diff options
Diffstat (limited to 'system')
20 files changed, 345 insertions, 204 deletions
diff --git a/system/COPYRIGHT b/system/COPYRIGHT index ef76b66f6b..ed06dcf69c 100644 --- a/system/COPYRIGHT +++ b/system/COPYRIGHT @@ -22,6 +22,17 @@ limitations under the License. %CopyrightEnd% --------------------------------------------------------------------------- +[stdlib, compiler] + +* assert.hrl is Copyright (C) 2004-1016 Richard Carlsson, Mickaël Rémond +* array.erl is Copyright (C) 2006-2016 Richard Carlsson and Ericsson AB +* gb_trees.erl is Copyright (C) 1999-2001 Sven-Olof Nyström, Richard Carlsson +* gb_sets.erl is Copyright (C) 1999-2001 Richard Carlsson, Sven-Olof Nyström +* proplists.erl is Copyright (C) 2000-2003 Richard Carlsson +* cerl{_trees,_clauses}.erl are Copyright (C) 1999-2002 Richard Carlsson +* cerl_inline.erl is Copyright (C) 1999-2002 Richard Carlsson + +--------------------------------------------------------------------------- [PCRE] PCRE LICENCE @@ -190,32 +201,108 @@ terms specified in this license. */ --------------------------------------------------------------------------- +[dialyzer] + +%% Copyright 1997-2016 Tobias Lindahl, Stavros Aronis, Kostis Sagonas, +%% Richard Carlsson, et al. +%% +%% Licensed under the Apache License, Version 2.0 (the "License"); +%% you may not use this file except in compliance with the License. +%% You may obtain a copy of the License at +%% +%% http://www.apache.org/licenses/LICENSE-2.0 +%% +%% Unless required by applicable law or agreed to in writing, software +%% distributed under the License is distributed on an "AS IS" BASIS, +%% WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +%% See the License for the specific language governing permissions and +%% limitations under the License. + +--------------------------------------------------------------------------- +[typer] + +%% Copyright 2006-2016 Bingwen He, Tobias Lindahl, Kostis Sagonas +%% +%% Licensed under the Apache License, Version 2.0 (the "License"); +%% you may not use this file except in compliance with the License. +%% You may obtain a copy of the License at +%% +%% http://www.apache.org/licenses/LICENSE-2.0 +%% +%% Unless required by applicable law or agreed to in writing, software +%% distributed under the License is distributed on an "AS IS" BASIS, +%% WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +%% See the License for the specific language governing permissions and +%% limitations under the License. + +--------------------------------------------------------------------------- +[hipe] + +%% Copyright 1997-2016 Erik Stenman (Johansson), Kostis Sagonas, +%% Richard Carlsson, Tobias Lindahl, Per Gustafsson, et al. +%% +%% Licensed under the Apache License, Version 2.0 (the "License"); +%% you may not use this file except in compliance with the License. +%% You may obtain a copy of the License at +%% +%% http://www.apache.org/licenses/LICENSE-2.0 +%% +%% Unless required by applicable law or agreed to in writing, software +%% distributed under the License is distributed on an "AS IS" BASIS, +%% WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +%% See the License for the specific language governing permissions and +%% limitations under the License. + +--------------------------------------------------------------------------- [edoc, syntax_tools] -%% ===================================================================== -%% This library is free software; you can redistribute it and/or modify -%% it under the terms of the GNU Lesser General Public License as -%% published by the Free Software Foundation; either version 2 of the -%% License, or (at your option) any later version. +%% Copyright 1997-2016 Richard Carlsson <[email protected]> +%% +%% Licensed under the Apache License, Version 2.0 (the "License"); you may +%% not use this file except in compliance with the License. You may obtain +%% a copy of the License at <http://www.apache.org/licenses/LICENSE-2.0> +%% +%% Unless required by applicable law or agreed to in writing, software +%% distributed under the License is distributed on an "AS IS" BASIS, +%% WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +%% See the License for the specific language governing permissions and +%% limitations under the License. %% -%% This library is distributed in the hope that it will be useful, but -%% WITHOUT ANY WARRANTY; without even the implied warranty of -%% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU -%% Lesser General Public License for more details. +%% Alternatively, you may use this file under the terms of the GNU Lesser +%% General Public License (the "LGPL") as published by the Free Software +%% Foundation; either version 2.1, or (at your option) any later version. +%% If you wish to allow use of your version of this file only under the +%% terms of the LGPL, you should delete the provisions above and replace +%% them with the notice and other provisions required by the LGPL; see +%% <http://www.gnu.org/licenses/>. If you do not delete the provisions +%% above, a recipient may use your version of this file under the terms of +%% either the Apache License or the LGPL. + +--------------------------------------------------------------------------- +[eunit] + +%% Copyright 2004-2016 Richard Carlsson <[email protected]>, +%% Mickaël Rémond <[email protected]> %% -%% You should have received a copy of the GNU Lesser General Public -%% License along with this library; if not, write to the Free Software -%% Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 -%% USA +%% Licensed under the Apache License, Version 2.0 (the "License"); you may +%% not use this file except in compliance with the License. You may obtain +%% a copy of the License at <http://www.apache.org/licenses/LICENSE-2.0> %% -%% $Id$ +%% Unless required by applicable law or agreed to in writing, software +%% distributed under the License is distributed on an "AS IS" BASIS, +%% WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +%% See the License for the specific language governing permissions and +%% limitations under the License. %% -%% @copyright 2001-2003 Richard Carlsson -%% @author Richard Carlsson <[email protected]> -%% [http://www.csd.uu.se/~richardc/] -%% @version {@vsn} -%% @end -%% ===================================================================== +%% Alternatively, you may use this file under the terms of the GNU Lesser +%% General Public License (the "LGPL") as published by the Free Software +%% Foundation; either version 2.1, or (at your option) any later version. +%% If you wish to allow use of your version of this file only under the +%% terms of the LGPL, you should delete the provisions above and replace +%% them with the notice and other provisions required by the LGPL; see +%% <http://www.gnu.org/licenses/>. If you do not delete the provisions +%% above, a recipient may use your version of this file under the terms of +%% either the Apache License or the LGPL. --------------------------------------------------------------------------- [leex] @@ -269,3 +356,4 @@ OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. --------------------------------------------------------------------------- +[typer]
\ No newline at end of file diff --git a/system/doc/design_principles/sup_princ.xml b/system/doc/design_principles/sup_princ.xml index 0a24e97950..478d1bf714 100644 --- a/system/doc/design_principles/sup_princ.xml +++ b/system/doc/design_principles/sup_princ.xml @@ -163,7 +163,9 @@ SupFlags = #{strategy => Strategy, ...}</code> SupFlags = #{intensity => MaxR, period => MaxT, ...}</code> <p>If more than <c>MaxR</c> number of restarts occur in the last <c>MaxT</c> seconds, the supervisor terminates all the child - processes and then itself.</p> + processes and then itself. + The termination reason for the supervisor itself in that case will be + <c>shutdown</c>.</p> <p>When the supervisor terminates, then the next higher-level supervisor takes some action. It either restarts the terminated supervisor or terminates itself.</p> @@ -173,6 +175,69 @@ SupFlags = #{intensity => MaxR, period => MaxT, ...}</code> <p>The keys <c>intensity</c> and <c>period</c> are optional in the supervisor flags map. If they are not given, they default to <c>1</c> and <c>5</c>, respectively.</p> + <section> + <title>Tuning the intensity and period</title> + <p>The default values are 1 restart per 5 seconds. This was chosen to + be safe for most systems, even with deep supervision hierarchies, + but you will probably want to tune the settings for your particular + use case.</p> + <p>First, the intensity decides how big bursts of restarts you want + to tolerate. For example, you might want to accept a burst of at + most 5 or 10 attempts, even within the same second, if it results + in a successful restart.</p> + <p>Second, you need to consider the sustained failure rate, if + crashes keep happening but not often enough to make the supervisor + give up. If you set intensity to 10 and set the period as low as 1, + the supervisor will allow child processes to keep restarting up to + 10 times per second, forever, filling your logs with crash reports + until someone intervenes manually.</p> + <p>You should therefore set the period to be long enough that you can + accept that the supervisor keeps going at that rate. For example, + if you have picked an intensity value of 5, then setting the period + to 30 seconds will give you at most one restart per 6 seconds for + any longer period of time, which means that your logs won't fill up + too quickly, and you will have a chance to observe the failures and + apply a fix.</p> + <p>These choices depend a lot on your problem domain. If you don't + have real time monitoring and ability to fix problems quickly, for + example in an embedded system, you might want to accept at most + one restart per minute before the supervisor should give up and + escalate to the next level to try to clear the error automatically. + On the other hand, if it is more important that you keep trying + even at a high failure rate, you might want a sustained rate of as + much as 1-2 restarts per second.</p> + <p>Avoiding common mistakes: + <list type="bulleted"> + <item> + <p>Do not forget to consider the burst rate. If you set intensity + to 1 and period to 6, it gives the same sustained error rate as + 5/30 or 10/60, but will not allow even 2 restart attempts in + quick succession. This is probably not what you wanted.</p> + </item> + <item> + <p>Do not set the period to a very high value if you want to + tolerate bursts. If you set intensity to 5 and period to 3600 + (one hour), the supervisor will allow a short burst of 5 + restarts, but then gives up if it sees another single restart + almost an hour later. You probably want to regard those crashes + as separate incidents, so setting the period to 5 or 10 minutes + will be more reasonable.</p> + </item> + <item> + <p>If your application has multiple levels of supervision, then + do not simply set the restart intensities to the same values on + all levels. Keep in mind that the total number of restarts + (before the top level supervisor gives up and terminates the + application) will be the product of the intensity values of all + the supervisors above the failing child process.</p> + <p>For example, if the top level allows 10 restarts, and the next + level also allows 10, a crashing child below that level will be + restarted 100 times, which is probably excessive. Allowing at + most 3 restarts for the top level supervisor might be a better + choice in this case.</p> + </item> + </list></p> + </section> </section> <section> diff --git a/system/doc/efficiency_guide/bench.erl b/system/doc/efficiency_guide/bench.erl index 1f60e858f6..a1be24b051 100644 --- a/system/doc/efficiency_guide/bench.erl +++ b/system/doc/efficiency_guide/bench.erl @@ -355,7 +355,7 @@ create_html_report(ResultList) -> {ok, OutputFile} = file:open("index.html", [write]), - %% Create the begining of the result html-file. + %% Create the beginning of the result html-file. Head = Title = "Benchmark Results", io:put_chars(OutputFile, "<html>\n"), io:put_chars(OutputFile, "<head>\n"), diff --git a/system/doc/efficiency_guide/binaryhandling.xml b/system/doc/efficiency_guide/binaryhandling.xml index 0295d18644..91fd9a7cd9 100644 --- a/system/doc/efficiency_guide/binaryhandling.xml +++ b/system/doc/efficiency_guide/binaryhandling.xml @@ -32,12 +32,9 @@ <file>binaryhandling.xml</file> </header> - <p>In R12B, the most natural way to construct and match binaries is - significantly faster than in earlier releases.</p> + <p>Binaries can be efficiently built in the following way:</p> - <p>To construct a binary, you can simply write as follows:</p> - - <p><em>DO</em> (in R12B) / <em>REALLY DO NOT</em> (in earlier releases)</p> + <p><em>DO</em></p> <code type="erl"><![CDATA[ my_list_to_binary(List) -> my_list_to_binary(List, <<>>). @@ -47,21 +44,13 @@ my_list_to_binary([H|T], Acc) -> my_list_to_binary([], Acc) -> Acc.]]></code> - <p>In releases before R12B, <c>Acc</c> is copied in every iteration. - In R12B, <c>Acc</c> is copied only in the first iteration and extra - space is allocated at the end of the copied binary. In the next iteration, - <c>H</c> is written into the extra space. When the extra space runs out, - the binary is reallocated with more extra space. The extra space allocated - (or reallocated) is twice the size of the - existing binary data, or 256, whichever is larger.</p> - - <p>The most natural way to match binaries is now the fastest:</p> + <p>Binaries can be efficiently matched like this:</p> - <p><em>DO</em> (in R12B)</p> + <p><em>DO</em></p> <code type="erl"><![CDATA[ my_binary_to_list(<<H,T/binary>>) -> [H|my_binary_to_list(T)]; -my_binary_to_list(<<>>) -> [].]]></code> +my_binary_to_list(<<>>) -> [].]]></code> <section> <title>How Binaries are Implemented</title> @@ -138,10 +127,7 @@ my_binary_to_list(<<>>) -> [].]]></code> pointer to the binary data. For each field that is matched out of a binary, the position in the match context is incremented.</p> - <p>In R11B, a match context was only used during a binary matching - operation.</p> - - <p>In R12B, the compiler tries to avoid generating code that + <p>The compiler tries to avoid generating code that creates a sub binary, only to shortly afterwards create a new match context and discard the sub binary. Instead of creating a sub binary, the match context is kept.</p> @@ -155,7 +141,7 @@ my_binary_to_list(<<>>) -> [].]]></code> <section> <title>Constructing Binaries</title> - <p>In R12B, appending to a binary or bitstring + <p>Appending to a binary or bitstring is specially optimized by the <em>runtime system</em>:</p> <code type="erl"><![CDATA[ @@ -292,7 +278,7 @@ Bin = <<Bin1,...>> %% Bin1 will be COPIED <p>Let us revisit the example in the beginning of the previous section:</p> - <p><em>DO</em> (in R12B)</p> + <p><em>DO</em></p> <code type="erl"><![CDATA[ my_binary_to_list(<<H,T/binary>>) -> [H|my_binary_to_list(T)]; @@ -304,15 +290,14 @@ my_binary_to_list(<<>>) -> [].]]></code> byte of the binary. 1 byte is matched out and the match context is updated to point to the second byte in the binary.</p> - <p>In R11B, at this point a - <seealso marker="#sub_binary">sub binary</seealso> - would be created. In R12B, - the compiler sees that there is no point in creating a sub binary, - because there will soon be a call to a function (in this case, + <p>At this point it would make sense to create a + <seealso marker="#sub_binary">sub binary</seealso>, + but in this particular example the compiler sees that + there will soon be a call to a function (in this case, to <c>my_binary_to_list/1</c> itself) that immediately will create a new match context and discard the sub binary.</p> - <p>Therefore, in R12B, <c>my_binary_to_list/1</c> calls itself + <p>Therefore <c>my_binary_to_list/1</c> calls itself with the match context instead of with a sub binary. The instruction that initializes the matching operation basically does nothing when it sees that it was passed a match context instead of a binary.</p> @@ -321,34 +306,10 @@ my_binary_to_list(<<>>) -> [].]]></code> the match context will simply be discarded (removed in the next garbage collection, as there is no longer any reference to it).</p> - <p>To summarize, <c>my_binary_to_list/1</c> in R12B only needs to create - <em>one</em> match context and no sub binaries. In R11B, if the binary - contains <em>N</em> bytes, <em>N+1</em> match contexts and <em>N</em> - sub binaries are created.</p> - - <p>In R11B, the fastest way to match binaries is as follows:</p> + <p>To summarize, <c>my_binary_to_list/1</c> only needs to create + <em>one</em> match context and no sub binaries.</p> - <p><em>DO NOT</em> (in R12B)</p> - <code type="erl"><![CDATA[ -my_complicated_binary_to_list(Bin) -> - my_complicated_binary_to_list(Bin, 0). - -my_complicated_binary_to_list(Bin, Skip) -> - case Bin of - <<_:Skip/binary,Byte,_/binary>> -> - [Byte|my_complicated_binary_to_list(Bin, Skip+1)]; - <<_:Skip/binary>> -> - [] - end.]]></code> - - <p>This function cleverly avoids building sub binaries, but it cannot - avoid building a match context in each recursion step. - Therefore, in both R11B and R12B, - <c>my_complicated_binary_to_list/1</c> builds <em>N+1</em> match - contexts. (In a future Erlang/OTP release, the compiler might be able - to generate code that reuses the match context.)</p> - - <p>Returning to <c>my_binary_to_list/1</c>, notice that the match context + <p>Notice that the match context in <c>my_binary_to_list/1</c> was discarded when the entire binary had been traversed. What happens if the iteration stops before it has reached the end of the binary? Will the optimization still work?</p> @@ -544,5 +505,15 @@ count3(<<>>, Count) -> Count.]]></code> not matched out.</p> </section> </section> + + <section> + <title>Historical Note</title> + + <p>Binary handling was significantly improved in R12B. Because + code that was efficient in R11B might not be efficient in R12B, + and vice versa, earlier revisions of this Efficiency Guide contained + some information about binary handling in R11B.</p> + </section> + </chapter> diff --git a/system/doc/efficiency_guide/commoncaveats.xml b/system/doc/efficiency_guide/commoncaveats.xml index ecfeff0349..94b1c0b222 100644 --- a/system/doc/efficiency_guide/commoncaveats.xml +++ b/system/doc/efficiency_guide/commoncaveats.xml @@ -148,10 +148,10 @@ multiple_setelement(T0) -> <p><c>size/1</c> returns the size for both tuples and binaries.</p> - <p>Using the new BIFs <c>tuple_size/1</c> and <c>byte_size/1</c>, introduced - in R12B, gives the compiler and the runtime system more opportunities for - optimization. Another advantage is that the new BIFs can help Dialyzer to - find more bugs in your program.</p> + <p>Using the BIFs <c>tuple_size/1</c> and <c>byte_size/1</c> + gives the compiler and the runtime system more opportunities for + optimization. Another advantage is that the BIFs give Dialyzer more + type information.</p> </section> <section> diff --git a/system/doc/efficiency_guide/functions.xml b/system/doc/efficiency_guide/functions.xml index 4a8248e65c..1d0f1f68b7 100644 --- a/system/doc/efficiency_guide/functions.xml +++ b/system/doc/efficiency_guide/functions.xml @@ -65,7 +65,7 @@ atom_map1(six) -> 6.</code> thus, quite efficient even if there are many values) to select which one of the first three clauses to execute (if any).</item> - <item>>If none of the first three clauses match, the fourth clause + <item>If none of the first three clauses match, the fourth clause match as a variable always matches.</item> <item>If the guard test <c>is_integer(Int)</c> succeeds, the fourth @@ -183,15 +183,6 @@ explicit_map_pairs(Map, Xs0, Ys0) -> A fun contains an (indirect) pointer to the function that implements the fun.</p> - <warning><p><em>Tuples are not fun(s)</em>. - A "tuple fun", <c>{Module,Function}</c>, is not a fun. - The cost for calling a "tuple fun" is similar to that - of <c>apply/3</c> or worse. - Using "tuple funs" is <em>strongly discouraged</em>, - as they might not be supported in a future Erlang/OTP release, - and because there exists a superior alternative from R10B, - namely the <c>fun Module:Function/Arity</c> syntax.</p></warning> - <p><c>apply/3</c> must look up the code for the function to execute in a hash table. It is therefore always slower than a direct call or a fun call.</p> diff --git a/system/doc/efficiency_guide/introduction.xml b/system/doc/efficiency_guide/introduction.xml index ca4a41c798..b650008ae8 100644 --- a/system/doc/efficiency_guide/introduction.xml +++ b/system/doc/efficiency_guide/introduction.xml @@ -46,14 +46,6 @@ to find out where the performance bottlenecks are and optimize only the bottlenecks. Let other code stay as clean as possible.</p> - <p>Fortunately, compiler and runtime optimizations introduced in - Erlang/OTP R12B makes it easier to write code that is both clean and - efficient. For example, the ugly workarounds needed in R11B and earlier - releases to get the most speed out of binary pattern matching are - no longer necessary. In fact, the ugly code is slower - than the clean code (because the clean code has become faster, not - because the uglier code has become slower).</p> - <p>This Efficiency Guide cannot really teach you how to write efficient code. It can give you a few pointers about what to avoid and what to use, and some understanding of how certain language features are implemented. diff --git a/system/doc/efficiency_guide/listhandling.xml b/system/doc/efficiency_guide/listhandling.xml index 2ebc877820..ec258d7c2a 100644 --- a/system/doc/efficiency_guide/listhandling.xml +++ b/system/doc/efficiency_guide/listhandling.xml @@ -90,7 +90,7 @@ tail_recursive_fib(N, Current, Next, Fibs) -> <p>Lists comprehensions still have a reputation for being slow. They used to be implemented using funs, which used to be slow.</p> - <p>In recent Erlang/OTP releases (including R12B), a list comprehension:</p> + <p>A list comprehension:</p> <code type="erl"><![CDATA[ [Expr(E) || E <- List]]]></code> @@ -102,7 +102,7 @@ tail_recursive_fib(N, Current, Next, Fibs) -> [Expr(E)|'lc^0'(Tail, Expr)]; 'lc^0'([], _Expr) -> [].</code> - <p>In R12B, if the result of the list comprehension will <em>obviously</em> + <p>If the result of the list comprehension will <em>obviously</em> not be used, a list will not be constructed. For example, in this code:</p> <code type="erl"><![CDATA[ @@ -131,6 +131,14 @@ some_function(...), 'lc^0'(Tail, Expr); 'lc^0'([], _Expr) -> [].</code> + <p>The compiler also understands that assigning to '_' means that + the value will not used. Therefore, the code in the following example + will also be optimized:</p> + + <code type="erl"><![CDATA[ +_ = [io:put_chars(E) || E <- List], +ok.]]></code> + </section> <section> @@ -209,11 +217,11 @@ some_function(...), <section> <title>Recursive List Functions</title> - <p>In Section 7.2, the following myth was exposed: + <p>In section about myths, the following myth was exposed: <seealso marker="myths#tail_recursive">Tail-Recursive Functions are Much Faster Than Recursive Functions</seealso>.</p> - <p>To summarize, in R12B there is usually not much difference between + <p>There is usually not much difference between a body-recursive list function and tail-recursive function that reverses the list at the end. Therefore, concentrate on writing beautiful code and forget about the performance of your list functions. In the diff --git a/system/doc/efficiency_guide/myths.xml b/system/doc/efficiency_guide/myths.xml index 5d3ad78b23..778cd06c09 100644 --- a/system/doc/efficiency_guide/myths.xml +++ b/system/doc/efficiency_guide/myths.xml @@ -24,7 +24,7 @@ The Initial Developer of the Original Code is Ericsson AB. </legalnotice> - <title>The Eight Myths of Erlang Performance</title> + <title>The Seven Myths of Erlang Performance</title> <prepared>Bjorn Gustavsson</prepared> <docno></docno> <date>2007-11-10</date> @@ -35,80 +35,33 @@ <marker id="myths"></marker> <p>Some truths seem to live on well beyond their best-before date, perhaps because "information" spreads faster from person-to-person - than a single release note that says, for example, that funs - have become faster.</p> + than a single release note that says, for example, that body-recursive + calls have become faster.</p> <p>This section tries to kill the old truths (or semi-truths) that have become myths.</p> <section> - <title>Myth: Funs are Slow</title> - <p>Funs used to be very slow, slower than <c>apply/3</c>. - Originally, funs were implemented using nothing more than - compiler trickery, ordinary tuples, <c>apply/3</c>, and a great - deal of ingenuity.</p> - - <p>But that is history. Funs was given its own data type - in R6B and was further optimized in R7B. - Now the cost for a fun call falls roughly between the cost for a call - to a local function and <c>apply/3</c>.</p> - </section> - - <section> - <title>Myth: List Comprehensions are Slow</title> - - <p>List comprehensions used to be implemented using funs, and in the - old days funs were indeed slow.</p> - - <p>Nowadays, the compiler rewrites list comprehensions into an ordinary - recursive function. Using a tail-recursive function with - a reverse at the end would be still faster. Or would it? - That leads us to the next myth.</p> - </section> - - <section> <title>Myth: Tail-Recursive Functions are Much Faster Than Recursive Functions</title> <p><marker id="tail_recursive"></marker>According to the myth, - recursive functions leave references - to dead terms on the stack and the garbage collector has to copy - all those dead terms, while tail-recursive functions immediately - discard those terms.</p> - - <p>That used to be true before R7B. In R7B, the compiler started - to generate code that overwrites references to terms that will never - be used with an empty list, so that the garbage collector would not - keep dead values any longer than necessary.</p> - - <p>Even after that optimization, a tail-recursive function is - still most of the times faster than a body-recursive function. Why?</p> - - <p>It has to do with how many words of stack that are used in each - recursive call. In most cases, a recursive function uses more words - on the stack for each recursion than the number of words a tail-recursive - would allocate on the heap. As more memory is used, the garbage - collector is invoked more frequently, and it has more work traversing - the stack.</p> - - <p>In R12B and later releases, there is an optimization that - in many cases reduces the number of words used on the stack in - body-recursive calls. A body-recursive list function and a - tail-recursive function that calls <seealso - marker="stdlib:lists#reverse/1">lists:reverse/1</seealso> at - the end will use the same amount of memory. - <c>lists:map/2</c>, <c>lists:filter/2</c>, list comprehensions, - and many other recursive functions now use the same amount of space - as their tail-recursive equivalents.</p> - - <p>So, which is faster? - It depends. On Solaris/Sparc, the body-recursive function seems to - be slightly faster, even for lists with a lot of elements. On the x86 - architecture, tail-recursion was up to about 30% faster.</p> - - <p>So, the choice is now mostly a matter of taste. If you really do need - the utmost speed, you must <em>measure</em>. You can no longer be - sure that the tail-recursive list function always is the fastest.</p> + using a tail-recursive function that builds a list in reverse + followed by a call to <c>lists:reverse/1</c> is faster than + a body-recursive function that builds the list in correct order; + the reason being that body-recursive functions use more memory than + tail-recursive functions.</p> + + <p>That was true to some extent before R12B. It was even more true + before R7B. Today, not so much. A body-recursive function + generally uses the same amount of memory as a tail-recursive + function. It is generally not possible to predict whether the + tail-recursive or the body-recursive version will be + faster. Therefore, use the version that makes your code cleaner + (hint: it is usually the body-recursive version).</p> + + <p>For a more thorough discussion about tail and body recursion, + see <url href="http://ferd.ca/erlang-s-tail-recursion-is-not-a-silver-bullet.html">Erlang's Tail Recursion is Not a Silver Bullet</url>.</p> <note><p>A tail-recursive function that does not need to reverse the list at the end is faster than a body-recursive function, @@ -199,6 +152,29 @@ vanilla_reverse([], Acc) -> <p>That was once true, but from R6B the BEAM compiler can see that a variable is not used.</p> + + <p>Similarly, trivial transformations on the source-code level + such as converting a <c>case</c> statement to clauses at the + top-level of the function seldom makes any difference to the + generated code.</p> + </section> + + <section> + <title>Myth: A NIF Always Speeds Up Your Program</title> + + <p>Rewriting Erlang code to a NIF to make it faster should be + seen as a last resort. It is only guaranteed to be dangerous, + but not guaranteed to speed up the program.</p> + + <p>Doing too much work in each NIF call will + <seealso marker="erts:erl_nif#WARNING">degrade responsiveness + of the VM</seealso>. Doing too little work may mean that + the gain of the faster processing in the NIF is eaten up by + the overhead of calling the NIF and checking the arguments.</p> + + <p>Be sure to read about + <seealso marker="erts:erl_nif#lengthy_work">Long-running NIFs</seealso> + before writing a NIF.</p> </section> </chapter> diff --git a/system/doc/efficiency_guide/part.xml b/system/doc/efficiency_guide/part.xml index 6e10a0c031..5673ddd320 100644 --- a/system/doc/efficiency_guide/part.xml +++ b/system/doc/efficiency_guide/part.xml @@ -39,5 +39,6 @@ <xi:include href="drivers.xml"/> <xi:include href="advanced.xml"/> <xi:include href="profiling.xml"/> + <xi:include href="retired_myths.xml"/> </part> diff --git a/system/doc/efficiency_guide/processes.xml b/system/doc/efficiency_guide/processes.xml index f2d9712f51..afa4325d8e 100644 --- a/system/doc/efficiency_guide/processes.xml +++ b/system/doc/efficiency_guide/processes.xml @@ -146,14 +146,14 @@ loop() -> <section> <title>Constant Pool</title> - <p>Constant Erlang terms (also called <em>literals</em>) are now + <p>Constant Erlang terms (also called <em>literals</em>) are kept in constant pools; each loaded module has its own pool. - The following function does no longer build the tuple every time + The following function does not build the tuple every time it is called (only to have it discarded the next time the garbage collector was run), but the tuple is located in the module's constant pool:</p> - <p><em>DO</em> (in R12B and later)</p> + <p><em>DO</em></p> <code type="erl"> days_in_month(M) -> element(M, {31,28,31,30,31,30,31,31,30,31,30,31}).</code> @@ -235,9 +235,7 @@ true return the same value. Sharing has been lost.</p> <p>In a future Erlang/OTP release, it might be implemented a - way to (optionally) preserve sharing. There are no plans to make - preserving of sharing the default behaviour, as that would - penalize the vast majority of Erlang applications.</p> + way to (optionally) preserve sharing.</p> </section> </section> @@ -261,10 +259,6 @@ true The estone benchmark, for example, is entirely sequential. So is the most common implementation of the "ring benchmark"; usually one process is active, while the others wait in a <c>receive</c> statement.</p> - - <p>The <seealso marker="percept:percept">percept</seealso> application - can be used to profile your application to see how much potential (or lack - thereof) it has for concurrency.</p> </section> </chapter> diff --git a/system/doc/efficiency_guide/retired_myths.xml b/system/doc/efficiency_guide/retired_myths.xml new file mode 100644 index 0000000000..37f46566cd --- /dev/null +++ b/system/doc/efficiency_guide/retired_myths.xml @@ -0,0 +1,63 @@ +<?xml version="1.0" encoding="utf-8" ?> +<!DOCTYPE chapter SYSTEM "chapter.dtd"> + +<chapter> + <header> + <copyright> + <year>2016</year> + <year>2016</year> + <holder>Ericsson AB, All Rights Reserved</holder> + </copyright> + <legalnotice> + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. + + The Initial Developer of the Original Code is Ericsson AB. + </legalnotice> + <marker id="retired_myths"/> + <title>Retired Myths</title> + <prepared>Bjorn Gustavsson</prepared> + <docno></docno> + <date>2016-06-07</date> + <rev></rev> + <file>retired_myths.xml</file> + </header> + + <p>We belive that the truth finally has caught with the following, + retired myths.</p> + + <section> + <title>Myth: Funs are Slow</title> + <p>Funs used to be very slow, slower than <c>apply/3</c>. + Originally, funs were implemented using nothing more than + compiler trickery, ordinary tuples, <c>apply/3</c>, and a great + deal of ingenuity.</p> + + <p>But that is history. Funs was given its own data type + in R6B and was further optimized in R7B. + Now the cost for a fun call falls roughly between the cost for a call + to a local function and <c>apply/3</c>.</p> + </section> + + <section> + <title>Myth: List Comprehensions are Slow</title> + + <p>List comprehensions used to be implemented using funs, and in the + old days funs were indeed slow.</p> + + <p>Nowadays, the compiler rewrites list comprehensions into an ordinary + recursive function. Using a tail-recursive function with + a reverse at the end would be still faster. Or would it? + That leads us to the myth that tail-recursive functions are faster + than body-recursive functions.</p> + </section> +</chapter> diff --git a/system/doc/reference_manual/character_set.xml b/system/doc/reference_manual/character_set.xml index d25f2c001d..f0f4c23608 100644 --- a/system/doc/reference_manual/character_set.xml +++ b/system/doc/reference_manual/character_set.xml @@ -32,9 +32,9 @@ <section> <title>Character Set</title> - <p>Since Erlang 4.8/OTP R5A, the syntax of Erlang tokens is extended to - allow the use of the full ISO-8859-1 (Latin-1) character set. This - is noticeable in the following ways:</p> + <p>The syntax of Erlang tokens allow the use of the full + ISO-8859-1 (Latin-1) character set. This is noticeable in the + following ways:</p> <list type="bulleted"> <item> <p>All the Latin-1 printable characters can be used and are diff --git a/system/doc/reference_manual/data_types.xml b/system/doc/reference_manual/data_types.xml index e63825b97d..107e403903 100644 --- a/system/doc/reference_manual/data_types.xml +++ b/system/doc/reference_manual/data_types.xml @@ -50,10 +50,7 @@ <item><em><c>base</c></em><c>#</c><em><c>value</c></em> <br></br> Integer with the base <em><c>base</c></em>, that must be an - integer in the range 2..36. <br></br> - - In Erlang 5.2/OTP R9B and earlier versions, the allowed range - is 2..16.</item> + integer in the range 2..36.</item> </list> <p><em>Examples:</em></p> <pre> diff --git a/system/doc/reference_manual/errors.xml b/system/doc/reference_manual/errors.xml index e764cf431f..3e2d306561 100644 --- a/system/doc/reference_manual/errors.xml +++ b/system/doc/reference_manual/errors.xml @@ -49,8 +49,7 @@ The Erlang programming language has built-in features for handling of run-time errors.</p> <p>A run-time error can also be emulated by calling - <c>erlang:error(Reason)</c> or <c>erlang:error(Reason, Args)</c> - (those appeared in Erlang 5.4/OTP-R10).</p> + <c>erlang:error(Reason)</c> or <c>erlang:error(Reason, Args)</c>.</p> <p>A run-time error is another name for an exception of class <c>error</c>. </p> @@ -79,7 +78,6 @@ <p>Exceptions are run-time errors or generated errors and are of three different classes, with different origins. The <seealso marker="expressions#try">try</seealso> expression - (new in Erlang 5.4/OTP R10B) can distinguish between the different classes, whereas the <seealso marker="expressions#catch">catch</seealso> expression cannot. They are described in @@ -94,7 +92,7 @@ <cell align="left" valign="middle"><c>error</c></cell> <cell align="left" valign="middle">Run-time error, for example, <c>1+a</c>, or the process called - <c>erlang:error/1,2</c> (new in Erlang 5.4/OTP R10B)</cell> + <c>erlang:error/1,2</c></cell> </row> <row> <cell align="left" valign="middle"><c>exit</c></cell> @@ -111,7 +109,7 @@ and a stack trace (which aids in finding the code location of the exception).</p> <p>The stack trace can be retrieved using - <c>erlang:get_stacktrace/0</c> (new in Erlang 5.4/OTP R10B) + <c>erlang:get_stacktrace/0</c> from within a <c>try</c> expression, and is returned for exceptions of class <c>error</c> from a <c>catch</c> expression.</p> <p>An exception of class <c>error</c> is also known as a run-time diff --git a/system/doc/reference_manual/expressions.xml b/system/doc/reference_manual/expressions.xml index 1a3d19aed1..acd1dec901 100644 --- a/system/doc/reference_manual/expressions.xml +++ b/system/doc/reference_manual/expressions.xml @@ -123,10 +123,9 @@ member(_Elem, []) -> or <c>receive</c> expression must be bound in all branches to have a value outside the expression. Otherwise they are regarded as 'unsafe' outside the expression.</p> - <p>For the <c>try</c> expression introduced in - Erlang 5.4/OTP R10B, variable scoping is limited so that + <p>For the <c>try</c> expression variable scoping is limited so that variables bound in the expression are always 'unsafe' outside - the expression. This is to be improved.</p> + the expression.</p> </section> <section> @@ -189,7 +188,6 @@ f([$p,$r,$e,$f,$i,$x | Str]) -> ...</pre> <pre> case {Value, Result} of {?THRESHOLD+1, ok} -> ...</pre> - <p>This feature was added in Erlang 5.0/OTP R7.</p> </section> </section> @@ -1348,8 +1346,8 @@ catch ExceptionBodyN end</code> <p>This is an enhancement of - <seealso marker="#catch">catch</seealso> that appeared in - Erlang 5.4/OTP R10B. It gives the possibility to:</p> + <seealso marker="#catch">catch</seealso>. + It gives the possibility to:</p> <list type="bulleted"> <item>Distinguish between different exception classes.</item> <item>Choose to handle only the desired ones.</item> diff --git a/system/doc/reference_manual/introduction.xml b/system/doc/reference_manual/introduction.xml index abb4ed407d..5701462443 100644 --- a/system/doc/reference_manual/introduction.xml +++ b/system/doc/reference_manual/introduction.xml @@ -80,8 +80,8 @@ <item>A <em>list</em> is any number of items. For example, an argument list can consist of zero, one, or more arguments.</item> </list> - <p>If a feature has been added recently, in Erlang 5.0/OTP R7 or - later, this is mentioned in the text.</p> + <p>If a feature has been added in R13A or later, + this is mentioned in the text.</p> </section> <section> diff --git a/system/doc/reference_manual/macros.xml b/system/doc/reference_manual/macros.xml index 5f24473557..b6c740dd10 100644 --- a/system/doc/reference_manual/macros.xml +++ b/system/doc/reference_manual/macros.xml @@ -286,7 +286,6 @@ t.erl:5: Warning: -warning("Macro VERSION not defined -- using default version." argument, is expanded to a string containing the tokens of the argument. This is similar to the <c>#arg</c> stringifying construction in C.</p> - <p>The feature was added in Erlang 5.0/OTP R7.</p> <p><em>Example:</em></p> <code type="none"> -define(TESTCALL(Call), io:format("Call ~s: ~w~n", [??Call, Call])). diff --git a/system/doc/reference_manual/records.xml b/system/doc/reference_manual/records.xml index 12a3e697cd..1eb13b353e 100644 --- a/system/doc/reference_manual/records.xml +++ b/system/doc/reference_manual/records.xml @@ -72,9 +72,9 @@ <pre> #Name{Field1=Expr1,...,FieldK=ExprK, _=ExprL}</pre> <p>Omitted fields then get the value of evaluating <c>ExprL</c> - instead of their default values. This feature was added in - Erlang 5.1/OTP R8 and is primarily intended to be used to create - patterns for ETS and Mnesia match functions.</p> + instead of their default values. This feature is primarily + intended to be used to create patterns for ETS and Mnesia match + functions.</p> <p><em>Example:</em></p> <pre> -record(person, {name, phone, address}). diff --git a/system/doc/tutorial/complex6_nif.c b/system/doc/tutorial/complex6_nif.c index b656ed43ce..f6c06e94f4 100644 --- a/system/doc/tutorial/complex6_nif.c +++ b/system/doc/tutorial/complex6_nif.c @@ -1,4 +1,4 @@ -#include "erl_nif.h" +#include <erl_nif.h> extern int foo(int x); extern int bar(int y); |