[Bug tree-optimization/101770] New: -Wmaybe-uninitialized false alarm with only locals in GNU diffutils

[Bug tree-optimization/101770] New: -Wmaybe-uninitialized false alarm with only locals in GNU diffutils

[eggert at cs dot ucla.edu]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101770

            Bug ID: 101770
           Summary: -Wmaybe-uninitialized false alarm with only locals in
                    GNU diffutils
           Product: gcc
           Version: 11.2.1
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: eggert at cs dot ucla.edu
  Target Milestone: ---

Created attachment 51256
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=51256&action=edit
False alarm with gcc -O2 -Wmaybe-uninitialized and only locals

I found this problem while compiling sdiff.c from GNU diffutils. The attached
is is a simplified version. It appears to be distinct from GCC bug 101768 since
it involves only local variables.

Compile it this way:

gcc -O2 -Wmaybe-uninitialized -S w.i

The output is:

w.i: In function ‘edit’:
w.i:50:18: warning: ‘cmd1’ may be used uninitialized in this function
[-Wmaybe-uninitialized]
   50 |           return !cmd1;
      |                  ^~~~~

This is a false alarm, because line 50 can be reached only if cmd0=='e', and if
cmd0=='e' then  the previous switch statement (inside the while loop - line 27)
guarantees that cmd1 has been initialized.

I briefly tried to simplify the example further, but failed. For example, if I
remove line 16 ("case '2':") the false alarm goes away.

This is with gcc (GCC) 11.2.1 20210728 (Red Hat 11.2.1-1) on x86-64.

[Bug tree-optimization/101770] -Wmaybe-uninitialized false alarm with only locals in GNU diffutils

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101770

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|                            |2021-08-04
             Blocks|                            |24639
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |NEW

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC is unable to prove that the conditions guarding the use match those of the
initialization.  The loop around the switch likely contributes to that.


Referenced Bugs:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=24639
[Bug 24639] [meta-bug] bug to track all Wuninitialized issues

[Bug tree-optimization/101770] -Wmaybe-uninitialized false alarm with only locals in GNU diffutils

[msebor at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101770

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |diagnostic
                 CC|                            |msebor at gcc dot gnu.org

--- Comment #2 from Martin Sebor <msebor at gcc dot gnu.org> ---
Here's output from the uninitialized pass patched to show in more detail what's
going on.  Cycles can contribute to false positives.  The chain length greater
messages also show possible reasons for false positives in an unpatched GCC
(the patched pass raises the limits to avoid them).  The note after the warning
shows the condition under which GCC determines the uninitialized variable is
used (it needs improvement).

Examining phi: cmd1_68 = PHI <cmd1_44(5), cmd1_12(4)>
cycle detected
cycle detected
cycle detected
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
cycle detected
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
cycle detected
cycle detected
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
cycle detected
cycle detected
predicate::predicate (def_bb = 6, use_bb = 3, func_t) initialized from 9
dep_chains:
        24 -> 8, 8 -> 9, 9 -> 26, 14 -> 15, 15 -> 32 
24 -> 8, 8 -> 9, 9 -> 26, 14 -> 15, 15 -> 18 
24 -> 8, 8 -> 9, 9 -> 26, 14 -> 15, 15 -> 19 
24 -> 8, 8 -> 10, 10 -> 11, 11 -> 27, 14 -> 15, 15 -> 32 
24 -> 8, 8 -> 10, 10 -> 11, 11 -> 27, 14 -> 15, 15 -> 18 
24 -> 8, 8 -> 10, 10 -> 11, 11 -> 27, 14 -> 15, 15 -> 19 
24 -> 8, 8 -> 10, 10 -> 13, 13 -> 30, 14 -> 15, 15 -> 32 
24 -> 8, 8 -> 10, 10 -> 13, 13 -> 30, 14 -> 15, 15 -> 18 
24 -> 8, 8 -> 10, 10 -> 13, 13 -> 30, 14 -> 15, 15 -> 19 
init_from_control_deps { { 24 -> 8, 8 -> 9, 9 -> 26, 14 -> 15, 15 -> 32 }, { 24
-> 8, 8 -> 9, 9 -> 26, 14 -> 15, 15 -> 18 }, { 24 -> 8, 8 -> 9, 9 -> 26, 14 ->
15, 15 -> 19 }, { 24 -> 8, 8 -> 10, 10 -> 11, 11 -> 27, 14 -> 15, 15 -> 32 }, {
24 -> 8, 8 -> 10, 10 -> 11, 11 -> 27, 14 -> 15, 15 -> 18 }, { 24 -> 8, 8 -> 10,
10 -> 11, 11 -> 27, 14 -> 15, 15 -> 19 }, { 24 -> 8, 8 -> 10, 10 -> 13, 13 ->
30, 14 -> 15, 15 -> 32 }, { 24 -> 8, 8 -> 10, 10 -> 13, 13 -> 30, 14 -> 15, 15
-> 18 }, { 24 -> 8, 8 -> 10, 10 -> 13, 13 -> 30, 14 -> 15, 15 -> 19 } }:
        (empty)
Found unguarded use in bb 3: cmd1_12 = PHI <cmd1_68(3), cmd1_19(D)(2)>
cycle detected
cycle detected
cycle detected
cycle detected
cycle detected
cycle detected
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
cycle detected
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
cycle detected
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
cycle detected
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
cycle detected
cycle detected
cycle detected
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
cycle detected
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
cycle detected
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
chain length greater than 5: 6
cycle detected
cycle detected
cycle detected
predicate::predicate (def_bb = 6, use_bb = 20, func_t) initialized from 3
dep_chains:
        24 -> 8, 8 -> 9, 9 -> 26, 14 -> 15, 15 -> 20 
24 -> 8, 8 -> 10, 10 -> 11, 11 -> 27, 14 -> 15, 15 -> 20 
24 -> 8, 8 -> 10, 10 -> 13, 13 -> 30, 14 -> 15, 15 -> 20 
init_from_control_deps { { 24 -> 8, 8 -> 9, 9 -> 26, 14 -> 15, 15 -> 20 }, { 24
-> 8, 8 -> 10, 10 -> 11, 11 -> 27, 14 -> 15, 15 -> 20 }, { 24 -> 8, 8 -> 10, 10
-> 13, 13 -> 30, 14 -> 15, 15 -> 20 } }:
        ((((MEM[(const char *)input_27 + -1B] != 101 && MEM[(const char
*)input_27 + -1B] <= 101) && MEM[(const char *)input_27 + -1B] == 50) &&
MEM[(const char *)input_74 + 1B] == 10) && _72 == 101 || ((((MEM[(const char
*)input_27 + -1B] != 101 && MEM[(const char *)input_27 + -1B] > 101) &&
MEM[(const char *)input_27 + -1B] <= 115) && MEM[(const char *)input_27 + -1B]
> 112) && MEM[(const char *)input_74 + 1B] == 10) && _72 == 101) ||
((((MEM[(const char *)input_27 + -1B] != 101 && MEM[(const char *)input_27 +
-1B] > 101) && MEM[(const char *)input_27 + -1B] > 115) && MEM[(const char
*)input_27 + -1B] == 118) && MEM[(const char *)input_74 + 1B] == 10) && _72 ==
101 (expanded)
        OR (AND (NOT (_2 == 101), NOT (_2 > 101), _2 == 50, _4 == 10, _72 ==
101), AND (NOT (_2 == 101), _2 > 101, NOT (_2 > 115), _2 > 112, _4 == 10, _72
== 101), AND (NOT (_2 == 101), _2 > 101, _2 > 115, _2 == 118, _4 == 10, _72 ==
101))
collect_phi_def_edges for cd_root 4 and cmd1_68 = PHI <cmd1_44(5), cmd1_12(4)>
        Found def edge 5 -> 6 for cd_root 4 and operand 0 of: cmd1_68 = PHI
<cmd1_44(5), cmd1_12(4)>
        Found def edge 3 -> 4 for cd_root 2 and operand 0 of: cmd1_12 = PHI
<cmd1_68(3), cmd1_19(D)(2)>
Found unguarded use in bb 20: _21 = cmd1_68 == 0;
pr101770.c: In function ‘edit’:
pr101770.c:50:18: warning: ‘cmd1’ may be used uninitialized in this function
[-Wmaybe-uninitialized]
   50 |           return !cmd1;
      |                  ^~~~~
pr101770.c:7:11: note: used when ‘*(input+-1) != 'e' && *(input+-1) <= 'e' &&
*(input+-1) == '2' && input[1] == 10 && _72 = PHI <118(30), 50(26), _2(27)> ==
'e' || *(input+-1) != 'e' && *(input+-1) > 'e' && *(input+-1) <= 's' &&
*(input+-1) > 'p' && input[1] == 10 && _72 = PHI <118(30), 50(26), _2(27)> ==
'e' || *(input+-1) != 'e' && *(input+-1) > 'e' && *(input+-1) > 's' &&
*(input+-1) == 'v' && input[1] == 10 && _72 = PHI <118(30), 50(26), _2(27)> ==
'e'’
    7 |       int cmd1;
      |           ^~~~
pr101770.c:7:11: note: ‘cmd1’ was declared here

[Bug tree-optimization/101770] -Wmaybe-uninitialized false alarm with only locals in GNU diffutils

[law at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101770

Jeffrey A. Law <law at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|NEW                         |RESOLVED
                 CC|                            |law at gcc dot gnu.org

--- Comment #3 from Jeffrey A. Law <law at gcc dot gnu.org> ---
Seems to be fixed on the trunk.

[Bug tree-optimization/101770] -Wmaybe-uninitialized false alarm with only locals in GNU diffutils

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101770

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

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

[Bug tree-optimization/101770] -Wmaybe-uninitialized false alarm with only locals in GNU diffutils

[eggert at cs dot ucla.edu]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101770

Paul Eggert <eggert at cs dot ucla.edu> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|FIXED                       |---
             Status|RESOLVED                    |REOPENED
            Version|11.2.1                      |13.1.1

--- Comment #4 from Paul Eggert <eggert at cs dot ucla.edu> ---
I seeing the bug with gcc (GCC) 13.1.1 20230511 (Red Hat 13.1.1-2) on x86-64
when compiling GNU diffutils, so although the bug was reported fixed on the
trunk last year, it appears that the fix hasn't propagated GCC 13 despite the
Target Milestone being 13.0.

The symptoms are:

$ gcc -O2 -Wmaybe-uninitialized -S w.i
w.i: In function ‘edit’:
w.i:50:18: warning: ‘cmd1’ may be used uninitialized [-Wmaybe-uninitialized]
   50 |           return !cmd1;
      |                  ^~~~~
w.i:7:11: note: ‘cmd1’ was declared here
    7 |       int cmd1;
      |           ^~~~

This appears to be the same bug as before so I am taking the liberty of
reopening the bug report.

[Bug tree-optimization/101770] -Wmaybe-uninitialized false alarm with only locals in GNU diffutils

[eggert at cs dot ucla.edu]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101770

--- Comment #5 from Paul Eggert <eggert at cs dot ucla.edu> ---
I can no longer reproduce the bug in bleeding-edge GNU diffutils, so this bug
is not so important in its own right - that is, it's merely that GCC 13.1.1
still mishandles w.i.