diff options
Diffstat (limited to 'erts/doc')
-rw-r--r-- | erts/doc/src/Makefile | 11 | ||||
-rw-r--r-- | erts/doc/src/atomics.xml | 183 | ||||
-rw-r--r-- | erts/doc/src/counters.xml | 142 | ||||
-rw-r--r-- | erts/doc/src/ref_man.xml | 2 | ||||
-rw-r--r-- | erts/doc/src/specs.xml | 2 |
5 files changed, 333 insertions, 7 deletions
diff --git a/erts/doc/src/Makefile b/erts/doc/src/Makefile index fccdd744f3..40f74b78ff 100644 --- a/erts/doc/src/Makefile +++ b/erts/doc/src/Makefile @@ -53,19 +53,16 @@ XML_REF3_EFILES = \ erl_tracer.xml \ init.xml \ persistent_term.xml \ + atomics.xml \ + counters.xml \ zlib.xml XML_REF3_FILES = \ + $(XML_REF3_EFILES) \ driver_entry.xml \ erl_nif.xml \ - erl_tracer.xml \ erl_driver.xml \ - erl_prim_loader.xml \ - erlang.xml \ - erts_alloc.xml \ - init.xml \ - persistent_term.xml \ - zlib.xml + erts_alloc.xml XML_PART_FILES = \ part.xml diff --git a/erts/doc/src/atomics.xml b/erts/doc/src/atomics.xml new file mode 100644 index 0000000000..3fca92fb97 --- /dev/null +++ b/erts/doc/src/atomics.xml @@ -0,0 +1,183 @@ +<?xml version="1.0" encoding="utf-8" ?> +<!DOCTYPE erlref SYSTEM "erlref.dtd"> + +<erlref> + <header> + <copyright> + <year>2018</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. + </legalnotice> + + <title>atomics</title> + </header> + <module>atomics</module> + <modulesummary>Atomic Functions</modulesummary> + <description> + <p>This module provides a set of functions to do atomic operations towards + mutable atomic variables. The implementation utilizes only + atomic hardware instructions without any software level locking, which makes + it very efficient for concurrent access. The atomics are organized into + arrays with the follwing semantics:</p> + <list type="bulleted"> + <item> + <p>Atomics are 64 bit integers.</p> + </item> + <item> + <p>Atomics can be represented as either signed or unsigned.</p> + </item> + <item> + <p>Atomics wrap around at overflow and underflow operations.</p> + </item> + <item> + <p>All operations guarantee atomicity. No intermediate results can be + seen. The result of one mutation can only be the input to one + following mutation.</p> + </item> + <item> + <p>All atomic operations are mutually ordered. If atomic B is updated + <em>after</em> atomic A, then that is how it will appear to any + concurrent readers. No one can read the new value of B and then read the + old value of A.</p> + </item> + <item> + <p>Indexes into atomic arrays are one-based. An atomic array of + arity N contains N atomics with index from 1 to N.</p> + </item> + </list> + </description> + + <datatypes> + <datatype> + <name name="atomics_ref"/> + <desc><p>Identifies an atomic array returned from + <seealso marker="#new/2"><c>new/2</c></seealso>.</p> + </desc> + </datatype> + </datatypes> + + <funcs> + <func> + <name name="new" arity="2"/> + <fsummary>Create atomic array</fsummary> + <desc> + <p>Create a new atomic array of <c><anno>Arity</anno></c> atomics.</p> + <p>Argument <c><anno>Opts</anno></c> is a list of the following possible + options:</p> + <taglist> + <tag><c>{signed, boolean()}</c></tag> + <item><p>Indicate if the elements of the array will be treated + as signed or unsigned integers. Default is <c>true</c> (signed).</p> + <p>The integer interval for signed atomics are from <c>-(1 bsl 63)</c> + to <c>(1 bsl 63)-1</c> and for unsigned atomics from <c>0</c> to <c>(1 + bsl 64)-1</c>.</p> + </item> + </taglist> + </desc> + </func> + + <func> + <name name="put" arity="3"/> + <fsummary>Set atomic value</fsummary> + <desc> + <p>Set atomic to <c><anno>Value</anno></c>.</p> + </desc> + </func> + + <func> + <name name="get" arity="2"/> + <fsummary>Read atomic value</fsummary> + <desc> + <p>Read atomic value.</p> + </desc> + </func> + + <func> + <name name="add" arity="3"/> + <fsummary>Add to atomic</fsummary> + <desc> + <p>Add <c><anno>Incr</anno></c> to atomic.</p> + </desc> + </func> + + <func> + <name name="add_get" arity="3"/> + <fsummary>Atomic add and get</fsummary> + <desc> + <p>Atomic addition and return of the result.</p> + </desc> + </func> + + <func> + <name name="sub" arity="3"/> + <fsummary>Subtract from atomic</fsummary> + <desc> + <p>Subtract <c><anno>Decr</anno></c> from atomic.</p> + </desc> + </func> + + <func> + <name name="sub_get" arity="3"/> + <fsummary>Atomic sub and get</fsummary> + <desc> + <p>Atomic subtraction and return of the result.</p> + </desc> + </func> + + <func> + <name name="exchange" arity="3"/> + <fsummary>Atomic exchange.</fsummary> + <desc> + <p>Atomically replaces the value of the atomic with + <c><anno>Desired</anno></c> and returns the value it held + previously.</p> + </desc> + </func> + + <func> + <name name="compare_exchange" arity="4"/> + <fsummary>Atomic compare and exchange.</fsummary> + <desc> + <p>Atomically compares the atomic with <c><anno>Expected</anno></c>, + and if those are equal, set atomic to <c><anno>Desired</anno></c>. + Returns <c>ok</c> if <c><anno>Desired</anno></c> was written. Returns + the actual atomic value if not equal to <c><anno>Expected</anno></c>.</p> + </desc> + </func> + + <func> + <name name="info" arity="1"/> + <fsummary>Get information about atomic array.</fsummary> + <desc> + <p>Return information about an atomic array in a map. The map + has the following keys:</p> + <taglist> + <tag><c>size</c></tag> + <item><p>The number of atomics in the array.</p></item> + <tag><c>max</c></tag> + <item><p>The highest possible value an atomic in this array can + hold.</p></item> + <tag><c>min</c></tag> + <item><p>The lowest possible value an atomic in this array can + hold.</p></item> + <tag><c>memory</c></tag> + <item><p>Approximate memory consumption for the array in + bytes.</p></item> + </taglist> + </desc> + </func> + + </funcs> +</erlref> diff --git a/erts/doc/src/counters.xml b/erts/doc/src/counters.xml new file mode 100644 index 0000000000..85eedfdadc --- /dev/null +++ b/erts/doc/src/counters.xml @@ -0,0 +1,142 @@ +<?xml version="1.0" encoding="utf-8" ?> +<!DOCTYPE erlref SYSTEM "erlref.dtd"> + +<erlref> + <header> + <copyright> + <year>2018</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. + </legalnotice> + + <title>counters</title> + </header> + <module>counters</module> + <modulesummary>Counter Functions</modulesummary> + <description> + <p>This module provides a set of functions to do operations towards + shared mutable counter variables. The implementation does not utilize any + software level locking, which makes it very efficient for concurrent + access. The counters are organized into arrays with the follwing + semantics:</p> + <list type="bulleted"> + <item> + <p>Counters are 64 bit signed integers.</p> + </item> + <item> + <p>Counters wrap around at overflow and underflow operations.</p> + </item> + <item><p>Counters are initialized to zero and can then only be written to + by adding or subtracting.</p> + </item> + <item> + <p>Write operations guarantee atomicity. No intermediate results can be + seen from a single write operation.</p> + </item> + <item> + <p>Two types of counter arrays can be created with options <c>atomics</c> or + <c>write_concurrency</c>. The <c>atomics</c> counters have good allround + performance with nice consistent semantics while + <c>write_concurrency</c> counters offers even better concurrent + write performance at the expense of some potential read + inconsistencies. See <seealso marker="#new/2"><c>new/2</c></seealso>.</p> + </item> + <item> + <p>Indexes into counter arrays are one-based. A counter array of + size N contains N counters with index from 1 to N.</p> + </item> + </list> + </description> + + <datatypes> + <datatype> + <name name="counters_ref"/> + <desc><p>Identifies a counter array returned from + <seealso marker="#new/2"><c>new/2</c></seealso>.</p> + </desc> + </datatype> + </datatypes> + + <funcs> + <func> + <name name="new" arity="2"/> + <fsummary>Create counter array</fsummary> + <desc> + <p>Create a new counter array of <c><anno>Size</anno></c> counters.</p> + <p>Argument <c><anno>Opts</anno></c> is a list of the following possible + options:</p> + <taglist> + <tag><c>atomics</c> (Default)</tag> + <item><p>Counters will be sequentially consistent. If write + operation A is done sequencially before write operation B, then a concurrent reader + may see none of them, only A, or both A and B. It cannot see only B.</p> + </item> + <tag><c>write_concurrency</c></tag> + <item><p>This is an optimization to achieve very efficient concurrent + write operations at the expense of potential read inconsistency and memory + consumption per counter.</p> + <p>Read operations may see sequentially inconsistent results with + regard to concurrent write operations. Even if write operation A is done + sequencially before write operation B, a concurrent reader may see any + combination of A and B, including only B. A read operation is only + guaranteed to see all writes done sequentially before the read. No writes + are ever lost, but will eventually all be seen.</p> + </item> + </taglist> + </desc> + </func> + + <func> + <name name="get" arity="2"/> + <fsummary>Read counter value</fsummary> + <desc> + <p>Read counter value.</p> + </desc> + </func> + + <func> + <name name="add" arity="3"/> + <fsummary>Add to counter</fsummary> + <desc> + <p>Add <c><anno>Incr</anno></c> to counter.</p> + </desc> + </func> + + <func> + <name name="sub" arity="3"/> + <fsummary>Subtract from counter</fsummary> + <desc> + <p>Subtract <c><anno>Decr</anno></c> from counter.</p> + </desc> + </func> + + <func> + <name name="info" arity="1"/> + <fsummary>Get information about counter array.</fsummary> + <desc> + <p>Return information about a counter array in a map. The map + has the following keys (at least):</p> + <taglist> + <tag><c>size</c></tag> + <item><p>The number of counters in the array.</p></item> + <tag><c>memory</c></tag> + <item><p>Approximate memory consumption for the array in + bytes.</p></item> + </taglist> + </desc> + </func> + + </funcs> +</erlref> diff --git a/erts/doc/src/ref_man.xml b/erts/doc/src/ref_man.xml index ff64aa86b8..a78aaa449e 100644 --- a/erts/doc/src/ref_man.xml +++ b/erts/doc/src/ref_man.xml @@ -50,5 +50,7 @@ <xi:include href="erts_alloc.xml"/> <xi:include href="erl_nif.xml"/> <xi:include href="erl_tracer.xml"/> + <xi:include href="atomics.xml"/> + <xi:include href="counters.xml"/> </application> diff --git a/erts/doc/src/specs.xml b/erts/doc/src/specs.xml index 1ab6cdc19e..0b943e6295 100644 --- a/erts/doc/src/specs.xml +++ b/erts/doc/src/specs.xml @@ -6,4 +6,6 @@ <xi:include href="../specs/specs_init.xml"/> <xi:include href="../specs/specs_persistent_term.xml"/> <xi:include href="../specs/specs_zlib.xml"/> + <xi:include href="../specs/specs_atomics.xml"/> + <xi:include href="../specs/specs_counters.xml"/> </specs> |