patch 2.0 patch #11

[Scott Sutherland]

In article <2781 at utastro.UUCP> werner at utastro.UUCP (Werner Uhrig) writes:

>    In article <3183 at cognos.UUCP>, glee at cognos.uucp (Godfrey Lee) writes:
>    > In article <560 at gamma.UUCP> jeb at gamma.UUCP (John Berry) writes:
>    > >I seem to have missed patch # 9  for patch 2.0. someone please mail me ...
> 
>    > I seem to have missed it too, please also send me a copy.
> 
>    the copy of patch-2.0 I had snarfed from somewhere was at patch-level 9
>    already and patches 10 and 11 went in smooth enough ....
> 
>    check your sources - you may have 9 already, too  !!??
> 
Larry Wall provides a nice automated retrieval system, which he
documents how to use in every patch, specifically (from patch #10)

	If you are missing previous patches they can be obtained from me:

	Larry Wall
	lwall at jpl-devvax.jpl.nasa.gov

	If you send a mail message of the following form it will greatly speed
	processing:

	Subject: Command
	@SH mailpatch PATH patch 2.0 LIST
		   ^ note the c

	where PATH is a return path FROM ME TO YOU in Internet notation, and
	LIST is the number of one or more patches you need, separated by spaces,
	commas, and/or hyphens.  Saying 35- says everything from 35 to the end.

	You can also get the patches via anonymous FTP from
	jpl-devvax.jpl.nasa.gov (128.149.8.43).



Just in case you had trouble with it, here's #9:

mailpatch speaking for lwall

System: patch version 2.0
Patch #: 9
Priority: LOW
Subject: Reversed new-style context diff can cause double free().
From: lwall at sdcrdcf.UUCP

Description:
	A new-style context diff that is reversed (-R) and that contains
	a hunk with an assumed old or new half is not freed properly.
	When there is a hunk-half assumed, the lines are duplicated
	internally by copying the pointers of the actual lines from the
	other half.  A pair of variables then tell the freeing routine
	not to free that range of fake lines since they will be freed when
	the other half hunk is freed.

	The problem is that the routine that swaps halves of the hunk didn't
	update the variables to point to where the fake lines were moved to.

Fix:	From rn, say "| patch -d DIR", where DIR is your patch source
	directory.  Outside of rn, say "cd DIR; patch <thisarticle".
	If you don't have the patch program, apply the following by hand,
	or get patch.

	If patch indicates that patchlevel is the wrong version, you may need
	to apply one or more previous patches, or the patch may already
	have been applied.  See the patchlevel.h file to find out what has or
	has not been applied.  In any event, don't continue with the patch.

	If you are missing previous patches they can be obtained from me:

	Larry Wall
	{allegra,burdvax,cbosgd,ihnp4,sdcsvax}!sdcrdcf!lwall

	If you send a mail message of the following form it will greatly speed
	processing:

	Subject: Command
	@SH mailpatch PATH patch 2.0 LIST
		   ^ note the c

	where PATH is a return path FROM ME TO YOU in bang notation, and LIST
	is the number of one or more patches you need, separated by spaces,
	commas, and/or hyphens.  Saying 35- says everything from 35 to the end.

Index: patchlevel.h
Prereq: 8
1c1
< #define PATCHLEVEL 8
---
> #define PATCHLEVEL 9
 
Index: pch.c
Prereq: 2.0.1.5
*** pch.c.old	Thu Jun  4 16:20:28 1987
*** pch.c	Thu Jun  4 16:20:43 1987
***************
*** 1,6 ****
! /* $Header: pch.c,v 2.0.1.5 87/01/30 22:47:42 lwall Exp $
   *
   * $Log:	pch.c,v $
   * Revision 2.0.1.5  87/01/30  22:47:42  lwall
   * Improved responses to mangled patches.
   * 
--- 1,9 ----
! /* $Header: pch.c,v 2.0.1.6 87/06/04 16:18:13 lwall Exp $
   *
   * $Log:	pch.c,v $
+  * Revision 2.0.1.6  87/06/04  16:18:13  lwall
+  * pch_swap didn't swap p_bfake and p_efake.
+  * 
   * Revision 2.0.1.5  87/01/30  22:47:42  lwall
   * Improved responses to mangled patches.
   * 
***************
*** 900,905 ****
--- 903,915 ----
      if (tp_char[i] == '\n') {		/* account for possible blank line */
  	blankline = TRUE;
  	i++;
+     }
+     if (p_efake >= 0) {			/* fix non-freeable ptr range */
+ 	n = p_end - i + 1;
+ 	if (p_efake > i)
+ 	    n = -n;
+ 	p_efake += n;
+ 	p_bfake += n;
      }
      for (n=0; i <= p_end; i++,n++) {
  	p_line[n] = tp_line[i];


	Scott Sutherland			
	scott at bu-ma.bu.edu
	Boston University Math Department