[Bug target/39323] New: MAX_OFILE_ALIGNMENT in elfos.h is too big

[Bug target/39323] New: MAX_OFILE_ALIGNMENT in elfos.h is too big

[hjl dot tools at gmail dot com]

struct tree_decl_common has

unsigned int align : 24;

And the maximum alignment we support is 1 << 23 bits (1 << 20 bytes). However,
config/elfos.h has

/* The biggest alignment supported by ELF in bits. 32-bit ELF 
   supports section alignment up to (0x80000000 * 8), while 
   64-bit ELF supports (0x8000000000000000 * 8). If this macro 
   is not defined, the default is the largest alignment supported 
   by 32-bit ELF and representable on a 32-bit host. Use this
   macro to limit the alignment which can be specified using
   the `__attribute__ ((aligned (N)))' construct.  */
#ifndef MAX_OFILE_ALIGNMENT
#define MAX_OFILE_ALIGNMENT (((unsigned int) 1 << 28) * 8)
#endif

If we ask for alignment greater than 1 << 20 bytes, we won't
get any warnings:

[hjl@gnu-17 gcc]$  cat x.c    
int foo __attribute__ ((aligned(1 << 21))) =  20;
int bar __attribute__ ((aligned(1 << 20))) =  20;
[hjl@gnu-17 gcc]$ ./xgcc -B./ -S x.i 
[hjl@gnu-17 gcc]$ cat x.s
        .file   "x.i"
.globl foo
        .data
        .type   foo, @object
        .size   foo, 4
foo:
        .long   20
.globl bar
        .align 1048576
        .type   bar, @object
        .size   bar, 4
bar:
        .long   20


-- 
           Summary: MAX_OFILE_ALIGNMENT in elfos.h is too big
           Product: gcc
           Version: 4.4.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: hjl dot tools at gmail dot com


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39323

[Bug target/39323] MAX_OFILE_ALIGNMENT in elfos.h is too big

[hjl dot tools at gmail dot com]

-- 

hjl dot tools at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.4.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39323

[Bug c/39323] MAX_OFILE_ALIGNMENT in elfos.h is too big

[hjl dot tools at gmail dot com]

------- Comment #1 from hjl dot tools at gmail dot com  2009-02-28 00:12 -------
Shouldn't handle_aligned_attribute check alignment overflow?


-- 

hjl dot tools at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|target                      |c


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39323

[Bug c/39323] MAX_OFILE_ALIGNMENT in elfos.h is too big

[hjl dot tools at gmail dot com]

------- Comment #2 from hjl dot tools at gmail dot com  2009-02-28 20:01 -------
A patch is posted at

http://gcc.gnu.org/ml/gcc-patches/2009-02/msg01274.html


-- 

hjl dot tools at gmail dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                URL|                            |http://gcc.gnu.org/ml/gcc-
                   |                            |patches/2009-
                   |                            |02/msg01274.html
   Target Milestone|4.4.0                       |4.5.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39323

[Bug c/39323] MAX_OFILE_ALIGNMENT in elfos.h is too big

[hjl at gcc dot gnu dot org]

------- Comment #3 from hjl at gcc dot gnu dot org  2009-03-07 16:32 -------
Subject: Bug 39323

Author: hjl
Date: Sat Mar  7 16:32:34 2009
New Revision: 144701

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=144701
Log:
gcc/

2009-03-07  H.J. Lu  <hongjiu.lu@intel.com>

        PR c/39323
        * c-common.c (handle_aligned_attribute): Properly check alignment
        overflow.  Use (1U << i) instead of (1 << i).

        * emit-rtl.c (get_mem_align_offset): Use "unsigned int" for
        align.

        * expr.h (get_mem_align_offset): Updated.

        * tree.h (tree_decl_common): Change align to "unsigned int" and
        move it before pointer_alias_set.

gcc/ada/

2009-03-07  H.J. Lu  <hongjiu.lu@intel.com>

        PR c/39323
        * gcc-interface/utils.c (create_field_decl): Use "unsigned int"
        on bit_align.

gcc/testsuite/

2009-03-07  H.J. Lu  <hongjiu.lu@intel.com>

        PR c/39323
        * gcc.dg/pr39323-1.c: New.
        * gcc.dg/pr39323-2.c: Likewise.
        * gcc.dg/pr39323-3.c: Likewise.

Added:
    branches/stack/gcc/ada/ChangeLog.stackalign
    branches/stack/gcc/testsuite/gcc.dg/pr39323-1.c
    branches/stack/gcc/testsuite/gcc.dg/pr39323-2.c
    branches/stack/gcc/testsuite/gcc.dg/pr39323-3.c
Modified:
    branches/stack/gcc/ChangeLog.stackalign
    branches/stack/gcc/ada/gcc-interface/utils.c
    branches/stack/gcc/c-common.c
    branches/stack/gcc/emit-rtl.c
    branches/stack/gcc/expr.h
    branches/stack/gcc/testsuite/ChangeLog.stackalign
    branches/stack/gcc/tree.h


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39323

[Bug c/39323] MAX_OFILE_ALIGNMENT in elfos.h is too big

[hjl at gcc dot gnu dot org]

------- Comment #4 from hjl at gcc dot gnu dot org  2009-03-27 22:33 -------
Subject: Bug 39323

Author: hjl
Date: Fri Mar 27 22:33:16 2009
New Revision: 145136

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=145136
Log:
gcc/

2009-03-27  H.J. Lu  <hongjiu.lu@intel.com>

        PR c/39323
        * c-common.c (handle_aligned_attribute): Properly check alignment
        overflow.  Use (1U << i) instead of (1 << i).

        * emit-rtl.c (get_mem_align_offset): Use "unsigned int" for
        align.

        * expr.h (get_mem_align_offset): Updated.

        * tree.h (tree_decl_common): Change align to "unsigned int" and
        move it before pointer_alias_set.

gcc/ada/

2009-03-27  H.J. Lu  <hongjiu.lu@intel.com>

        PR c/39323
        * gcc-interface/utils.c (create_field_decl): Use "unsigned int"
        on bit_align.

gcc/testsuite/

2009-03-27  H.J. Lu  <hongjiu.lu@intel.com>

        PR c/39323
        * gcc.dg/pr39323-1.c: New.
        * gcc.dg/pr39323-2.c: Likewise.
        * gcc.dg/pr39323-3.c: Likewise.

Added:
    trunk/gcc/testsuite/gcc.dg/pr39323-1.c
    trunk/gcc/testsuite/gcc.dg/pr39323-2.c
    trunk/gcc/testsuite/gcc.dg/pr39323-3.c
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/ada/ChangeLog
    trunk/gcc/ada/gcc-interface/utils.c
    trunk/gcc/c-common.c
    trunk/gcc/emit-rtl.c
    trunk/gcc/expr.h
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/tree.h


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39323

[Bug target/39323] MAX_OFILE_ALIGNMENT in elfos.h is too big

[janis at gcc dot gnu dot org]

------- Comment #5 from janis at gcc dot gnu dot org  2009-04-02 16:41 -------
The alignment checks in tests pr39323-2.c and pr39323-3.c fail on
powerpc*-linux, where the value in .align is log2 of the alignment, not the
alignment itself.    For an alignment of (1<<28) we get ".align 28", for an
alignment of 32 we get ".align 5".  That might be true of some other ELF
targets as well.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39323

[Bug target/39323] MAX_OFILE_ALIGNMENT in elfos.h is too big

[pinskia at gcc dot gnu dot org]

------- Comment #6 from pinskia at gcc dot gnu dot org  2009-04-16 18:34 -------
Stop setting the target milestone unless it is a regression.


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|4.5.0                       |---


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39323

[Bug target/39323] MAX_OFILE_ALIGNMENT in elfos.h is too big

[uros at gcc dot gnu dot org]

------- Comment #7 from uros at gcc dot gnu dot org  2009-04-28 09:48 -------
Subject: Bug 39323

Author: uros
Date: Tue Apr 28 09:48:04 2009
New Revision: 146881

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=146881
Log:
        PR c/39323
        * config/alpha/elf.h (MAX_OFILE_ALIGNMENT): Sync with elfos.h

testsuite/ChangeLog:

        PR c/39323
        * gcc.dg/pr39323-2.c: Also scan for alignment in log2 format.
        * gcc.dg/pr39323-3.c: Ditto.


Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/config/alpha/elf.h
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/testsuite/gcc.dg/pr39323-2.c
    trunk/gcc/testsuite/gcc.dg/pr39323-3.c


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39323

[Bug target/39323] MAX_OFILE_ALIGNMENT in elfos.h is too big

[ubizjak at gmail dot com]

------- Comment #8 from ubizjak at gmail dot com  2009-04-28 09:49 -------
(In reply to comment #5)
> The alignment checks in tests pr39323-2.c and pr39323-3.c fail on
> powerpc*-linux, where the value in .align is log2 of the alignment, not the
> alignment itself.    For an alignment of (1<<28) we get ".align 28", for an
> alignment of 32 we get ".align 5".  That might be true of some other ELF
> targets as well.

This is fixed in rev 146881.


-- 

ubizjak at gmail dot com changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=39323