cvs-cvs
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Cvs-cvs] Changes to ccvs/windows-NT/stamp-chi


From: Derek Robert Price
Subject: [Cvs-cvs] Changes to ccvs/windows-NT/stamp-chi
Date: Thu, 22 Sep 2005 23:42:43 -0400

Index: ccvs/windows-NT/stamp-chi
diff -u ccvs/windows-NT/stamp-chi:1.81 ccvs/windows-NT/stamp-chi:1.82
--- ccvs/windows-NT/stamp-chi:1.81      Tue Sep 20 02:49:55 2005
+++ ccvs/windows-NT/stamp-chi   Fri Sep 23 03:42:42 2005
@@ -1294,6 +1294,12 @@
    it because _open_osfhandle doesn't work.  */
 #define NO_SOCKET_TO_FD 1
 
+/* Is this true on NT?  Seems like I remember reports that NT 3.51 has
+   problems with 200K writes (of course, the issue of large writes is
+   moot since the use of buffer.c ensures that writes will only be as big
+   as the buffers).  */
+#define SEND_NEVER_PARTIAL 1
+
 /* Stop server macro */
 #define SHUTDOWN_SERVER wnt_shutdown_server
 
@@ -1324,11 +1330,21 @@
 /* Macro name tells the story */
 #define SYSTEM_INITIALIZE(pargc,pargv) woe32_init_winsock()
 
-/* Is this true on NT?  Seems like I remember reports that NT 3.51 has
-   problems with 200K writes (of course, the issue of large writes is
-   moot since the use of buffer.c ensures that writes will only be as big
-   as the buffers).  */
-#define SEND_NEVER_PARTIAL 1
+/*
+ * According to GNU conventions, we should avoid referencing any macro
+ * containing "WIN" as a reference to Microsoft Windows, as we would like to
+ * avoid any implication that we consider Microsoft Windows any sort of "win".
+ *
+ * FIXME: As of 2003-06-09, folks on the GNULIB project were discussing
+ * defining a configure macro to define WOE32 appropriately.  If they ever do
+ * write such a beast, we should use it, though in most cases it would be
+ * preferable to avoid referencing any OS or compiler anyhow, per Autoconf
+ * convention, and reference only tested features of the system.
+ *
+ * re FIXME: This definition would still be necessary since systems which run
+ * configure do not use this config.h and vice-versa.
+ */
+#define WOE32 1
 
 /* Define POSIX name to Microsoft name */
 #define dup _dup
@@ -1336,6 +1352,9 @@
 /* getpagesize is missing on Windows, 4096 does the right thing. */
 #define getpagesize() 4096
 
+/* Define to a substitute for the `lstat' function. */
+#define lstat stat
+
 /* Define POSIX name to Microsoft name */
 #define popen _popen
 
@@ -1348,6 +1367,9 @@
    now we put it here.  */
 #define same_file(s,t) (-1)
 
+/* Define to a substitute for the stat function. */
+#define stat wnt_stat
+
 /* ====================== Typedef Declarations Follow ===================== */
 
 /* ====================== Function Prototypes Follow ====================== */
@@ -1375,25 +1397,3 @@
 
 /* #define stat wnt_stat and #define lstat wnt_stat */
 int wnt_stat (const char *file, struct wnt_stat *sb);
-
-/* Define to a substitute for the `lstat' function. */
-#define lstat stat
-
-/* Define to a substitute for the stat function. */
-#define stat wnt_stat
-
-/*
- * According to GNU conventions, we should avoid referencing any macro
- * containing "WIN" as a reference to Microsoft Windows, as we would like to
- * avoid any implication that we consider Microsoft Windows any sort of "win".
- *
- * FIXME: As of 2003-06-09, folks on the GNULIB project were discussing
- * defining a configure macro to define WOE32 appropriately.  If they ever do
- * write such a beast, we should use it, though in most cases it would be
- * preferable to avoid referencing any OS or compiler anyhow, per Autoconf
- * convention, and reference only tested features of the system.
- *
- * re FIXME: This definition would still be necessary since systems which run
- * configure do not use this config.h and vice-versa.
- */
-#define WOE32 1




reply via email to

[Prev in Thread] Current Thread [Next in Thread]