<?xml version="1.0" encoding="latin1" ?>
<!DOCTYPE chapter SYSTEM "chapter.dtd">
<chapter>
<header>
<copyright>
<year>2004</year><year>2009</year>
<holder>Ericsson AB. All Rights Reserved.</holder>
</copyright>
<legalnotice>
The contents of this file are subject to the Erlang Public License,
Version 1.1, (the "License"); you may not use this file except in
compliance with the License. You should have received a copy of the
Erlang Public License along with this software. If not, it can be
retrieved online at http://www.erlang.org/.
Software distributed under the License is distributed on an "AS IS"
basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See
the License for the specific language governing rights and limitations
under the License.
</legalnotice>
<title>Kernel Release Notes</title>
<prepared></prepared>
<docno></docno>
<date></date>
<rev></rev>
<file>notes.xml</file>
</header>
<p>This document describes the changes made to the Kernel application.</p>
<section><title>Kernel 2.13.4</title>
<section><title>Fixed Bugs and Malfunctions</title>
<list>
<item>
<p>A link in <c>pg2(3)</c> has been fixed. (Thanks to
Christophe Romain.)</p>
<p>
Own Id: OTP-8198</p>
</item>
<item>
<p>
A ticker process could potentially be blocked
indefinitely trying to send a tick to a node not
responding. If this happened, the connection would not be
brought down as it should.</p>
<p>
Own Id: OTP-8218</p>
</item>
<item>
<p>A bug in <c>pg2</c> when members who died did not
leave process groups has been fixed. (Thanks to Matthew
Dempsky.)</p>
<p>
Own Id: OTP-8259</p>
</item>
</list>
</section>
<section><title>Improvements and New Features</title>
<list>
<item>
<p>
The documentation is now built with open source tools
(xsltproc and fop) that exists on most platforms. One
visible change is that the frames are removed.</p>
<p>
Own Id: OTP-8201</p>
</item>
<item>
<p>
The top directory in archive files does not need to have
a <c>-vsn</c> suffix anymore. For example if the archive
file has the name like <c>mnesia-4.4.7.ez</c> the top
directory in the archive can either be named
<c>mnesia</c> or <c>mnesia-4.4.7</c>. If the archive file
has a name like <c>mnesia.ez</c> the top directory in the
archive must be named <c>mnesia</c> as earlier.</p>
<p>
Own Id: OTP-8266</p>
</item>
<item>
<p>The -on_load() directive can be used to run a function
when a module is loaded. It is documented in the section
about code loading in the Reference Manual.</p>
<p>
Own Id: OTP-8295</p>
</item>
</list>
</section>
</section>
<section><title>Kernel 2.13.3</title>
<section><title>Improvements and New Features</title>
<list>
<item>
<p> The DNS resolver client inet_res has been rewritten,
documented and released. See inet_res(3) and Erts User's
Guide: Inet configuration. </p><p> It can formally not be
incompatible with respect to earlier versions since there
was no earlier official version. However it was used
before and some details have changed. </p><p>
Configuration now initializes from /etc/resolv.conf and
/etc/hosts on all unix platforms regardless of which
distribution mode the node is started in. The directory
(/etc) these files are supposed to reside in can be
changed via an environment variable. These configuration
file locations can also be changed in the inet
configuration. The files are monitored for change and
re-read, which makes a few resolver configuration
variables out of application control. The /etc/hosts
entries have now their own cache table that is shadowed
(with lookup method 'file' is used) by the application
configured host entries. This problem (that inet_res
configuration only worked for distribution mode long
names) was among other reported by Matthew O'Gorman many
moons ago. </p><p> The lookup methods are still 'native'
only per default. Resolver configuration is done on all
Unix platforms just to get a usable configuration for
direct calls to inet_res. </p><p> The functions
<c>inet_res:nslookup/3..5</c> and
<c>inet_res:nnslookup/4..4</c> are no longer recommended
to use, instead use <c>inet_res:lookup/3..5</c> and
<c>inet_res:resolve/3..5</c> which provide clearer
argument types and the possibility to override options in
the call. </p><p> Users of previous unsupported versions
of inet_res have included internal header files to get to
the internal record definitions in order to examine DNS
replies. This is still unsupported and there are access
functions in inet_dns to use instead. These are
documented in inet_res(3). </p><p> Bug fix: a compression
reference loop would make DNS message decoding loop
forever. Problem reported by Florian Weimer. </p><p> Bug
fix and patch suggestion by Sergei Golovan: configuring
IPv6 nameservers did not work. His patch (as he warned)
created many UDP sockets; one per nameserver. This has
been fixed in the released version. </p><p> Improvement:
<c>inet_res</c> is now EDNS0 capable. The current
implementation is simple and does not probe and cache
EDNS info for nameservers, which a fully capable
implementation probably should do. EDNS has to be enabled
via resolver configuration, and if a nameserver replies
that it does not support EDNS, <c>inet_res</c> falls back
to a regular DNS query. </p><p> Improvement: now
<c>inet_res</c> automatically falls back to TCP if it
gets a truncated answer from a nameserver. </p><p>
Warning: some of the ancient and exotic record types
handled by <c>inet_res</c> and <c>inet_dns</c> are not
supported by current versions of BIND, so they could not
be tested after the rewrite, with reasonable effort, e.g
MD, MF, NULL, and SPF. The risk for bugs in these
particular records is still low since their code is
mostly shared with other tested record types. </p>
<p>
*** POTENTIAL INCOMPATIBILITY ***</p>
<p>
Own Id: OTP-7955 Aux Id: OTP-7107 OTP-6852 </p>
</item>
<item>
<p>
A TCP socket with option <c>{packet,4}</c> could crash
the emulator if it received a packet header with a very
large size value (>2Gb). The same bug caused
<c>erlang:decode_packet/3</c> to return faulty values.
(Thanks to Georgos Seganos.)</p>
<p>
Own Id: OTP-8102</p>
</item>
<item>
<p>
The file module has now a read_line/1 function similar to
the io:get_line/2, but with byte oriented semantics. The
function file:read_line/1 works for raw files as well,
but for good performance it is recommended to use it
together with the 'read_ahead' option for raw file
access.</p>
<p>
Own Id: OTP-8108</p>
</item>
</list>
</section>
</section>
<section><title>Kernel 2.13.2</title>
<section><title>Fixed Bugs and Malfunctions</title>
<list>
<item>
<p>
A bug when doing io:get_line (among other calls) from a
file opened with encoding other than latin1, causing
false unicode errors to occur, is now corrected.</p>
<p>
Own Id: OTP-7974</p>
</item>
</list>
</section>
<section><title>Improvements and New Features</title>
<list>
<item>
<p>
Added functionality to get higher resolution timestamp
from system. The erlang:now function returns a timestamp
that's not always consistent with the actual operating
system time (due to resilience against large time changes
in the operating system). The function os:timestamp/0 is
added to get a similar timestamp as the one being
returned by erlang:now, but untouched by Erlangs time
correcting and smoothing algorithms. The timestamp
returned by os:timestamp is always consistent with the
operating systems view of time, like the calendar
functions for getting wall clock time, but with higher
resolution. Example of usage can be found in the os
manual page.</p>
<p>
Own Id: OTP-7971</p>
</item>
</list>
</section>
</section>
<section><title>Kernel 2.13.1</title>
<section><title>Fixed Bugs and Malfunctions</title>
<list>
<item>
<p>
Many concurrent calls to <c>os:cmd/1</c> will only block
one scheduler thread at a time, making an smp emulator
more responsive if the OS is slow forking processes.</p>
<p>
Own Id: OTP-7890 Aux Id: seq11219 </p>
</item>
<item>
<p>
Fixed hanging early RPC that did IO operation during node
start.</p>
<p>
Own Id: OTP-7903 Aux Id: seq11224 </p>
</item>
<item>
<p>
The error behavior of gen_tcp and gen_udp has been
corrected. gen_tcp:connect/3,4 and gen_udp:send/4 now
returns {error,eafnosupport} for conflicting destination
address versus socket address family. Other corner cases
for IP address string host names combined with not using
the native (OS) resolver (which is not default) has also
been changed to return {error,nxdomain} instead of
{error,einval}. Those changes just may surprise old
existing code. gen_tcp:listen/2 and gen_udp:open/2 now
fails for conflicting local address versus socket address
family instead of trying to use an erroneous address.
Problem reported by Per Hedeland.</p>
<p>
*** POTENTIAL INCOMPATIBILITY ***</p>
<p>
Own Id: OTP-7929</p>
</item>
</list>
</section>
<section><title>Improvements and New Features</title>
<list>
<item>
<p>
Several glitches and performance issues in the Unicode
and I/O-system implementation of R13A have been
corrected.</p>
<p>
Own Id: OTP-7896 Aux Id: OTP-7648 OTP-7887 </p>
</item>
<item>
<p>
The unsupported DNS resolver client inet_res has now been
improved to handle NAPTR queries.</p>
<p>
Own Id: OTP-7925 Aux Id: seq11231 </p>
</item>
</list>
</section>
</section>
<section><title>Kernel 2.13</title>
<section><title>Fixed Bugs and Malfunctions</title>
<list>
<item>
<p>
The old Erlang DNS resolver inet_res has been corrected
to handle TXT records with more than one character
string. Patch courtesy of Geoff Cant.</p>
<p>
Own Id: OTP-7588</p>
</item>
<item>
<p>When chunk reading a disk log opened in read_only
mode, bad terms could crash the disk log process.</p>
<p>
Own Id: OTP-7641 Aux Id: seq11090 </p>
</item>
<item>
<p>
<c>gen_tcp:send()</c> did sometimes (only observed on
Solaris) return <c>{error,enotconn}</c> instead of the
expected <c>{error,closed}</c> as the peer socket had
been explicitly closed.</p>
<p>
Own Id: OTP-7647</p>
</item>
<item>
<p>
The gen_sctp option sctp_peer_addr_params,
#sctp_paddrparams{address={IP,Port} was erroneously
decoded in the inet driver. This bug has now been
corrected.</p>
<p>
Own Id: OTP-7755</p>
</item>
</list>
</section>
<section><title>Improvements and New Features</title>
<list>
<item>
<p>
Erlang programs can now access STDERR on platforms where
such a file descriptor is available by using the
io_server 'standard_error', i.e.
io:format(standard_error,"~s~n",[ErrorMessage]),</p>
<p>
Own Id: OTP-6688</p>
</item>
<item>
<p>
The format of the string returned by
<c>erlang:system_info(system_version)</c> (as well as the
first message when Erlang is started) has changed. The
string now contains the both the OTP version number as
well as the erts version number.</p>
<p>
Own Id: OTP-7649</p>
</item>
<item>
<p>As of this version, the global name server no longer
supports nodes running Erlang/OTP R10B.</p>
<p>
Own Id: OTP-7661</p>
</item>
<item>
<p>
A <c>{nodedown, Node}</c> message passed by the
<c>net_kernel:monitor_nodes/X</c> functionality is now
guaranteed to be sent after <c>Node</c> has been removed
from the result returned by <c>erlang:nodes/Y</c>.</p>
<p>
Own Id: OTP-7725</p>
</item>
<item>
<p>The deprecated functions <c>erlang:fault/1</c>,
<c>erlang:fault/2</c>, and <c>file:rawopen/2</c> have
been removed.</p>
<p>
*** POTENTIAL INCOMPATIBILITY ***</p>
<p>
Own Id: OTP-7812</p>
</item>
<item>
<p>
Nodes belonging to different independent clusters can now
co-exist on the same host with the help of a new
environment variable setting ERL_EPMD_PORT.</p>
<p>
Own Id: OTP-7826</p>
</item>
<item>
<p>The copyright notices have been updated.</p>
<p>
Own Id: OTP-7851</p>
</item>
</list>
</section>
</section>
<section><title>Kernel 2.12.5.1</title>
<section><title>Fixed Bugs and Malfunctions</title>
<list>
<item>
<p>When chunk reading a disk log opened in read_only
mode, bad terms could crash the disk log process.</p>
<p>
Own Id: OTP-7641 Aux Id: seq11090 </p>
</item>
<item>
<p>
Calling <c>gen_tcp:send()</c> from several processes on
socket with option <c>send_timeout</c> could lead to much
longer timeout than specified. The solution is a new
socket option <c>{send_timeout_close,true}</c> that will
do automatic close on timeout. Subsequent calls to send
will then immediately fail due to the closed connection.</p>
<p>
Own Id: OTP-7731 Aux Id: seq11161 </p>
</item>
</list>
</section>
</section>
<section><title>Kernel 2.12.5</title>
<section><title>Fixed Bugs and Malfunctions</title>
<list>
<item>
<p>The documentation of <c>rpc:pmap/3</c> has been
corrected. (Thanks to Kirill Zaborski.)</p>
<p>
Own Id: OTP-7537</p>
</item>
<item>
<p>
The listen socket used for the distributed Erlang
protocol now uses the socket option 'reuseaddr', which is
useful when you force the listen port number using kernel
options 'inet_dist_listen_min' and 'inet_dist_listen_max'
and restarts a node with open connections.</p>
<p>
Own Id: OTP-7563</p>
</item>
<item>
<p>
Fixed memory leak of unclosed TCP-ports. A gen_tcp:send()
followed by a failing gen_tcp:recv() could in some cases
cause the port to linger after being closed.</p>
<p>
Own Id: OTP-7615</p>
</item>
</list>
</section>
<section><title>Improvements and New Features</title>
<list>
<item>
<p>Processes spawned using <c>proc_lib</c> (including
<c>gen_server</c> and other library modules that use
<c>proc_lib</c>) no longer keep the entire argument list
for the initial call, but only the arity.</p>
<p>Also, if <c>proc_lib:spawn/1</c> is used to spawn a
fun, the actual fun is not kept, but only module,
function name, and arity of the function that implements
the fun.</p>
<p>The reason for the change is that keeping the initial
fun (or a fun in an argument list), would prevent
upgrading the code for the module. A secondary reason is
that keeping the fun and function arguments could waste a
significant amount of memory.</p>
<p>The drawback with the change is that the crash reports
will provide less precise information about the initial
call (only <c>Module:Function/Arity</c> instead of
<c>Module:Function(Arguments)</c>). The function
<c>proc_lib:initial_call/1</c> still returns a list, but
each argument has been replaced with a dummy atom.</p>
<p>
Own Id: OTP-7531 Aux Id: seq11036 </p>
</item>
<item>
<p>
<c>io:get_line/1</c> when reading from standard input is
now substantially faster. There are also some minor
performance improvements in <c>io:get_line/1</c> when
reading from any file opened in binary mode. (Thanks to
Fredrik Svahn.)</p>
<p>
Own Id: OTP-7542</p>
</item>
<item>
<p>
There is now experimental support for loading of code
from archive files. See the documentation of <c>code</c>,
<c>init</c>, <c>erl_prim_loader </c> and <c>escript</c>
for more info.</p>
<p>
The error handling of <c>escripts</c> has been improved.</p>
<p>
An <c>escript</c> may now set explicit arguments to the
emulator, such as <c>-smp enabled</c>.</p>
<p>
An <c>escript</c> may now contain a precompiled beam
file.</p>
<p>
An <c>escript</c> may now contain an archive file
containing one or more applications (experimental).</p>
<p>
The internal module <c>code_aux</c> has been removed.</p>
<p>
Own Id: OTP-7548 Aux Id: otp-6622 </p>
</item>
<item>
<p>
<c>code:is_sticky/1</c> is now documented. (Thanks to
Vlad Dumitrescu.)</p>
<p>
Own Id: OTP-7561</p>
</item>
<item>
<p>
In the job control mode, the "s" and "r" commands now
take an optional argument to specify which shell to
start. (Thanks to Robert Virding.)</p>
<p>
Own Id: OTP-7617</p>
</item>
<item>
<p>
<c>net_adm:world/0,1</c> could crash if called in an
emulator that has not been started with either the
<c>-sname</c> or <c>-name</c> option; now it will return
an empty list. (Thanks to Edwin Fine.)</p>
<p>
Own Id: OTP-7618</p>
</item>
</list>
</section>
</section>
<section><title>Kernel 2.12.4</title>
<section><title>Fixed Bugs and Malfunctions</title>
<list>
<item>
<p>
Large files are now handled on Windows, where the
filesystem supports it.</p>
<p>
Own Id: OTP-7410</p>
</item>
</list>
</section>
<section><title>Improvements and New Features</title>
<list>
<item>
<p>
New BIF <c>erlang:decode_packet/3</c> that extracts a
protocol packet from a binary. Similar to the socket
option <c>{packet, Type}</c>. Also documented the socket
packet type <c>http</c> and made it official.
<em>NOTE</em>: The tuple format for <c>http</c> packets
sent from an active socket has been changed in an
incompatible way.</p>
<p>
*** POTENTIAL INCOMPATIBILITY ***</p>
<p>
Own Id: OTP-7404</p>
</item>
<item>
<p>
Setting the <c>{active,once}</c> for a socket (using
inets:setopts/2) is now specially optimized (because the
<c>{active,once}</c> option is typically used much more
frequently than other options).</p>
<p>
Own Id: OTP-7520</p>
</item>
</list>
</section>
</section>
<section><title>Kernel 2.12.3</title>
<section><title>Fixed Bugs and Malfunctions</title>
<list>
<item>
<p>
SCTP_ADDR_CONFIRMED events are now handled by gen_sctp.</p>
<p>
Own Id: OTP-7276</p>
</item>
<item>
<p>When leaving a process group with <c>pg2:leave/2</c>
the process was falsely assumed to be a member of the
group. This bug has been fixed.</p>
<p>
Own Id: OTP-7277</p>
</item>
<item>
<p>
In the Erlang shell, using up and down arrow keys, the
wrong previous command could sometimes be retrieved.</p>
<p>
Own Id: OTP-7278</p>
</item>
<item>
<p>
The documentation for <c>erlang:trace/3</c> has been
corrected.</p>
<p>
Own Id: OTP-7279 Aux Id: seq10927 </p>
</item>
<item>
<p>
In the SMP emulator, there was small risk that
<c>code:purge(Mod)</c> would kill a process that was
running code in <c>Mod</c> and unload the module
<c>Mod</c> before the process had terminated.
<c>code:purge(Mod)</c> now waits for confirmation (using
<c>erlang:monitor/2</c>) that the process has been killed
before proceeding.</p>
<p>
Own Id: OTP-7282</p>
</item>
<item>
<p>
<c>zlib:inflate</c> failed when the size of the inflated
data was an exact multiple of the internal buffer size
(4000 bytes by default).</p>
<p>
Own Id: OTP-7359</p>
</item>
</list>
</section>
<section><title>Improvements and New Features</title>
<list>
<item>
<p>
Additional library directories can now be specified in
the environment variable ERL_LIBS. See the manual page
for the <c>code</c> module. (Thanks to Serge Aleynikov.)</p>
<p>
Own Id: OTP-6940</p>
</item>
<item>
<p>
crypto and zlib drivers improved to allow concurrent smp
access.</p>
<p>
Own Id: OTP-7262</p>
</item>
<item>
<p>
There is a new function <c>init:stop/1</c> which can be
used to shutdown the system cleanly AND generate a
non-zero exit status or crash dump. (Thanks to Magnus
Froberg.)</p>
<p>
Own Id: OTP-7308</p>
</item>
<item>
<p>
The <c>hide</c> option for <c>open_port/2</c> is now
documented. (Thanks to Richard Carlsson.)</p>
<p>
Own Id: OTP-7358</p>
</item>
</list>
</section>
</section>
<section><title>Kernel 2.12.2.1</title>
<section><title>Improvements and New Features</title>
<list>
<item>
<p>
<c>os:cmd/1</c> on unix platforms now use <c>/bin/sh</c>
as shell instead of looking for <c>sh</c> in the
<c>PATH</c> environment.</p>
<p>
Own Id: OTP-7283</p>
</item>
</list>
</section>
</section>
<section><title>Kernel 2.12.2</title>
<section><title>Fixed Bugs and Malfunctions</title>
<list>
<item>
<p>A bug caused by a race condition involving
<c>disk_log</c> and <c>pg2</c> has been fixed.</p>
<p>
Own Id: OTP-7209 Aux Id: seq10890 </p>
</item>
<item>
<p>The beta testing module <c>gen_sctp</c> now supports
active mode as stated in the documentation. Active mode
is still rather untested, and there are some issues about
what should be the right semantics for
<c>gen_sctp:connect/5</c>. In particular: should it be
blocking or non-blocking or choosable. There is a high
probability it will change semantics in a (near) future
patch.</p> <p>Try it, give comments and send in bug
reports!</p>
<p>
Own Id: OTP-7225</p>
</item>
</list>
</section>
<section><title>Improvements and New Features</title>
<list>
<item>
<p><c>erlang:system_info/1</c> now accepts the
<c>logical_processors</c>, and <c>debug_compiled</c>
arguments. For more info see the, <c>erlang(3)</c>
documentation.</p> <p>The scale factor returned by
<c>test_server:timetrap_scale_factor/0</c> is now also
effected if the emulator uses a larger amount of
scheduler threads than the amount of logical processors
on the system. </p>
<p>
Own Id: OTP-7175</p>
</item>
<item>
<p>
Updated the documentation for
<c>erlang:function_exported/3</c> and <c>io:format/2</c>
functions to no longer state that those functions are
kept mainly for backwards compatibility.</p>
<p>
Own Id: OTP-7186</p>
</item>
<item>
<p>
A process executing the <c>processes/0</c> BIF can now be
preempted by other processes during its execution. This
in order to disturb the rest of the system as little as
possible. The returned result is, of course, still a
consistent snapshot of existing processes at a time
during the call to <c>processes/0</c>.</p>
<p>
The documentation of the <c>processes/0</c> BIF and the
<c>is_process_alive/1</c> BIF have been updated in order
to clarify the difference between an existing process and
a process that is alive.</p>
<p>
Own Id: OTP-7213</p>
</item>
<item>
<p><c>tuple_size/1</c> and <c>byte_size/1</c> have been
substituted for <c>size/1</c> in the documentation.</p>
<p>
Own Id: OTP-7244</p>
</item>
</list>
</section>
</section>
<section><title>Kernel 2.12.1.2</title>
<section><title>Improvements and New Features</title>
<list>
<item>
<p>The <c>{allocator_sizes, Alloc}</c> and
<c>alloc_util_allocators</c> arguments are now accepted
by <c>erlang:system_info/1</c>. For more information see
the <c>erlang(3)</c> documentation.</p>
<p>
Own Id: OTP-7167</p>
</item>
</list>
</section>
</section>
<section><title>Kernel 2.12.1.1</title>
<section><title>Fixed Bugs and Malfunctions</title>
<list>
<item>
<p>
Fixed a problem in group that could cause the ssh server
to lose answers or hang.</p>
<p>
Own Id: OTP-7185 Aux Id: seq10871 </p>
</item>
</list>
</section>
</section>
<section><title>Kernel 2.12.1</title>
<section><title>Fixed Bugs and Malfunctions</title>
<list>
<item>
<p>
file:read/2 and file:consult_stream/1,3 did not use an
empty prompt on I/O devices. This bug has now been
corrected.</p>
<p>
Own Id: OTP-7013</p>
</item>
<item>
<p>
The sctp driver has been updated to work against newer
lksctp packages e.g 1.0.7 that uses the API spelling
change adaption -> adaptation. Older lksctp (1.0.6) still
work. The erlang API in gen_sctp.erl and inet_sctp.hrl
now spells 'adaptation' regardless of the underlying C
API.</p>
<p>
*** POTENTIAL INCOMPATIBILITY ***</p>
<p>
Own Id: OTP-7120</p>
</item>
</list>
</section>
<section><title>Improvements and New Features</title>
<list>
<item>
<p>The documentation has been updated so as to reflect
the last updates of the Erlang shell as well as the minor
modifications of the control sequence <c>p</c> of the
<c>io_lib</c> module.</p> <p>Superfluous empty lines have
been removed from code examples and from Erlang shell
examples.</p>
<p>
Own Id: OTP-6944 Aux Id: OTP-6554, OTP-6911 </p>
</item>
<item>
<p><c>tuple_size/1</c> and <c>byte_size/1</c> have been
substituted for <c>size/1</c>.</p>
<p>
Own Id: OTP-7009</p>
</item>
</list>
</section>
</section>
<section><title>Kernel 2.12</title>
<section><title>Fixed Bugs and Malfunctions</title>
<list>
<item>
<p>
A bug for raw files when reading 0 bytes returning 'eof'
instead of empty data has been corrected.</p>
<p>
Own Id: OTP-6291 Aux Id: OTP-6967 </p>
</item>
<item>
<p>
A bug in gen_udp:fdopen reported by David Baird and also
found by Dialyzer has been fixed.</p>
<p>
Own Id: OTP-6836 Aux Id: OTP-6594 </p>
</item>
<item>
<p>
Calling <c>error_logger:tty(true)</c> multiple times does
not give multiple error log printouts.</p>
<p>
Own Id: OTP-6884 Aux Id: seq10767 </p>
</item>
<item>
<p>The global name server now ignores <c>nodeup</c>
messages when the command line flag <c>-connect_all
false</c> has been used. (Thanks to Trevor
Woollacott.)</p>
<p>
Own Id: OTP-6931</p>
</item>
<item>
<p>file:write_file/3, file:write/2 and file:read/2 could
crash (contrary to documentation) for odd enough file
system problems, e.g write to full file system. This bug
has now been corrected.</p> <p>In this process the file
module has been rewritten to produce better error codes.
Posix error codes now originate from the OS file system
calls or are generated only for very similar causes (for
example 'enomem' is generated if a memory allocation
fails, and 'einval' is generated if the file handle in
Erlang is a file handle but currently invalid).</p>
<p>More Erlang-ish error codes are now generated. For
example <c>{error,badarg}</c> is now returned from
<c>file:close/1</c> if the argument is not of a file
handle type. See file(3).</p> <p>The possibility to write
a single byte using <c>file:write/2</c> instead of a list
or binary of one byte, contradictory to the
documentation, has been removed.</p>
<p>
*** POTENTIAL INCOMPATIBILITY ***</p>
<p>
Own Id: OTP-6967 Aux Id: OTP-6597 OTP-6291 </p>
</item>
<item>
<p>
Monitor messages produced by the system monitor
functionality, and garbage collect trace messages could
contain erroneous heap and/or stack sizes when the actual
heaps and/or stacks were huge.</p>
<p>
As of erts version 5.6 the <c>large_heap</c> option to
<c>erlang:system_monitor/[1,2]</c> has been modified. The
monitor message is sent if the sum of the sizes of all
memory blocks allocated for all heap generations is equal
to or larger than the specified size. Previously the
monitor message was sent if the memory block allocated
for the youngest generation was equal to or larger than
the specified size.</p>
<p>
*** POTENTIAL INCOMPATIBILITY ***</p>
<p>
Own Id: OTP-6974 Aux Id: seq10796 </p>
</item>
<item>
<p>
<c>inet:getopts/2</c> returned random values on Windows
Vista.</p>
<p>
Own Id: OTP-7003</p>
</item>
</list>
</section>
<section><title>Improvements and New Features</title>
<list>
<item>
<p>
Minor documentation corrections for file:pread/2 and
file:pread/3.</p>
<p>
Own Id: OTP-6853</p>
</item>
<item>
<p>
The deprecated functions <c>file:file_info/1</c>,
<c>init:get_flag/1</c>, <c>init:get_flags/0</c>, and
<c>init:get_args/0</c> have been removed.</p>
<p>
*** POTENTIAL INCOMPATIBILITY ***</p>
<p>
Own Id: OTP-6886</p>
</item>
<item>
<p>
Contract directives for modules in Kernel and STDLIB.</p>
<p>
Own Id: OTP-6895</p>
</item>
<item>
<p>The functions io:columns/0, io:columns/1, io:rows/0
and io:rows/1 are added to allow the user to get
information about the terminal geometry. The shell takes
some advantage of this when formatting output. For
regular files and other io-devices where height and width
are not applicable, the functions return
{error,enotsup}.</p>
<p>Potential incompatibility: If one has written a custom
io-handler, the handler has to either return an error or
take care of io-requests regarding terminal height and
width. Usually that is no problem as io-handlers, as a
rule of thumb, should give an error reply when receiving
unknown io-requests, instead of crashing.</p>
<p>
*** POTENTIAL INCOMPATIBILITY ***</p>
<p>
Own Id: OTP-6933</p>
</item>
<item>
<p>
The undocumented and unsupported functions
<c>inet:ip_to_bytes/1</c>, <c>inet:ip4_to_bytes/1</c>,
<c>inet:ip6_to_bytes/1</c>, and
<c>inet:bytes_to_ip6/16</c> have been removed.</p>
<p>
Own Id: OTP-6938</p>
</item>
<item>
<p>
Added new checksum combine functions to <c>zlib</c>. And
fixed a bug in <c>zlib:deflate</c>. Thanks Matthew
Dempsky.</p>
<p>
Own Id: OTP-6970</p>
</item>
<item>
<p>
The <c>spawn_monitor/1</c> and <c>spawn_monitor/3</c> BIFs
are now auto-imported (i.e. they no longer need an
<c>erlang:</c> prefix).</p>
<p>
Own Id: OTP-6975</p>
</item>
<item>
<p>All functions in the <c>code</c> module now fail with
an exception if they are called with obviously bad
arguments, such as a tuple when an atom was expected.
Some functions now also fail for undocumented argument
types (for instance, <c>ensure_loaded/1</c> now only
accepts an atom as documented; it used to accept a string
too).</p>
<p><c>Dialyzer</c> will generally emit warnings for any
calls that use undocumented argument types. Even if the
call happens to still work in R12B, you should correct
your code. A future release will adhere to the
documentation.</p>
<p>
*** POTENTIAL INCOMPATIBILITY ***</p>
<p>
Own Id: OTP-6983</p>
</item>
</list>
</section>
</section>
<section><title>Kernel 2.11.5.2</title>
<section><title>Fixed Bugs and Malfunctions</title>
<list>
<item>
<p>
The kernel parameter dist_auto_connect once could fail to
block a node if massive parallel sends were issued
during a transient failure of network communication</p>
<p>
Own Id: OTP-6893 Aux Id: seq10753 </p>
</item>
</list>
</section>
</section>
<section><title>Kernel 2.11.5.1</title>
<section><title>Fixed Bugs and Malfunctions</title>
<list>
<item>
<p>
The internal (rarely used) DNS resolver has been modified
to not use the domain search list when asked to resolve
an absolute name; a name with a terminating dot. There
was also a bug causing it to create malformed DNS queries
for absolute names that has been corrected, correction
suggested by Scott Lystig Fritchie. The code has also
been corrected to look up cached RRs in the same search
order as non-cached, now allows having the root domain
among the search domains, and can now actually do a zone
transfer request.</p>
<p>
*** POTENTIAL INCOMPATIBILITY ***</p>
<p>
Own Id: OTP-6806 Aux Id: seq10714 EABln35459 </p>
</item>
<item>
<p>
zlib:close/1 would leave an EXIT message in the message
queue if the calling process had the trap_exit flag
enabled.</p>
<p>
Own Id: OTP-6811</p>
</item>
</list>
</section>
<section><title>Improvements and New Features</title>
<list>
<item>
<p>The documentation of <c>process_flag(priority,
Level)</c> has been updated, see the <c>erlang(3)</c>
documentation. </p>
<p>
Own Id: OTP-6745 Aux Id: OTP-6715 </p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.11.5</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>The shell has been updated to fix the following flaws:
Shell process exit left you with an unresponsive initial
shell if not using oldshell. Starting a restricted shell
with a nonexisting callback module resulted in a shell
where no commands could be used, not even init:stop/0.
Fun's could not be used as parameters to local shell
functions (in shell_default or user_default) when
restricted_shell was active.</p>
<p>Own Id: OTP-6537</p>
</item>
<item>
<p>The undocumented feature gen_tcp:fdopen/2 was broken
in R11B-4. It is now fixed again.</p>
<p>Own Id: OTP-6615</p>
</item>
<item>
<p>Corrected cancellation of timers in three places in the
inet_res module. (Problem found by Dialyzer.)</p>
<p>Own Id: OTP-6676</p>
</item>
</list>
</section>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p>Corrected protocol layer flue for socket options
SO_LINGER, SO_SNDBUF and SO_RCVBUF, for SCTP.</p>
<p>Own Id: OTP-6625 Aux Id: OTP-6336 </p>
</item>
<item>
<p>The behaviour of the inet option {active,once} on peer
close is improved and documented.</p>
<p>Own Id: OTP-6681</p>
</item>
<item>
<p>The inet option send_timeout for connection oriented
sockets is added to allow for timeouts in communicating
send requests to the underlying TCP stack.</p>
<p>Own Id: OTP-6684 Aux Id: seq10637 OTP-6681 </p>
</item>
<item>
<p>Minor Makefile changes.</p>
<p>Own Id: OTP-6689 Aux Id: OTP-6742 </p>
</item>
<item>
<p>The documentation of <c>process_flag(priority, Level)</c> has been updated, see the <c>erlang(3)</c>
documentation. </p>
<p>Own Id: OTP-6715</p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.11.4.2</title>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p>process_flag/2 accepts the new flag <c>sensitive</c>.</p>
<p>Own Id: OTP-6592 Aux Id: seq10555 </p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.11.4.1</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>A bug in gen_udp:open that broke the 'fd' option has been
fixed.</p>
<p>Own Id: OTP-6594 Aux Id: seq10619 </p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.11.4</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>Added a warning to the documentation for the
<c>error_logger</c> functions <c>error_msg/1,2</c>,
<c>warning_msg/1,2</c> and <c>info_msg/1,2</c> that
calling these function with bad arguments can crash the
standard event handler.</p>
<p>Own Id: OTP-4575 Aux Id: seq7693 </p>
</item>
<item>
<p>A bug in <c>inet_db</c> concerning getting the resolver
option <c>retry</c> has been corrected.</p>
<p>Own Id: OTP-6380 Aux Id: seq10534 </p>
</item>
<item>
<p>Names registered by calling
<c>global:register_name()</c> or
<c>global:re_register_name()</c> were not always
unregistered when the registering or registered process
died. This bug has been fixed.</p>
<p>Own Id: OTP-6428</p>
</item>
<item>
<p>When setting the kernel configuration parameter
<c>error_logger</c> to <c>false</c>, the documentation
stated that "No error logger handler is installed". This
is true, but error logging is not turned off, as the
initial, primitive error logger event handler is kept,
printing raw event messages to tty.</p>
<p>Changing this behavior can be viewed as a backward
incompatible change. Instead a new value <c>silent</c>
for the configuration parameter has been added, which
ensures that error logging is completely turned off.</p>
<p>Own Id: OTP-6445</p>
</item>
<item>
<p>Clarified the documentation for <c>code:lib_dir/1</c> and
<c>code:priv_dir/1</c>. The functions traverse the names
of the code path, they do not search the actual
directories.</p>
<p>Own Id: OTP-6466</p>
</item>
<item>
<p><c>io:setopts</c> returned <c>{error,badarg}</c>, when
called with only an <c>expand_fun</c> argument. (Thanks to
igwan.)</p>
<p>Own Id: OTP-6508</p>
</item>
</list>
</section>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p>An interface towards the SCTP Socket API Extensions
has been implemented.It is an Open Source patch courtesy
of Serge Aleynikov and Leonid Timochouk. The Erlang code
parts has been adapted by the OTP team, changing the
Erlang API somewhat.</p>
<p>The Erlang interface consists of the module
<c>gen_sctp</c> and an include file
<c>-include_lib("kernel/include/inet_sctp.hrl").</c> for
option record definitions. The <c>gen_sctp</c> module is
documented.</p>
<p>The delivered Open Source patch, before the OTP team
rewrites, was written according to
<url href="http://tools.ietf.org/html/draft-ietf-tsvwg-sctpsocket-13">http://tools.ietf.org/html/draft-ietf-tsvwg-sctpsocket-13</url>
and was claimed to work fine, tested on Linux Fedora Core
5.0 (kernel 2.6.15-2054 or later) and on Solaris 10 and
11. The OTP team rewrites used the same standard document
but might have accidentally broken some functionality. If
so, it will soon be patched to working state. The tricky
parts in C and the general design has essentially not
changed. During the rewrites the code was hand tested on
SuSE Linux Enterprise Server 10, and briefly on Solaris
10. Feedbach on code and docs is very much
appreciated.</p>
<p>The SCTP interface is in beta state. It has only been
hand tested and has no automatic test suites in OTP
meaning everything is most certainly not tested. Socket
active mode is broken. IPv6 is not tested. The documentation
has been reworked due to the API changes,
but has not been proofread after this.</p>
<p>Thank you from the OTP team to Serge Aleynikov and
Leonid Timochouk for a valuable contribution. We hope we
have not messed it up too much.</p>
<p>Own Id: OTP-6336</p>
</item>
<item>
<p>A <c>{minor_version,Version}</c> option is now recognized
by <c>term_to_binary/2</c>. {minor_version,1} will cause
floats to be encoded in an exact and more space-efficient
way compared to the previous encoding.</p>
<p>Own Id: OTP-6434</p>
</item>
<item>
<p>Monitoring of nodes has been improved. Now the following
properties apply to
<c>net_kernel:monitor_nodes/[1,2]</c>:</p>
<list type="bulleted">
<item><c>nodeup</c> messages will be delivered before delivery
of any message from the remote node passed through the
newly established connection. </item>
<item><c>nodedown</c> messages will not be delivered until all
messages from the remote node that have been passed
through the connection have been delivered. </item>
<item>Subscriptions can also be made before the
<c>net_kernel</c> server has been started. </item>
</list>
<p>Own Id: OTP-6481</p>
</item>
<item>
<p>Setting and getting socket options in a "raw" fashion is
now allowed. Using this feature will inevitably produce
non portable code, but will allow setting ang getting
arbitrary uncommon options on TCP stacks that do have
them.</p>
<p>Own Id: OTP-6519</p>
</item>
<item>
<p>Dialyzer warnings have been eliminated.</p>
<p>Own Id: OTP-6523</p>
</item>
<item>
<p>The documentation for <c>file:delete/1</c> and
<c>file:set_cwd/1</c> has been updated to clarify what
happens if the input arguments are of an incorrect type.</p>
<p>Own Id: OTP-6535</p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.11.3.1</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>An erroneous packet size could be used for the first
messages passed through a newly established connection
between two Erlang nodes. This could cause messages to be
discarded, or termination of the connection.</p>
<p>Own Id: OTP-6473</p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.11.3</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>On Unix, the <c>unix:cmd/1</c> function could leave an
'EXIT' message in the message queue for the calling
process That problem was more likely to happen in an SMP
emulator.</p>
<p>Own Id: OTP-6368</p>
</item>
</list>
</section>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p>More interfaces are added in erl_ddll, to support
different usage scenarios.</p>
<p>Own Id: OTP-6307 Aux Id: OTP-6234 </p>
</item>
<item>
<p>Locks set by calling <c>global:set_lock()</c> were not
always deleted when the locking process died. This bug
has been fixed.</p>
<p>Own Id: OTP-6341 Aux Id: seq10445 </p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.11.2</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>Behavior in case of disappeared nodes when using he
dist_auto_connect once got changed in R11B-1. The
timeouts regarding normal distributed operations is now
reverted to the old (pre R11B-1).</p>
<p>Own Id: OTP-6258 Aux Id: OTP-6200, seq10449 </p>
</item>
<item>
<p>Start-up problems for the internal process used by the
<c>inet:gethostbyname()</c> functions were eliminated. If
the internal process (<c>inet_gethost_native</c>) had not
previously been started, and if several processes at the
same time called one of the <c>inet:gethostbyname()</c>
functions, the calls could fail.</p>
<p>Own Id: OTP-6286</p>
</item>
</list>
</section>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p>Code cleanup: the old internal obsolete file_server has
been removed. It was only used when communicating with R7
and older nodes.</p>
<p>Own Id: OTP-6245</p>
</item>
<item>
<p>Trying to open a non-existent or badly formed disk log
no longer results in a crash report. In particular,
<c>ets:file2tab/1</c> reports no error when the argument
is not a well-formed disk log file. (The return value has
not been changed, it is still an error tuple.)</p>
<p>Own Id: OTP-6278 Aux Id: seq10421 </p>
</item>
<item>
<p>There are new BIFs <c>erlang:spawn_monitor/1,3</c>,
and the new option <c>monitor</c> for
<c>spawn_opt/2,3,4,5</c>.</p>
<p>The <c>observer_backend</c> module has been updated to
handle the new BIFs.</p>
<p>Own Id: OTP-6281</p>
</item>
<item>
<p>To help Dialyzer find more bugs, many functions in the
Kernel and STDLIB applications now only accept arguments
of the type that is documented.</p>
<p>For instance, the functions <c>lists:prefix/2</c> and
<c>lists:suffix/2</c> are documented to only accept lists
as their arguments, but they actually accepted anything
and returned <c>false</c>. That has been changed so that
the functions cause an exception if one or both arguments
are not lists.</p>
<p>Also, the <c>string:strip/3</c> function is documented
to take a character argument that is a character to strip
from one or both ends of the string. Given a list instead
of a character, it used to do nothing, but will now cause
an exception.</p>
<p>Dialyzer will find most cases where those functions
are passed arguments of the wrong type.</p>
<p>*** POTENTIAL INCOMPATIBILITY ***</p>
<p>Own Id: OTP-6295</p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.11.1.1</title>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p>There is now an option read_packets for UDP sockets that
sets the maximum number of UDP packets that will be read
for each invocation of the socket driver.</p>
<p>Own Id: OTP-6249 Aux Id: seq10452 </p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.11.1</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>In R11B-0, the erl_ddll server process is always started.
Despite that, the configuration parameter
<c>start_ddll</c> for the Kernel application was still
obeyed, which would cause the erl_ddll server to be
started TWICE (and the system shutting down as a result).
In this release, <c>start_ddll</c> is no longer used and
its documentation has been removed.</p>
<p>Own Id: OTP-6163</p>
</item>
<item>
<p>The kernel option {dist_auto_connect,once} could block
out nodes that had never been connected, causing
persistent partitioning of networks. Furthermore, partial
restarts of networks could cause inconsistent global name
databases. Both problems are now solved.</p>
<p>Own Id: OTP-6200 Aux Id: seq10377 </p>
</item>
</list>
</section>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p>Late arriving tcp_closed and udp_closed messages are now
removed from the message queue of a process calling
gen_tcp:close/1, gen_udp:close/1, and inet:close/1.</p>
<p>Own Id: OTP-6197</p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.11</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>When repairing a disk log with a corrupt index file
(caused by for instance a hard disk failure) the old
contents of the index file is kept unmodified. This will
make repeated attempts to open the disk log fail every
time.</p>
<p>Own Id: OTP-5558 Aux Id: seq9823 </p>
</item>
<item>
<p>Previously <c>unlink/1</c> and <c>erlang:demonitor/2</c>
behaved completely asynchronous. This had one undesirable
effect, though. You could never know when you were
guaranteed <em>not</em> to be affected by a link that you
had unlinked or a monitor that you had demonitored.</p>
<p>The new behavior of <c>unlink/1</c> and
<c>erlang:demonitor/2</c> can be viewed as two operations
performed atomically. Asynchronously send an unlink
signal or a demonitor signal, and ignore any future
results of the link or monitor.</p>
<p><em>NOTE</em>: This change can cause some obscure code
to fail which previously did not. For example, the
following code might hang:</p>
<code type="none">
Mon = erlang:monitor(process, Pid),
%% ...
exit(Pid, bang),
erlang:demonitor(Mon),
receive
{'DOWN', Mon, process, Pid, _} -> ok
%% We were previously guaranteed to get a down message
%% (since we exited the process ourself), so we could
%% in this case leave out:
%% after 0 -> ok
end,
</code>
<p>*** POTENTIAL INCOMPATIBILITY ***</p>
<p>Own Id: OTP-5772</p>
</item>
<item>
<p>The behavior when an application fails to start and
possibly causes the runtime system to halt has been
cleaned up, including fixing some minor bugs.</p>
<p><c>application_controller</c> should now always terminate
with a non-nested string, meaning the slogan in an
<c>erl_crash.dump</c> should always be easy to read.</p>
<p><c>init</c> now makes sure that the slogan passed to
<c>erlang:halt/1</c> does not exceed the maximum allowed
length.</p>
<p>Redundant calls to <c>list_to_atom/1</c> has been removed
from the primitive <c>error_logger</c> event handler.
(Thanks Serge Aleynikov for pointing this out).</p>
<p>The changes only affects the contents of the error
messages and crashdump file slogan.</p>
<p>Own Id: OTP-5964</p>
</item>
<item>
<p>The <c>erl_ddll</c> server is now started when OTP is
started and placed under the Kernel supervisor. This
fixes several minor issues. It used to be started on
demand.</p>
<p>The documentation for the <c>start</c> and <c>stop</c>
functions in the <c>erl_ddll</c> module has been removed,
as those functions are not meant to be used by other
applications.</p>
<p>Furthermore, the <c>erl_ddll:stop/1</c> function no longer
terminates the <c>erl_ddll</c> server, as that would
terminate the entire runtime system.</p>
<p>Own Id: OTP-6033</p>
</item>
</list>
</section>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p>Removed some unused functions from
<c>application_master</c>.</p>
<p>Own Id: OTP-3889</p>
</item>
<item>
<p>Global no longer allows the registration of a process
under more than one name. If the old (buggy) behavior is
desired the Kernel application variable
<c>global_multi_name_action</c> can be given the value
<c>allow</c>.</p>
<p>Own Id: OTP-5640 Aux Id: OTP-5603</p>
</item>
<item>
<p>The (slightly misleading) warnings that was shown when
the <c>erlang.erl</c> file was compiled has been
eliminated.</p>
<p>Own Id: OTP-5947</p>
</item>
<item>
<p>The <c>auth</c> module API is deprecated.</p>
<p>Own Id: OTP-6037</p>
</item>
<item>
<p>Added <c>erlang:demonitor/2</c>, making it possible to at
the same time flush a received <c>'DOWN'</c> message, if
there is one. See <c>erlang(3)</c>.</p>
<p>Own Id: OTP-6100 Aux Id: OTP-5772 </p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.10.13</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>Large files (more than 2 GBytes) are now handled on
Solaris 8.</p>
<p>Own Id: OTP-5849 Aux Id: seq10157</p>
</item>
<item>
<p>During startup, a garbage <c>{'DOWN', ...}</c> message was
left by <c>inet_gethost_native</c>, that caused problems
for the starting code server.</p>
<p>Own Id: OTP-5978 Aux Id: OTP-5974</p>
</item>
</list>
</section>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p><c>global</c> now makes several attempts to connect nodes
when maintaining the fully connected network. More than one
attempt is sometimes needed under very heavy load.</p>
<p>Own Id: OTP-5889</p>
</item>
<item>
<p><c>erl_epmd</c> now explicitly sets the timeout to
<c>infinity</c> when calling <c>gen_server:call</c>. The
old timeout of 15 seconds could time out under very heavy
load.</p>
<p>Own Id: OTP-5959</p>
</item>
<item>
<p>Corrected the start of code server to use reference-tagged
tuples to ensure that an unexpected message sent to
the parent process does not cause a halt of the system.
Also removed the useless <c>start/*</c> functions in both
<c>code.erl</c> and <c>code_server.erl</c> and no longer
exports the <c>init</c> function from
<c>code_server.erl</c>.</p>
<p>Own Id: OTP-5974 Aux Id: seq10243, OTP-5978</p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.10.12</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>A bug in <c>global</c> has been fixed: the locker process
added <c>nonode@nohost</c> to the list of nodes to lock.
This could happen before any nodes got known to the global
name server. Depending on net configuration the symptom was
a delay.</p>
<p>Own Id: OTP-5792 Aux Id: OTP-5563</p>
</item>
<item>
<p>If an <c>.app</c> file is missing, the error reason
returned by <c>application:load/1</c> has been corrected
to <c>{"no such file or directory", "FILE.app"}</c>,
instead of the less informative <c>{"unknown POSIX error","FILE.app"}</c>.</p>
<p>Own Id: OTP-5809</p>
</item>
<item>
<p>Bug fixes: <c>disk_log:accessible_logs/0</c> no longer
reports all <c>pg2</c> process groups as distributed disk
logs; <c>disk_log:pid2name/1</c> did not recognize
processes of distributed disk logs.</p>
<p>Own Id: OTP-5810</p>
</item>
<item>
<p>The functions <c>file:consult/1</c>,
<c>file:path_consult/2</c>, <c>file:eval/1,2</c>,
<c>file:path_eval/2,3</c>, <c>file:script/1,2</c>,
<c>file:path_script/2,3</c> now return correct line
numbers in error tuples.</p>
<p>Own Id: OTP-5814</p>
</item>
<item>
<p>If there were user-defined variables in the boot script,
and their values were not provided using
the <c>-boot_var</c> option, the emulator would refuse to
start with a confusing error message. Corrected to show a
clear, understandable message.</p>
<p>The <c>prim_file</c> module was modified to not depend
on the <c>lists</c> module, to make it possible to start
the emulator using a user-defined loader. (Thanks to
Martin Bjorklund.)</p>
<p>Own Id: OTP-5828 Aux Id: seq10151</p>
</item>
<item>
<p>Minor corrections in the description of open modes.
(Thanks to Richard Carlsson.)</p>
<p>Own Id: OTP-5856</p>
</item>
</list>
</section>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p><c>application_controller</c> now terminates with the
actual error reason, instead of <c>shutdown</c>. This
means that the crash dump now should be somewhat more
informative, in the case where the runtime system is
terminated due to an error in an application.</p>
<p>Example: If the (permanent) application <c>app1</c> fails
to start, the slogan now will be: "<c>Kernel pid terminated (application_controller) ({application_start_failure,app1,{shutdown, {app1,start,[normal,[]]}}})</c>"</p>
<p>rather than the previous "<c>Kernel pid terminated (application_controller) (shutdown)</c>".</p>
<p>Own Id: OTP-5811</p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.10.11.1</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>Timers could sometimes timeout too early. This bug has
now been fixed.</p>
<p>Automatic cancellation of timers created by
<c>erlang:send_after(Time,</c> pid(), Msg), and
<c>erlang:start_timer(Time,</c> pid(), Msg) has been
introduced.
Timers created with the receiver specified by a pid, will
automatically be cancelled when the receiver exits. For
more information see the <c>erlang(3)</c> man page.</p>
<p>In order to be able to maintain a larger amount of timers
without increasing the maintenance cost, the internal
timer wheel and bif timer table have been enlarged.</p>
<p>Also a number of minor bif timer optimizations have been
implemented.</p>
<p>Own Id: OTP-5795 Aux Id: OTP-5090, seq8913, seq10139,
OTP-5782</p>
</item>
</list>
</section>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p>Documentation improvements:</p>
<p>- documentation for <c>erlang:link/1</c> corrected</p>
<p>- command line flag <c>-code_path_cache</c> added</p>
<p>- <c>erl</c> command line flags clarifications</p>
<p>- <c>net_kernel(3)</c> clarifications</p>
<p>Own Id: OTP-5847</p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.10.11</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>Several bug fixes and improvements in the global name
registration facility (see <c>global(3)</c>):</p>
<list type="bulleted">
<item>the name resolving procedure did not always unlink no
longer registered processes;</item>
<item>the global name could sometimes hang when a
<c>nodedown</c> was immediately followed by a
<c>nodeup</c>;</item>
<item>global names were not always unregistered when a node
went down;</item>
<item>it is now possible to set and delete locks at
the same time as the global name server is resolving
names--the handling of global locks has been separated
from registration of global names;</item>
</list>
<p>As of this version, <c>global</c> no longer supports nodes
running Erlang/OTP R7B or earlier.</p>
<p>*** POTENTIAL INCOMPATIBILITY ***</p>
<p>Own Id: OTP-5563</p>
</item>
<item>
<p>The functions <c>global:set_lock/3</c> and
<c>global:trans/4</c> now accept the value <c>0</c>
(zero) of the <c>Retries</c> argument.</p>
<p>Own Id: OTP-5737</p>
</item>
<item>
<p>The <c>inet:getaddr(Addr, Family)</c> no longer
validates the <c>Addr</c> argument if it is a 4 or 8
tuple containing the IP address, except for the size of
the tuple and that it contains integers in the correct
range.</p>
<p>The reason for the change is that validation could
cause the following sequence of calls to fail:</p>
<p><c>{ok,Addr} = inet:getaddr(localhost, inet6), gen_tcp:connect(Addr, 7, [inet6])</c></p>
<p>Own Id: OTP-5743</p>
</item>
</list>
</section>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p>The previously undocumented and UNSUPPORTED <c>zlib</c>
module has been updated in an incompatible way and many
bugs have been corrected. It is now also documented.</p>
<p>*** POTENTIAL INCOMPATIBILITY ***</p>
<p>Own Id: OTP-5715</p>
</item>
<item>
<p>Added <c>application</c> interface functions
<c>which_applications/1</c>, <c>set_env/4</c> and
<c>unset_env/3</c>, which take an additional
<c>Timeout</c> argument. To be used in situations where
the standard gen_server timeout (5000ms) is not adequate.</p>
<p>Own Id: OTP-5724 Aux Id: seq10083</p>
</item>
<item>
<p>Improved documentation regarding synchronized start of
applications with included applications (using start
phases and <c>application_starter</c>).</p>
<p>Own Id: OTP-5754</p>
</item>
<item>
<p>New socket options <c>priority</c> and <c>tos</c> for
platforms that support them (currently only Linux).</p>
<p>Own Id: OTP-5756</p>
</item>
<item>
<p>The global name server has been optimized when it comes
to maintaining a fully connected network.</p>
<p>Own Id: OTP-5770</p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.10.10.1</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>The native resolver has gotten an control API for
extended debugging and soft restart. It is:
<c>inet_gethost_native:control(Control)</c> <br></br>
<c>Control = {debug_level,Level} | soft_restart</c> <br></br>
<c>Level = integer() in the range 0-4</c>.</p>
<p>Own Id: OTP-5751 Aux Id: EABln25013</p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.10.10</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>If several processes (at the same node) simultaneously
tried to start the same distributed application, this
could lead to <c>application:start</c> returning an
erroneous value, or even hang.</p>
<p>Own Id: OTP-5606 Aux Id: seq9838</p>
</item>
</list>
</section>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p>The manual pages for most of the Kernel and some of
the STDLIB modules have been updated, in particular
regarding type definitions.</p>
<p>The documentation of the return value for
<c>erts:info/1</c> has been corrected.</p>
<p>The documentation for <c>erlang:statistics/1</c> now
lists all possible arguments.</p>
<p>Own Id: OTP-5360</p>
</item>
<item>
<p>When the native resolver fails a <c>gethostbyaddr</c>
lookup, <c>nxdomain</c> should be returned. There should be
no attempt to fallback on a routine that succeeds if only
the syntax of the IP address is valid. This has been fixed.</p>
<p>Own Id: OTP-5598 Aux Id: OTP-5576</p>
</item>
<item>
<p>Replaced some tuple funs with the new <c>fun M:F/A</c>
construct.</p>
<p>The high-order functions in the <c>lists</c> module no
longer accept bad funs under any circumstances.
'<c>lists:map(bad_fun, [])</c>' used to return
'<c>[]</c>' but now causes an exception.</p>
<p>Unused, broken compatibility code in the <c>ets</c>
module was removed. (Thanks to Dialyzer.)</p>
<p>Eliminated 5 discrepancies found by Dialyzer in the
Appmon application.</p>
<p>Own Id: OTP-5633</p>
</item>
<item>
<p>The possibility to have comments following the list of
tuples in a config file (file specified with
the <c>-config</c> flag) has been added.</p>
<p>Own Id: OTP-5661 Aux Id: seq10003</p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.10.9</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>'<c>erl -config sys.config</c>' would fail to start if
the <c>sys.config</c> file did not contain any whitespace
at all after the dot. (Thanks to Anders Nygren.)</p>
<p>Own Id: OTP-5543</p>
</item>
<item>
<p>A bug regarding tcp sockets which results in hanging
<c>gen_tcp:send/2</c> has been corrected. To encounter
this bug you needed one process that read from a socket,
one that wrote more date than the reader read out so the
sender got suspended, and then the reader closed the
socket. (Reported and diagnosed by Alexey Shchepin.)</p>
<p>Corrected a bug in the (undocumented and unsupported)
option <c>{packet,http}</c> for <c>gen_tcp.</c>
(Thanks to Claes Wikstrom and Luke Gorrie.)</p>
<p>Updated the documentation regarding the second argument to
<c>gen_tcp:recv/2</c>, the <c>Length</c> to receive.</p>
<p>Own Id: OTP-5582 Aux Id: seq9839</p>
</item>
</list>
</section>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p>At startup, the Erlang resolver hosts table was used to
look up the name of the local (and possibly stand alone)
host. This was incorrect. The configured resolver method
is now used for this purpose.</p>
<p>Own Id: OTP-5393</p>
</item>
<item>
<p>The <c>erlang:port_info/1</c> BIF is now documented. Minor
corrections of the documentation for
<c>erlang:port_info/2</c>.</p>
<p>Added a note to the documentation of the <c>math</c> module
that all functions are not available on all platforms.</p>
<p>Added more information about the <c>+c</c> option in
the <c>erl</c> man page in the ERTS documentation.</p>
<p>Own Id: OTP-5555</p>
</item>
<item>
<p>The new <c>fun M:F/A</c> construct creates a fun that
refers to the latest version of <c>M:F/A.</c> This syntax is
meant to replace tuple funs <c>{M,F}</c> which have many
problems.</p>
<p>The new type test <c>is_function(Fun,A)</c> (which may be
used in guards) test whether <c>Fun</c> is a fun that can be
applied with <c>A</c> arguments. (Currently, <c>Fun</c> can
also be a tuple fun.)</p>
<p>Own Id: OTP-5584</p>
</item>
<item>
<p>According to the documentation <c>global</c> implements
the equivalent of <c>register/2</c>, which returns
<c>badarg</c> if a process is already registered. As it
turns out there is no check in <c>global</c> if a process is
registered under more than one name. If some process is
accidentaly or by design given several names, it is
possible that the name registry becomes inconsistent due
to the way the resolve function is called when name
clashes are discovered (see <c>register_name/3</c> in
<c>global(3)</c>).</p>
<p>In OTP R11B <c>global</c> will not allow the registration of
a process under more than one name. To help finding code
where <c>no</c> will be returned, a Kernel application
variable, <c>global_multi_name_action</c>, is hereby
introduced. Depending on its value (<c>info</c>,
<c>warning</c>, or <c>error</c>), messages are sent to
the error logger when <c>global</c> discovers that some
process is given more than one name. The variable only
affects the node where it is defined.</p>
<p>Own Id: OTP-5603</p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.10.8</title>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p>In case of a DNS lookup loop, <c>inet_db:getbyname</c> ends
up building an infinite list. This has been fixed.</p>
<p>Own Id: OTP-5449</p>
</item>
<item>
<p>When doing an <c>inet6</c> name lookup on an IPv4 address
it was possible to get an address on IPv4 format back. This
has been corrected. Some other minor inconsistencies
regarding IPv6 name lookup have also been corrected.</p>
<p>Own Id: OTP-5576</p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.10.7</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>Under certain circumstances the <c>net_kernel</c> could
emit spurious nodedown messages. This bug has been fixed.</p>
<p>Own Id: OTP-5396</p>
</item>
<item>
<p>Removed description of the <c>keep_zombies</c>
configuration parameter in the <c>kernel</c> man page.</p>
<p>Own Id: OTP-5497</p>
</item>
</list>
</section>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p>Eliminated Dialyzer warnings (caused by dead code) in
the <c>init</c> and <c>prim_file</c> modules.</p>
<p>Own Id: OTP-5496</p>
</item>
<item>
<p><c>inet_config</c> now also checks the environment variable
<c>ERL_INETRC</c> for a possible user configuration file.
See the ERTS User's Guide for details.</p>
<p>Own Id: OTP-5512</p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.10.6</title>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p>The <c>c</c> option for the <c>+B</c> flag has been
introduced which makes it possible to use Ctrl-C
(Ctrl-Break on Windows) to interrupt the shell process
rather than to invoke the emulator break handler. All new
<c>+B</c> options are also supported on Windows (werl) as
of now. Furthermore, Ctrl-C on Windows has now been
reserved for copying text (what Ctrl-Ins was used for
previously). Ctrl-Break should be used for break handling.
Lastly, the documentation of the system flags has been
updated.</p>
<p>Own Id: OTP-5388</p>
</item>
<item>
<p>The possibility to start the Erlang shell in parallel
with the rest of the system was reintroduced for backwards
compatibility in STDLIB 1.13.1. The flag to be used for
this is now called <c>async_shell_start</c> and has
been documented. New shells started from the JCL menu are
not synchronized with <c>init</c> anymore. This makes it
possible to start a new shell (e.g. for debugging purposes)
even if the initial shell has not come up.</p>
<p>Own Id: OTP-5406 Aux Id: OTP-5218</p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.10.5</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>Documentation for <c>erlang:binary_to_float/1</c> deleted.
The BIF itself was removed several releases ago.</p>
<p>Updated documentation for <c>apply/2</c> and
<c>apply/3</c>.</p>
<p>Own Id: OTP-5391</p>
</item>
</list>
</section>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p><c>net_kernel:monitor_nodes/2</c> which takes a flag and an
option list has been added. By use of
<c>net_kernel:monitor_nodes/2</c> one can subscribe for
<c>nodeup/nodedown</c> messages with extra information. It
is now possible to monitor hidden nodes, and get
<c>nodedown</c> reason. See the <c>net_kernel(3)</c>
documentation for more information.</p>
<p>Own Id: OTP-5374</p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.10.4</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>The application master for an application now terminates
the application faster, which reduces the risk for
timeouts in other parts of the system.</p>
<p>Own Id: OTP-5363 Aux Id: EABln19084</p>
</item>
<item>
<p>A BIF <c>erlang:raise/3</c> has been added. See the manual
for details. It is intended for internal system programming
only, advanced error handling.</p>
<p>Own Id: OTP-5376 Aux Id: OTP-5257</p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.10.3</title>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p>With the <c>-eval</c> flag (<c>erl -eval Expr</c>), an
arbitrary expression can be evaluated during system
initialization. This is documented in <c>init(3)</c>.</p>
<p>Own Id: OTP-5260</p>
</item>
<item>
<p>The unsupported and undocumented modules <c>socks5</c>,
<c>socks5_auth</c>, <c>socks5_tcp</c>, and <c>socks5_udp</c>
have been removed.</p>
<p>Own Id: OTP-5266</p>
</item>
</list>
</section>
</section>
<section>
<title>Kernel 2.10.1</title>
<section>
<title>Fixed Bugs and Malfunctions</title>
<list type="bulleted">
<item>
<p>The Pman 'trace shell' functionality was broken and has
now been fixed. Furthermore, Pman could not correctly
find the pid of the active shell if more than one shell
process was running on the node. This has also been
corrected.</p>
<p>Own Id: OTP-5191</p>
</item>
<item>
<p>The documentation for the <c>auth:open/1</c> function
which no longer exists has been removed. (Thanks to
Miguel Barreiro.)</p>
<p>Own Id: OTP-5208</p>
</item>
<item>
<p>Corrected the <c>crc32/3</c> function in the undocumented
and unsupported <c>zlib</c> module.</p>
<p>Own Id: OTP-5227</p>
</item>
</list>
</section>
<section>
<title>Improvements and New Features</title>
<list type="bulleted">
<item>
<p>You can now start Erlang with the <c>-rsh</c> flag which
gives you a remote initial shell instead of a local one.
Example:</p>
<pre>
erl -sname this_node -rsh other_node@other_host
</pre>
<p>Own Id: OTP-5210</p>
</item>
<item>
<p>If <c>/etc/hosts</c> specified two hosts with the same IP
address (on separate lines), only the last host would be
registered by inet_db during inet configuration. This has
been corrected now so that both aliases are registered
with the same IP address.</p>
<p>Own Id: OTP-5212 Aux Id: seq7128</p>
</item>
<item>
<p>The documentation for BIFs that take I/O lists have
been clarified. Those are <c>list_to_binary/1</c>,
<c>port_command/2</c>, <c>port_control/3</c>.</p>
<p>Documentation for all <c>is_*</c> BIFs (such as
<c>is_atom/1</c>) has been added.</p>
<p>Removed the documentation for
<c>erlang:float_to_binary/2</c> which was removed from
the run-time system several releases ago.</p>
<p>Own Id: OTP-5222</p>
</item>
</list>
</section>
</section>
</chapter>
