From 06496fba0af8c45b06a397d45ba9c7e6995296fe Mon Sep 17 00:00:00 2001 From: Sebastian Huber Date: Fri, 4 Sep 2015 13:09:17 +0200 Subject: smp: Documentation --- doc/user/conf.t | 22 +++++++++++----------- doc/user/glossary.texi | 5 +++-- doc/user/smp.t | 4 ++-- 3 files changed, 16 insertions(+), 15 deletions(-) (limited to 'doc/user') diff --git a/doc/user/conf.t b/doc/user/conf.t index b19dbd9e82..ec9b1f8172 100644 --- a/doc/user/conf.t +++ b/doc/user/conf.t @@ -4203,26 +4203,26 @@ guidance. For guidance on the configuration macros, please examine Deterministic Priority Scheduler. @c -@c === Configuring Clustered/Partitioned Schedulers === +@c === Configuring Clustered Schedulers === @c -@subsection Configuring Clustered/Partitioned Schedulers +@subsection Configuring Clustered Schedulers -Clustered/partitioned scheduling helps to control the worst-case latencies in -the system. The goal is to reduce the amount of shared state in the system and -thus prevention of lock contention. Modern multi-processor systems tend to -have several layers of data and instruction caches. With clustered/partitioned -scheduling it is possible to honour the cache topology of a system and thus -avoid expensive cache synchronization traffic. +Clustered scheduling helps to control the worst-case latencies in a +multi-processor system. The goal is to reduce the amount of shared state in +the system and thus prevention of lock contention. Modern multi-processor +systems tend to have several layers of data and instruction caches. With +clustered scheduling it is possible to honour the cache topology of a system +and thus avoid expensive cache synchronization traffic. We have clustered scheduling in case the set of processors of a system is partitioned into non-empty pairwise-disjoint subsets. These subsets are called clusters. Clusters with a cardinality of one are partitions. Each cluster is -owned by exactly one scheduler instance. In order to use clustered/partitioned +owned by exactly one scheduler instance. In order to use clustered scheduling the application designer has to answer two questions. @enumerate -@item How is the set of processors partitioned into clusters/partitions? -@item Which scheduler is used for which cluster/partition? +@item How is the set of processors partitioned into clusters? +@item Which scheduler is used for which cluster? @end enumerate @subheading CONFIGURATION: diff --git a/doc/user/glossary.texi b/doc/user/glossary.texi index cff7028cb7..f0beeab4d7 100644 --- a/doc/user/glossary.texi +++ b/doc/user/glossary.texi @@ -92,7 +92,8 @@ array in that it is not limited to a predefined size. @item cluster We have clustered scheduling in case the set of processors of a system is partitioned into non-empty pairwise disjoint subsets. These subsets are called -@dfn{clusters}. Each cluster is owned by exactly one scheduler instance. +@dfn{clusters}. Clusters with a cardinality of one are partitions. Each +cluster is owned by exactly one scheduler instance. @item coalesce The process of merging adjacent holes into @@ -647,7 +648,7 @@ A @dfn{scheduler instance} is a scheduling algorithm with a corresponding context to store its internal state. Each processor in the system is owned by at most one scheduler instance. The processor to scheduler instance assignment is determined at application configuration time. @xref{Configuring a System -Configuring Clustered/Partitioned Schedulers}. +Configuring Clustered Schedulers}. @item segments Variable sized memory blocks allocated diff --git a/doc/user/smp.t b/doc/user/smp.t index 1b4849ad4f..0c3b58021e 100644 --- a/doc/user/smp.t +++ b/doc/user/smp.t @@ -589,8 +589,8 @@ the processor set of this scheduler is empty @subheading DESCRIPTION: Identifies a scheduler by its name. The scheduler name is determined by the -scheduler configuration. @xref{Configuring a System Configuring -Clustered/Partitioned Schedulers}. +scheduler configuration. @xref{Configuring a System Configuring Clustered +Schedulers}. @subheading NOTES: -- cgit v1.2.3