From alex@kapran.bitmcnit.bryansk.su Thu Jul 12 08:06:23 2001 Return-Path: Received: from rbn-gw.bgtu.debryansk.ru (rbn-gw.bgtu.debryansk.ru [62.76.89.2]) by hub.freebsd.org (Postfix) with ESMTP id A846C37B401 for ; Thu, 12 Jul 2001 08:05:57 -0700 (PDT) (envelope-from alex@kapran.bitmcnit.bryansk.su) Received: from server.bitmcnit.bryansk.su (root@bitmcnit.bryansk.su [192.168.121.2]) by rbn-gw.bgtu.debryansk.ru (8.11.2/8.11.2) with ESMTP id f6CF1qD25457; Thu, 12 Jul 2001 19:01:52 +0400 Received: (from uucp@localhost) by server.bitmcnit.bryansk.su (8.9.3/8.9.3) with UUCP id SAA15873; Thu, 12 Jul 2001 18:58:30 +0400 Received: (from alex@localhost) by kapran.bitmcnit.bryansk.su (8.11.4/8.11.4) id f6CExeq01850; Thu, 12 Jul 2001 18:59:40 +0400 (MSD) (envelope-from alex) Message-Id: <200107121459.f6CExeq01850@kapran.bitmcnit.bryansk.su> Date: Thu, 12 Jul 2001 18:59:40 +0400 (MSD) From: Alex Kapranoff To: FreeBSD-gnats-submit@freebsd.org Cc: honig@sprynet.com Subject: DocBook conversion of doc/articles/ipsec-must X-Send-Pr-Version: 3.113 X-GNATS-Notify: >Number: 28916 >Category: docs >Synopsis: DocBook conversion of doc/articles/ipsec-must >Confidential: no >Severity: non-critical >Priority: high >Responsible: freebsd-doc >State: closed >Quarter: >Keywords: >Date-Required: >Class: change-request >Submitter-Id: current-users >Arrival-Date: Thu Jul 12 08:10:01 PDT 2001 >Closed-Date: Sun Jul 29 02:47:45 PDT 2001 >Last-Modified: Sun Jul 29 02:48:00 PDT 2001 >Originator: Alex Kapranoff >Release: FreeBSD 5.0-CURRENT i386 >Organization: Inner Mongolia >Environment: System: FreeBSD kapran.bitmcnit.bryansk.su 5.0-CURRENT FreeBSD 5.0-CURRENT #5: Sun Jun 24 22:56:52 MSD 2001 root@kapran.bitmcnit.bryansk.su:/usr/src/sys/compile/KAPRAN i386 >Description: I added some content (mostly removing obsolete info and providing additional links) along with converting the text to DocBook. A review would be appreciated. >How-To-Repeat: >Fix: # This is a shell archive. Save it in a file, remove anything before # this line, and then unpack it by entering "sh file". Note, it may # create directories; files and directories will be owned by you and # have default permissions. # # This archive contains: # # Makefile # article.sgml # echo x - Makefile sed 's/^X//' >Makefile << 'END-of-Makefile' X# $FreeBSD: doc/en_US.ISO8859-1/articles/ipsec-must/Makefile,v 1.1 2000/06/26 09:10:24 nik Exp $ X XDOC?= article X XFORMATS?= html X XINSTALL_COMPRESSED?=gz XINSTALL_ONLY_COMPRESSED?= X XSRCS= article.sgml X XDOC_PREFIX?= ${.CURDIR}/../../.. X X.include "${DOC_PREFIX}/share/mk/doc.project.mk" END-of-Makefile echo x - article.sgml sed 's/^X//' >article.sgml << 'END-of-article.sgml' X X%man; X]> X X X X
X X Independent Verification of IPSec Functionality in FreeBSD X X X David X Honig X X X
honig@sprynet.com
X
X
X X 3 May 1999 X X X You installed IPsec and it seems to be working. How do you X know? I describe a method for experimentally verifying that IPsec is X working. X X
X X X The Problem X X First, let's assume you have X installed IPsec. How do you know X it's working? Sure, your X connection won't work if its misconfigured, and it will work X when you finally get it right. &man.netstat.1; will list it. X But can you independently confirm it? X X X X The Solution X X First, some crypto-relevant info theory: X X X X encrypted data is uniformly distributed, i.e., has maximal X entropy per symbol; X X X X raw, uncompressed data is typically redundant, i.e., has X sub-maximal entropy. X X X X Suppose you could measure the entropy of the data to- and X from- your network interface. Then you could see the difference X between unencrypted data and encrypted data. This would be true X even if some of the data in encrypted mode was X not encrypted---as the outermost IP header must be, if the X packet is to be routable. X X X MUST X X Ueli Maurer's Universal Statistical Test for Random X Bit Generators( X MUST) quickly measures the entropy X of a sample. It uses a compression-like algorithm. The code is given below for a variant X which measures successive (~quarter megabyte) chunks of a X file. X X X X Tcpdump X X We also need a way to capture the raw network data. A X program called &man.tcpdump.1; lets you do this, if you have X enabled the Berkeley Packet Filter X interface in your kernel's config X file. X X The command X X X tcpdump -c 4000 -s 10000 -w dumpfile.bin X X X will capture 4000 raw packets to X dumpfile.bin. Up to 10,000 bytes per X packet will be captured in this example. X X X X The Experiment X X Here's the experiment. X X X X Open a window to an IPsec host and another window to an X insecure host. X X X X Now start capturing X packets. X X X X In the secure window, run the UNIX X command &man.yes.1;, which will stream the y X character. After a while, stop this. Switch to the X insecure window, and repeat. After a while, stop. X X X X Now run MUST on the X captured packets. You should see something like the X following. The important thing to note is that the secure X connection has 93% (6.7) of the expected value (7.18), and X the normal connection has 29% (2.1) of the X expected value. X X X&prompt.user; tcpdump -c 4000 -s 10000 -w ipsecdemo.bin X&prompt.user; uliscan ipsecdemo.bin X XUliscan 21 Dec 98 XL=8 256 258560 XMeasuring file ipsecdemo.bin XInit done XExpected value for L=8 is 7.1836656 X6.9396 -------------------------------------------------------- X6.6177 ----------------------------------------------------- X6.4100 --------------------------------------------------- X2.1101 ----------------- X2.0838 ----------------- X2.0983 ----------------- X X X X X X X Caveat X X This experiment shows that IPsec does X seem to be distributing the payload data X uniformly, as encryption should. However, X the experiment described here can not X detect many possible flaws in a system (none of which do I have X any evidence for). These include poor key generation or X exchange, data or keys being visible to others, use of weak X algorithms, kernel subversion, etc. Study the source; know the X code. X X X X IPsec---Definition X X Internet Protocol security extensions to IPv4; required for X IPv6. A protocol for negotiating encryption and authentication X at the IP (host-to-host) level. SSL secures only one application X socket; SSH secures only a login; X PGP secures only a specified file or X message. IPsec encrypts everything between two hosts. X X X X Installing IPsec X X Most of the modern versions of FreeBSD have IPsec support X in their base source. So you'll probably will need to include X option in your kernel config and, after X kernel rebuild and reinstall, configure IPsec connections using X &man.setkey.8; command. X X A comprehensive guide on running IPsec on FreeBSD is X provided in FreeBSD X Handbook. X X X X usr/src/sys/i386/conf/KERNELNAME X X This needs to be present in the kernel config file in order X to be able to capture network data with &man.tcpdump.1;. Be sure X to run &man.config.8; after adding this, and rebuild and X reinstall. X X Xdevice bpf X X X X X Maurer's Universal Statistical Test (for block size=8 X bits) X X You can find the same code at X this link. X X X/* X ULISCAN.c ---blocksize of 8 X X 1 Oct 98 X 1 Dec 98 X 21 Dec 98 uliscan.c derived from ueli8.c X X This version has // comments removed for Sun cc X X This implements Ueli M Maurer's "Universal Statistical Test for Random X Bit Generators" using L=8 X X Accepts a filename on the command line; writes its results, with other X info, to stdout. X X Handles input file exhaustion gracefully. X X Ref: J. Cryptology v 5 no 2, 1992 pp 89-105 X also on the web somewhere, which is where I found it. X X -David Honig X honig@sprynet.com X X Usage: X ULISCAN filename X outputs to stdout X*/ X X#define L 8 X#define V (1<<L) X#define Q (10*V) X#define K (100 *Q) X#define MAXSAMP (Q + K) X X#include <stdio.h> X#include <math.h> X Xint main(argc, argv) Xint argc; Xchar **argv; X{ X FILE *fptr; X int i,j; X int b, c; X int table[V]; X double sum = 0.0; X int iproduct = 1; X int run; X X extern double log(/* double x */); X X printf("Uliscan 21 Dec 98 \nL=%d %d %d \n", L, V, MAXSAMP); X X if (argc < 2) { X printf("Usage: Uliscan filename\n"); X exit(-1); X } else { X printf("Measuring file %s\n", argv[1]); X } X X fptr = fopen(argv[1],"rb"); X X if (fptr == NULL) { X printf("Can't find %s\n", argv[1]); X exit(-1); X } X X for (i = 0; i < V; i++) { X table[i] = 0; X } X X for (i = 0; i < Q; i++) { X b = fgetc(fptr); X table[b] = i; X } X X printf("Init done\n"); X X printf("Expected value for L=8 is 7.1836656\n"); X X run = 1; X X while (run) { X sum = 0.0; X iproduct = 1; X X if (run) X for (i = Q; run && i < Q + K; i++) { X j = i; X b = fgetc(fptr); X X if (b < 0) X run = 0; X X if (run) { X if (table[b] > j) X j += K; X X sum += log((double)(j-table[b])); X X table[b] = i; X } X } X X if (!run) X printf("Premature end of file; read %d blocks.\n", i - Q); X X sum = (sum/((double)(i - Q))) / log(2.0); X printf("%4.4f ", sum); X X for (i = 0; i < (int)(sum*8.0 + 0.50); i++) X printf("-"); X X printf("\n"); X X /* refill initial table */ X if (0) { X for (i = 0; i < Q; i++) { X b = fgetc(fptr); X if (b < 0) { X run = 0; X } else { X table[b] = i; X } X } X } X } X} X X X
END-of-article.sgml exit >Release-Note: >Audit-Trail: From: Dima Dorfman To: Alex Kapranoff Cc: FreeBSD-gnats-submit@freebsd.org, honig@sprynet.com Subject: Re: docs/28916: DocBook conversion of doc/articles/ipsec-must Date: Fri, 13 Jul 2001 07:51:40 -0700 Alex Kapranoff writes: > >Description: > I added some content (mostly removing obsolete info and > providing additional links) along with converting the text to > DocBook. A review would be appreciated. Could you please (1) separate the content changes from the DocBook conversion, and (2) send this in the form of a diff against the old version. (1) because content changes must be separate from markup changes (if not for `cvs diff` convenience, for translators), and (2) because sharballs for files already in the repository aren't very convenient to work with. From: Alex Kapranoff To: Dima Dorfman Cc: FreeBSD-gnats-submit@FreeBSD.ORG, honig@sprynet.com Subject: Re: docs/28916: DocBook conversion of doc/articles/ipsec-must Date: Mon, 16 Jul 2001 08:39:22 +0400 * Dima Dorfman [July 13 2001, 18:51]: > Alex Kapranoff writes: > > >Description: > > I added some content (mostly removing obsolete info and > > providing additional links) along with converting the text to > > DocBook. A review would be appreciated. > > Could you please > > (1) separate the content changes from the DocBook conversion, and > (2) send this in the form of a diff against the old version. > > (1) because content changes must be separate from markup changes (if > not for `cvs diff` convenience, for translators), and (2) because > sharballs for files already in the repository aren't very convenient > to work with. Yes, my fault. Thanks for the reminder. Below is the content diff for translators (there's a ja_JA.eucJP translation). And markup diff in this case is neither human-comprehendable nor space-saving. The main reason for me not to submit changes in the diff form was that it won't help anybody. You can easily generate it with `cvs diff', however, and see that 95% of lines are changed and therefore included in the diff (twice). And why do you say that sharballs are less convenient to work with? Seems that it's true only if the diff is readable. --- /usr/doc/en_US.ISO8859-1/articles/ipsec-must/article.sgml Wed Jun 13 18:16:55 2001 +++ article.html Mon Jul 16 08:22:26 2001 @@ -2,12 +2,12 @@ - Independent Verification of IPSec Functionality in FreeBSD + Independent Verification of IPsec Functionality in FreeBSD -

Independent Verification of IPsec Functionality Under FreeBSD 3.0

+

Independent Verification of IPsec Functionality in FreeBSD

You installed IPsec and it seems to be working.  How do you know? I describe a method for experimentally verifying @@ -27,12 +27,12 @@

  1. -

    Encrypted data is uniformly distributed, ie, has maximal entropy - per symbol.

    +

    encrypted data is uniformly distributed, i.e., has maximal entropy + per symbol;

  2. -

    Raw, uncompressed data is typically redundant, i.e., has +

    raw, uncompressed data is typically redundant, i.e., has sub-maximal entropy.

@@ -40,16 +40,17 @@

Suppose you could measure the entropy of the data to- and from- your network interface. Then you could see the difference between unencrypted data and encrypted data. This would be true even if some of the data - in "encrypted mode" was not encrypted ---as the outermost IP header must + in "encrypted mode" was not encrypted---as the outermost IP header must be, if the packet is to be routable.

MUST

Ueli Maurer's "Universal Statistical Test for Random Bit Generators" - ("MUST") quickly measures the entropy of a sample. It uses a - compression-like algorithm. The code is given below for a variant which measures successive - (~quarter megabyte) chunks of a file.

+ (MUST) + quickly measures the entropy of a sample. It uses a + compression-like algorithm. The code is given below for a variant which measures successive + (~quarter megabyte) chunks of a file.

Tcpdump

@@ -103,15 +104,15 @@

This experiment shows that IPsec does seem to be distributing the payload data uniformly, as encryption should. However, the - experiment described here can not detect many possible flaws in a + experiment described here can not detect many possible flaws in a system (none of which do I have any evidence for). These include poor key generation or exchange, data or keys being visible to others, use of weak algorithms, kernel subversion, etc. Study the source; know the code.

-

IPsec -Definition

+

IPsec---Definition

-

Internet Protocol security extensions to IP v 4; required for IP v6. A +

Internet Protocol security extensions to IPv4; required for IPv6. A protocol for negotiating encryption and authentication at the IP (host-to-host) level. SSL secures only one application socket; SSH secures only a login; PGP secures only a specified file or @@ -119,49 +120,34 @@

Installing IPsec

-

Starting from the BSD 3.0 stable release,

+

Most of the modern versions of FreeBSD have IPsec support + in their base source. So you'll probably will need to + include IPSEC option in your kernel config + and, after kernel rebuild and reinstall, configure IPsec + connections using setkey command.

-
    -
  1. -

    install IPsec v0.04, rebuild, reinstall

    -
  2. -
  3. -

    run the administration tools (e.g, ipsecadm) and distribute - keys (or use Photuris for key exchange)

    -
  4. - -
  5. -

    set the routes (rt) up appropriately

    -
  6. -
- -

You may want to make an "ipsec_setup" script containing the - ipsecadm and rt commands which establish your IPsec - tunnel. You can run this script automatically at boottime from your - /etc/rc.local The ipsec_setup script will have to contain at - least two ipsecadm commands and one rt command to be - useful.

+

A comprehensive guide on running IPsec on FreeBSD is + provided in FreeBSD + Handbook.

usr/src/sys/i386/conf/KERNELNAME

-

This needs to be present in the kernel config file in order to run - IPsec. After adding it, run config, etc. and rebuild and +

This needs to be present in the kernel config file in order to be able + to capture network data with tcpdump. + Be sure to run config after adding this, and rebuild and reinstall.

-
# The `bpfilter' pseudo-device enables the Berkeley Packet Filter. Be
 -# aware of the legal and administrative consequences of enabling this
 -# option. Heh heh. The number of devices determines the maximum number of
 -# simultaneous BPF clients programs runnable.
 -pseudo-device bpfilter 2 #Berkeley packet filter
 -
 -# IPSEC
 -options IPSEC
 -options "MD5"
 -pseudo-device enc 1
+
device	bpf
 +

Maurer's Universal Statistical Test (for block size=8 bits)

+ +

You can find the same code at + this link.


To: Dima Dorfman 
Cc: freebsd-doc@FreeBSD.ORG, freebsd-gnats-submit@FreeBSD.ORG
Subject: Re: docs/28916: DocBook conversion of doc/articles/ipsec-must
Date: Sat, 21 Jul 2001 10:42:33 +0400

 * Dima Dorfman  [July 19 2001, 15:57]:
 > Alex Kapranoff  writes:
 > >    And why do you say that sharballs are less convenient to work with?
 > >  Seems that it's true only if the diff is readable.
 > 
 > Well, for one it *would* be readable, at least for the Makefile.  Two,
 
   Okay, now try to read the diff below. ;)
 
 > it'd be nice to know that you wouldn't be overwriting other people's
 > changes (e.g., chern made a spelling fix, and if I just unshar'd your
 > files it'd be overwritten).  And three, with a diff I can save the
 > e-mail to a file then pass it through patch; I can't just pass a shar
 > archive through sh because of the cruft above the archive (okay, okay,
 > I'm lazy :-) ).
 
   Points taken.
 
 > >  --- /usr/doc/en_US.ISO8859-1/articles/ipsec-must/article.sgml	Wed Jun13 18:16:55 2001
 > >  +++ article.html	Mon Jul 16 08:22:26 2001
 > 
 > I've applied this.  Now that that's done, could you send me a diff
 > that converts this mess to DocBook?  Thanks!
 
   Sorry for confusion. With this patch applied the PR could be closed.
 This is all obtained from FreeBSD Russian Documentation Project.
 
 diff -u /usr/doc/en_US.ISO8859-1/articles/ipsec-must/Makefile ./Makefile
 --- /usr/doc/en_US.ISO8859-1/articles/ipsec-must/Makefile	Mon Jun 26 13:10:24 2000
 +++ ./Makefile	Thu Jul 12 18:55:10 2001
 @@ -2,8 +2,6 @@
  
  DOC?= article
  
 -DOCFORMAT= html
 -
  FORMATS?= html
  
  INSTALL_COMPRESSED?=gz
 diff -u /usr/doc/en_US.ISO8859-1/articles/ipsec-must/article.sgml ./article.sgml
 --- /usr/doc/en_US.ISO8859-1/articles/ipsec-must/article.sgml	Fri Jul 20 18:55:28 2001
 +++ ./article.sgml	Sat Jul 21 10:39:56 2001
 @@ -1,92 +1,138 @@
 -
 +
 +
 +
 +%man;
 +]>
 +
 +
+ + Independent Verification of IPSec Functionality in FreeBSD + + + David + Honig + + +
honig@sprynet.com
+
+
+ + 3 May 1999 + + + You installed IPsec and it seems to be working. How do you + know? I describe a method for experimentally verifying that IPsec is + working. + +
+ + + The Problem + + First, let's assume you have + installed IPsec. How do you know + it's working? Sure, your + connection won't work if its misconfigured, and it will work + when you finally get it right. &man.netstat.1; will list it. + But can you independently confirm it? + + + + The Solution + + First, some crypto-relevant info theory: + + + + encrypted data is uniformly distributed, i.e., has maximal + entropy per symbol; + + + + raw, uncompressed data is typically redundant, i.e., has + sub-maximal entropy. + + + + Suppose you could measure the entropy of the data to- and + from- your network interface. Then you could see the difference + between unencrypted data and encrypted data. This would be true + even if some of the data in encrypted mode was + not encrypted---as the outermost IP header must be, if the + packet is to be routable. + + + MUST + + Ueli Maurer's Universal Statistical Test for Random + Bit Generators( + MUST) quickly measures the entropy + of a sample. It uses a compression-like algorithm. The code is given below for a variant + which measures successive (~quarter megabyte) chunks of a + file. + + + + Tcpdump + + We also need a way to capture the raw network data. A + program called &man.tcpdump.1; lets you do this, if you have + enabled the Berkeley Packet Filter + interface in your kernel's config + file. + + The command + + + tcpdump -c 4000 -s 10000 -w dumpfile.bin + + + will capture 4000 raw packets to + dumpfile.bin. Up to 10,000 bytes per + packet will be captured in this example. + + + + The Experiment + + Here's the experiment. + + + + Open a window to an IPsec host and another window to an + insecure host. + + + + Now start capturing + packets. + + + + In the secure window, run the UNIX + command &man.yes.1;, which will stream the y + character. After a while, stop this. Switch to the + insecure window, and repeat. After a while, stop. + + + + Now run MUST on the + captured packets. You should see something like the + following. The important thing to note is that the secure + connection has 93% (6.7) of the expected value (7.18), and + the normal connection has 29% (2.1) of the + expected value. + + +&prompt.user; tcpdump -c 4000 -s 10000 -w ipsecdemo.bin +&prompt.user; uliscan ipsecdemo.bin Uliscan 21 Dec 98 L=8 256 258560 @@ -98,58 +144,75 @@ 6.4100 --------------------------------------------------- 2.1101 ----------------- 2.0838 ----------------- -2.0983 -----------------
- -

Caveat

- -

This experiment shows that IPsec does seem to be distributing the - payload data uniformly, as encryption should. However, the - experiment described here cannotdetect many possible flaws in a - system (none of which do I have any evidence for). These include poor - key generation or exchange, data or keys being visible to others, use of - weak algorithms, kernel subversion, etc. Study the source; know the - code.

- -

IPsec---Definition

- -

Internet Protocol security extensions to IPv4; required for IPv6. A - protocol for negotiating encryption and authentication at the IP - (host-to-host) level. SSL secures only one application socket; SSH - secures only a login; PGP secures only a specified file or - message. IPsec encrypts everything between two hosts.

- -

Installing IPsec

- -

Most of the modern versions of FreeBSD have IPsec support - in their base source. So you'll probably will need to - include IPSEC option in your kernel config - and, after kernel rebuild and reinstall, configure IPsec - connections using setkey command.

- - -

A comprehensive guide on running IPsec on FreeBSD is - provided in FreeBSD - Handbook. - -

usr/src/sys/i386/conf/KERNELNAME

- -

This needs to be present in the kernel config file in order to be able - to capture network data with tcpdump. - Be sure to run config after adding this, and rebuild and - reinstall.

- -
device	bpf
 -
- -

Maurer's Universal Statistical Test (for block - size=8 bits)

- -

You can find the same code at - this link.

+2.0983 ----------------- + + + + + + + Caveat + + This experiment shows that IPsec does + seem to be distributing the payload data + uniformly, as encryption should. However, + the experiment described here cannot + detect many possible flaws in a system (none of which do I have + any evidence for). These include poor key generation or + exchange, data or keys being visible to others, use of weak + algorithms, kernel subversion, etc. Study the source; know the + code. + + + + IPsec---Definition + + Internet Protocol security extensions to IPv4; required for + IPv6. A protocol for negotiating encryption and authentication + at the IP (host-to-host) level. SSL secures only one application + socket; SSH secures only a login; + PGP secures only a specified file or + message. IPsec encrypts everything between two hosts. + + + + Installing IPsec + + Most of the modern versions of FreeBSD have IPsec support + in their base source. So you'll probably will need to include + option in your kernel config and, after + kernel rebuild and reinstall, configure IPsec connections using + &man.setkey.8; command. + + A comprehensive guide on running IPsec on FreeBSD is + provided in FreeBSD + Handbook. + + + + usr/src/sys/i386/conf/KERNELNAME + + This needs to be present in the kernel config file in order + to be able to capture network data with &man.tcpdump.1;. Be sure + to run &man.config.8; after adding this, and rebuild and + reinstall. + + +device bpf + + + + + Maurer's Universal Statistical Test (for block size=8 + bits) + + You can find the same code at + this link. -

 +/*
    ULISCAN.c   ---blocksize of 8
  
    1 Oct 98
 @@ -178,13 +241,13 @@
  */
  
  #define L 8
 -#define V (1<
 -#include 
 +#include <stdio.h>
 +#include <math.h>
  
  int main(argc, argv)
  int argc;
 @@ -202,7 +265,7 @@
  
    printf("Uliscan 21 Dec 98 \nL=%d %d %d \n", L, V, MAXSAMP);
  
 -  if (argc < 2) {
 +  if (argc < 2) {
      printf("Usage: Uliscan filename\n");
      exit(-1);
    } else {
 @@ -216,11 +279,11 @@
      exit(-1);
    }
  
 -  for (i = 0; i < V; i++) {
 +  for (i = 0; i < V; i++) {
      table[i] = 0;
    }
  
 -  for (i = 0; i < Q; i++) {
 +  for (i = 0; i < Q; i++) {
      b = fgetc(fptr);
      table[b] = i;
    }
 @@ -236,15 +299,15 @@
      iproduct = 1;
  
      if (run)
 -      for (i = Q; run && i < Q + K; i++) {
 +      for (i = Q; run && i < Q + K; i++) {
          j = i;
          b = fgetc(fptr);
  
 -        if (b < 0)
 +        if (b < 0)
            run = 0;
  
          if (run) {
 -          if (table[b] > j)
 +          if (table[b] > j)
              j += K;
  
            sum += log((double)(j-table[b]));
 @@ -259,16 +322,16 @@
      sum = (sum/((double)(i - Q))) /  log(2.0);
      printf("%4.4f ", sum);
  
 -    for (i = 0; i < (int)(sum*8.0 + 0.50); i++)
 +    for (i = 0; i < (int)(sum*8.0 + 0.50); i++)
        printf("-");
  
      printf("\n");
  
      /* refill initial table */
      if (0) {
 -      for (i = 0; i < Q; i++) {
 +      for (i = 0; i < Q; i++) {
          b = fgetc(fptr);
 -        if (b < 0) {
 +        if (b < 0) {
            run = 0;
          } else {
            table[b] = i;
 @@ -276,8 +339,7 @@
        }
      }
    }
 -}]]>
- - - - +} + +
+ -- Alex Kapranoff, Voice: +7(0832)791845 We've lived 201 days in the brand new millenium... State-Changed-From-To: open->closed State-Changed-By: dd State-Changed-When: Sun Jul 29 02:47:45 PDT 2001 State-Changed-Why: Committed, thanks! http://www.FreeBSD.org/cgi/query-pr.cgi?pr=28916 >Unformatted: