[Bug bootstrap/95940] New: sparc64-linux bootstrap with gcc-9.3 broken

[Bug bootstrap/95940] New: sparc64-linux bootstrap with gcc-9.3 broken

[hp at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95940

            Bug ID: 95940
           Summary: sparc64-linux bootstrap with gcc-9.3 broken
           Product: gcc
           Version: 11.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: bootstrap
          Assignee: unassigned at gcc dot gnu.org
          Reporter: hp at gcc dot gnu.org
  Target Milestone: ---

As seen on gcc202 at LAST_UPDATED:
Fri Jun 26 21:36:32 CEST 2020
Fri Jun 26 19:36:32 UTC 2020 (revision
0fce12c0920:2873383514d:0801f419440c14f6772b28f763ad7d40f7f7a580)
(a.k.a. r11-1677-g0801f419440c14f6772b28f763ad7d40f7f7a580)
using no configure options and no make options:

/home/hp/combcheck/o0/./prev-gcc/xg++ -B/home/hp/combcheck/o0/./prev-gcc/
-B/usr/local/sparc64-unknown-linux-gnu/bin/ -nostdinc++
-B/home/hp/combcheck/o0/prev-sparc64-unknown-linux-gnu/libstdc++-v3/src/.libs
-B/home/hp/combcheck/o0/prev-sparc64-unknown-linux-gnu/libstdc++-v3/libsupc++/.libs

-I/home/hp/combcheck/o0/prev-sparc64-unknown-linux-gnu/libstdc++-v3/include/sparc64-unknown-linux-gnu
 -I/home/hp/combcheck/o0/prev-sparc64-unknown-linux-gnu/libstdc++-v3/include 
-I/home/hp/combcheck/gcc/libstdc++-v3/libsupc++
-L/home/hp/combcheck/o0/prev-sparc64-unknown-linux-gnu/libstdc++-v3/src/.libs
-L/home/hp/combcheck/o0/prev-sparc64-unknown-linux-gnu/libstdc++-v3/libsupc++/.libs
 -fno-PIE -c   -g -O2 -fno-checking -gtoggle -DIN_GCC     -fno-exceptions
-fno-rtti -fasynchronous-unwind-tables -W -Wall -Wno-narrowing -Wwrite-strings
-Wcast-qual -Wno-error=format-diag -Wmissing-format-attribute
-Woverloaded-virtual -pedantic -Wno-long-long -Wno-variadic-macros
-Wno-overlength-strings -Werror -fno-common  -DHAVE_CONFIG_H -I. -I.
-I/home/hp/combcheck/gcc/gcc -I/home/hp/combcheck/gcc/gcc/.
-I/home/hp/combcheck/gcc/gcc/../include
-I/home/hp/combcheck/gcc/gcc/../libcpp/include 
-I/home/hp/combcheck/gcc/gcc/../libdecnumber
-I/home/hp/combcheck/gcc/gcc/../libdecnumber/dpd -I../libdecnumber
-I/home/hp/combcheck/gcc/gcc/../libbacktrace   -o tree-ssanames.o -MT
tree-ssanames.o -MMD -MP -MF ./.deps/tree-ssanames.TPo
/home/hp/combcheck/gcc/gcc/tree-ssanames.c
In function ?wide_int get_nonzero_bits(const_tree)?,
    inlined from ?wide_int get_nonzero_bits(const_tree)? at
/home/hp/combcheck/gcc/gcc/tree-ssanames.c:514:1:
/home/hp/combcheck/gcc/gcc/tree-ssanames.c:527:46: error:
?<anonymous>.generic_wide_int<wide_int_storage>::<anonymous>.wide_int_storage::val[1]?
may be used uninitialized [-Werror=maybe-uninitialized]
  527 |     | (HOST_WIDE_INT) pi->misalign, precision);
      |                                              ^
/home/hp/combcheck/gcc/gcc/tree-ssanames.c: In function ?wide_int
get_nonzero_bits(const_tree)?:
/home/hp/combcheck/gcc/gcc/tree-ssanames.c:527:46: note: ?<anonymous>? declared
here
  527 |     | (HOST_WIDE_INT) pi->misalign, precision);
      |                                              ^
In function ?wide_int get_nonzero_bits(const_tree)?,
    inlined from ?wide_int get_nonzero_bits(const_tree)? at
/home/hp/combcheck/gcc/gcc/tree-ssanames.c:514:1:
/home/hp/combcheck/gcc/gcc/tree-ssanames.c:527:46: error:
?<anonymous>.generic_wide_int<wide_int_storage>::<anonymous>.wide_int_storage::val[2]?
may be used uninitialized [-Werror=maybe-uninitialized]
  527 |     | (HOST_WIDE_INT) pi->misalign, precision);
      |                                              ^
/home/hp/combcheck/gcc/gcc/tree-ssanames.c: In function ?wide_int
get_nonzero_bits(const_tree)?:
/home/hp/combcheck/gcc/gcc/tree-ssanames.c:527:46: note: ?<anonymous>? declared
here
  527 |     | (HOST_WIDE_INT) pi->misalign, precision);
      |                                              ^
In function ?wide_int get_nonzero_bits(const_tree)?,
    inlined from ?wide_int get_nonzero_bits(const_tree)? at
/home/hp/combcheck/gcc/gcc/tree-ssanames.c:514:1:
/home/hp/combcheck/gcc/gcc/tree-ssanames.c:528:37: error:
?<anonymous>.generic_wide_int<wide_int_storage>::<anonymous>.wide_int_storage::val[1]?
may be used uninitialized [-Werror=maybe-uninitialized]
  528 |       return wi::shwi (-1, precision);
      |                                     ^
/home/hp/combcheck/gcc/gcc/tree-ssanames.c: In function ?wide_int
get_nonzero_bits(const_tree)?:
/home/hp/combcheck/gcc/gcc/tree-ssanames.c:528:37: note: ?<anonymous>? declared
here
  528 |       return wi::shwi (-1, precision);
      |                                     ^
In function ?wide_int get_nonzero_bits(const_tree)?,
    inlined from ?wide_int get_nonzero_bits(const_tree)? at
/home/hp/combcheck/gcc/gcc/tree-ssanames.c:514:1:
/home/hp/combcheck/gcc/gcc/tree-ssanames.c:528:37: error:
?<anonymous>.generic_wide_int<wide_int_storage>::<anonymous>.wide_int_storage::val[2]?
may be used uninitialized [-Werror=maybe-uninitialized]
  528 |       return wi::shwi (-1, precision);
      |                                     ^
/home/hp/combcheck/gcc/gcc/tree-ssanames.c: In function ?wide_int
get_nonzero_bits(const_tree)?:
/home/hp/combcheck/gcc/gcc/tree-ssanames.c:528:37: note: ?<anonymous>? declared
here
  528 |       return wi::shwi (-1, precision);
      |                                     ^
In function ?wide_int get_nonzero_bits(const_tree)?,
    inlined from ?wide_int get_nonzero_bits(const_tree)? at
/home/hp/combcheck/gcc/gcc/tree-ssanames.c:514:1:
/home/hp/combcheck/gcc/gcc/tree-ssanames.c:533:35: error:
?<anonymous>.generic_wide_int<wide_int_storage>::<anonymous>.wide_int_storage::val[1]?
may be used uninitialized [-Werror=maybe-uninitialized]
  533 |     return wi::shwi (-1, precision);
      |                                   ^
/home/hp/combcheck/gcc/gcc/tree-ssanames.c: In function ?wide_int
get_nonzero_bits(const_tree)?:
/home/hp/combcheck/gcc/gcc/tree-ssanames.c:533:35: note: ?<anonymous>? declared
here
  533 |     return wi::shwi (-1, precision);
      |                                   ^
In function ?wide_int get_nonzero_bits(const_tree)?,
    inlined from ?wide_int get_nonzero_bits(const_tree)? at
/home/hp/combcheck/gcc/gcc/tree-ssanames.c:514:1:
/home/hp/combcheck/gcc/gcc/tree-ssanames.c:533:35: error:
?<anonymous>.generic_wide_int<wide_int_storage>::<anonymous>.wide_int_storage::val[2]?
may be used uninitialized [-Werror=maybe-uninitialized]
  533 |     return wi::shwi (-1, precision);
      |                                   ^
/home/hp/combcheck/gcc/gcc/tree-ssanames.c: In function ?wide_int
get_nonzero_bits(const_tree)?:
/home/hp/combcheck/gcc/gcc/tree-ssanames.c:533:35: note: ?<anonymous>? declared
here
  533 |     return wi::shwi (-1, precision);
      |                                   ^
cc1plus: all warnings being treated as errors
make[3]: *** [Makefile:1124: tree-ssanames.o] Error 1
make[3]: Leaving directory '/home/hp/combcheck/o0/gcc'
make[2]: *** [Makefile:4783: all-stage2-gcc] Error 2
make[2]: Leaving directory '/home/hp/combcheck/o0'
make[1]: *** [Makefile:24260: stage2-bubble] Error 2
make[1]: Leaving directory '/home/hp/combcheck/o0'
make: *** [Makefile:998: all] Error 2

[Bug bootstrap/95940] [11 Regression] sparc64-linux bootstrap with gcc-9.3 broken

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95940

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |11.0
           Keywords|                            |diagnostic
            Summary|sparc64-linux bootstrap     |[11 Regression]
                   |with gcc-9.3 broken         |sparc64-linux bootstrap
                   |                            |with gcc-9.3 broken
                 CC|                            |msebor at gcc dot gnu.org

[Bug bootstrap/95940] [11 Regression] sparc64-linux bootstrap with gcc-9.3 broken

[msebor at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95940

Martin Sebor <msebor at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|                            |2020-06-29
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |WAITING

--- Comment #1 from Martin Sebor <msebor at gcc dot gnu.org> ---
Can you provide a test case or a translation unit?

[Bug bootstrap/95940] [11 Regression] sparc64-linux bootstrap with gcc-9.3 broken

[hp at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95940

--- Comment #2 from Hans-Peter Nilsson <hp at gcc dot gnu.org> ---
(In reply to Martin Sebor from comment #1)
> Can you provide a test case or a translation unit?

You're aware that the report is for bootstrap stage2 on a CompileFarm machine?
>From the look of it, something is already miscompiled.
I'm not sure how e.g. tree-ssanames.ii by itself would help, but I'll attach
it.

[Bug bootstrap/95940] [11 Regression] sparc64-linux bootstrap with gcc-9.3 broken

[hp at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95940

--- Comment #3 from Hans-Peter Nilsson <hp at gcc dot gnu.org> ---
Created attachment 48810
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=48810&action=edit
tree-ssanames.ii.gz

Beware, gzippped file.

Repeat with /home/hp/combcheck/o0/./prev-gcc/cc1plus -fpreprocessed
tree-ssanames.ii -quiet -dumpbase tree-ssanames.c -dumpbase-ext .c -mcpu=v9 -g
-gtoggle -O2 -Wextra -Wall -Wno-narrowing -Wwrite-strings -Wcast-qual
-Wno-error=format-diag -Wsuggest-attribute=format -Woverloaded-virtual
-Wpedantic -Wno-long-long -Wno-variadic-macros -Wno-overlength-strings -Werror
-version -fno-PIE -fno-checking -fno-exceptions -fno-rtti
-fasynchronous-unwind-tables -fno-common -o tree-ssanames.s

(in stage 2)

[Bug bootstrap/95940] [11 Regression] bootstrap broken by -Wmaybe-unintialized warnings

[ebotcazou at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95940

Martin Sebor <msebor at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |NEW

Eric Botcazou <ebotcazou at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[11 Regression]             |[11 Regression] bootstrap
                   |sparc64-linux bootstrap     |broken by
                   |with gcc-9.3 broken         |-Wmaybe-unintialized
                   |                            |warnings
                 CC|                            |ebotcazou at gcc dot gnu.org

--- Comment #4 from Martin Sebor <msebor at gcc dot gnu.org> ---
The first -Wmaybe-unintialized warning is issued for the read in this
statement:

  SR.1076_294 = D.75512.D.16218.val[1];

in the following basic block:

  <bb 24> [local count: 130803720]:
  # SR.268_98 = PHI <_104(23), _75(39)>
  D.75553 ={v} {CLOBBER};
  D.75512 ={v} {CLOBBER};
  xi ={v} {CLOBBER};
  SR.1076_294 = D.75512.D.16218.val[1];   <<< warning
  SR.1077_296 = D.75512.D.16218.val[2];   <<< warning
  D.75512 ={v} {CLOBBER};
  D.75513 ={v} {CLOBBER};
  goto <bb 33>; [100.00%]

Since D.75512 is clobbered just prior to the read from it I think the warning
is justified.  Subsequent warnings follow a similar pattern:

  <bb 26> [local count: 349243402]:
  # SR.268_123 = PHI <_129(25), -1(41)>
  D.75575 ={v} {CLOBBER};
  D.75514 ={v} {CLOBBER};
  xi ={v} {CLOBBER};
  SR.1076_230 = D.75514.D.16218.val[1];   <<< warning
  SR.1077_105 = D.75514.D.16218.val[2];   <<< warning
  D.75514 ={v} {CLOBBER};
  D.75515 ={v} {CLOBBER};
  goto <bb 33>; [100.00%]
  ...
  <bb 31> [local count: 68586742]:
  # SR.268_148 = PHI <_154(30), -1(40)>
  D.75597 ={v} {CLOBBER};
  D.75516 ={v} {CLOBBER};
  xi ={v} {CLOBBER};
  SR.1076_219 = D.75516.D.16218.val[1];   <<< warning
  SR.1077_220 = D.75516.D.16218.val[2];   <<< warning
  D.75516 ={v} {CLOBBER};
  D.75517 ={v} {CLOBBER};
  goto <bb 33>; [100.00%]

--- Comment #5 from Martin Sebor <msebor at gcc dot gnu.org> ---
The first -Wmaybe-unintialized warning is issued for the read in this
statement:

  SR.1076_294 = D.75512.D.16218.val[1];

in the following basic block:

  <bb 24> [local count: 130803720]:
  # SR.268_98 = PHI <_104(23), _75(39)>
  D.75553 ={v} {CLOBBER};
  D.75512 ={v} {CLOBBER};
  xi ={v} {CLOBBER};
  SR.1076_294 = D.75512.D.16218.val[1];   <<< warning
  SR.1077_296 = D.75512.D.16218.val[2];   <<< warning
  D.75512 ={v} {CLOBBER};
  D.75513 ={v} {CLOBBER};
  goto <bb 33>; [100.00%]

Since D.75512 is clobbered just prior to the read from it I think the warning
is justified.  Subsequent warnings follow a similar pattern:

  <bb 26> [local count: 349243402]:
  # SR.268_123 = PHI <_129(25), -1(41)>
  D.75575 ={v} {CLOBBER};
  D.75514 ={v} {CLOBBER};
  xi ={v} {CLOBBER};
  SR.1076_230 = D.75514.D.16218.val[1];   <<< warning
  SR.1077_105 = D.75514.D.16218.val[2];   <<< warning
  D.75514 ={v} {CLOBBER};
  D.75515 ={v} {CLOBBER};
  goto <bb 33>; [100.00%]
  ...
  <bb 31> [local count: 68586742]:
  # SR.268_148 = PHI <_154(30), -1(40)>
  D.75597 ={v} {CLOBBER};
  D.75516 ={v} {CLOBBER};
  xi ={v} {CLOBBER};
  SR.1076_219 = D.75516.D.16218.val[1];   <<< warning
  SR.1077_220 = D.75516.D.16218.val[2];   <<< warning
  D.75516 ={v} {CLOBBER};
  D.75517 ={v} {CLOBBER};
  goto <bb 33>; [100.00%]

--- Comment #6 from Eric Botcazou <ebotcazou at gcc dot gnu.org> ---
> From the look of it, something is already miscompiled.

No, not at all, it's just warnings turned into errors.

[Bug bootstrap/95940] [11 Regression] bootstrap broken by -Wmaybe-unintialized warnings

[hp at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95940

--- Comment #7 from Hans-Peter Nilsson <hp at gcc dot gnu.org> ---
(In reply to Eric Botcazou from comment #6)
> > From the look of it, something is already miscompiled.
> 
> No, not at all, it's just warnings turned into errors.

Not obvious, but I see from the comments that people seem to be on the right
track.  Some target-specific thing under the hood (not obvious to me in the
source) in tree-ssanames.c provoking this?

FWIW, this problem did not manifest on neither ppc64le, x86_64 nor aarch64,
same source. (All *-linux.)

[Bug bootstrap/95940] [11 Regression] bootstrap broken by -Wmaybe-unintialized warnings

[ebotcazou at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95940

Eric Botcazou <ebotcazou at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |DUPLICATE

--- Comment #8 from Eric Botcazou <ebotcazou at gcc dot gnu.org> ---
> Not obvious, but I see from the comments that people seem to be on the right
> track.  Some target-specific thing under the hood (not obvious to me in the
> source) in tree-ssanames.c provoking this?

Obvious for those who maintain the SPARC compiler. ;-)  The warnings have been
there for months in the 10.x series.  In fact it's a duplicate.

*** This bug has been marked as a duplicate of bug 92002 ***