[Bug debug/105586] New: [11/12/13 Regression] -fcompare-debug failure (length) with -O2 -fno-if-conversion -mtune=power4 -fno-guess-branch-probability

[Bug debug/105586] New: [11/12/13 Regression] -fcompare-debug failure (length) with -O2 -fno-if-conversion -mtune=power4 -fno-guess-branch-probability

[zsojka at seznam dot cz]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105586

            Bug ID: 105586
           Summary: [11/12/13 Regression] -fcompare-debug failure (length)
                    with -O2 -fno-if-conversion -mtune=power4
                    -fno-guess-branch-probability
           Product: gcc
           Version: 13.0
            Status: UNCONFIRMED
          Keywords: compare-debug-failure
          Severity: normal
          Priority: P3
         Component: debug
          Assignee: unassigned at gcc dot gnu.org
          Reporter: zsojka at seznam dot cz
  Target Milestone: ---
              Host: x86_64-pc-linux-gnu
            Target: powerpc64le-unknown-linux-gnu

Created attachment 52968
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=52968&action=edit
reduced testcase

Compiler output:
$ powerpc64le-unknown-linux-gnu-gcc -O2 -fno-if-conversion -mtune=power4
-fno-guess-branch-probability -fcompare-debug testcase.c 
powerpc64le-unknown-linux-gnu-gcc: error: testcase.c: '-fcompare-debug' failure
(length)

$ diff -u *gkd
--- a-testcase.c.gkd    2022-05-12 21:05:45.139502516 +0200
+++ a-testcase.gk.c.gkd 2022-05-12 21:05:45.172835666 +0200
@@ -128,21 +128,23 @@
         (and:SI (reg:SI 10 10 [orig:136 u128_1 ] [136])
             (const_int 8 [0x8]))) "testcase.c":14:78# {andsi3_mask}
      (nil))
-(insn:TI # 0 0 (set (reg:DI 3 3 [orig:161 u64_4 ] [161])
-        (ashift:DI (reg:DI 3 3 [147])
-            (reg:SI 10 10 [160]))) "testcase.c":14:7# {ashldi3}
-     (expr_list:REG_DEAD (reg:SI 10 10 [160])
-        (nil)))
-(insn # 0 0 (set (reg:DI 30 30 [162])
+(insn:TI # 0 0 (set (reg:DI 30 30 [162])
         (plus:DI (reg/v:DI 30 30 [orig:145 b ] [145])
             (reg/v:DI 9 9 [orig:134 u64_3 ] [134]))) "testcase.c":15:17#
{*adddi3}
      (expr_list:REG_DEAD (reg/v:DI 9 9 [orig:134 u64_3 ] [134])
         (nil)))
+(insn # 0 0 (set (reg:DI 9 9 [orig:161 u64_4 ] [161])
+        (ashift:DI (reg:DI 3 3 [147])
+            (reg:SI 10 10 [160]))) "testcase.c":14:7# {ashldi3}
+     (expr_list:REG_DEAD (reg:SI 10 10 [160])
+        (expr_list:REG_DEAD (reg:DI 3 3 [147])
+            (nil))))
 (insn # 0 0 (set (reg:DI 3 3 [orig:163 u64_r ] [163])
-        (plus:DI (reg:DI 3 3 [orig:161 u64_4 ] [161])
+        (plus:DI (reg:DI 9 9 [orig:161 u64_4 ] [161])
             (reg:DI 30 30 [162]))) "testcase.c":15:7# {*adddi3}
      (expr_list:REG_DEAD (reg:DI 30 30 [162])
-        (nil)))
+        (expr_list:REG_DEAD (reg:DI 9 9 [orig:161 u64_4 ] [161])
+            (nil))))
 (insn # 0 0 (set (reg:SI 3 3 [orig:164 u64_r ] [164])
         (sign_extend:SI (reg:HI 3 3 [orig:163 u64_r ] [163])))
"testcase.c":16:7# {*extendhisi2}
      (nil))

$ powerpc64le-unknown-linux-gnu-gcc -v
Using built-in specs.
COLLECT_GCC=/repo/gcc-trunk/binary-latest-powerpc64le/bin/powerpc64le-unknown-linux-gnu-gcc
COLLECT_LTO_WRAPPER=/repo/gcc-trunk/binary-trunk-r13-333-20220511161459-g25addf8352e-checking-yes-rtl-df-extra-powerpc64le/bin/../libexec/gcc/powerpc64le-unknown-linux-gnu/13.0.0/lto-wrapper
Target: powerpc64le-unknown-linux-gnu
Configured with: /repo/gcc-trunk//configure --enable-languages=c,c++
--enable-valgrind-annotations --disable-nls --enable-checking=yes,rtl,df,extra
--with-cloog --with-ppl --with-isl
--with-sysroot=/usr/powerpc64le-unknown-linux-gnu --build=x86_64-pc-linux-gnu
--host=x86_64-pc-linux-gnu --target=powerpc64le-unknown-linux-gnu
--with-ld=/usr/bin/powerpc64le-unknown-linux-gnu-ld
--with-as=/usr/bin/powerpc64le-unknown-linux-gnu-as --disable-libstdcxx-pch
--prefix=/repo/gcc-trunk//binary-trunk-r13-333-20220511161459-g25addf8352e-checking-yes-rtl-df-extra-powerpc64le
Thread model: posix
Supported LTO compression algorithms: zlib zstd
gcc version 13.0.0 20220511 (experimental) (GCC)

[Bug debug/105586] [11/12/13 Regression] -fcompare-debug failure (length) with -O2 -fno-if-conversion -mtune=power4 -fno-guess-branch-probability

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105586

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |11.4

[Bug debug/105586] [11/12/13 Regression] -fcompare-debug failure (length) with -O2 -fno-if-conversion -mtune=power4 -fno-guess-branch-probability

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105586

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |NEW
                 CC|                            |jakub at gcc dot gnu.org,
                   |                            |linkw at gcc dot gnu.org,
                   |                            |segher at gcc dot gnu.org
   Last reconfirmed|                            |2022-05-13

--- Comment #1 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
The differences start during sched1 pass.
Could be the backend (rs6000_sched*, rs6000_variable_issue*) or the scheduler.

[Bug debug/105586] [11/12/13 Regression] -fcompare-debug failure (length) with -O2 -fno-if-conversion -mtune=power4 -fno-guess-branch-probability

[segher at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105586

--- Comment #2 from Segher Boessenkool <segher at gcc dot gnu.org> ---
We have

+(debug_insn 11 10 81 2 (var_location:QI u8_1 (mem/c:QI (plus:DI (unspec:DI [
+                    (symbol_ref:DI ("*.LANCHOR0") [flags 0x182])
+                    (reg:DI 2 2)
+                ] UNSPEC_TOCREL)
+            (const_int 3 [0x3])) [1 g+3 S1 A8])) "../105586.c":11:8 -1
+     (nil))

(the first such unspec in the file), and sched1 does

+;;       0--> b  0: i  11 loc [unspec[`*.LANCHOR0',%2] 47+0x3]   
:nothing:GENERAL_REGS+0(0)FLOAT_REGS+0(0)CR_REGS+0(0)SPECIAL_REGS+0(0)

with it.  Things in debug_insns should not influence code generation.

[Bug debug/105586] [11/12/13 Regression] -fcompare-debug failure (length) with -O2 -fno-if-conversion -mtune=power4 -fno-guess-branch-probability

[linkw at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105586

Kewen Lin <linkw at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jskumari at gcc dot gnu.org,
                   |                            |pthaugen at linux dot ibm.com

--- Comment #3 from Kewen Lin <linkw at gcc dot gnu.org> ---
btw, it doesn't need the cross build for reproduction, but if the native cpu
isn't power8, it needs the explicit option -mcpu=power8, otherwise cpu
power9/10 makes this issue gone.

[Bug debug/105586] [11/12/13 Regression] -fcompare-debug failure (length) with -O2 -fno-if-conversion -mtune=power4 -fno-guess-branch-probability

[jskumari at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105586

Surya Kumari Jangala <jskumari at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Assignee|unassigned at gcc dot gnu.org      |jskumari at gcc dot gnu.org

--- Comment #4 from Surya Kumari Jangala <jskumari at gcc dot gnu.org> ---
I will look into the issue.

[Bug rtl-optimization/105586] [11/12/13 Regression] -fcompare-debug failure (length) with -O2 -fno-if-conversion -mtune=power4 -fno-guess-branch-probability

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105586

--- Comment #5 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Surya Kumari Jangala
<jskumari@gcc.gnu.org>:

https://gcc.gnu.org/g:bec35caafae8db0278e1d037a7ba00c3e6899bbd

commit r13-2525-gbec35caafae8db0278e1d037a7ba00c3e6899bbd
Author: Surya Kumari Jangala <jskumari@linux.ibm.com>
Date:   Tue Aug 2 23:00:03 2022 -0500

    sched1: Fix -fcompare-debug issue in schedule_region [PR105586]

    In schedule_region(), a basic block that does not contain any real insns
    is not scheduled and the dfa state at the entry of the bb is not copied
    to the fallthru basic block. However a DEBUG insn is treated as a real
    insn, and if a bb contains non-real insns and a DEBUG insn, it's dfa
    state is copied to the fallthru bb. This was resulting in
    -fcompare-debug failure as the incoming dfa state of the fallthru block
    is different with -g. We should always copy the dfa state of a bb to
    it's fallthru bb even if the bb does not contain real insns.

    2022-08-22  Surya Kumari Jangala  <jskumari@linux.ibm.com>

    gcc/
            PR rtl-optimization/105586
            * sched-rgn.cc (save_state_for_fallthru_edge): New function.
            (schedule_region): Use it for all blocks.

    gcc/testsuite/
            PR rtl-optimization/105586
            * gcc.target/powerpc/pr105586.c: New test.

[Bug rtl-optimization/105586] [11/12 Regression] -fcompare-debug failure (length) with -O2 -fno-if-conversion -mtune=power4 -fno-guess-branch-probability

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105586

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
      Known to work|                            |13.0
           Priority|P3                          |P2
      Known to fail|13.0                        |
            Summary|[11/12/13 Regression]       |[11/12 Regression]
                   |-fcompare-debug failure     |-fcompare-debug failure
                   |(length) with -O2           |(length) with -O2
                   |-fno-if-conversion          |-fno-if-conversion
                   |-mtune=power4               |-mtune=power4
                   |-fno-guess-branch-probabili |-fno-guess-branch-probabili
                   |ty                          |ty

--- Comment #6 from Richard Biener <rguenth at gcc dot gnu.org> ---
Fixed on trunk sofar.

[Bug rtl-optimization/105586] [11/12 Regression] -fcompare-debug failure (length) with -O2 -fno-if-conversion -mtune=power4 -fno-guess-branch-probability

[bergner at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105586

Peter Bergner <bergner at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED

--- Comment #7 from Peter Bergner <bergner at gcc dot gnu.org> ---
(In reply to CVS Commits from comment #5)
> The master branch has been updated by Surya Kumari Jangala
> <jskumari@gcc.gnu.org>:
> 
> https://gcc.gnu.org/g:bec35caafae8db0278e1d037a7ba00c3e6899bbd
> 
> commit r13-2525-gbec35caafae8db0278e1d037a7ba00c3e6899bbd
> Author: Surya Kumari Jangala <jskumari@linux.ibm.com>
> Date:   Tue Aug 2 23:00:03 2022 -0500
> 
>     sched1: Fix -fcompare-debug issue in schedule_region [PR105586]

Surya, I think this has baked on trunk long enough, so if you have approval to
backport, I think you can do that now.

[Bug rtl-optimization/105586] [11/12 Regression] -fcompare-debug failure (length) with -O2 -fno-if-conversion -mtune=power4 -fno-guess-branch-probability

[segher at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105586

--- Comment #8 from Segher Boessenkool <segher at gcc dot gnu.org> ---
Please do that together with the follow-up fix only?  PR107171.

[Bug rtl-optimization/105586] [11/12 Regression] -fcompare-debug failure (length) with -O2 -fno-if-conversion -mtune=power4 -fno-guess-branch-probability

[segher at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105586

--- Comment #9 from Segher Boessenkool <segher at gcc dot gnu.org> ---
I read <https://gcc.gnu.org/pipermail/gcc-patches/2022-September/601914.html>
as approval to backport, fwiw :-)

[Bug rtl-optimization/105586] [11/12 Regression] -fcompare-debug failure (length) with -O2 -fno-if-conversion -mtune=power4 -fno-guess-branch-probability

[jskumari at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105586

--- Comment #10 from Surya Kumari Jangala <jskumari at gcc dot gnu.org> ---
(In reply to Segher Boessenkool from comment #9)
> I read <https://gcc.gnu.org/pipermail/gcc-patches/2022-September/601914.html>
> as approval to backport, fwiw :-)

I read that as: Since it is *not* a regression, no need to backport.

[Bug rtl-optimization/105586] [11/12 Regression] -fcompare-debug failure (length) with -O2 -fno-if-conversion -mtune=power4 -fno-guess-branch-probability

[segher at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105586

--- Comment #11 from Segher Boessenkool <segher at gcc dot gnu.org> ---
Ah.  It's a nasty bug, so "safe side" is to *do* fix it, in my reading.  But
you
can also say "not doing anything" (so, staying susceptible to the bug) is
safer.
It happens rather infrequently after all.

Just follow up, ask for clarification?

[Bug rtl-optimization/105586] [11/12 Regression] -fcompare-debug failure (length) with -O2 -fno-if-conversion -mtune=power4 -fno-guess-branch-probability

[jskumari at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105586

--- Comment #12 from Surya Kumari Jangala <jskumari at gcc dot gnu.org> ---
Richard has clarified here
(https://gcc.gnu.org/pipermail/gcc-patches/2022-November/605386.html) that
backporting is not required.

[Bug rtl-optimization/105586] [11/12 Regression] -fcompare-debug failure (length) with -O2 -fno-if-conversion -mtune=power4 -fno-guess-branch-probability

[jskumari at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105586

Surya Kumari Jangala <jskumari at gcc dot gnu.org> changed:

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

--- Comment #13 from Surya Kumari Jangala <jskumari at gcc dot gnu.org> ---
Closing the bug as it is fixed on trunk and we don't plan to backport it.

[Bug rtl-optimization/105586] [11/12 Regression] -fcompare-debug failure (length) with -O2 -fno-if-conversion -mtune=power4 -fno-guess-branch-probability

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=105586

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|11.4                        |13.0