summaryrefslogtreecommitdiff
path: root/doc/Xserver-spec.xml
diff options
context:
space:
mode:
Diffstat (limited to 'doc/Xserver-spec.xml')
-rw-r--r--doc/Xserver-spec.xml10
1 files changed, 1 insertions, 9 deletions
diff --git a/doc/Xserver-spec.xml b/doc/Xserver-spec.xml
index 37fd2b29a..2bf4fb3b7 100644
--- a/doc/Xserver-spec.xml
+++ b/doc/Xserver-spec.xml
@@ -1324,10 +1324,6 @@ This re-enables X request processing for the specified client.
void
FatalError(char *f, ...)
-
- void
- Error(str)
- char *str;
</programlisting></blockquote>
You should write these three routines to provide for diagnostic output
from the dix and ddx layers, although implementing them to produce no
@@ -1335,11 +1331,7 @@ output will not affect the correctness of your server. ErrorF() and
FatalError() take a printf() type of format specification in the first
argument and an implementation-dependent number of arguments following
that. Normally, the formats passed to ErrorF() and FatalError()
-should be terminated with a newline. Error() provides an os interface
-for printing out the string passed as an argument followed by a
-meaningful explanation of the last system error. Normally the string
-does not contain a newline, and it is only called by the ddx layer.
-In the sample implementation, Error() uses the perror() function.
+should be terminated with a newline.
</para>
<para>
After printing the message arguments, FatalError() must be implemented