[Bug target/45250] New: [4.6 Regression] FAIL: tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check_nan.cc

[Bug target/45250] New: [4.6 Regression] FAIL: tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check_nan.cc

[danglin at gcc dot gnu dot org]

Executing on host: /home/dave/gnu/gcc/objdir/./gcc/g++ -shared-libgcc
-B/home/dave/gnu/gcc/objdir/./gcc -nostdinc++
-L/home/dave/gnu/gcc/objdir/hppa-linux/libstdc++-v3/src
-L/home/dave/gnu/gcc/objdir/hppa-linux/libstdc++-v3/src/.libs
-B/home/dave/opt/gnu/gcc/gcc-4.6.0/hppa-linux/bin/
-B/home/dave/opt/gnu/gcc/gcc-4.6.0/hppa-linux/lib/ -isystem
/home/dave/opt/gnu/gcc/gcc-4.6.0/hppa-linux/include -isystem
/home/dave/opt/gnu/gcc/gcc-4.6.0/hppa-linux/sys-include
-B/home/dave/gnu/gcc/objdir/hppa-linux/./libstdc++-v3/src/.libs -g -O2
-D_GLIBCXX_ASSERT -fmessage-length=0 -ffunction-sections -fdata-sections -g -O2
-D_GNU_SOURCE -g -O2 -D_GNU_SOURCE -DLOCALEDIR="." -nostdinc++
-I/home/dave/gnu/gcc/objdir/hppa-linux/libstdc++-v3/include/hppa-linux
-I/home/dave/gnu/gcc/objdir/hppa-linux/libstdc++-v3/include
-I/home/dave/gnu/gcc/gcc/libstdc++-v3/libsupc++
-I/home/dave/gnu/gcc/gcc/libstdc++-v3/include/backward
-I/home/dave/gnu/gcc/gcc/libstdc++-v3/testsuite/util
/home/dave/gnu/gcc/gcc/libstdc++-v3/testsuite/tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check_nan.cc
   -include bits/stdc++.h ./libtestc++.a -Wl,--gc-sections  -lm   -o
./check_nan.exe    (timeout = 600)
spawn /home/dave/gnu/gcc/objdir/./gcc/g++ -shared-libgcc
-B/home/dave/gnu/gcc/objdir/./gcc -nostdinc++
-L/home/dave/gnu/gcc/objdir/hppa-linux/libstdc++-v3/src
-L/home/dave/gnu/gcc/objdir/hppa-linux/libstdc++-v3/src/.libs
-B/home/dave/opt/gnu/gcc/gcc-4.6.0/hppa-linux/bin/
-B/home/dave/opt/gnu/gcc/gcc-4.6.0/hppa-linux/lib/ -isystem
/home/dave/opt/gnu/gcc/gcc-4.6.0/hppa-linux/include -isystem
/home/dave/opt/gnu/gcc/gcc-4.6.0/hppa-linux/sys-include
-B/home/dave/gnu/gcc/objdir/hppa-linux/./libstdc++-v3/src/.libs -g -O2
-D_GLIBCXX_ASSERT -fmessage-length=0 -ffunction-sections -fdata-sections -g -O2
-D_GNU_SOURCE -g -O2 -D_GNU_SOURCE -DLOCALEDIR="." -nostdinc++
-I/home/dave/gnu/gcc/objdir/hppa-linux/libstdc++-v3/include/hppa-linux
-I/home/dave/gnu/gcc/objdir/hppa-linux/libstdc++-v3/include
-I/home/dave/gnu/gcc/gcc/libstdc++-v3/libsupc++
-I/home/dave/gnu/gcc/gcc/libstdc++-v3/include/backward
-I/home/dave/gnu/gcc/gcc/libstdc++-v3/testsuite/util
/home/dave/gnu/gcc/gcc/libstdc++-v3/testsuite/tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check_nan.cc
-include bits/stdc++.h ./libtestc++.a -Wl,--gc-sections -lm -o ./check_nan.exe
In file included from
/home/dave/gnu/gcc/objdir/hppa-linux/libstdc++-v3/include/tr1/cmath:95:0,
                 from
/home/dave/gnu/gcc/gcc/libstdc++-v3/testsuite/tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check_nan.cc:25:
/home/dave/gnu/gcc/objdir/hppa-linux/libstdc++-v3/include/tr1/poly_laguerre.tcc:
In function '_Tp std::tr1::__detail::__poly_laguerre_large_n(unsigned int,
_Tpa, _Tp) [with _Tpa = unsigned int, _Tp = long double]':
/home/dave/gnu/gcc/objdir/hppa-linux/libstdc++-v3/include/tr1/poly_laguerre.tcc:106:5:
internal compiler error: in simplify_subreg, at simplify-rtx.c:5129
Please submit a full bug report,

Similar fails are:
FAIL:
tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check_value
.cc (test for excess errors)
FAIL: tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/compile.cc 
(test for excess errors)
FAIL:
tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/compile_2.c
c (test for excess errors)
FAIL: tr1/5_numerical_facilities/special_functions/18_laguerre/check_nan.cc
(tes
t for excess errors)
FAIL: tr1/5_numerical_facilities/special_functions/18_laguerre/check_value.cc
(t
est for excess errors)
FAIL: tr1/5_numerical_facilities/special_functions/18_laguerre/compile.cc (test 
for excess errors)
FAIL: tr1/5_numerical_facilities/special_functions/18_laguerre/compile_2.cc
(tes
t for excess errors)

According to testsuite results, 162412 was ok and 162417 fails.

Don't know how far backtrace can be trusted.  Line numbers are whacky.

Breakpoint 2, 0x00552e94 in simplify_subreg (outermode=DFmode, op=0x60293440, 
    innermode=DFmode, byte=16) at ../../gcc/gcc/simplify-rtx.c:215
215           && MEM_EXPR (x)
(gdb) p debug_rtx (op)
(const_double:DF -2147483648 [0x80000000] 1.0e+0 [0x0.8p+1])
$1 = void
(gdb) bt
#0  0x00552e94 in simplify_subreg (outermode=DFmode, op=0x60293440, 
    innermode=DFmode, byte=16) at ../../gcc/gcc/simplify-rtx.c:215
#1  0x0055422c in avoid_constant_pool_reference (x=0x41547130)
    at ../../gcc/gcc/simplify-rtx.c:215
#2  0x0051a0e8 in commutative_operand_precedence (op=<value optimized out>)
    at ../../gcc/gcc/rtlanal.c:1617
#3  0x0051a22c in swap_commutative_operands_p (x=<value optimized out>, 
    y=0x41547130) at ../../gcc/gcc/rtlanal.c:1617
#4  0x0054aef8 in simplify_rtx (x=0x41547100)
    at ../../gcc/gcc/simplify-rtx.c:215
#5  0x002ec2b0 in cselib_expand_value_rtx_1 (orig=<value optimized out>, 
    evd=0xbff01f48, max_depth=<value optimized out>)
    at ../../gcc/gcc/cselib.c:2314
#6  0x002ed340 in cselib_expand_value_rtx_cb (orig=<value optimized out>, 
    regs_active=<value optimized out>, max_depth=<value optimized out>, 
    cb=<value optimized out>, data=0xbff01d48) at ../../gcc/gcc/cselib.c:2314
#7  0x0071dc64 in vt_expand_loc_callback (x=0xb20818, regs=0xb0fc50, 
    max_depth=8, data=0xbff01d48) at ../../gcc/gcc/var-tracking.c:4558
#8  0x002ec1a0 in cselib_expand_value_rtx_1 (orig=0xb20818, evd=0xbff01d48, 
    max_depth=11667480) at ../../gcc/gcc/cselib.c:2314
#9  0x002ed340 in cselib_expand_value_rtx_cb (orig=<value optimized out>, 
    regs_active=<value optimized out>, max_depth=<value optimized out>, 
    cb=<value optimized out>, data=0xbff01d48) at ../../gcc/gcc/cselib.c:2314
---Type <return> to continue, or q <return> to quit---
#10 0x00711034 in vt_expand_loc (loc=<value optimized out>, 
    vars=<value optimized out>) at ../../gcc/gcc/var-tracking.c:4558
#11 0x00714608 in emit_note_insn_var_location (varp=0xb15930, 
    data=<value optimized out>) at ../../gcc/gcc/var-tracking.c:4558
#12 0x008d9528 in htab_traverse_noresize (htab=<value optimized out>, 
    callback=@0x9ff9be: 0x714158 <emit_note_insn_var_location>, 
    info=0xbff01b48) at ../../gcc/libiberty/hashtab.c:543
#13 0x0071c3b4 in emit_notes_for_changes (insn=0x415379d8, 
    where=EMIT_NOTE_BEFORE_INSN, vars=<value optimized out>)
    at ../../gcc/gcc/var-tracking.c:4558
#14 0x0071c70c in vt_emit_notes () at ../../gcc/gcc/var-tracking.c:4558
#15 0x0071ffac in variable_tracking_main ()
    at ../../gcc/gcc/var-tracking.c:4558
#16 0x004b9ae4 in execute_one_pass (pass=0x9fc1a0) at ./gt-passes.h:44

Failed assertion:
  gcc_assert (byte < GET_MODE_SIZE (innermode));


-- 
           Summary: [4.6 Regression] FAIL:
                    tr1/5_numerical_facilities/special_functions/01_assoc_la
                    guerre/check_nan.cc
           Product: gcc
           Version: 4.6.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: danglin at gcc dot gnu dot org
 GCC build triplet: ppa-unknown-linux-gnu
  GCC host triplet: hppa-unknown-linux-gnu
GCC target triplet: ppa-unknown-linux-gnu


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45250

[Bug target/45250] [4.6 Regression] FAIL: tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check_nan.cc

[danglin at gcc dot gnu dot org]

------- Comment #1 from danglin at gcc dot gnu dot org  2010-08-19 18:26 -------
Introduced by following change:

2010-07-22  Martin Jambor  <mjambor@suse.cz>

        PR tree-optimization/44891
        * tree-sra.c: Include gimple-pretty-print.h.
        (replace_uses_with_default_def_ssa_name): Renamed to
        get_repl_default_def_ssa_name, return the new SSA name instead of
        replacing the old one.
        (sra_modify_assign): Dump a message when removing a load, if the LHS
        is an SSA_NAME, do not do any propagation, just set the RHS to a
        default definition SSA NAME, type convert if necessary.
        * Makefile.in (tree-sra.o): Add gimple-pretty-print.h to dependencies.


-- 

danglin at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |mjambor at suse dot cz


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45250

[Bug target/45250] [4.6 Regression] FAIL: tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check_nan.cc

[danglin at gcc dot gnu dot org]

------- Comment #2 from danglin at gcc dot gnu dot org  2010-08-19 18:33 -------
(In reply to comment #1)
> Introduced by following change:
> 2010-07-22  Martin Jambor  <mjambor@suse.cz>
>         PR tree-optimization/44891
>         * tree-sra.c: Include gimple-pretty-print.h.
>         (replace_uses_with_default_def_ssa_name): Renamed to
>         get_repl_default_def_ssa_name, return the new SSA name instead of
>         replacing the old one.
>         (sra_modify_assign): Dump a message when removing a load, if the LHS
>         is an SSA_NAME, do not do any propagation, just set the RHS to a
>         default definition SSA NAME, type convert if necessary.
>         * Makefile.in (tree-sra.o): Add gimple-pretty-print.h to dependencies.

Sorry Martin, I must be blind...
http://gcc.gnu.org/ml/gcc-testresults/2010-08/msg01959.html


-- 

danglin at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|mjambor at suse dot cz      |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45250

[Bug target/45250] [4.6 Regression] FAIL: tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check_nan.cc

[danglin at gcc dot gnu dot org]

------- Comment #3 from danglin at gcc dot gnu dot org  2010-08-27 12:50 -------
Introduced in revision 162337:

2010-07-20  Jakub Jelinek  <jakub@redhat.com>

        * var-tracking.c (vt_expand_loc, vt_expand_loc_dummy): Bump maximum
        depth to 8 from 5.


-- 

danglin at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jakub at redhat dot com


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45250

[Bug target/45250] [4.6 Regression] FAIL: tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check_nan.cc

[rguenth at gcc dot gnu dot org]

-- 

rguenth at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.6.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45250

[Bug target/45250] [4.6 Regression] FAIL: tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check_nan.cc

[jakub at gcc dot gnu dot org]

------- Comment #4 from jakub at gcc dot gnu dot org  2010-08-30 16:06 -------
Please provide preprocessed source.  Sounds like my patch uncovered a latent
issue.


-- 

jakub at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45250

[Bug target/45250] [4.6 Regression] FAIL: tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check_nan.cc

[dave at hiauly1 dot hia dot nrc dot ca]

------- Comment #5 from dave at hiauly1 dot hia dot nrc dot ca  2010-08-30 17:23 -------
Subject: Re:  [4.6 Regression] FAIL:
        tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check
        _nan.cc

> Please provide preprocessed source.  Sounds like my patch uncovered a latent
> issue.

Attached.


------- Comment #6 from dave at hiauly1 dot hia dot nrc dot ca  2010-08-30 17:23 -------
Created an attachment (id=21594)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=21594&action=view)


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45250

[Bug target/45250] [4.6 Regression] FAIL: tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check_nan.cc

[jakub at gcc dot gnu dot org]

------- Comment #7 from jakub at gcc dot gnu dot org  2010-08-31 19:48 -------
Created an attachment (id=21614)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=21614&action=view)
gcc46-pr45250.patch

The problem is that the PA backend has quite lame setup, where
FRAME_POINTER_REGNUM is the same as HARD_FRAME_POINTER_REGNUM and
ARG_POINTER_REGNUM, thus the replacements var-tracking is doing in order to
decrease size of loclists and use DW_OP_fbreg where possible, aren't reliable,
as the testcase shows, because the same hard register can be used for arbitrary
other data.  The following patch fixes it by not doing the replacements on PA
and similar targets at all when doing VALUE based var-tracking.  Ideally PA
backend would be fixed up to use different (fixed) hard regno, made up and
always eliminated, for FRAME_POINTER_REGNUM.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45250

[Bug target/45250] [4.6 Regression] FAIL: tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check_nan.cc

[dave at hiauly1 dot hia dot nrc dot ca]

------- Comment #8 from dave at hiauly1 dot hia dot nrc dot ca  2010-09-01 22:11 -------
Subject: Re:  [4.6 Regression] FAIL:
        tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check
        _nan.cc

On Tue, 31 Aug 2010, jakub at gcc dot gnu dot org wrote:

> 
> 
> ------- Comment #7 from jakub at gcc dot gnu dot org  2010-08-31 19:48 -------
> Created an attachment (id=21614)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=21614&action=view)
>  --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=21614&action=view)
> gcc46-pr45250.patch
> 
> The problem is that the PA backend has quite lame setup, where
> FRAME_POINTER_REGNUM is the same as HARD_FRAME_POINTER_REGNUM and
> ARG_POINTER_REGNUM, thus the replacements var-tracking is doing in order to
> decrease size of loclists and use DW_OP_fbreg where possible, aren't reliable,
> as the testcase shows, because the same hard register can be used for arbitrary
> other data.  The following patch fixes it by not doing the replacements on PA
> and similar targets at all when doing VALUE based var-tracking.  Ideally PA
> backend would be fixed up to use different (fixed) hard regno, made up and
> always eliminated, for FRAME_POINTER_REGNUM.

The change fixes the PR on hppa-unknown-linux-gnu.  I'll look at
implementing the above this weekend.

Thanks very much for looking at this problem.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45250

[Bug target/45250] [4.6 Regression] FAIL: tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check_nan.cc

[danglin at gcc dot gnu dot org]

------- Comment #9 from danglin at gcc dot gnu dot org  2010-09-08 23:32 -------
Subject: Bug 45250

Author: danglin
Date: Wed Sep  8 23:32:06 2010
New Revision: 164036

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=164036
Log:
        PR target/45250
        * config/pa/pa.md (nonlocal_goto): Restore hard frame pointer using
        hard_frame_pointer_rtx instead of virtual_stack_vars_rtx.
        (builtin_longjmp): Likewise.
        (allocate_stack): Use hard_frame_pointer_rtx instead of
        frame_pointer_rtx.
        * config/pa/pa-protos.h (pa_initial_elimination_offset): Declare.
        * config/pa/pa.c (pa_internal_arg_pointer): Declare.
        (pa_can_eliminate): Likewise.
        (TARGET_INTERNAL_ARG_POINTER): Define.
        (TARGET_CAN_ELIMINATE): Define.
        (hppa_expand_prologue): Use hard frame pointer instead of soft frame
        pointer.
        (hppa_expand_epilogue, pa_eh_return_handler_rtx): Likewise.
        (pa_internal_arg_pointer, pa_can_eliminate,
        pa_initial_elimination_offset): New.
        * config/pa/pa.h (FRAME_POINTER_REGNUM): Set to new general register.
        (HARD_FRAME_POINTER_REGNUM): Set to register three.
        (INITIAL_FRAME_POINTER_OFFSET): Delete.
        (ELIMINABLE_REGS, INITIAL_ELIMINATION_OFFSET, DWARF_FRAME_REGISTERS):
        Define.
        (DWARF_ALT_FRAME_RETURN_COLUMN, REGNO_OK_FOR_INDEX_P,
        REGNO_OK_FOR_BASE_P): Update to include soft frame pointer.
        * config/pa/pa32-regs.h (FIRST_PSEUDO_REGISTER): Increase by one.
        (FIXED_REGISTERS, CALL_USED_REGISTERS, REG_ALLOC_ORDER,
        REG_CLASS_CONTENTS, REGNO_REG_CLASS, REGISTER_NAMES): Update to include
        new soft frame pointer.
        * config/pa/pa64-regs.h: Likewise.


Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/config/pa/pa-protos.h
    trunk/gcc/config/pa/pa.c
    trunk/gcc/config/pa/pa.h
    trunk/gcc/config/pa/pa.md
    trunk/gcc/config/pa/pa32-regs.h
    trunk/gcc/config/pa/pa64-regs.h


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45250

[Bug target/45250] [4.6 Regression] FAIL: tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check_nan.cc

[rguenth at gcc dot gnu dot org]

------- Comment #10 from rguenth at gcc dot gnu dot org  2010-09-21 13:17 -------
Hm, fixed?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45250

[Bug target/45250] [4.6 Regression] FAIL: tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check_nan.cc

[dave at hiauly1 dot hia dot nrc dot ca]

------- Comment #11 from dave at hiauly1 dot hia dot nrc dot ca  2010-09-21 15:16 -------
Subject: Re:  [4.6 Regression] FAIL:
tr1/5_numerical_facilities/special_functions/01_assoc_laguerre/check_nan.cc

> Hm, fixed?

It's fixed on the target.  I left it open because I wasn't sure whether
the variable tracking issue should be addressed or not.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45250