            |1
   Last reconfirmed|0000-00-00 00:00:00         |2004-04-12 13:00:04
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14650

[Bug target/14650] Testsuite failures with -mabi=64 in compile/20031023-1.c, 20031023-2.c and 20031023-3.c

[rsandifo at redhat dot com]

------- Additional Comments From rsandifo at redhat dot com  2004-04-12 15:37 -------
Subject: Re:  Testsuite failures with -mabi=64 in
 compile/20031023-1.c, 20031023-2.c and 20031023-3.c

"pinskia at gcc dot gnu dot org" <gcc-bugzilla@gcc.gnu.org> writes:
> Changing back to new as this is definitely a bug.

Please explain.  As per the previous comments, this is just a limitation
of the MIPSpro assembler, it isn't a GCC bug.

Richard


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14650

[Bug target/14650] Testsuite failures with -mabi=64 in compile/20031023-1.c, 20031023-2.c and 20031023-3.c

[pinskia at physics dot uc dot edu]

------- Additional Comments From pinskia at physics dot uc dot edu  2004-04-12 15:40 -------
Subject: Re:  Testsuite failures with -mabi=64 in compile/20031023-1.c, 20031023-2.c and 20031023-3.c


On Apr 12, 2004, at 11:38, Richard Sandiford wrote:

> "pinskia at gcc dot gnu dot org" <gcc-bugzilla@gcc.gnu.org> writes:
>> Changing back to new as this is definitely a bug.
>
> Please explain.  As per the previous comments, this is just a 
> limitation
> of the MIPSpro assembler, it isn't a GCC bug.

Yes it is a limitation of the assembler but it was in WAITING which was
just not the right state for this bug, maybe suspended is more the right
state here.

Thanks,
Andrew Pinski



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14650

[Bug target/14650] Testsuite failures with -mabi=64 in compile/20031023-1.c, 20031023-2.c and 20031023-3.c

[ghazi at caip dot rutgers dot edu]

------- Additional Comments From ghazi at caip dot rutgers dot edu  2004-04-12 15:56 -------
Subject: Re:  Testsuite failures with -mabi=64 in compile/20031023-1.c, 20031023-2.c and 20031023-3.c

Although it's not a GCC bug, still I hate spurious testsuite noise.  I
have a patch to solve this waiting for review.
http://gcc.gnu.org/ml/gcc-patches/2004-03/msg01966.html

On reflection my patch may be overkill, and we no longer need it
anyway for attr-ctor1.C, etc.  Perhaps a better way to solve the
problem would be to add a check_*_available procedure in
testsuite/lib/target-supports.exp as was recently done for named
sections.

Richard - Can you suggest a minimal C testcase that passes with GAS
but fails with the MIPSpro n64 assembler to check for >4GB frames?
I'll try adding that to the testsuite harness.

If that's overkill, perhaps I could just bypass these tests for n64
regardless of whether we're using GAS.

Thoughts?

		--Kaveh


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14650

[Bug target/14650] Testsuite failures with -mabi=64 in compile/20031023-1.c, 20031023-2.c and 20031023-3.c

[rsandifo at redhat dot com]

------- Additional Comments From rsandifo at redhat dot com  2004-04-14 19:35 -------
Subject: Re:  Testsuite failures with -mabi=64 in
 compile/20031023-1.c, 20031023-2.c and 20031023-3.c

"ghazi at caip dot rutgers dot edu" <gcc-bugzilla@gcc.gnu.org> writes:
> Richard - Can you suggest a minimal C testcase that passes with GAS
> but fails with the MIPSpro n64 assembler to check for >4GB frames?
> I'll try adding that to the testsuite harness.

Would an assembler testcase be OK?  A minimal-ish example is:

	.ent	foo
foo:
	.frame	$fp,5000000000,$31
	.end	foo

assembled with "as -64".  For MIPSpro, you get:

as: Error: foo.s, line 3: Overflow: 5000000000
      .frame $fp,5000000000,$31

but it works OK with binutils.

> If that's overkill, perhaps I could just bypass these tests for n64
> regardless of whether we're using GAS.

Well, the tests only pass for other ABIs because they're a no-op when
sizeof (long) < 8.  If we're going to disable them for n64, we might as
well go the whole hog and disable them for mips-sgi-irix*, or perhaps
for all MIPS targets.

And maybe that's a reasonable thing to do.  The tests did show up a bug
in the MIPS backend ("long" being used instead of HOST_WIDE_INT), but it
seems very unlikely that any real MIPS code would want to use stack
frames >4Gb.

Richard


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14650

[Bug target/14650] Testsuite failures with -mabi=64 in compile/20031023-1.c, 20031023-2.c and 20031023-3.c

[rsandifo at gcc dot gnu dot org]

------- Additional Comments From rsandifo at gcc dot gnu dot org  2004-08-20 10:25 -------
"Fixed" by requiring GNU as.  I doubt that this is important
enough to warrant a 3.4-specific patch to the testsuite.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED
   Target Milestone|---                         |3.5.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=14650