https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88680
Bug ID: 88680
Summary: [9 Regression] bogus -Wtype-limits for constant
expressions after r267272
Product: gcc
Version: 9.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c++
Assignee: unassigned at gcc dot gnu.org
Reporter: msebor at gcc dot gnu.org
Target Milestone: ---
-Wtype-limits is documented to
Warn if a comparison is always true or always false due to the limited
range of the data type, but do not warn for constant expressions.
According to the documentation the following test case should be accepted with
no warning, but since r267272 triggers one (Clang does not warn):
$ cat t.C && gcc -S -Wtype-limits t.C
const unsigned n = 8;
static_assert (n >= 0 && n % 2 == 0, "");
t.C:3:18: warning: comparison of unsigned expression >= 0 is always true
[-Wtype-limits]
3 | static_assert (n >= 0 && n % 2 == 0, "");
| ~~^~~~
This change apparently causes a large number of new warnings in Firefox builds.
According to my breakdown of Honza's build here at the following link, there
are over 20 thousand such warnings:
https://hg.mozilla.org/try/rev/e1e0472c3f68e47b9741d7814ef4417759cde24c
Diagnostic Count Unique Files
-Wtype-limits 20882 10441 604
-Wmultistatement-macros 1552 5 2
-Wcoverage-mismatch 793 225 141
-Wclass-memaccess 346 72 46
-Wsign-compare 292 127 55
-Wnarrowing 272 2 1
-Wmissing-profile 245 212 207
-Wattributes 212 1 1
-Wint-in-bool-context 140 3 2
-Wdeprecated-copy 104 22 10
-Wmaybe-uninitialized 103 67 54
