<?xml version="1.0"?>
<?xml-stylesheet type="text/xsl" href="/rss.xsl.xml"?>
<rss version="2.0" xmlns:dc="http://purl.org/dc/elements/1.1/">
<channel>
    <title>Changes in Makefile</title>
    <link>http://nxr.netbsd.org/rss/src/usr.bin/ftp/Makefile</link>
    <description></description>
    <language>en</language>
    <copyright>Copyright 2005</copyright>
    <generator>Java</generator>
    
<item>
    <title>ftp: fix some -Wformat-overflow issues<br/><br/>Remove -Wno-format-overflow and fix the issues.<br/>Detected by gcc 11 compiling tnftp.</title>
    <description>/src/usr.bin/ftp/Makefile - 1.45</description>
    <pubDate>Sun Feb 08 09:00:54 UTC 2026</pubDate>
    <dc:creator>lukem</dc:creator>
</item>

<item>
    <title>openssl: Stop dragging libpthread into every libcrypto user.<br/><br/>With the new _NETBSD_PTHREAD_CREATE_WEAK, libcrypto can have a call<br/>to pthread_create without linking against libpthread, so it is not<br/>necessary to link all libcrypto users against libpthread too.<br/><br/>Applications that link against libcrypto but not libpthread can't use<br/>threaded OpenSSL functionality, of course -- OSSL_set_max_threads<br/>will always fail in such applications.<br/><br/>This reverts all the makefile churn that I found for unnecessary<br/>libpthread linkage since the openssl 3.5 import.<br/><br/>PR lib/59685: libcrypto should not depend on libpthread</title>
    <description>/src/usr.bin/ftp/Makefile - 1.44</description>
    <pubDate>Sun Oct 19 01:00:00 UTC 2025</pubDate>
    <dc:creator>riastradh</dc:creator>
</item>

<item>
    <title>m68* triggers a bad-maybe clobbered warning, avoid it.<br/><br/>ftp.c:779:23: error: variable 'rc' might be clobbered by 'longjmp' or 'vfork' [-Werror=clobbered]<br/>  779 |                 off_t rc;<br/>      |                       ^~</title>
    <description>/src/usr.bin/ftp/Makefile - 1.43</description>
    <pubDate>Wed Sep 24 01:27:43 UTC 2025</pubDate>
    <dc:creator>mrg</dc:creator>
</item>

<item>
    <title>sun2 and GCC 14 gets a maybe clobbered wrong here, i think.</title>
    <description>/src/usr.bin/ftp/Makefile - 1.42</description>
    <pubDate>Sat Sep 20 18:59:22 UTC 2025</pubDate>
    <dc:creator>mrg</dc:creator>
</item>

<item>
    <title>branches:  1.41.2;<br/>crypto needs threads</title>
    <description>/src/usr.bin/ftp/Makefile - 1.41</description>
    <pubDate>Sat Jul 19 00:31:45 UTC 2025</pubDate>
    <dc:creator>christos</dc:creator>
</item>

<item>
    <title>branches:  1.40.2;<br/>bsd.own.mk: rename GCC_NO_* to CC_WNO_*<br/><br/>Rename compiler-warning-disable variables from<br/>	GCC_NO_warning<br/>to<br/>	CC_WNO_warning<br/>where warning is the full warning name as used by the compiler.<br/><br/>GCC_NO_IMPLICIT_FALLTHRU is CC_WNO_IMPLICIT_FALLTHROUGH<br/><br/>Using the convention CC_compilerflag, where compilerflag<br/>is based on the full compiler flag name.</title>
    <description>/src/usr.bin/ftp/Makefile - 1.40</description>
    <pubDate>Sat Jun 03 09:09:16 UTC 2023</pubDate>
    <dc:creator>lukem</dc:creator>
</item>

<item>
    <title>use fetch_*() for I/O with SMALLPROG / !WITH_SSL builds<br/><br/>Adapt the SMALLPROG / -UWITH_SSL build to also use the fetch_*()<br/>methods from ssl.c, instead of using stdio, as stdio isn't robust<br/>when using interruptable signals.<br/><br/>Disable ssl-specific support in the fetch_*() methods if WITH_SSL<br/>isn't defined, so SMALLPROG still doesn't have ssl support (as expected).<br/><br/>The resulting SMALLPROG binary is slightly larger than before<br/>(e.g., 157KiB vs 153KiB on amd64).<br/><br/>Set version to 20210603 for this fix and the SO_KEEPALIVE fix for PR 56129.<br/><br/>PR install/56219</title>
    <description>/src/usr.bin/ftp/Makefile - 1.39</description>
    <pubDate>Thu Jun 03 10:23:33 UTC 2021</pubDate>
    <dc:creator>lukem</dc:creator>
</item>

<item>
    <title>branches:  1.38.4;<br/>add support for new GCC 9 warnings that may be too much to fix<br/>right now.  new address-of-packed-member and format-overflow<br/>warnings have new GCC_NO_ADDR_OF_PACKED_MEMBER amd<br/>GCC_NO_FORMAT_OVERFLOW variables to remove these warnings.<br/><br/>apply to a bunch of the tree.  mostly, these are real bugs that<br/>should be fixed, but in many cases, only by removing the 'packed'<br/>attribute from some structure that doesn't really need it.  (i<br/>looked at many different ones, and while perhaps 60-80% were<br/>already properly aligned, it wasn't clear to me that the uses<br/>were always coming from sane data vs network alignment, so it<br/>doesn't seem safe to remove packed without careful research for<br/>each affect struct.)  clang already warned (and was not erroring)<br/>for many of these cases, but gcc picked up dozens more.</title>
    <description>/src/usr.bin/ftp/Makefile - 1.38</description>
    <pubDate>Sun Sep 06 07:20:31 UTC 2020</pubDate>
    <dc:creator>mrg</dc:creator>
</item>

<item>
    <title>branches:  1.37.2;  1.37.12;<br/>Remove MKCRYPTO option.<br/><br/>Originally, MKCRYPTO was introduced because the United States<br/>classified cryptography as a munition and restricted its export.  The<br/>export controls were substantially relaxed fifteen years ago, and are<br/>essentially irrelevant for software with published source code.<br/><br/>In the intervening time, nobody bothered to remove the option after<br/>its motivation -- the US export restriction -- was eliminated.  I'm<br/>not aware of any other operating system that has a similar option; I<br/>expect it is mainly out of apathy for churn that we still have it.<br/>Today, cryptography is an essential part of modern computing -- you<br/>can't use the internet responsibly without cryptography.<br/><br/>The position of the TNF board of directors is that TNF makes no<br/>representation that MKCRYPTO=no satisfies any country's cryptography<br/>regulations.<br/><br/>My personal position is that the availability of cryptography is a<br/>basic human right; that any local laws restricting it to a privileged<br/>few are fundamentally immoral; and that it is wrong for developers to<br/>spend effort crippling cryptography to work around such laws.<br/><br/>As proposed on tech-crypto, tech-security, and tech-userlevel to no<br/>objections:<br/><br/>https://mail-index.netbsd.org/tech-crypto/2017/05/06/msg000719.html<br/>https://mail-index.netbsd.org/tech-security/2017/05/06/msg000928.html<br/>https://mail-index.netbsd.org/tech-userlevel/2017/05/06/msg010547.html<br/><br/>P.S.  Reviewing all the uses of MKCRYPTO in src revealed a lot of<br/>*bad* crypto that was conditional on it, e.g. DES in telnet...  That<br/>should probably be removed too, but on the grounds that it is bad,<br/>not on the grounds that it is (nominally) crypto.</title>
    <description>/src/usr.bin/ftp/Makefile - 1.37</description>
    <pubDate>Sun May 21 15:28:43 UTC 2017</pubDate>
    <dc:creator>riastradh</dc:creator>
</item>

<item>
    <title>PR/47276: Add https support</title>
    <description>/src/usr.bin/ftp/Makefile - 1.36</description>
    <pubDate>Fri Dec 21 18:07:36 UTC 2012</pubDate>
    <dc:creator>christos</dc:creator>
</item>

<item>
    <title>branches:  1.35.2;  1.35.4;  1.35.8;<br/>fix gcc-4.5 warnings</title>
    <description>/src/usr.bin/ftp/Makefile - 1.35</description>
    <pubDate>Sun Aug 14 00:58:15 UTC 2011</pubDate>
    <dc:creator>christos</dc:creator>
</item>

<item>
    <title>apply some -Wno-error and/or -fno-strict-aliasing.<br/><br/><br/>all of this should be looked at closer, but some of them are not<br/>very trivial.</title>
    <description>/src/usr.bin/ftp/Makefile - 1.34</description>
    <pubDate>Wed Jun 22 02:49:44 UTC 2011</pubDate>
    <dc:creator>mrg</dc:creator>
</item>

<item>
    <title>Userland now builds and uses terminfo instead of termcap.<br/><br/>OK: core@, jdc@</title>
    <description>/src/usr.bin/ftp/Makefile - 1.33</description>
    <pubDate>Wed Feb 03 15:34:45 UTC 2010</pubDate>
    <dc:creator>roy</dc:creator>
</item>

<item>
    <title>Add new Makefile knob, USE_FORT, which extends USE_SSP by turning on the<br/>FORTIFY_SOURCE feature of libssp, thus checking the size of arguments to<br/>various string and memory copy and set functions (as well as a few system<br/>calls and other miscellany) where known at function entry.  RedHat has<br/>evidently built all "core system packages" with this option for some time.<br/><br/>This option should be used at the top of Makefiles (or Makefile.inc where<br/>this is used for subdirectories) but after any setting of LIB.<br/><br/>This is only useful for userland code, and cannot be used in libc or in<br/>any code which includes the libc internals, because it overrides certain<br/>libc functions with macros.  Some effort has been made to make USE_FORT=yes<br/>work correctly for a full-system build by having the bsd.sys.mk logic<br/>disable the feature where it should not be used (libc, libssp iteself,<br/>the kernel) but no attempt has been made to build the entire system with<br/>USE_FORT and doing so will doubtless expose numerous bugs and misfeatures.<br/><br/>Adjust the system build so that all programs and libraries that are setuid,<br/>directly handle network data (including serial comm data), perform<br/>authentication, or appear likely to have (or have a history of having)<br/>data-driven bugs (e.g. file(1)) are built with USE_FORT=yes by default,<br/>with the exception of libc, which cannot use USE_FORT and thus uses<br/>only USE_SSP by default.  Tested on i386 with no ill results; USE_FORT=no<br/>per-directory or in a system build will disable if desired.</title>
    <description>/src/usr.bin/ftp/Makefile - 1.32</description>
    <pubDate>Mon May 28 00:06:26 UTC 2007</pubDate>
    <dc:creator>tls</dc:creator>
</item>

<item>
    <title>Add NO_USAGE and NO_DEBUG so that we can fit in the floppies again.</title>
    <description>/src/usr.bin/ftp/Makefile - 1.31</description>
    <pubDate>Wed Jun 29 02:31:19 UTC 2005</pubDate>
    <dc:creator>christos</dc:creator>
</item>

<item>
    <title>branches:  1.30.2;<br/>As dsl points out, sysinst uses the progress bar so put it back on SMALLPROG</title>
    <description>/src/usr.bin/ftp/Makefile - 1.30</description>
    <pubDate>Fri Feb 11 15:13:28 UTC 2005</pubDate>
    <dc:creator>jmc</dc:creator>
</item>

<item>
    <title>Add NO_PROGRESS for -DSMALL</title>
    <description>/src/usr.bin/ftp/Makefile - 1.29</description>
    <pubDate>Thu Feb 10 16:00:06 UTC 2005</pubDate>
    <dc:creator>jmc</dc:creator>
</item>

<item>
    <title>Only compile in IPv6 support if ${USE_INET6} != "no"<br/><br/>MKINET6 is for providing IPv6 infrastructure.<br/>USE_INET6 is for compiling IPv6 support into the programs (needs MKINET6).</title>
    <description>/src/usr.bin/ftp/Makefile - 1.28</description>
    <pubDate>Mon Jan 10 02:58:59 UTC 2005</pubDate>
    <dc:creator>lukem</dc:creator>
</item>

<item>
    <title>Save approximately 8K by not including http authentication, extended status<br/>messages and help strings when the appropriate options are set.</title>
    <description>/src/usr.bin/ftp/Makefile - 1.27</description>
    <pubDate>Sun Jun 06 01:37:41 UTC 2004</pubDate>
    <dc:creator>christos</dc:creator>
</item>

<item>
    <title>Abstract out progress bar support into progressbar.[ch], for inclusion<br/>in external programs (conditionalized on -DSTANDALONE_PROGRESS).<br/><br/>The following moved from util.c to progressbar.c:<br/>  alarmtimer(), progressmeter(), psummary(), ptransfer(),<br/>  xsignal(), xsignal_restart()<br/><br/>The following moved from extern.h and ftp_var.h to progressbar.h:<br/>  STALLTIME, verbose, fromatty, progress, quit_time, ttywidth</title>
    <description>/src/usr.bin/ftp/Makefile - 1.26</description>
    <pubDate>Tue Jan 21 16:08:06 UTC 2003</pubDate>
    <dc:creator>jhawk</dc:creator>
</item>
</channel></rss>

