Re: please unblock xorg-server 2:1.1.1-18

[Steve Langasek <vorlon@debian.org>]

Hi Julien,

On Sat, Feb 17, 2007 at 01:46:18PM +0100, Julien Cristau wrote:
> I uploaded xorg-server 2:1.1.1-18 to fix a bug which made xserver-xephyr
> unusable on 64bit architectures (segfault on startup, #405928, which is
> currently severity:important, but could be considered grave, I guess).
> The patch is stolen from FC6.  Please let it transition to etch.

Well, we've discussed previous versions of this patch before; I still don't
understand how this:

+@@ -77,9 +77,10 @@
+
+ static int            HostXWantDamageDebug = 0;
+
+-extern KeySym         EphyrKeymap[];
++/* these two need to be 32-bits. do not say KeySym, that way lies madness */
++extern unsigned int   EphyrKeymap[];
++extern unsigned int   kdKeymap[];
+
+-extern KeySym       kdKeymap[];
+ extern int          kdMinScanCode;
+ extern int          kdMaxScanCode;
+ extern int          kdMinKeyCode;

can possibly be correct without a change to the variable declarations as
well.  Was EphyrKeymap really declared as unsigned int[] elsewhere?

Actually, looking at the source, EpyhrKeymap is never defined anywhere else
-- nor used, so it's not an issue.

But kdKeymap is declared in kdrive/src/kinput.c as:

KeySym              kdKeymap[KD_MAX_LENGTH * KD_MAX_WIDTH];

and referenced several other places as KeySym[], not unsigned int[].

So does this patch really fix anything at all?  And if so... how?

-- 
Steve Langasek                   Give me a lever long enough and a Free OS
Debian Developer                   to set it on, and I can move the world.
vorlon@debian.org                                   http://www.debian.org/