summaryrefslogtreecommitdiffstats
path: root/spec/glossary
diff options
context:
space:
mode:
authorRichi Dubey <richidubey@gmail.com>2020-06-30 11:40:01 +0530
committerSebastian Huber <sebastian.huber@embedded-brains.de>2020-06-30 08:59:59 +0200
commit26c1df5c52538e5b304ea12ae4069f04a6b02a3f (patch)
treef7fdc72bdd523d7ec85e663762f369f5922cb74e /spec/glossary
parentUpdate rtems-docs submodule (diff)
downloadrtems-central-26c1df5c52538e5b304ea12ae4069f04a6b02a3f.tar.bz2
Added APA terms in glossary
Diffstat (limited to 'spec/glossary')
-rw-r--r--spec/glossary/apa.yml17
-rw-r--r--spec/glossary/strongapa.yml20
-rw-r--r--spec/glossary/weakapa.yml17
3 files changed, 54 insertions, 0 deletions
diff --git a/spec/glossary/apa.yml b/spec/glossary/apa.yml
new file mode 100644
index 00000000..c132c169
--- /dev/null
+++ b/spec/glossary/apa.yml
@@ -0,0 +1,17 @@
+SPDX-License-Identifier: CC-BY-SA-4.0 OR BSD-2-Clause
+copyrights:
+- Copyright (C) 2020 Richi Dubey (richidubey@gmail.com)
+enabled-by: true
+glossary-type: term
+links:
+- role: glossary-member
+ uid: ../glossary-general
+term: APA
+text: |
+ This term is an acronym for Arbitrary Processor Affinity. APA schedulers
+ allow a thread to have an arbitrary affinity to a processor set, rather than
+ a restricted mapping to only one processor of the set or the ability to run
+ on all processors of the set.
+
+ It has two variants, ${weakapa:/term} and ${strongapa:/term}.
+type: glossary
diff --git a/spec/glossary/strongapa.yml b/spec/glossary/strongapa.yml
new file mode 100644
index 00000000..e8c55a9a
--- /dev/null
+++ b/spec/glossary/strongapa.yml
@@ -0,0 +1,20 @@
+SPDX-License-Identifier: CC-BY-SA-4.0 OR BSD-2-Clause
+copyrights:
+- Copyright (C) 2020 Richi Dubey (richidubey@gmail.com)
+enabled-by: true
+glossary-type: term
+links:
+- role: glossary-member
+ uid: ../glossary-general
+term: Strong APA
+text: |
+ Strong APA is a specialization of ${apa:/term}. Schedulers which implement
+ strong APA recursively searches for a processor in the ${thread:/term}'s
+ affinity set, whenever a thread becomes ready for execution, followed by the
+ processors in the affinity set of threads that are assigned the processor
+ present in the ready thread's affinity set. This is done to find a thread to
+ processor mapping that does not violate the priority ordering and to provide
+ a thread to processor mapping with a higher total priority of the threads
+ allocated a processor. Similar analysis is done when a thread blocks. See
+ also :cite:`Cerqueira:2014:LPA`.
+type: glossary
diff --git a/spec/glossary/weakapa.yml b/spec/glossary/weakapa.yml
new file mode 100644
index 00000000..87e92cc7
--- /dev/null
+++ b/spec/glossary/weakapa.yml
@@ -0,0 +1,17 @@
+SPDX-License-Identifier: CC-BY-SA-4.0 OR BSD-2-Clause
+copyrights:
+- Copyright (C) 2020 Richi Dubey (richidubey@gmail.com)
+enabled-by: true
+glossary-type: term
+links:
+- role: glossary-member
+ uid: ../glossary-general
+term: Weak APA
+text: |
+ Weak APA is a specialization of ${apa:/term}. This refers to Linux's push
+ and pull implementation of APA model. When a ${thread:/term} becomes ready
+ for execution, it is allocated a processor if there is an idle processor, or
+ a processor executing a lower priority thread in its affinity set. Unlike
+ ${strongapa:/term}, no thread is migrated from its processor to find a thread
+ to processor mapping. See also :cite:`Cerqueira:2014:LPA`.
+type: glossary