* [PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning
@ 2023-09-28 16:37 Dimitry Andric
2023-10-26 17:00 ` [PING][PATCH] " Dimitry Andric
2024-01-30 16:19 ` [PATCH] " Jonathan Wakely
0 siblings, 2 replies; 17+ messages in thread
From: Dimitry Andric @ 2023-09-28 16:37 UTC (permalink / raw)
To: gcc-patches; +Cc: Gerald Pfeifer
Ref: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111632
When building gcc's C++ sources against recent libc++, the poisoning of
the ctype macros due to including safe-ctype.h before including C++
standard headers such as <list>, <map>, etc, causes many compilation
errors, similar to:
In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
In file included from /home/dim/src/gcc/master/gcc/system.h:233:
In file included from /usr/include/c++/v1/vector:321:
In file included from
/usr/include/c++/v1/__format/formatter_bool.h:20:
In file included from
/usr/include/c++/v1/__format/formatter_integral.h:32:
In file included from /usr/include/c++/v1/locale:202:
/usr/include/c++/v1/__locale:546:5: error: '__abi_tag__' attribute
only applies to structs, variables, functions, and namespaces
546 | _LIBCPP_INLINE_VISIBILITY
| ^
/usr/include/c++/v1/__config:813:37: note: expanded from macro
'_LIBCPP_INLINE_VISIBILITY'
813 | # define _LIBCPP_INLINE_VISIBILITY _LIBCPP_HIDE_FROM_ABI
| ^
/usr/include/c++/v1/__config:792:26: note: expanded from macro
'_LIBCPP_HIDE_FROM_ABI'
792 |
__attribute__((__abi_tag__(_LIBCPP_TOSTRING(
_LIBCPP_VERSIONED_IDENTIFIER))))
| ^
In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
In file included from /home/dim/src/gcc/master/gcc/system.h:233:
In file included from /usr/include/c++/v1/vector:321:
In file included from
/usr/include/c++/v1/__format/formatter_bool.h:20:
In file included from
/usr/include/c++/v1/__format/formatter_integral.h:32:
In file included from /usr/include/c++/v1/locale:202:
/usr/include/c++/v1/__locale:547:37: error: expected ';' at end of
declaration list
547 | char_type toupper(char_type __c) const
| ^
/usr/include/c++/v1/__locale:553:48: error: too many arguments
provided to function-like macro invocation
553 | const char_type* toupper(char_type* __low, const
char_type* __high) const
| ^
/home/dim/src/gcc/master/gcc/../include/safe-ctype.h:146:9: note:
macro 'toupper' defined here
146 | #define toupper(c) do_not_use_toupper_with_safe_ctype
| ^
This is because libc++ uses different transitive includes than
libstdc++, and some of those transitive includes pull in various ctype
declarations (typically via <locale>).
There was already a special case for including <string> before
safe-ctype.h, so move the rest of the C++ standard header includes to
the same location, to fix the problem.
Signed-off-by: Dimitry Andric <dimitry@andric.com>
---
gcc/system.h | 39 ++++++++++++++++++---------------------
1 file changed, 18 insertions(+), 21 deletions(-)
diff --git a/gcc/system.h b/gcc/system.h
index e924152ad4c..7a516b11438 100644
--- a/gcc/system.h
+++ b/gcc/system.h
@@ -194,27 +194,8 @@ extern int fprintf_unlocked (FILE *, const char *, ...);
#undef fread_unlocked
#undef fwrite_unlocked
-/* Include <string> before "safe-ctype.h" to avoid GCC poisoning
- the ctype macros through safe-ctype.h */
-
-#ifdef __cplusplus
-#ifdef INCLUDE_STRING
-# include <string>
-#endif
-#endif
-
-/* There are an extraordinary number of issues with <ctype.h>.
- The last straw is that it varies with the locale. Use libiberty's
- replacement instead. */
-#include "safe-ctype.h"
-
-#include <sys/types.h>
-
-#include <errno.h>
-
-#if !defined (errno) && defined (HAVE_DECL_ERRNO) && !HAVE_DECL_ERRNO
-extern int errno;
-#endif
+/* Include C++ standard headers before "safe-ctype.h" to avoid GCC
+ poisoning the ctype macros through safe-ctype.h */
#ifdef __cplusplus
#if defined (INCLUDE_ALGORITHM) || !defined (HAVE_SWAP_IN_UTILITY)
@@ -229,6 +210,9 @@ extern int errno;
#ifdef INCLUDE_SET
# include <set>
#endif
+#ifdef INCLUDE_STRING
+# include <string>
+#endif
#ifdef INCLUDE_VECTOR
# include <vector>
#endif
@@ -245,6 +229,19 @@ extern int errno;
# include <type_traits>
#endif
+/* There are an extraordinary number of issues with <ctype.h>.
+ The last straw is that it varies with the locale. Use libiberty's
+ replacement instead. */
+#include "safe-ctype.h"
+
+#include <sys/types.h>
+
+#include <errno.h>
+
+#if !defined (errno) && defined (HAVE_DECL_ERRNO) && !HAVE_DECL_ERRNO
+extern int errno;
+#endif
+
/* Some of glibc's string inlines cause warnings. Plus we'd rather
rely on (and therefore test) GCC's string builtins. */
#define __NO_STRING_INLINES
--
2.42.0
^ permalink raw reply [flat|nested] 17+ messages in thread
* [PING][PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning
2023-09-28 16:37 [PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning Dimitry Andric
@ 2023-10-26 17:00 ` Dimitry Andric
2023-10-29 13:08 ` Sam James
2024-01-30 16:19 ` [PATCH] " Jonathan Wakely
1 sibling, 1 reply; 17+ messages in thread
From: Dimitry Andric @ 2023-10-26 17:00 UTC (permalink / raw)
To: gcc-patches; +Cc: Gerald Pfeifer
Ping. It would be nice to get this QoL fix in.
-Dimitry
> On 28 Sep 2023, at 18:37, Dimitry Andric <dimitry@andric.com> wrote:
>
> Ref: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111632
>
> When building gcc's C++ sources against recent libc++, the poisoning of
> the ctype macros due to including safe-ctype.h before including C++
> standard headers such as <list>, <map>, etc, causes many compilation
> errors, similar to:
>
> In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
> In file included from /home/dim/src/gcc/master/gcc/system.h:233:
> In file included from /usr/include/c++/v1/vector:321:
> In file included from
> /usr/include/c++/v1/__format/formatter_bool.h:20:
> In file included from
> /usr/include/c++/v1/__format/formatter_integral.h:32:
> In file included from /usr/include/c++/v1/locale:202:
> /usr/include/c++/v1/__locale:546:5: error: '__abi_tag__' attribute
> only applies to structs, variables, functions, and namespaces
> 546 | _LIBCPP_INLINE_VISIBILITY
> | ^
> /usr/include/c++/v1/__config:813:37: note: expanded from macro
> '_LIBCPP_INLINE_VISIBILITY'
> 813 | # define _LIBCPP_INLINE_VISIBILITY _LIBCPP_HIDE_FROM_ABI
> | ^
> /usr/include/c++/v1/__config:792:26: note: expanded from macro
> '_LIBCPP_HIDE_FROM_ABI'
> 792 |
> __attribute__((__abi_tag__(_LIBCPP_TOSTRING(
> _LIBCPP_VERSIONED_IDENTIFIER))))
> | ^
> In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
> In file included from /home/dim/src/gcc/master/gcc/system.h:233:
> In file included from /usr/include/c++/v1/vector:321:
> In file included from
> /usr/include/c++/v1/__format/formatter_bool.h:20:
> In file included from
> /usr/include/c++/v1/__format/formatter_integral.h:32:
> In file included from /usr/include/c++/v1/locale:202:
> /usr/include/c++/v1/__locale:547:37: error: expected ';' at end of
> declaration list
> 547 | char_type toupper(char_type __c) const
> | ^
> /usr/include/c++/v1/__locale:553:48: error: too many arguments
> provided to function-like macro invocation
> 553 | const char_type* toupper(char_type* __low, const
> char_type* __high) const
> | ^
> /home/dim/src/gcc/master/gcc/../include/safe-ctype.h:146:9: note:
> macro 'toupper' defined here
> 146 | #define toupper(c) do_not_use_toupper_with_safe_ctype
> | ^
>
> This is because libc++ uses different transitive includes than
> libstdc++, and some of those transitive includes pull in various ctype
> declarations (typically via <locale>).
>
> There was already a special case for including <string> before
> safe-ctype.h, so move the rest of the C++ standard header includes to
> the same location, to fix the problem.
>
> Signed-off-by: Dimitry Andric <dimitry@andric.com>
> ---
> gcc/system.h | 39 ++++++++++++++++++---------------------
> 1 file changed, 18 insertions(+), 21 deletions(-)
>
> diff --git a/gcc/system.h b/gcc/system.h
> index e924152ad4c..7a516b11438 100644
> --- a/gcc/system.h
> +++ b/gcc/system.h
> @@ -194,27 +194,8 @@ extern int fprintf_unlocked (FILE *, const char *, ...);
> #undef fread_unlocked
> #undef fwrite_unlocked
>
> -/* Include <string> before "safe-ctype.h" to avoid GCC poisoning
> - the ctype macros through safe-ctype.h */
> -
> -#ifdef __cplusplus
> -#ifdef INCLUDE_STRING
> -# include <string>
> -#endif
> -#endif
> -
> -/* There are an extraordinary number of issues with <ctype.h>.
> - The last straw is that it varies with the locale. Use libiberty's
> - replacement instead. */
> -#include "safe-ctype.h"
> -
> -#include <sys/types.h>
> -
> -#include <errno.h>
> -
> -#if !defined (errno) && defined (HAVE_DECL_ERRNO) && !HAVE_DECL_ERRNO
> -extern int errno;
> -#endif
> +/* Include C++ standard headers before "safe-ctype.h" to avoid GCC
> + poisoning the ctype macros through safe-ctype.h */
>
> #ifdef __cplusplus
> #if defined (INCLUDE_ALGORITHM) || !defined (HAVE_SWAP_IN_UTILITY)
> @@ -229,6 +210,9 @@ extern int errno;
> #ifdef INCLUDE_SET
> # include <set>
> #endif
> +#ifdef INCLUDE_STRING
> +# include <string>
> +#endif
> #ifdef INCLUDE_VECTOR
> # include <vector>
> #endif
> @@ -245,6 +229,19 @@ extern int errno;
> # include <type_traits>
> #endif
>
> +/* There are an extraordinary number of issues with <ctype.h>.
> + The last straw is that it varies with the locale. Use libiberty's
> + replacement instead. */
> +#include "safe-ctype.h"
> +
> +#include <sys/types.h>
> +
> +#include <errno.h>
> +
> +#if !defined (errno) && defined (HAVE_DECL_ERRNO) && !HAVE_DECL_ERRNO
> +extern int errno;
> +#endif
> +
> /* Some of glibc's string inlines cause warnings. Plus we'd rather
> rely on (and therefore test) GCC's string builtins. */
> #define __NO_STRING_INLINES
> --
> 2.42.0
>
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PING][PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning
2023-10-26 17:00 ` [PING][PATCH] " Dimitry Andric
@ 2023-10-29 13:08 ` Sam James
2024-01-20 9:17 ` Sam James
0 siblings, 1 reply; 17+ messages in thread
From: Sam James @ 2023-10-29 13:08 UTC (permalink / raw)
To: Dimitry Andric; +Cc: Gerald Pfeifer, gcc-patches, Jeff Law
Dimitry Andric <dimitry@andric.com> writes:
> Ping. It would be nice to get this QoL fix in.
>
Yes please - we've been using this in Gentoo since around when it was
first posted. No complaints.
I cannot approve but it looks good to me.
> -Dimitry
>
>> On 28 Sep 2023, at 18:37, Dimitry Andric <dimitry@andric.com> wrote:
>>
>> Ref: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111632
>>
>> When building gcc's C++ sources against recent libc++, the poisoning of
>> the ctype macros due to including safe-ctype.h before including C++
>> standard headers such as <list>, <map>, etc, causes many compilation
>> errors, similar to:
>>
>> In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
>> In file included from /home/dim/src/gcc/master/gcc/system.h:233:
>> In file included from /usr/include/c++/v1/vector:321:
>> In file included from
>> /usr/include/c++/v1/__format/formatter_bool.h:20:
>> In file included from
>> /usr/include/c++/v1/__format/formatter_integral.h:32:
>> In file included from /usr/include/c++/v1/locale:202:
>> /usr/include/c++/v1/__locale:546:5: error: '__abi_tag__' attribute
>> only applies to structs, variables, functions, and namespaces
>> 546 | _LIBCPP_INLINE_VISIBILITY
>> | ^
>> /usr/include/c++/v1/__config:813:37: note: expanded from macro
>> '_LIBCPP_INLINE_VISIBILITY'
>> 813 | # define _LIBCPP_INLINE_VISIBILITY _LIBCPP_HIDE_FROM_ABI
>> | ^
>> /usr/include/c++/v1/__config:792:26: note: expanded from macro
>> '_LIBCPP_HIDE_FROM_ABI'
>> 792 |
>> __attribute__((__abi_tag__(_LIBCPP_TOSTRING(
>> _LIBCPP_VERSIONED_IDENTIFIER))))
>> | ^
>> In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
>> In file included from /home/dim/src/gcc/master/gcc/system.h:233:
>> In file included from /usr/include/c++/v1/vector:321:
>> In file included from
>> /usr/include/c++/v1/__format/formatter_bool.h:20:
>> In file included from
>> /usr/include/c++/v1/__format/formatter_integral.h:32:
>> In file included from /usr/include/c++/v1/locale:202:
>> /usr/include/c++/v1/__locale:547:37: error: expected ';' at end of
>> declaration list
>> 547 | char_type toupper(char_type __c) const
>> | ^
>> /usr/include/c++/v1/__locale:553:48: error: too many arguments
>> provided to function-like macro invocation
>> 553 | const char_type* toupper(char_type* __low, const
>> char_type* __high) const
>> | ^
>> /home/dim/src/gcc/master/gcc/../include/safe-ctype.h:146:9: note:
>> macro 'toupper' defined here
>> 146 | #define toupper(c) do_not_use_toupper_with_safe_ctype
>> | ^
>>
>> This is because libc++ uses different transitive includes than
>> libstdc++, and some of those transitive includes pull in various ctype
>> declarations (typically via <locale>).
>>
>> There was already a special case for including <string> before
>> safe-ctype.h, so move the rest of the C++ standard header includes to
>> the same location, to fix the problem.
>>
>> Signed-off-by: Dimitry Andric <dimitry@andric.com>
>> ---
>> gcc/system.h | 39 ++++++++++++++++++---------------------
>> 1 file changed, 18 insertions(+), 21 deletions(-)
>>
>> diff --git a/gcc/system.h b/gcc/system.h
>> index e924152ad4c..7a516b11438 100644
>> --- a/gcc/system.h
>> +++ b/gcc/system.h
>> @@ -194,27 +194,8 @@ extern int fprintf_unlocked (FILE *, const char *, ...);
>> #undef fread_unlocked
>> #undef fwrite_unlocked
>>
>> -/* Include <string> before "safe-ctype.h" to avoid GCC poisoning
>> - the ctype macros through safe-ctype.h */
>> -
>> -#ifdef __cplusplus
>> -#ifdef INCLUDE_STRING
>> -# include <string>
>> -#endif
>> -#endif
>> -
>> -/* There are an extraordinary number of issues with <ctype.h>.
>> - The last straw is that it varies with the locale. Use libiberty's
>> - replacement instead. */
>> -#include "safe-ctype.h"
>> -
>> -#include <sys/types.h>
>> -
>> -#include <errno.h>
>> -
>> -#if !defined (errno) && defined (HAVE_DECL_ERRNO) && !HAVE_DECL_ERRNO
>> -extern int errno;
>> -#endif
>> +/* Include C++ standard headers before "safe-ctype.h" to avoid GCC
>> + poisoning the ctype macros through safe-ctype.h */
>>
>> #ifdef __cplusplus
>> #if defined (INCLUDE_ALGORITHM) || !defined (HAVE_SWAP_IN_UTILITY)
>> @@ -229,6 +210,9 @@ extern int errno;
>> #ifdef INCLUDE_SET
>> # include <set>
>> #endif
>> +#ifdef INCLUDE_STRING
>> +# include <string>
>> +#endif
>> #ifdef INCLUDE_VECTOR
>> # include <vector>
>> #endif
>> @@ -245,6 +229,19 @@ extern int errno;
>> # include <type_traits>
>> #endif
>>
>> +/* There are an extraordinary number of issues with <ctype.h>.
>> + The last straw is that it varies with the locale. Use libiberty's
>> + replacement instead. */
>> +#include "safe-ctype.h"
>> +
>> +#include <sys/types.h>
>> +
>> +#include <errno.h>
>> +
>> +#if !defined (errno) && defined (HAVE_DECL_ERRNO) && !HAVE_DECL_ERRNO
>> +extern int errno;
>> +#endif
>> +
>> /* Some of glibc's string inlines cause warnings. Plus we'd rather
>> rely on (and therefore test) GCC's string builtins. */
>> #define __NO_STRING_INLINES
>> --
>> 2.42.0
>>
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PING][PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning
2023-10-29 13:08 ` Sam James
@ 2024-01-20 9:17 ` Sam James
2024-01-30 15:07 ` Sam James
0 siblings, 1 reply; 17+ messages in thread
From: Sam James @ 2024-01-20 9:17 UTC (permalink / raw)
To: Sam James; +Cc: Dimitry Andric, Gerald Pfeifer, gcc-patches, Jeff Law
Sam James <sam@gentoo.org> writes:
> Dimitry Andric <dimitry@andric.com> writes:
>
>> Ping. It would be nice to get this QoL fix in.
>>
>
> Yes please - we've been using this in Gentoo since around when it was
> first posted. No complaints.
>
> I cannot approve but it looks good to me.
Ping.
>
>> -Dimitry
>>
>>> On 28 Sep 2023, at 18:37, Dimitry Andric <dimitry@andric.com> wrote:
>>>
>>> Ref: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111632
>>>
>>> When building gcc's C++ sources against recent libc++, the poisoning of
>>> the ctype macros due to including safe-ctype.h before including C++
>>> standard headers such as <list>, <map>, etc, causes many compilation
>>> errors, similar to:
>>>
>>> In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
>>> In file included from /home/dim/src/gcc/master/gcc/system.h:233:
>>> In file included from /usr/include/c++/v1/vector:321:
>>> In file included from
>>> /usr/include/c++/v1/__format/formatter_bool.h:20:
>>> In file included from
>>> /usr/include/c++/v1/__format/formatter_integral.h:32:
>>> In file included from /usr/include/c++/v1/locale:202:
>>> /usr/include/c++/v1/__locale:546:5: error: '__abi_tag__' attribute
>>> only applies to structs, variables, functions, and namespaces
>>> 546 | _LIBCPP_INLINE_VISIBILITY
>>> | ^
>>> /usr/include/c++/v1/__config:813:37: note: expanded from macro
>>> '_LIBCPP_INLINE_VISIBILITY'
>>> 813 | # define _LIBCPP_INLINE_VISIBILITY _LIBCPP_HIDE_FROM_ABI
>>> | ^
>>> /usr/include/c++/v1/__config:792:26: note: expanded from macro
>>> '_LIBCPP_HIDE_FROM_ABI'
>>> 792 |
>>> __attribute__((__abi_tag__(_LIBCPP_TOSTRING(
>>> _LIBCPP_VERSIONED_IDENTIFIER))))
>>> | ^
>>> In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
>>> In file included from /home/dim/src/gcc/master/gcc/system.h:233:
>>> In file included from /usr/include/c++/v1/vector:321:
>>> In file included from
>>> /usr/include/c++/v1/__format/formatter_bool.h:20:
>>> In file included from
>>> /usr/include/c++/v1/__format/formatter_integral.h:32:
>>> In file included from /usr/include/c++/v1/locale:202:
>>> /usr/include/c++/v1/__locale:547:37: error: expected ';' at end of
>>> declaration list
>>> 547 | char_type toupper(char_type __c) const
>>> | ^
>>> /usr/include/c++/v1/__locale:553:48: error: too many arguments
>>> provided to function-like macro invocation
>>> 553 | const char_type* toupper(char_type* __low, const
>>> char_type* __high) const
>>> | ^
>>> /home/dim/src/gcc/master/gcc/../include/safe-ctype.h:146:9: note:
>>> macro 'toupper' defined here
>>> 146 | #define toupper(c) do_not_use_toupper_with_safe_ctype
>>> | ^
>>>
>>> This is because libc++ uses different transitive includes than
>>> libstdc++, and some of those transitive includes pull in various ctype
>>> declarations (typically via <locale>).
>>>
>>> There was already a special case for including <string> before
>>> safe-ctype.h, so move the rest of the C++ standard header includes to
>>> the same location, to fix the problem.
>>>
>>> Signed-off-by: Dimitry Andric <dimitry@andric.com>
>>> ---
>>> gcc/system.h | 39 ++++++++++++++++++---------------------
>>> 1 file changed, 18 insertions(+), 21 deletions(-)
>>>
>>> diff --git a/gcc/system.h b/gcc/system.h
>>> index e924152ad4c..7a516b11438 100644
>>> --- a/gcc/system.h
>>> +++ b/gcc/system.h
>>> @@ -194,27 +194,8 @@ extern int fprintf_unlocked (FILE *, const char *, ...);
>>> #undef fread_unlocked
>>> #undef fwrite_unlocked
>>>
>>> -/* Include <string> before "safe-ctype.h" to avoid GCC poisoning
>>> - the ctype macros through safe-ctype.h */
>>> -
>>> -#ifdef __cplusplus
>>> -#ifdef INCLUDE_STRING
>>> -# include <string>
>>> -#endif
>>> -#endif
>>> -
>>> -/* There are an extraordinary number of issues with <ctype.h>.
>>> - The last straw is that it varies with the locale. Use libiberty's
>>> - replacement instead. */
>>> -#include "safe-ctype.h"
>>> -
>>> -#include <sys/types.h>
>>> -
>>> -#include <errno.h>
>>> -
>>> -#if !defined (errno) && defined (HAVE_DECL_ERRNO) && !HAVE_DECL_ERRNO
>>> -extern int errno;
>>> -#endif
>>> +/* Include C++ standard headers before "safe-ctype.h" to avoid GCC
>>> + poisoning the ctype macros through safe-ctype.h */
>>>
>>> #ifdef __cplusplus
>>> #if defined (INCLUDE_ALGORITHM) || !defined (HAVE_SWAP_IN_UTILITY)
>>> @@ -229,6 +210,9 @@ extern int errno;
>>> #ifdef INCLUDE_SET
>>> # include <set>
>>> #endif
>>> +#ifdef INCLUDE_STRING
>>> +# include <string>
>>> +#endif
>>> #ifdef INCLUDE_VECTOR
>>> # include <vector>
>>> #endif
>>> @@ -245,6 +229,19 @@ extern int errno;
>>> # include <type_traits>
>>> #endif
>>>
>>> +/* There are an extraordinary number of issues with <ctype.h>.
>>> + The last straw is that it varies with the locale. Use libiberty's
>>> + replacement instead. */
>>> +#include "safe-ctype.h"
>>> +
>>> +#include <sys/types.h>
>>> +
>>> +#include <errno.h>
>>> +
>>> +#if !defined (errno) && defined (HAVE_DECL_ERRNO) && !HAVE_DECL_ERRNO
>>> +extern int errno;
>>> +#endif
>>> +
>>> /* Some of glibc's string inlines cause warnings. Plus we'd rather
>>> rely on (and therefore test) GCC's string builtins. */
>>> #define __NO_STRING_INLINES
>>> --
>>> 2.42.0
>>>
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PING][PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning
2024-01-20 9:17 ` Sam James
@ 2024-01-30 15:07 ` Sam James
2024-01-30 20:27 ` Sam James
0 siblings, 1 reply; 17+ messages in thread
From: Sam James @ 2024-01-30 15:07 UTC (permalink / raw)
To: Sam James
Cc: Dimitry Andric, Gerald Pfeifer, gcc-patches, Jeff Law,
Arsen Arsenović
[-- Attachment #1: Type: text/plain, Size: 5843 bytes --]
Sam James <sam@gentoo.org> writes:
> Sam James <sam@gentoo.org> writes:
>
>> Dimitry Andric <dimitry@andric.com> writes:
>>
>>> Ping. It would be nice to get this QoL fix in.
>>>
>>
>> Yes please - we've been using this in Gentoo since around when it was
>> first posted. No complaints.
>>
>> I cannot approve but it looks good to me.
>
> Ping.
Ping. This is straightforward, I just can't merge it myself.
>
>>
>>> -Dimitry
>>>
>>>> On 28 Sep 2023, at 18:37, Dimitry Andric <dimitry@andric.com> wrote:
>>>>
>>>> Ref: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111632
>>>>
>>>> When building gcc's C++ sources against recent libc++, the poisoning of
>>>> the ctype macros due to including safe-ctype.h before including C++
>>>> standard headers such as <list>, <map>, etc, causes many compilation
>>>> errors, similar to:
>>>>
>>>> In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
>>>> In file included from /home/dim/src/gcc/master/gcc/system.h:233:
>>>> In file included from /usr/include/c++/v1/vector:321:
>>>> In file included from
>>>> /usr/include/c++/v1/__format/formatter_bool.h:20:
>>>> In file included from
>>>> /usr/include/c++/v1/__format/formatter_integral.h:32:
>>>> In file included from /usr/include/c++/v1/locale:202:
>>>> /usr/include/c++/v1/__locale:546:5: error: '__abi_tag__' attribute
>>>> only applies to structs, variables, functions, and namespaces
>>>> 546 | _LIBCPP_INLINE_VISIBILITY
>>>> | ^
>>>> /usr/include/c++/v1/__config:813:37: note: expanded from macro
>>>> '_LIBCPP_INLINE_VISIBILITY'
>>>> 813 | # define _LIBCPP_INLINE_VISIBILITY _LIBCPP_HIDE_FROM_ABI
>>>> | ^
>>>> /usr/include/c++/v1/__config:792:26: note: expanded from macro
>>>> '_LIBCPP_HIDE_FROM_ABI'
>>>> 792 |
>>>> __attribute__((__abi_tag__(_LIBCPP_TOSTRING(
>>>> _LIBCPP_VERSIONED_IDENTIFIER))))
>>>> | ^
>>>> In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
>>>> In file included from /home/dim/src/gcc/master/gcc/system.h:233:
>>>> In file included from /usr/include/c++/v1/vector:321:
>>>> In file included from
>>>> /usr/include/c++/v1/__format/formatter_bool.h:20:
>>>> In file included from
>>>> /usr/include/c++/v1/__format/formatter_integral.h:32:
>>>> In file included from /usr/include/c++/v1/locale:202:
>>>> /usr/include/c++/v1/__locale:547:37: error: expected ';' at end of
>>>> declaration list
>>>> 547 | char_type toupper(char_type __c) const
>>>> | ^
>>>> /usr/include/c++/v1/__locale:553:48: error: too many arguments
>>>> provided to function-like macro invocation
>>>> 553 | const char_type* toupper(char_type* __low, const
>>>> char_type* __high) const
>>>> | ^
>>>> /home/dim/src/gcc/master/gcc/../include/safe-ctype.h:146:9: note:
>>>> macro 'toupper' defined here
>>>> 146 | #define toupper(c) do_not_use_toupper_with_safe_ctype
>>>> | ^
>>>>
>>>> This is because libc++ uses different transitive includes than
>>>> libstdc++, and some of those transitive includes pull in various ctype
>>>> declarations (typically via <locale>).
>>>>
>>>> There was already a special case for including <string> before
>>>> safe-ctype.h, so move the rest of the C++ standard header includes to
>>>> the same location, to fix the problem.
>>>>
>>>> Signed-off-by: Dimitry Andric <dimitry@andric.com>
>>>> ---
>>>> gcc/system.h | 39 ++++++++++++++++++---------------------
>>>> 1 file changed, 18 insertions(+), 21 deletions(-)
>>>>
>>>> diff --git a/gcc/system.h b/gcc/system.h
>>>> index e924152ad4c..7a516b11438 100644
>>>> --- a/gcc/system.h
>>>> +++ b/gcc/system.h
>>>> @@ -194,27 +194,8 @@ extern int fprintf_unlocked (FILE *, const char *, ...);
>>>> #undef fread_unlocked
>>>> #undef fwrite_unlocked
>>>>
>>>> -/* Include <string> before "safe-ctype.h" to avoid GCC poisoning
>>>> - the ctype macros through safe-ctype.h */
>>>> -
>>>> -#ifdef __cplusplus
>>>> -#ifdef INCLUDE_STRING
>>>> -# include <string>
>>>> -#endif
>>>> -#endif
>>>> -
>>>> -/* There are an extraordinary number of issues with <ctype.h>.
>>>> - The last straw is that it varies with the locale. Use libiberty's
>>>> - replacement instead. */
>>>> -#include "safe-ctype.h"
>>>> -
>>>> -#include <sys/types.h>
>>>> -
>>>> -#include <errno.h>
>>>> -
>>>> -#if !defined (errno) && defined (HAVE_DECL_ERRNO) && !HAVE_DECL_ERRNO
>>>> -extern int errno;
>>>> -#endif
>>>> +/* Include C++ standard headers before "safe-ctype.h" to avoid GCC
>>>> + poisoning the ctype macros through safe-ctype.h */
>>>>
>>>> #ifdef __cplusplus
>>>> #if defined (INCLUDE_ALGORITHM) || !defined (HAVE_SWAP_IN_UTILITY)
>>>> @@ -229,6 +210,9 @@ extern int errno;
>>>> #ifdef INCLUDE_SET
>>>> # include <set>
>>>> #endif
>>>> +#ifdef INCLUDE_STRING
>>>> +# include <string>
>>>> +#endif
>>>> #ifdef INCLUDE_VECTOR
>>>> # include <vector>
>>>> #endif
>>>> @@ -245,6 +229,19 @@ extern int errno;
>>>> # include <type_traits>
>>>> #endif
>>>>
>>>> +/* There are an extraordinary number of issues with <ctype.h>.
>>>> + The last straw is that it varies with the locale. Use libiberty's
>>>> + replacement instead. */
>>>> +#include "safe-ctype.h"
>>>> +
>>>> +#include <sys/types.h>
>>>> +
>>>> +#include <errno.h>
>>>> +
>>>> +#if !defined (errno) && defined (HAVE_DECL_ERRNO) && !HAVE_DECL_ERRNO
>>>> +extern int errno;
>>>> +#endif
>>>> +
>>>> /* Some of glibc's string inlines cause warnings. Plus we'd rather
>>>> rely on (and therefore test) GCC's string builtins. */
>>>> #define __NO_STRING_INLINES
>>>> --
>>>> 2.42.0
>>>>
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 377 bytes --]
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning
2023-09-28 16:37 [PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning Dimitry Andric
2023-10-26 17:00 ` [PING][PATCH] " Dimitry Andric
@ 2024-01-30 16:19 ` Jonathan Wakely
1 sibling, 0 replies; 17+ messages in thread
From: Jonathan Wakely @ 2024-01-30 16:19 UTC (permalink / raw)
To: Dimitry Andric; +Cc: gcc-patches, Gerald Pfeifer
On 28/09/23 18:37 +0200, Dimitry Andric wrote:
>Ref: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111632
>
>When building gcc's C++ sources against recent libc++, the poisoning of
>the ctype macros due to including safe-ctype.h before including C++
>standard headers such as <list>, <map>, etc, causes many compilation
>errors, similar to:
>
> In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
> In file included from /home/dim/src/gcc/master/gcc/system.h:233:
> In file included from /usr/include/c++/v1/vector:321:
> In file included from
> /usr/include/c++/v1/__format/formatter_bool.h:20:
> In file included from
> /usr/include/c++/v1/__format/formatter_integral.h:32:
> In file included from /usr/include/c++/v1/locale:202:
> /usr/include/c++/v1/__locale:546:5: error: '__abi_tag__' attribute
> only applies to structs, variables, functions, and namespaces
> 546 | _LIBCPP_INLINE_VISIBILITY
> | ^
> /usr/include/c++/v1/__config:813:37: note: expanded from macro
> '_LIBCPP_INLINE_VISIBILITY'
> 813 | # define _LIBCPP_INLINE_VISIBILITY _LIBCPP_HIDE_FROM_ABI
> | ^
> /usr/include/c++/v1/__config:792:26: note: expanded from macro
> '_LIBCPP_HIDE_FROM_ABI'
> 792 |
> __attribute__((__abi_tag__(_LIBCPP_TOSTRING(
> _LIBCPP_VERSIONED_IDENTIFIER))))
> | ^
> In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
> In file included from /home/dim/src/gcc/master/gcc/system.h:233:
> In file included from /usr/include/c++/v1/vector:321:
> In file included from
> /usr/include/c++/v1/__format/formatter_bool.h:20:
> In file included from
> /usr/include/c++/v1/__format/formatter_integral.h:32:
> In file included from /usr/include/c++/v1/locale:202:
> /usr/include/c++/v1/__locale:547:37: error: expected ';' at end of
> declaration list
> 547 | char_type toupper(char_type __c) const
> | ^
> /usr/include/c++/v1/__locale:553:48: error: too many arguments
> provided to function-like macro invocation
> 553 | const char_type* toupper(char_type* __low, const
> char_type* __high) const
> | ^
> /home/dim/src/gcc/master/gcc/../include/safe-ctype.h:146:9: note:
> macro 'toupper' defined here
> 146 | #define toupper(c) do_not_use_toupper_with_safe_ctype
> | ^
>
>This is because libc++ uses different transitive includes than
>libstdc++, and some of those transitive includes pull in various ctype
>declarations (typically via <locale>).
>
>There was already a special case for including <string> before
>safe-ctype.h, so move the rest of the C++ standard header includes to
>the same location, to fix the problem.
I can't approve this either, but FWIW it looks good to me.
>Signed-off-by: Dimitry Andric <dimitry@andric.com>
>---
>gcc/system.h | 39 ++++++++++++++++++---------------------
>1 file changed, 18 insertions(+), 21 deletions(-)
>
>diff --git a/gcc/system.h b/gcc/system.h
>index e924152ad4c..7a516b11438 100644
>--- a/gcc/system.h
>+++ b/gcc/system.h
>@@ -194,27 +194,8 @@ extern int fprintf_unlocked (FILE *, const char *, ...);
>#undef fread_unlocked
>#undef fwrite_unlocked
>
>-/* Include <string> before "safe-ctype.h" to avoid GCC poisoning
>- the ctype macros through safe-ctype.h */
>-
>-#ifdef __cplusplus
>-#ifdef INCLUDE_STRING
>-# include <string>
>-#endif
>-#endif
>-
>-/* There are an extraordinary number of issues with <ctype.h>.
>- The last straw is that it varies with the locale. Use libiberty's
>- replacement instead. */
>-#include "safe-ctype.h"
>-
>-#include <sys/types.h>
>-
>-#include <errno.h>
>-
>-#if !defined (errno) && defined (HAVE_DECL_ERRNO) && !HAVE_DECL_ERRNO
>-extern int errno;
>-#endif
>+/* Include C++ standard headers before "safe-ctype.h" to avoid GCC
>+ poisoning the ctype macros through safe-ctype.h */
>
>#ifdef __cplusplus
>#if defined (INCLUDE_ALGORITHM) || !defined (HAVE_SWAP_IN_UTILITY)
>@@ -229,6 +210,9 @@ extern int errno;
>#ifdef INCLUDE_SET
># include <set>
>#endif
>+#ifdef INCLUDE_STRING
>+# include <string>
>+#endif
>#ifdef INCLUDE_VECTOR
># include <vector>
>#endif
>@@ -245,6 +229,19 @@ extern int errno;
># include <type_traits>
>#endif
>
>+/* There are an extraordinary number of issues with <ctype.h>.
>+ The last straw is that it varies with the locale. Use libiberty's
>+ replacement instead. */
>+#include "safe-ctype.h"
>+
>+#include <sys/types.h>
>+
>+#include <errno.h>
>+
>+#if !defined (errno) && defined (HAVE_DECL_ERRNO) && !HAVE_DECL_ERRNO
>+extern int errno;
>+#endif
>+
>/* Some of glibc's string inlines cause warnings. Plus we'd rather
> rely on (and therefore test) GCC's string builtins. */
>#define __NO_STRING_INLINES
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PING][PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning
2024-01-30 15:07 ` Sam James
@ 2024-01-30 20:27 ` Sam James
0 siblings, 0 replies; 17+ messages in thread
From: Sam James @ 2024-01-30 20:27 UTC (permalink / raw)
To: Sam James
Cc: Dimitry Andric, Gerald Pfeifer, gcc-patches, Jeff Law,
Arsen Arsenović
Sam James <sam@gentoo.org> writes:
> [[PGP Signed Part:Undecided]]
>
> Sam James <sam@gentoo.org> writes:
>
>> Sam James <sam@gentoo.org> writes:
>>
>>> Dimitry Andric <dimitry@andric.com> writes:
>>>
>>>> Ping. It would be nice to get this QoL fix in.
>>>>
>>>
>>> Yes please - we've been using this in Gentoo since around when it was
>>> first posted. No complaints.
>>>
>>> I cannot approve but it looks good to me.
>>
>> Ping.
>
> Ping. This is straightforward, I just can't merge it myself.
>
>>
We discussed it on IRC and Iain is going to test it on Darwin, I'm going
to test it on the cfarm on platforms the patch was not already exposed
to, after Jeff raised some concerns.
I also have an additional hunk or two I'm testing for JIT.
>>>
>>>> -Dimitry
>>>>
>>>>> On 28 Sep 2023, at 18:37, Dimitry Andric <dimitry@andric.com> wrote:
>>>>>
>>>>> Ref: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111632
>>>>>
>>>>> When building gcc's C++ sources against recent libc++, the poisoning of
>>>>> the ctype macros due to including safe-ctype.h before including C++
>>>>> standard headers such as <list>, <map>, etc, causes many compilation
>>>>> errors, similar to:
>>>>>
>>>>> In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
>>>>> In file included from /home/dim/src/gcc/master/gcc/system.h:233:
>>>>> In file included from /usr/include/c++/v1/vector:321:
>>>>> In file included from
>>>>> /usr/include/c++/v1/__format/formatter_bool.h:20:
>>>>> In file included from
>>>>> /usr/include/c++/v1/__format/formatter_integral.h:32:
>>>>> In file included from /usr/include/c++/v1/locale:202:
>>>>> /usr/include/c++/v1/__locale:546:5: error: '__abi_tag__' attribute
>>>>> only applies to structs, variables, functions, and namespaces
>>>>> 546 | _LIBCPP_INLINE_VISIBILITY
>>>>> | ^
>>>>> /usr/include/c++/v1/__config:813:37: note: expanded from macro
>>>>> '_LIBCPP_INLINE_VISIBILITY'
>>>>> 813 | # define _LIBCPP_INLINE_VISIBILITY _LIBCPP_HIDE_FROM_ABI
>>>>> | ^
>>>>> /usr/include/c++/v1/__config:792:26: note: expanded from macro
>>>>> '_LIBCPP_HIDE_FROM_ABI'
>>>>> 792 |
>>>>> __attribute__((__abi_tag__(_LIBCPP_TOSTRING(
>>>>> _LIBCPP_VERSIONED_IDENTIFIER))))
>>>>> | ^
>>>>> In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
>>>>> In file included from /home/dim/src/gcc/master/gcc/system.h:233:
>>>>> In file included from /usr/include/c++/v1/vector:321:
>>>>> In file included from
>>>>> /usr/include/c++/v1/__format/formatter_bool.h:20:
>>>>> In file included from
>>>>> /usr/include/c++/v1/__format/formatter_integral.h:32:
>>>>> In file included from /usr/include/c++/v1/locale:202:
>>>>> /usr/include/c++/v1/__locale:547:37: error: expected ';' at end of
>>>>> declaration list
>>>>> 547 | char_type toupper(char_type __c) const
>>>>> | ^
>>>>> /usr/include/c++/v1/__locale:553:48: error: too many arguments
>>>>> provided to function-like macro invocation
>>>>> 553 | const char_type* toupper(char_type* __low, const
>>>>> char_type* __high) const
>>>>> | ^
>>>>> /home/dim/src/gcc/master/gcc/../include/safe-ctype.h:146:9: note:
>>>>> macro 'toupper' defined here
>>>>> 146 | #define toupper(c) do_not_use_toupper_with_safe_ctype
>>>>> | ^
>>>>>
>>>>> This is because libc++ uses different transitive includes than
>>>>> libstdc++, and some of those transitive includes pull in various ctype
>>>>> declarations (typically via <locale>).
>>>>>
>>>>> There was already a special case for including <string> before
>>>>> safe-ctype.h, so move the rest of the C++ standard header includes to
>>>>> the same location, to fix the problem.
>>>>>
>>>>> Signed-off-by: Dimitry Andric <dimitry@andric.com>
>>>>> ---
>>>>> gcc/system.h | 39 ++++++++++++++++++---------------------
>>>>> 1 file changed, 18 insertions(+), 21 deletions(-)
>>>>>
>>>>> diff --git a/gcc/system.h b/gcc/system.h
>>>>> index e924152ad4c..7a516b11438 100644
>>>>> --- a/gcc/system.h
>>>>> +++ b/gcc/system.h
>>>>> @@ -194,27 +194,8 @@ extern int fprintf_unlocked (FILE *, const char *, ...);
>>>>> #undef fread_unlocked
>>>>> #undef fwrite_unlocked
>>>>>
>>>>> -/* Include <string> before "safe-ctype.h" to avoid GCC poisoning
>>>>> - the ctype macros through safe-ctype.h */
>>>>> -
>>>>> -#ifdef __cplusplus
>>>>> -#ifdef INCLUDE_STRING
>>>>> -# include <string>
>>>>> -#endif
>>>>> -#endif
>>>>> -
>>>>> -/* There are an extraordinary number of issues with <ctype.h>.
>>>>> - The last straw is that it varies with the locale. Use libiberty's
>>>>> - replacement instead. */
>>>>> -#include "safe-ctype.h"
>>>>> -
>>>>> -#include <sys/types.h>
>>>>> -
>>>>> -#include <errno.h>
>>>>> -
>>>>> -#if !defined (errno) && defined (HAVE_DECL_ERRNO) && !HAVE_DECL_ERRNO
>>>>> -extern int errno;
>>>>> -#endif
>>>>> +/* Include C++ standard headers before "safe-ctype.h" to avoid GCC
>>>>> + poisoning the ctype macros through safe-ctype.h */
>>>>>
>>>>> #ifdef __cplusplus
>>>>> #if defined (INCLUDE_ALGORITHM) || !defined (HAVE_SWAP_IN_UTILITY)
>>>>> @@ -229,6 +210,9 @@ extern int errno;
>>>>> #ifdef INCLUDE_SET
>>>>> # include <set>
>>>>> #endif
>>>>> +#ifdef INCLUDE_STRING
>>>>> +# include <string>
>>>>> +#endif
>>>>> #ifdef INCLUDE_VECTOR
>>>>> # include <vector>
>>>>> #endif
>>>>> @@ -245,6 +229,19 @@ extern int errno;
>>>>> # include <type_traits>
>>>>> #endif
>>>>>
>>>>> +/* There are an extraordinary number of issues with <ctype.h>.
>>>>> + The last straw is that it varies with the locale. Use libiberty's
>>>>> + replacement instead. */
>>>>> +#include "safe-ctype.h"
>>>>> +
>>>>> +#include <sys/types.h>
>>>>> +
>>>>> +#include <errno.h>
>>>>> +
>>>>> +#if !defined (errno) && defined (HAVE_DECL_ERRNO) && !HAVE_DECL_ERRNO
>>>>> +extern int errno;
>>>>> +#endif
>>>>> +
>>>>> /* Some of glibc's string inlines cause warnings. Plus we'd rather
>>>>> rely on (and therefore test) GCC's string builtins. */
>>>>> #define __NO_STRING_INLINES
>>>>> --
>>>>> 2.42.0
>>>>>
>
> [[End of PGP Signed Part]]
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning
2024-04-03 9:01 ` Iain Sandoe
@ 2024-04-03 9:38 ` Richard Biener
0 siblings, 0 replies; 17+ messages in thread
From: Richard Biener @ 2024-04-03 9:38 UTC (permalink / raw)
To: Iain Sandoe; +Cc: GCC Patches, Gerald Pfeifer
On Wed, 3 Apr 2024, Iain Sandoe wrote:
> Hi Richard,
>
> > On 7 Mar 2024, at 13:40, FX Coudert <fxcoudert@gmail.com> wrote:
> >
> >> I think it's an obvious change ...
> >
> > Thanks, pushed.
> >
> > Dimitry, I suggest you post the second patch for review.
>
> Given that the two patches here (for https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111632) were considered obvious - and are needed on release branches.
>
> OK for backporting?
OK.
> (Gerald has volunteered to do the earlier ones, I have already made/tested the gcc-13 case)
>
> thanks
> Iain
>
>
--
Richard Biener <rguenther@suse.de>
SUSE Software Solutions Germany GmbH,
Frankenstrasse 146, 90461 Nuernberg, Germany;
GF: Ivo Totev, Andrew McDonald, Werner Knoblich; (HRB 36809, AG Nuernberg)
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning
2024-03-07 13:40 ` FX Coudert
@ 2024-04-03 9:01 ` Iain Sandoe
2024-04-03 9:38 ` Richard Biener
0 siblings, 1 reply; 17+ messages in thread
From: Iain Sandoe @ 2024-04-03 9:01 UTC (permalink / raw)
To: Richard Biener; +Cc: GCC Patches, Gerald Pfeifer
Hi Richard,
> On 7 Mar 2024, at 13:40, FX Coudert <fxcoudert@gmail.com> wrote:
>
>> I think it's an obvious change ...
>
> Thanks, pushed.
>
> Dimitry, I suggest you post the second patch for review.
Given that the two patches here (for https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111632) were considered obvious - and are needed on release branches.
OK for backporting?
(Gerald has volunteered to do the earlier ones, I have already made/tested the gcc-13 case)
thanks
Iain
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning
2024-03-07 9:51 ` Richard Biener
@ 2024-03-07 13:40 ` FX Coudert
2024-04-03 9:01 ` Iain Sandoe
0 siblings, 1 reply; 17+ messages in thread
From: FX Coudert @ 2024-03-07 13:40 UTC (permalink / raw)
To: Richard Biener; +Cc: Iain Sandoe, Dimitry Andric, Sam James, GCC Patches
> I think it's an obvious change ...
Thanks, pushed.
Dimitry, I suggest you post the second patch for review.
FX
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning
2024-03-07 7:11 ` Iain Sandoe
@ 2024-03-07 9:51 ` Richard Biener
2024-03-07 13:40 ` FX Coudert
0 siblings, 1 reply; 17+ messages in thread
From: Richard Biener @ 2024-03-07 9:51 UTC (permalink / raw)
To: Iain Sandoe; +Cc: Dimitry Andric, FX Coudert, Sam James, GCC Patches
On Thu, 7 Mar 2024, Iain Sandoe wrote:
> Hi Dimitry, folks,
>
> > On 6 Mar 2024, at 23:02, Dimitry Andric <dimitry@andric.com> wrote:
> >
> > On 6 Mar 2024, at 15:57, FX Coudert <fxcoudert@gmail.com> wrote:
> >>
> >>> Hmm I recall trying it and finding a problem - was there some different fix applied
> >>> in the end?
> >>
> >> The bug is still open, I don?t think a patch was applied, and I don?t find any email to the list stating what the problem could be.
> >
> > The original patch (https://gcc.gnu.org/bugzilla/attachment.cgi?id=56010) still applies to the master branch.
>
> I have retested this on various Darwin versions and confirm that it fixes the bootstrap fail on x86_64-darwin23 and works OK on other versions (including 32b hosts).
>
> +1 for applying this soon.
I think it's an obvious change ...
Richard.
> ====
>
> the second patch really needs to be posted separately to make review easier (if we were not in stage 4, I?d say it?s ?obvious? anyway).
>
> thanks
> Iain
>
>
> > It turned out there is also a related problem in libcc1plugin.cc and libcp1plugin.cc <http://libcp1plugin.cc/>, which is fixed by https://gcc.gnu.org/bugzilla/attachment.cgi?id=57639 :
> >
> > commit 49222b98ac8e30a4a042ada0ece3d7df93f049d2
> > Author: Dimitry Andric <dimitry@andric.com>
> > Date: 2024-03-06T23:46:27+01:00
> >
> > Fix libcc1plugin and libc1plugin to use INCLUDE_VECTOR before including
> > system.h, instead of directly including <vector>, to avoid running into
> > poisoned identifiers.
> >
> > diff --git a/libcc1/libcc1plugin.cc b/libcc1/libcc1plugin.cc
> > index 72d17c3b81c..e64847466f4 100644
> > --- a/libcc1/libcc1plugin.cc
> > +++ b/libcc1/libcc1plugin.cc
> > @@ -32,6 +32,7 @@
> > #undef PACKAGE_VERSION
> >
> > #define INCLUDE_MEMORY
> > +#define INCLUDE_VECTOR
> > #include "gcc-plugin.h"
> > #include "system.h"
> > #include "coretypes.h"
> > @@ -69,8 +70,6 @@
> > #include "gcc-c-interface.h"
> > #include "context.hh"
> >
> > -#include <vector>
> > -
> > using namespace cc1_plugin;
> >
> >
> > diff --git a/libcc1/libcp1plugin.cc b/libcc1/libcp1plugin.cc
> > index 0eff7c68d29..da68c5d0ac1 100644
> > --- a/libcc1/libcp1plugin.cc
> > +++ b/libcc1/libcp1plugin.cc
> > @@ -33,6 +33,7 @@
> > #undef PACKAGE_VERSION
> >
> > #define INCLUDE_MEMORY
> > +#define INCLUDE_VECTOR
> > #include "gcc-plugin.h"
> > #include "system.h"
> > #include "coretypes.h"
> > @@ -71,8 +72,6 @@
> > #include "rpc.hh"
> > #include "context.hh"
> >
> > -#include <vector>
> > -
> > using namespace cc1_plugin;
> >
> >
> >
>
>
--
Richard Biener <rguenther@suse.de>
SUSE Software Solutions Germany GmbH,
Frankenstrasse 146, 90461 Nuernberg, Germany;
GF: Ivo Totev, Andrew McDonald, Werner Knoblich; (HRB 36809, AG Nuernberg)
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning
2024-03-06 23:02 ` Dimitry Andric
@ 2024-03-07 7:11 ` Iain Sandoe
2024-03-07 9:51 ` Richard Biener
0 siblings, 1 reply; 17+ messages in thread
From: Iain Sandoe @ 2024-03-07 7:11 UTC (permalink / raw)
To: Dimitry Andric; +Cc: FX Coudert, Sam James, Richard Biener, GCC Patches
Hi Dimitry, folks,
> On 6 Mar 2024, at 23:02, Dimitry Andric <dimitry@andric.com> wrote:
>
> On 6 Mar 2024, at 15:57, FX Coudert <fxcoudert@gmail.com> wrote:
>>
>>> Hmm I recall trying it and finding a problem - was there some different fix applied
>>> in the end?
>>
>> The bug is still open, I don’t think a patch was applied, and I don’t find any email to the list stating what the problem could be.
>
> The original patch (https://gcc.gnu.org/bugzilla/attachment.cgi?id=56010) still applies to the master branch.
I have retested this on various Darwin versions and confirm that it fixes the bootstrap fail on x86_64-darwin23 and works OK on other versions (including 32b hosts).
+1 for applying this soon.
====
the second patch really needs to be posted separately to make review easier (if we were not in stage 4, I’d say it’s ‘obvious’ anyway).
thanks
Iain
> It turned out there is also a related problem in libcc1plugin.cc and libcp1plugin.cc <http://libcp1plugin.cc/>, which is fixed by https://gcc.gnu.org/bugzilla/attachment.cgi?id=57639 :
>
> commit 49222b98ac8e30a4a042ada0ece3d7df93f049d2
> Author: Dimitry Andric <dimitry@andric.com>
> Date: 2024-03-06T23:46:27+01:00
>
> Fix libcc1plugin and libc1plugin to use INCLUDE_VECTOR before including
> system.h, instead of directly including <vector>, to avoid running into
> poisoned identifiers.
>
> diff --git a/libcc1/libcc1plugin.cc b/libcc1/libcc1plugin.cc
> index 72d17c3b81c..e64847466f4 100644
> --- a/libcc1/libcc1plugin.cc
> +++ b/libcc1/libcc1plugin.cc
> @@ -32,6 +32,7 @@
> #undef PACKAGE_VERSION
>
> #define INCLUDE_MEMORY
> +#define INCLUDE_VECTOR
> #include "gcc-plugin.h"
> #include "system.h"
> #include "coretypes.h"
> @@ -69,8 +70,6 @@
> #include "gcc-c-interface.h"
> #include "context.hh"
>
> -#include <vector>
> -
> using namespace cc1_plugin;
>
>
> diff --git a/libcc1/libcp1plugin.cc b/libcc1/libcp1plugin.cc
> index 0eff7c68d29..da68c5d0ac1 100644
> --- a/libcc1/libcp1plugin.cc
> +++ b/libcc1/libcp1plugin.cc
> @@ -33,6 +33,7 @@
> #undef PACKAGE_VERSION
>
> #define INCLUDE_MEMORY
> +#define INCLUDE_VECTOR
> #include "gcc-plugin.h"
> #include "system.h"
> #include "coretypes.h"
> @@ -71,8 +72,6 @@
> #include "rpc.hh"
> #include "context.hh"
>
> -#include <vector>
> -
> using namespace cc1_plugin;
>
>
>
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning
2024-03-06 14:57 ` FX Coudert
@ 2024-03-06 23:02 ` Dimitry Andric
2024-03-07 7:11 ` Iain Sandoe
0 siblings, 1 reply; 17+ messages in thread
From: Dimitry Andric @ 2024-03-06 23:02 UTC (permalink / raw)
To: FX Coudert; +Cc: Iain Sandoe, Sam James, GCC Patches
On 6 Mar 2024, at 15:57, FX Coudert <fxcoudert@gmail.com> wrote:
>
>> Hmm I recall trying it and finding a problem - was there some different fix applied
>> in the end?
>
> The bug is still open, I don’t think a patch was applied, and I don’t find any email to the list stating what the problem could be.
The original patch (https://gcc.gnu.org/bugzilla/attachment.cgi?id=56010) still applies to the master branch. It turned out there is also a related problem in libcc1plugin.cc and libcp1plugin.cc <http://libcp1plugin.cc/>, which is fixed by https://gcc.gnu.org/bugzilla/attachment.cgi?id=57639 :
commit 49222b98ac8e30a4a042ada0ece3d7df93f049d2
Author: Dimitry Andric <dimitry@andric.com>
Date: 2024-03-06T23:46:27+01:00
Fix libcc1plugin and libc1plugin to use INCLUDE_VECTOR before including
system.h, instead of directly including <vector>, to avoid running into
poisoned identifiers.
diff --git a/libcc1/libcc1plugin.cc b/libcc1/libcc1plugin.cc
index 72d17c3b81c..e64847466f4 100644
--- a/libcc1/libcc1plugin.cc
+++ b/libcc1/libcc1plugin.cc
@@ -32,6 +32,7 @@
#undef PACKAGE_VERSION
#define INCLUDE_MEMORY
+#define INCLUDE_VECTOR
#include "gcc-plugin.h"
#include "system.h"
#include "coretypes.h"
@@ -69,8 +70,6 @@
#include "gcc-c-interface.h"
#include "context.hh"
-#include <vector>
-
using namespace cc1_plugin;
\f
diff --git a/libcc1/libcp1plugin.cc b/libcc1/libcp1plugin.cc
index 0eff7c68d29..da68c5d0ac1 100644
--- a/libcc1/libcp1plugin.cc
+++ b/libcc1/libcp1plugin.cc
@@ -33,6 +33,7 @@
#undef PACKAGE_VERSION
#define INCLUDE_MEMORY
+#define INCLUDE_VECTOR
#include "gcc-plugin.h"
#include "system.h"
#include "coretypes.h"
@@ -71,8 +72,6 @@
#include "rpc.hh"
#include "context.hh"
-#include <vector>
-
using namespace cc1_plugin;
\f
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning
2024-03-06 14:35 ` Iain Sandoe
@ 2024-03-06 14:57 ` FX Coudert
2024-03-06 23:02 ` Dimitry Andric
0 siblings, 1 reply; 17+ messages in thread
From: FX Coudert @ 2024-03-06 14:57 UTC (permalink / raw)
To: Iain Sandoe; +Cc: Sam James, GCC Patches, dimitry
> Hmm I recall trying it and finding a problem - was there some different fix applied
> in the end?
The bug is still open, I don’t think a patch was applied, and I don’t find any email to the list stating what the problem could be.
FX
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning
2024-03-06 13:54 ` Sam James
@ 2024-03-06 14:35 ` Iain Sandoe
2024-03-06 14:57 ` FX Coudert
0 siblings, 1 reply; 17+ messages in thread
From: Iain Sandoe @ 2024-03-06 14:35 UTC (permalink / raw)
To: Sam James; +Cc: FX Coudert, GCC Patches, dimitry
> On 6 Mar 2024, at 13:54, Sam James <sam@gentoo.org> wrote:
>
> FX Coudert <fxcoudert@gmail.com> writes:
>
>> I would like to patch this patch from September 2023:
>> https://gcc.gnu.org/pipermail/gcc-patches/2023-September/631611.html
>>
>> This bug is now hitting macOS in the latest version of Xcode (it was originally seen on freebsd).
>> I confirm that the patch is restoring bootstrap on x86_64-apple-darwin23
>
> Iain hit an issue with it and I never got a chance to look into how to
> fix it.
Hmm I recall trying it and finding a problem - was there some different fix applied
in the end?
Iain
>
>>
>> OK to push?
>> FX
>>
>>
>>
>>
>>
>>> Ref: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111632
>>>
>>> When building gcc's C++ sources against recent libc++, the poisoning of
>>> the ctype macros due to including safe-ctype.h before including C++
>>> standard headers such as <list>, <map>, etc, causes many compilation
>>> errors, similar to:
>>>
>>> In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
>>> In file included from /home/dim/src/gcc/master/gcc/system.h:233:
>>> In file included from /usr/include/c++/v1/vector:321:
>>> In file included from
>>> /usr/include/c++/v1/__format/formatter_bool.h:20:
>>> In file included from
>>> /usr/include/c++/v1/__format/formatter_integral.h:32:
>>> In file included from /usr/include/c++/v1/locale:202:
>>> /usr/include/c++/v1/__locale:546:5: error: '__abi_tag__' attribute
>>> only applies to structs, variables, functions, and namespaces
>>> 546 | _LIBCPP_INLINE_VISIBILITY
>>> | ^
>>> /usr/include/c++/v1/__config:813:37: note: expanded from macro
>>> '_LIBCPP_INLINE_VISIBILITY'
>>> 813 | # define _LIBCPP_INLINE_VISIBILITY _LIBCPP_HIDE_FROM_ABI
>>> | ^
>>> /usr/include/c++/v1/__config:792:26: note: expanded from macro
>>> '_LIBCPP_HIDE_FROM_ABI'
>>> 792 |
>>> __attribute__((__abi_tag__(_LIBCPP_TOSTRING(
>>> _LIBCPP_VERSIONED_IDENTIFIER))))
>>> | ^
>>> In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
>>> In file included from /home/dim/src/gcc/master/gcc/system.h:233:
>>> In file included from /usr/include/c++/v1/vector:321:
>>> In file included from
>>> /usr/include/c++/v1/__format/formatter_bool.h:20:
>>> In file included from
>>> /usr/include/c++/v1/__format/formatter_integral.h:32:
>>> In file included from /usr/include/c++/v1/locale:202:
>>> /usr/include/c++/v1/__locale:547:37: error: expected ';' at end of
>>> declaration list
>>> 547 | char_type toupper(char_type __c) const
>>> | ^
>>> /usr/include/c++/v1/__locale:553:48: error: too many arguments
>>> provided to function-like macro invocation
>>> 553 | const char_type* toupper(char_type* __low, const
>>> char_type* __high) const
>>> | ^
>>> /home/dim/src/gcc/master/gcc/../include/safe-ctype.h:146:9: note:
>>> macro 'toupper' defined here
>>> 146 | #define toupper(c) do_not_use_toupper_with_safe_ctype
>>> | ^
>>>
>>> This is because libc++ uses different transitive includes than
>>> libstdc++, and some of those transitive includes pull in various ctype
>>> declarations (typically via <locale>).
>>>
>>> There was already a special case for including <string> before
>>> safe-ctype.h, so move the rest of the C++ standard header includes to
>>> the same location, to fix the problem.
>>>
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning
2024-03-06 13:42 FX Coudert
@ 2024-03-06 13:54 ` Sam James
2024-03-06 14:35 ` Iain Sandoe
0 siblings, 1 reply; 17+ messages in thread
From: Sam James @ 2024-03-06 13:54 UTC (permalink / raw)
To: FX Coudert; +Cc: gcc-patches, Iain Sandoe, dimitry
[-- Attachment #1: Type: text/plain, Size: 3109 bytes --]
FX Coudert <fxcoudert@gmail.com> writes:
> I would like to patch this patch from September 2023:
> https://gcc.gnu.org/pipermail/gcc-patches/2023-September/631611.html
>
> This bug is now hitting macOS in the latest version of Xcode (it was originally seen on freebsd).
> I confirm that the patch is restoring bootstrap on x86_64-apple-darwin23
Iain hit an issue with it and I never got a chance to look into how to
fix it.
>
> OK to push?
> FX
>
>
>
>
>
>> Ref: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111632
>>
>> When building gcc's C++ sources against recent libc++, the poisoning of
>> the ctype macros due to including safe-ctype.h before including C++
>> standard headers such as <list>, <map>, etc, causes many compilation
>> errors, similar to:
>>
>> In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
>> In file included from /home/dim/src/gcc/master/gcc/system.h:233:
>> In file included from /usr/include/c++/v1/vector:321:
>> In file included from
>> /usr/include/c++/v1/__format/formatter_bool.h:20:
>> In file included from
>> /usr/include/c++/v1/__format/formatter_integral.h:32:
>> In file included from /usr/include/c++/v1/locale:202:
>> /usr/include/c++/v1/__locale:546:5: error: '__abi_tag__' attribute
>> only applies to structs, variables, functions, and namespaces
>> 546 | _LIBCPP_INLINE_VISIBILITY
>> | ^
>> /usr/include/c++/v1/__config:813:37: note: expanded from macro
>> '_LIBCPP_INLINE_VISIBILITY'
>> 813 | # define _LIBCPP_INLINE_VISIBILITY _LIBCPP_HIDE_FROM_ABI
>> | ^
>> /usr/include/c++/v1/__config:792:26: note: expanded from macro
>> '_LIBCPP_HIDE_FROM_ABI'
>> 792 |
>> __attribute__((__abi_tag__(_LIBCPP_TOSTRING(
>> _LIBCPP_VERSIONED_IDENTIFIER))))
>> | ^
>> In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
>> In file included from /home/dim/src/gcc/master/gcc/system.h:233:
>> In file included from /usr/include/c++/v1/vector:321:
>> In file included from
>> /usr/include/c++/v1/__format/formatter_bool.h:20:
>> In file included from
>> /usr/include/c++/v1/__format/formatter_integral.h:32:
>> In file included from /usr/include/c++/v1/locale:202:
>> /usr/include/c++/v1/__locale:547:37: error: expected ';' at end of
>> declaration list
>> 547 | char_type toupper(char_type __c) const
>> | ^
>> /usr/include/c++/v1/__locale:553:48: error: too many arguments
>> provided to function-like macro invocation
>> 553 | const char_type* toupper(char_type* __low, const
>> char_type* __high) const
>> | ^
>> /home/dim/src/gcc/master/gcc/../include/safe-ctype.h:146:9: note:
>> macro 'toupper' defined here
>> 146 | #define toupper(c) do_not_use_toupper_with_safe_ctype
>> | ^
>>
>> This is because libc++ uses different transitive includes than
>> libstdc++, and some of those transitive includes pull in various ctype
>> declarations (typically via <locale>).
>>
>> There was already a special case for including <string> before
>> safe-ctype.h, so move the rest of the C++ standard header includes to
>> the same location, to fix the problem.
>>
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 377 bytes --]
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning
@ 2024-03-06 13:42 FX Coudert
2024-03-06 13:54 ` Sam James
0 siblings, 1 reply; 17+ messages in thread
From: FX Coudert @ 2024-03-06 13:42 UTC (permalink / raw)
To: gcc-patches; +Cc: Iain Sandoe, dimitry
I would like to patch this patch from September 2023:
https://gcc.gnu.org/pipermail/gcc-patches/2023-September/631611.html
This bug is now hitting macOS in the latest version of Xcode (it was originally seen on freebsd).
I confirm that the patch is restoring bootstrap on x86_64-apple-darwin23
OK to push?
FX
> Ref: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111632
>
> When building gcc's C++ sources against recent libc++, the poisoning of
> the ctype macros due to including safe-ctype.h before including C++
> standard headers such as <list>, <map>, etc, causes many compilation
> errors, similar to:
>
> In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
> In file included from /home/dim/src/gcc/master/gcc/system.h:233:
> In file included from /usr/include/c++/v1/vector:321:
> In file included from
> /usr/include/c++/v1/__format/formatter_bool.h:20:
> In file included from
> /usr/include/c++/v1/__format/formatter_integral.h:32:
> In file included from /usr/include/c++/v1/locale:202:
> /usr/include/c++/v1/__locale:546:5: error: '__abi_tag__' attribute
> only applies to structs, variables, functions, and namespaces
> 546 | _LIBCPP_INLINE_VISIBILITY
> | ^
> /usr/include/c++/v1/__config:813:37: note: expanded from macro
> '_LIBCPP_INLINE_VISIBILITY'
> 813 | # define _LIBCPP_INLINE_VISIBILITY _LIBCPP_HIDE_FROM_ABI
> | ^
> /usr/include/c++/v1/__config:792:26: note: expanded from macro
> '_LIBCPP_HIDE_FROM_ABI'
> 792 |
> __attribute__((__abi_tag__(_LIBCPP_TOSTRING(
> _LIBCPP_VERSIONED_IDENTIFIER))))
> | ^
> In file included from /home/dim/src/gcc/master/gcc/gensupport.cc:23:
> In file included from /home/dim/src/gcc/master/gcc/system.h:233:
> In file included from /usr/include/c++/v1/vector:321:
> In file included from
> /usr/include/c++/v1/__format/formatter_bool.h:20:
> In file included from
> /usr/include/c++/v1/__format/formatter_integral.h:32:
> In file included from /usr/include/c++/v1/locale:202:
> /usr/include/c++/v1/__locale:547:37: error: expected ';' at end of
> declaration list
> 547 | char_type toupper(char_type __c) const
> | ^
> /usr/include/c++/v1/__locale:553:48: error: too many arguments
> provided to function-like macro invocation
> 553 | const char_type* toupper(char_type* __low, const
> char_type* __high) const
> | ^
> /home/dim/src/gcc/master/gcc/../include/safe-ctype.h:146:9: note:
> macro 'toupper' defined here
> 146 | #define toupper(c) do_not_use_toupper_with_safe_ctype
> | ^
>
> This is because libc++ uses different transitive includes than
> libstdc++, and some of those transitive includes pull in various ctype
> declarations (typically via <locale>).
>
> There was already a special case for including <string> before
> safe-ctype.h, so move the rest of the C++ standard header includes to
> the same location, to fix the problem.
>
^ permalink raw reply [flat|nested] 17+ messages in thread
end of thread, other threads:[~2024-04-03 9:38 UTC | newest]
Thread overview: 17+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-09-28 16:37 [PATCH] Include safe-ctype.h after C++ standard headers, to avoid over-poisoning Dimitry Andric
2023-10-26 17:00 ` [PING][PATCH] " Dimitry Andric
2023-10-29 13:08 ` Sam James
2024-01-20 9:17 ` Sam James
2024-01-30 15:07 ` Sam James
2024-01-30 20:27 ` Sam James
2024-01-30 16:19 ` [PATCH] " Jonathan Wakely
2024-03-06 13:42 FX Coudert
2024-03-06 13:54 ` Sam James
2024-03-06 14:35 ` Iain Sandoe
2024-03-06 14:57 ` FX Coudert
2024-03-06 23:02 ` Dimitry Andric
2024-03-07 7:11 ` Iain Sandoe
2024-03-07 9:51 ` Richard Biener
2024-03-07 13:40 ` FX Coudert
2024-04-03 9:01 ` Iain Sandoe
2024-04-03 9:38 ` Richard Biener
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).