Notes |
|
(0029626)
|
Brad King
|
2012-06-07 13:03
|
|
For SunPro >= 5.9 why "define inline __inline__" instead of just leaving it undefined? |
|
|
(0029627)
|
Daniel Richard G.
|
2012-06-07 13:40
|
|
Hmm. I thought that was so things would continue to work in strict-ANSI mode (as with g++), but I just tried the 5.10 compiler using -Xc, and it accepts plain old "inline" just as well.
I guess leaving it undefined should be fine, then :-] |
|
|
(0029628)
|
Brad King
|
2012-06-07 13:47
|
|
How about this?
diff --git a/Utilities/cmlibarchive/libarchive/archive_endian.h b/Utilities/cmlibarchive/libarchive/archive_endian.h
index c3c78b3..bbf58fd 100644
--- a/Utilities/cmlibarchive/libarchive/archive_endian.h
+++ b/Utilities/cmlibarchive/libarchive/archive_endian.h
@@ -45,11 +45,14 @@
* - SGI MIPSpro
* - Microsoft Visual C++ 6.0 (supposedly newer versions too)
* - IBM VisualAge 6 (XL v6)
+ * - Sun WorkShop C (SunPro) before 5.9
*/
#if defined(__WATCOMC__) || defined(__sgi) || defined(__hpux) || defined(__BORLANDC__)
#define inline
#elif defined(__IBMC__) && __IBMC__ < 700
#define inline
+#elif defined(__SUNPRO_C) && __SUNPRO_C < 0x590
+#define inline
#elif defined(_MSC_VER) || defined(__osf__)
#define inline __inline
#endif
|
|
|
(0029634)
|
Daniel Richard G.
|
2012-06-07 15:00
|
|
Yep, that gets cmlibarchive built. |
|
|
(0029639)
|
Brad King
|
2012-06-07 15:25
|
|
|
|
(0031446)
|
David Cole
|
2012-11-05 14:33
|
|
Closing resolved issues that have not been updated in more than 4 months. |
|