pan-users
[Top][All Lists]
Advanced

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

Re: [Pan-users] Re: make errors


From: David Shochat
Subject: Re: [Pan-users] Re: make errors
Date: Thu, 19 Jun 2008 06:09:06 -0400


On Jun 18, 2008, at 11:34 PM, walt wrote:

David Shochat wrote:

On Jun 18, 2008, at 8:27 PM, walt wrote:


How about:

char *p;
...
int *q = (int *) p;
q++;
p = (char *) q;

That would have the same net effect and is accepted by gcc 4.2.3, which
does not accept the original construct.

Well, here is my first attempt at implementing your suggestion, but it
produces exactly the same compiler error as the original code. Perhaps
in the morning when I'm more awake I'll do better.

(snip)


-                               *((unsigned long int *) tmp)++ =
+                               *qtmp = *((unsigned long int *) b1)++;
+                               qtmp++;
+                               tmp = (char *)qtmp;
+/*                             *((unsigned long int *) tmp)++ =
                                       *((unsigned long int *) b1)++;
+*/
                       }
                       else
                       {
                               --n2;
-                               *((unsigned long int *) tmp)++ =
+                               *qtmp = *((unsigned long int *) b2)++;
+                               qtmp++;
+                               tmp = (char *)qtmp;
+/*                             *((unsigned long int *) tmp)++ =
                                       *((unsigned long int *) b2)++;
+*/

I think that should be:
qtmp = (unsigned long int *) tmp;
qb2 = (unsigned long int *) b2;
*qtmp++ = *qb2++;
tmp = (char *)qtmp;
b2 = (char *) qb2;
Not sure about the last line: I'm assuming that b2 is a char*.
Is this code from old pan or new pan? Maybe I'll try doing the actual patch.
-- David






reply via email to

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