* [Bug testsuite/111883] Wstringop-overflow-6.C FAILs with -std=c++26
2023-10-19 17:32 [Bug testsuite/111883] New: Wstringop-overflow-6.C FAILs with -std=c++26 mpolacek at gcc dot gnu.org
@ 2023-10-19 20:15 ` redi at gcc dot gnu.org
2023-10-19 20:29 ` jakub at gcc dot gnu.org
` (6 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: redi at gcc dot gnu.org @ 2023-10-19 20:15 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111883
--- Comment #1 from Jonathan Wakely <redi at gcc dot gnu.org> ---
I think Jakub wrote that code, but it looks like we just want the explicit
casts. I can add those.
^ permalink raw reply [flat|nested] 9+ messages in thread* [Bug testsuite/111883] Wstringop-overflow-6.C FAILs with -std=c++26
2023-10-19 17:32 [Bug testsuite/111883] New: Wstringop-overflow-6.C FAILs with -std=c++26 mpolacek at gcc dot gnu.org
2023-10-19 20:15 ` [Bug testsuite/111883] " redi at gcc dot gnu.org
@ 2023-10-19 20:29 ` jakub at gcc dot gnu.org
2023-10-19 20:41 ` mpolacek at gcc dot gnu.org
` (5 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: jakub at gcc dot gnu.org @ 2023-10-19 20:29 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111883
--- Comment #2 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Why doesn't it fail with -std=c++23 though? Was there some C++26 change I'm
not aware of?
In the to_chars cases, we already use float(__value) casts in the
_Float16/__bfloat16_t cases (but others too), so I think we just want to add
explicit casts also to all the from_chars
__value = __val;
lines (or at least the _Float16/__bfloat16_t cases).
^ permalink raw reply [flat|nested] 9+ messages in thread* [Bug testsuite/111883] Wstringop-overflow-6.C FAILs with -std=c++26
2023-10-19 17:32 [Bug testsuite/111883] New: Wstringop-overflow-6.C FAILs with -std=c++26 mpolacek at gcc dot gnu.org
2023-10-19 20:15 ` [Bug testsuite/111883] " redi at gcc dot gnu.org
2023-10-19 20:29 ` jakub at gcc dot gnu.org
@ 2023-10-19 20:41 ` mpolacek at gcc dot gnu.org
2023-10-20 11:27 ` jakub at gcc dot gnu.org
` (4 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: mpolacek at gcc dot gnu.org @ 2023-10-19 20:41 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111883
--- Comment #3 from Marek Polacek <mpolacek at gcc dot gnu.org> ---
Did you mean like the following? I have no idea if that's correct but is
suppresses the warnings I see.
In C++23 I don't see the code in the .ii file at all, so it doesn't warn.
--- a/libstdc++-v3/include/std/charconv
+++ b/libstdc++-v3/include/std/charconv
@@ -684,7 +684,7 @@ namespace __detail
from_chars_result __res
= __from_chars_float16_t(__first, __last, __val, __fmt);
if (__res.ec == errc{})
- __value = __val;
+ __value = _Float16(__val);
return __res;
}
#endif
@@ -697,7 +697,7 @@ namespace __detail
float __val;
from_chars_result __res = from_chars(__first, __last, __val, __fmt);
if (__res.ec == errc{})
- __value = __val;
+ __value = _Float32(__val);
return __res;
}
#endif
@@ -710,7 +710,7 @@ namespace __detail
double __val;
from_chars_result __res = from_chars(__first, __last, __val, __fmt);
if (__res.ec == errc{})
- __value = __val;
+ __value = _Float64(__val);
return __res;
}
#endif
@@ -723,7 +723,7 @@ namespace __detail
long double __val;
from_chars_result __res = from_chars(__first, __last, __val, __fmt);
if (__res.ec == errc{})
- __value = __val;
+ __value = _Float128(__val);
return __res;
}
#elif defined(__STDCPP_FLOAT128_T__) && defined(_GLIBCXX_HAVE_FLOAT128_MATH)
@@ -739,7 +739,7 @@ namespace __detail
__extension__ __ieee128 __val;
from_chars_result __res = from_chars(__first, __last, __val, __fmt);
if (__res.ec == errc{})
- __value = __val;
+ __value = _Float128(__val);
return __res;
}
#else
@@ -760,7 +760,7 @@ namespace __detail
from_chars_result __res
= __from_chars_bfloat16_t(__first, __last, __val, __fmt);
if (__res.ec == errc{})
- __value = __val;
+ __value = __gnu_cxx::__bfloat16_t(__val);
return __res;
}
#endif
^ permalink raw reply [flat|nested] 9+ messages in thread* [Bug testsuite/111883] Wstringop-overflow-6.C FAILs with -std=c++26
2023-10-19 17:32 [Bug testsuite/111883] New: Wstringop-overflow-6.C FAILs with -std=c++26 mpolacek at gcc dot gnu.org
` (2 preceding siblings ...)
2023-10-19 20:41 ` mpolacek at gcc dot gnu.org
@ 2023-10-20 11:27 ` jakub at gcc dot gnu.org
2023-10-20 12:04 ` redi at gcc dot gnu.org
` (3 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: jakub at gcc dot gnu.org @ 2023-10-20 11:27 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111883
--- Comment #4 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Yeah, that is what I meant. Note, I do see the from_chars inlines even with
-std=c++23
in preprocessed dump when including <charconv>, the difference is in
#if __cplusplus > 202302L
# include <charconv>
#endif
in bits/basic_string.h (and the trigger is -Wsystem-headers).
^ permalink raw reply [flat|nested] 9+ messages in thread* [Bug testsuite/111883] Wstringop-overflow-6.C FAILs with -std=c++26
2023-10-19 17:32 [Bug testsuite/111883] New: Wstringop-overflow-6.C FAILs with -std=c++26 mpolacek at gcc dot gnu.org
` (3 preceding siblings ...)
2023-10-20 11:27 ` jakub at gcc dot gnu.org
@ 2023-10-20 12:04 ` redi at gcc dot gnu.org
2023-10-20 13:13 ` mpolacek at gcc dot gnu.org
` (2 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: redi at gcc dot gnu.org @ 2023-10-20 12:04 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111883
Jonathan Wakely <redi at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |NEW
Ever confirmed|0 |1
Last reconfirmed| |2023-10-20
--- Comment #5 from Jonathan Wakely <redi at gcc dot gnu.org> ---
Ah yes, std::to_string(float) needs to use std::to_chars in C++26, so includes
<charconv>.
The patch in comment 3 looks correct, feel free to push it if it passes tests.
^ permalink raw reply [flat|nested] 9+ messages in thread* [Bug testsuite/111883] Wstringop-overflow-6.C FAILs with -std=c++26
2023-10-19 17:32 [Bug testsuite/111883] New: Wstringop-overflow-6.C FAILs with -std=c++26 mpolacek at gcc dot gnu.org
` (4 preceding siblings ...)
2023-10-20 12:04 ` redi at gcc dot gnu.org
@ 2023-10-20 13:13 ` mpolacek at gcc dot gnu.org
2023-10-20 16:36 ` cvs-commit at gcc dot gnu.org
2023-10-20 16:37 ` mpolacek at gcc dot gnu.org
7 siblings, 0 replies; 9+ messages in thread
From: mpolacek at gcc dot gnu.org @ 2023-10-20 13:13 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111883
Marek Polacek <mpolacek at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Assignee|unassigned at gcc dot gnu.org |mpolacek at gcc dot gnu.org
Status|NEW |ASSIGNED
--- Comment #6 from Marek Polacek <mpolacek at gcc dot gnu.org> ---
Thanks, I'll do a full test cycle.
^ permalink raw reply [flat|nested] 9+ messages in thread* [Bug testsuite/111883] Wstringop-overflow-6.C FAILs with -std=c++26
2023-10-19 17:32 [Bug testsuite/111883] New: Wstringop-overflow-6.C FAILs with -std=c++26 mpolacek at gcc dot gnu.org
` (5 preceding siblings ...)
2023-10-20 13:13 ` mpolacek at gcc dot gnu.org
@ 2023-10-20 16:36 ` cvs-commit at gcc dot gnu.org
2023-10-20 16:37 ` mpolacek at gcc dot gnu.org
7 siblings, 0 replies; 9+ messages in thread
From: cvs-commit at gcc dot gnu.org @ 2023-10-20 16:36 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111883
--- Comment #7 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The trunk branch has been updated by Marek Polacek <mpolacek@gcc.gnu.org>:
https://gcc.gnu.org/g:1d9263f292c3efc4e37c3a88117388e86d19b954
commit r14-4799-g1d9263f292c3efc4e37c3a88117388e86d19b954
Author: Marek Polacek <polacek@redhat.com>
Date: Thu Oct 19 17:48:44 2023 -0400
libstdc++: add casts to from_chars in <charconv> [PR111883]
This fixes
.../charconv: In function 'std::from_chars_result std::from_chars(const
char*, const char*, _Float16&, chars_format)':
.../charconv:687:17: warning: converting to '_Float16' from 'float' with
greater conversion rank
687 | __value = __val;
| ^~~~~
.../charconv: In function 'std::from_chars_result std::from_chars(const
char*, const char*, __gnu_cxx::__bfloat16_t&, chars_format)':
.../charconv:763:17: warning: converting to '__gnu_cxx::__bfloat16_t' {aka
'__bf16'} from 'float' with greater conversion rank
763 | __value = __val;
| ^~~~~
which was breaking a test:
FAIL: g++.dg/warn/Wstringop-overflow-6.C -std=gnu++26 (test for excess
errors)
PR testsuite/111883
libstdc++-v3/ChangeLog:
* include/std/charconv (from_chars): Add explicit casts.
^ permalink raw reply [flat|nested] 9+ messages in thread* [Bug testsuite/111883] Wstringop-overflow-6.C FAILs with -std=c++26
2023-10-19 17:32 [Bug testsuite/111883] New: Wstringop-overflow-6.C FAILs with -std=c++26 mpolacek at gcc dot gnu.org
` (6 preceding siblings ...)
2023-10-20 16:36 ` cvs-commit at gcc dot gnu.org
@ 2023-10-20 16:37 ` mpolacek at gcc dot gnu.org
7 siblings, 0 replies; 9+ messages in thread
From: mpolacek at gcc dot gnu.org @ 2023-10-20 16:37 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111883
Marek Polacek <mpolacek at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |RESOLVED
Resolution|--- |FIXED
--- Comment #8 from Marek Polacek <mpolacek at gcc dot gnu.org> ---
Should be fixed now.
^ permalink raw reply [flat|nested] 9+ messages in thread