gwakeup: Clarify buffer sizing in g_wakeup_signal()

The code in g_wakeup_signal() is currently correct: it writes a 64-bit
counter increment value if the FD is an eventfd, and writes an arbitrary
8-bit value if using a normal pipe.

However, the reasoning behind these buffer sizes is not clear, and the
mismatch between the allocated buffer size and the length passed to
write() in the pipe case could be mistaken for a bug.

Coverity issue: #1159490

https://bugzilla.gnome.org/show_bug.cgi?id=732002

diff --git a/glib/gwakeup.c b/glib/gwakeup.c
index 7125a4b..c0f1ba0 100644 (file)
--- a/glib/gwakeup.c
+++ b/glib/gwakeup.c
@@ -227,19 +227,26 @@ g_wakeup_acknowledge (GWakeup *wakeup)
 void
 g_wakeup_signal (GWakeup *wakeup)
 {
-  guint64 one = 1;
   int res;
 
   if (wakeup->fds[1] == -1)
     {
+      guint64 one = 1;
+
+      /* eventfd() case. It requires a 64-bit counter increment value to be
+       * written. */
       do
         res = write (wakeup->fds[0], &one, sizeof one);
       while (G_UNLIKELY (res == -1 && errno == EINTR));
     }
   else
     {
+      guint8 one = 1;
+
+      /* Non-eventfd() case. Only a single byte needs to be written, and it can
+       * have an arbitrary value. */
       do
-        res = write (wakeup->fds[1], &one, 1);
+        res = write (wakeup->fds[1], &one, sizeof one);
       while (G_UNLIKELY (res == -1 && errno == EINTR));
     }
 }