[Bug debug/103808] New: [12 Regression] '-fcompare-debug' failure (length) w/ -O2 -ftrapv

[Bug debug/103808] New: [12 Regression] '-fcompare-debug' failure (length) w/ -O2 -ftrapv

[asolokha at gmx dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103808

            Bug ID: 103808
           Summary: [12 Regression] '-fcompare-debug' failure (length) w/
                    -O2 -ftrapv
           Product: gcc
           Version: 12.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: debug
          Assignee: unassigned at gcc dot gnu.org
          Reporter: asolokha at gmx dot com
  Target Milestone: ---
            Target: x86_64-unknown-linux-gnu

gcc 12.0.0 20211219 snapshot (g:fcbf94a5be9e0c1ecad92da773a6632b86b7f70a) fails
-fcompare-debug check when compiling the following testcase w/ -O2 -ftrapv:

void
foo (__int128 x, int y)
{
  for (;;)
    {
      __int128 a, b;

      x |= !!y;
      a = x + 1;
      b = y ? ++y : ++x;
      y = a < b;
      if (x >> 2)
        y *= 2;

      if (y == b)
        __builtin_unreachable ();
    }
}

% x86_64-unknown-linux-gnu-gcc-12.0.0 -O2 -fcompare-debug -ftrapv -c ddz9adk2.c
x86_64-unknown-linux-gnu-gcc-12.0.0: error: ddz9adk2.c: '-fcompare-debug'
failure (length)

--- ddz9adk2.c.gkd      2021-12-23 00:04:28.297518259 +0700
+++ ddz9adk2.gk.c.gkd   2021-12-23 00:04:28.387512439 +0700
@@ -6,11 +6,11 @@
 (note # 0 0 [bb 2] NOTE_INSN_BASIC_BLOCK)
 (note # 0 0 NOTE_INSN_DELETED)
 (note # 0 0 NOTE_INSN_DELETED)
-(insn/f:TI # 0 0 2 (set (mem:DI (pre_dec:DI (reg/f:DI 7 sp)) [  S8 A8])
+(insn/f:TI # 0 0 2 (set (mem:DI (pre_dec:DI (reg/f:DI 7 sp [ x+8 ])) [  S8
A8])
         (reg:DI 42 r14)) "ddz9adk2.c":3:1# {*pushdi2_rex64}
      (expr_list:REG_DEAD (reg:DI 42 r14)
         (nil)))
-(insn/f:TI # 0 0 2 (set (mem:DI (pre_dec:DI (reg/f:DI 7 sp)) [  S8 A8])
+(insn/f:TI # 0 0 2 (set (mem:DI (pre_dec:DI (reg/f:DI 7 sp [ x+8 ])) [  S8
A8])
         (reg:DI 41 r13)) "ddz9adk2.c":3:1# {*pushdi2_rex64}
      (expr_list:REG_DEAD (reg:DI 41 r13)
         (nil)))
@@ -18,7 +18,7 @@
         (reg:DI 4 si [123])) "ddz9adk2.c":3:1# {*movdi_internal}
      (expr_list:REG_DEAD (reg:DI 4 si [123])
         (nil)))
-(insn/f:TI # 0 0 2 (set (mem:DI (pre_dec:DI (reg/f:DI 7 sp)) [  S8 A8])
+(insn/f:TI # 0 0 2 (set (mem:DI (pre_dec:DI (reg/f:DI 7 sp [ x+8 ])) [  S8
A8])
         (reg:DI 40 r12)) "ddz9adk2.c":3:1# {*pushdi2_rex64}
      (expr_list:REG_DEAD (reg:DI 40 r12)
         (nil)))
@@ -26,11 +26,11 @@
         (reg:DI 5 di [122])) "ddz9adk2.c":3:1# {*movdi_internal}
      (expr_list:REG_DEAD (reg:DI 5 di [122])
         (nil)))
-(insn/f:TI # 0 0 2 (set (mem:DI (pre_dec:DI (reg/f:DI 7 sp)) [  S8 A8])
+(insn/f:TI # 0 0 2 (set (mem:DI (pre_dec:DI (reg/f:DI 7 sp [ x+8 ])) [  S8
A8])
         (reg:DI 6 bp)) "ddz9adk2.c":3:1# {*pushdi2_rex64}
      (expr_list:REG_DEAD (reg:DI 6 bp)
         (nil)))
-(insn/f:TI # 0 0 2 (set (mem:DI (pre_dec:DI (reg/f:DI 7 sp)) [  S8 A8])
+(insn/f:TI # 0 0 2 (set (mem:DI (pre_dec:DI (reg/f:DI 7 sp [ x+8 ])) [  S8
A8])
         (reg:DI 3 bx)) "ddz9adk2.c":3:1# {*pushdi2_rex64}
      (expr_list:REG_DEAD (reg:DI 3 bx)
         (nil)))

[Bug middle-end/103808] [12 Regression] '-fcompare-debug' failure (length) w/ -O2 -ftrapv

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103808

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|debug                       |middle-end
   Target Milestone|---                         |12.0
           Keywords|                            |ice-on-valid-code,
                   |                            |wrong-debug

[Bug target/103808] [12 Regression] '-fcompare-debug' failure (length) w/ -O2 -ftrapv

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103808

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|middle-end                  |target

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
t233.gk.c.327r.mach is where the difference shows up so the bug is inside
ix86_reorg somewhere.

[Bug target/103808] [12 Regression] '-fcompare-debug' failure (length) w/ -O2 -ftrapv

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103808

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
           Keywords|                            |needs-bisection
   Last reconfirmed|                            |2021-12-22
     Ever confirmed|0                           |1

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Confirmed.

[Bug target/103808] [12 Regression] '-fcompare-debug' failure (length) w/ -O2 -ftrapv

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103808

--- Comment #3 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
I think this is a latent bug exposed by my match.pd patch (r12-5392).

In GCC 11.2.0 we had:
  y_18 = iftmp.0_10 > a_14 ? 2 : 0;

While on the trunk we have:
  y_18 = prephitmp_29 * 2;

But doing this:

void
foo (__int128 x, int y)
{
  for (;;)
    {
      __int128 a, b;

      x |= !!y;
      a = x + 1;
      b = y ? ++y : ++x;
      y = a < b;
      asm("":"+r"(y));
      if (x >> 2)
        y *= 2;

      if (y == b)
        __builtin_unreachable ();
    }
}

Which causes the IR to be the same out of 11 and 12, does not produce compare
debug in 11 but does in 12.

[Bug target/103808] [12 Regression] '-fcompare-debug' failure (length) w/ -O2 -ftrapv since r12-5944-ga7acb6dca941db2b

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103808

Martin Liška <marxin at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|needs-bisection             |
            Summary|[12 Regression]             |[12 Regression]
                   |'-fcompare-debug' failure   |'-fcompare-debug' failure
                   |(length) w/ -O2 -ftrapv     |(length) w/ -O2 -ftrapv
                   |                            |since
                   |                            |r12-5944-ga7acb6dca941db2b
                 CC|                            |marxin at gcc dot gnu.org,
                   |                            |vmakarov at gcc dot gnu.org

--- Comment #4 from Martin Liška <marxin at gcc dot gnu.org> ---
Example from #c3 started to fail with r12-5944-ga7acb6dca941db2b.

[Bug target/103808] [12 Regression] '-fcompare-debug' failure (length) w/ -O2 -ftrapv since r12-5944-ga7acb6dca941db2b

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103808

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jakub at gcc dot gnu.org

--- Comment #5 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
It starts during var-tracking and not ix86_reorg.
The problem is we have:
(debug_insn 110 103 104 4 (var_location:TI D#2 (reg:TI 6 bp [orig:111 x ]
[111])) -1
     (nil))
and
(debug_insn 25 24 26 3 (var_location:TI x (concatn/v:TI [
            (reg:DI 6 bp [orig:111 x ] [111])
            (subreg:DI (debug_expr:TI D#2) 8)
        ])) "pr103808.c":8:9 -1
     (nil))
I guess it is originally from a paradoxical subreg or what, but RA can't
because of debug insns decide differently and not allocate pseudo 111 into bp
register.
When var-tracking.c attempts to simplify the (subreg:DI (reg:TI 6 bp [orig:111
x ] [111]) 8)
simplify-rtx.c calls gen_rtx_REG_offset which calls gen_rtx_REG and that
returns stack_pointer_rtx and later the function changes REG_ATTRS on that
shared rtx.
That is just wrong.
So, IMHO either gen_rtx_REG_offset should be changed to call gen_raw_REG
instead of gen_rtx_REG, or alternatively it shouldn't call update_reg_offset if
gen_rtx_REG returned one of the shared rtxes
({{frame,hard_frame,arg,return_address,stack}_pointer,pic_offset_table}_rtx).

[Bug debug/103808] [12 Regression] '-fcompare-debug' failure (length) w/ -O2 -ftrapv since r12-5944-ga7acb6dca941db2b

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103808

--- Comment #6 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Jakub Jelinek <jakub@gcc.gnu.org>:

https://gcc.gnu.org/g:52441f18db7e051c16f2e6e12f415b93cb5450c4

commit r12-6153-g52441f18db7e051c16f2e6e12f415b93cb5450c4
Author: Jakub Jelinek <jakub@redhat.com>
Date:   Thu Dec 30 14:35:06 2021 +0100

    emit-rtl: Fix a -fcompare-debug issue due to var-tracking [PR103808]

    We get a -fcompare-debug FAIL on the following testcase.  The problem is
    that during cprop we get when a TImode pseudo holding x is being
    constructed:
    (debug_insn 111 59 103 7 (var_location:TI D#2 (clobber (const_int 0 [0])))
-1
         (nil))
    (insn 103 111 110 7 (clobber (reg/v:TI 89 [ x ])) "pr103808.c":8:9 -1
         (nil))
    (debug_insn 110 103 104 7 (var_location:TI D#2 (subreg:TI (reg:DI 111 [ x
]) 0)) -1
         (nil))
    (insn 104 110 109 7 (set (subreg:DI (reg/v:TI 89 [ x ]) 0)
            (reg:DI 111 [ x ])) "pr103808.c":8:9 80 {*movdi_internal}
         (expr_list:REG_DEAD (reg:DI 111 [ x ])
            (nil)))
    Now, during RA that paradoxical subreg in a debug insn obviously can't
    affect where pseudo 111 is allocated and RA puts it into the bp register,
    so we have:
    (debug_insn 110 111 109 4 (var_location:TI D#2 (reg:TI 6 bp [orig:111 x ]
[111])) -1
         (nil))
    Now, during var-tracking when we for:
    (debug_insn 25 23 26 3 (var_location:TI x (concatn/v:TI [
                (reg:DI 6 bp [orig:111 x ] [111])
                (subreg:DI (debug_expr:TI D#2) 8)
            ])) "pr103808.c":8:9 -1
         (nil))
    try to simplify the highpart subreg of bp, gen_rtx_REG_offset is called in:
          if (HARD_REGISTER_NUM_P (final_regno))
            {
              rtx x = gen_rtx_REG_offset (op, outermode, final_regno,
                                          subreg_memory_offset (outermode,
                                                                innermode,
byte));
    and that unfortunately sets REG_ATTRS on stack_pointer_rtx, because
    gen_rtx_REG_offset uses gen_rtx_REG which for Pmode STACK_POINTER_REGNUM
    returns stack_pointer_rtx rather than newly created register.
    The clobbering of REG_ATTRS on the shared stack_pointer_rtx then shows up
    in the dumps as (reg/f:DI 7 sp [ x+8 ]) instead of (reg/f:DI 7 sp)
    that shows up without var-tracking.
    Clobbering of REG_ATTRS on the shared *_pointer_rtx looks just wrong.
    So, IMHO either simplify_gen_subreg -> gen_rtx_REG_offset should call
    gen_raw_REG to make sure we get a new non-shared REG we can set REG_ATTRS
    on, or we should make sure that we don't overwrite the REG_ATTRS on the
    shared REGs (but then simplify_gen_subreg shouldn't try to overwrite
    ORIGINAL_REGNO on those either).
    For non-DEBUG_INSNs, I'd hope this never happens, the RA shouldn't allocate
    multi-word regs overlapping with stack pointer, hard frame pointer etc.

    2021-12-30  Jakub Jelinek  <jakub@redhat.com>

            PR debug/103808
            * emit-rtl.c (gen_rtx_REG_offset): Use gen_raw_REG instead of
            gen_rtx_REG.

            * gcc.dg/pr103808.c: New test.

[Bug debug/103808] [12 Regression] '-fcompare-debug' failure (length) w/ -O2 -ftrapv since r12-5944-ga7acb6dca941db2b

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=103808

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
           Assignee|unassigned at gcc dot gnu.org      |jakub at gcc dot gnu.org
         Resolution|---                         |FIXED

--- Comment #7 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Fixed.