Mailing List Archive

Portability patch
Index: gcrypt.h
===================================================================
RCS file: /cvs/gnupg/libgcrypt/src/gcrypt.h,v
retrieving revision 1.101
diff -u -p -u -w -r1.101 gcrypt.h
--- gcrypt.h 31 Jul 2003 06:36:55 -0000 1.101
+++ gcrypt.h 8 Aug 2003 12:27:01 -0000
@@ -91,7 +91,7 @@ typedef gpg_error_t gcry_error_t;
typedef gpg_err_code_t gcry_err_code_t;
typedef gpg_err_source_t gcry_err_source_t;

-static __inline__ gcry_error_t
+static GPG_ERR_INLINE gcry_error_t
gcry_err_make (gcry_err_source_t source, gcry_err_code_t code)
{
return gpg_err_make (source, code);
@@ -103,20 +103,20 @@ gcry_err_make (gcry_err_source_t source,
#define GCRY_ERR_SOURCE_DEFAULT GPG_ERR_SOURCE_USER_1
#endif

-static __inline__ gcry_error_t
+static GPG_ERR_INLINE gcry_error_t
gcry_error (gcry_err_code_t code)
{
return gcry_err_make (GCRY_ERR_SOURCE_DEFAULT, code);
}

-static __inline__ gcry_err_code_t
+static GPG_ERR_INLINE gcry_err_code_t
gcry_err_code (gcry_error_t err)
{
return gpg_err_code (err);
}


-static __inline__ gcry_err_source_t
+static GPG_ERR_INLINE gcry_err_source_t
gcry_err_source (gcry_error_t err)
{
return gpg_err_source (err);
Re: Portability patch [ In reply to ]
Hi,

this patch can not go in. GPG_ERR_INLINE is not part of the public
interface. Sorry for the confusion, I will change this to make it more
clear by prefixing the symbol with underscores.

libgcrypt must use its own inline magic, or use macros. I will discuss this
with Moritz.

Thanks,
Marcus

--
`Rhubarb is no Egyptian god.' GNU http://www.gnu.org marcus@gnu.org
Marcus Brinkmann The Hurd http://www.gnu.org/software/hurd/
Marcus.Brinkmann@ruhr-uni-bochum.de
http://www.marcus-brinkmann.de/
Re: Portability patch [ In reply to ]
Marcus Brinkmann <Marcus.Brinkmann@ruhr-uni-bochum.de> writes:

> Hi,

Hello.

> this patch can not go in. GPG_ERR_INLINE is not part of the public
> interface. Sorry for the confusion, I will change this to make it more
> clear by prefixing the symbol with underscores.

Ah, thanks.

> libgcrypt must use its own inline magic, or use macros. I will discuss this
> with Moritz.

Whatever solution compile makes me happy.