kernel/Kconfig.preempt

   1
   2 choice
   3         prompt "Preemption Model"
   4         default PREEMPT_NONE
   5
   6 config PREEMPT_NONE
   7         bool "No Forced Preemption (Server)"
   8         help
   9           This is the traditional Linux preemption model, geared towards
  10           throughput. It will still provide good latencies most of the
  11           time, but there are no guarantees and occasional longer delays
  12           are possible.
  13
  14           Select this option if you are building a kernel for a server or
  15           scientific/computation system, or if you want to maximize the
  16           raw processing power of the kernel, irrespective of scheduling
  17           latencies.
  18
  19 config PREEMPT_VOLUNTARY
  20         bool "Voluntary Kernel Preemption (Desktop)"
  21         help
  22           This option reduces the latency of the kernel by adding more
  23           "explicit preemption points" to the kernel code. These new
  24           preemption points have been selected to reduce the maximum
  25           latency of rescheduling, providing faster application reactions,
  26           at the cost of slightly lower throughput.
  27
  28           This allows reaction to interactive events by allowing a
  29           low priority process to voluntarily preempt itself even if it
  30           is in kernel mode executing a system call. This allows
  31           applications to run more 'smoothly' even when the system is
  32           under load.
  33
  34           Select this if you are building a kernel for a desktop system.
  35
  36 config PREEMPT
  37         bool "Preemptible Kernel (Low-Latency Desktop)"
  38         help
  39           This option reduces the latency of the kernel by making
  40           all kernel code (that is not executing in a critical section)
  41           preemptible.  This allows reaction to interactive events by
  42           permitting a low priority process to be preempted involuntarily
  43           even if it is in kernel mode executing a system call and would
  44           otherwise not be about to reach a natural preemption point.
  45           This allows applications to run more 'smoothly' even when the
  46           system is under load, at the cost of slightly lower throughput
  47           and a slight runtime overhead to kernel code.
  48
  49           Select this if you are building a kernel for a desktop or
  50           embedded system with latency requirements in the milliseconds
  51           range.
  52
  53 endchoice
  54
  55 choice
  56         prompt "RCU Implementation"
  57         default CLASSIC_RCU
  58
  59 config CLASSIC_RCU
  60         bool "Classic RCU"
  61         help
  62           This option selects the classic RCU implementation that is
  63           designed for best read-side performance on non-realtime
  64           systems.
  65
  66           Select this option if you are unsure.
  67
  68 config TREE_RCU
  69         bool "Tree-based hierarchical RCU"
  70         help
  71           This option selects the RCU implementation that is
  72           designed for very large SMP system with hundreds or
  73           thousands of CPUs.
  74
  75 config PREEMPT_RCU
  76         bool "Preemptible RCU"
  77         depends on PREEMPT
  78         help
  79           This option reduces the latency of the kernel by making certain
  80           RCU sections preemptible. Normally RCU code is non-preemptible, if
  81           this option is selected then read-only RCU sections become
  82           preemptible. This helps latency, but may expose bugs due to
  83           now-naive assumptions about each RCU read-side critical section
  84           remaining on a given CPU through its execution.
  85
  86 endchoice
  87
  88 config RCU_TRACE
  89         bool "Enable tracing for RCU"
  90         depends on TREE_RCU || PREEMPT_RCU
  91         help
  92           This option provides tracing in RCU which presents stats
  93           in debugfs for debugging RCU implementation.
  94
  95           Say Y here if you want to enable RCU tracing
  96           Say N if you are unsure.
  97
  98 config RCU_FANOUT
  99         int "Tree-based hierarchical RCU fanout value"
 100         range 2 64 if 64BIT
 101         range 2 32 if !64BIT
 102         depends on TREE_RCU
 103         default 64 if 64BIT
 104         default 32 if !64BIT
 105         help
 106           This option controls the fanout of hierarchical implementations
 107           of RCU, allowing RCU to work efficiently on machines with
 108           large numbers of CPUs.  This value must be at least the cube
 109           root of NR_CPUS, which allows NR_CPUS up to 32,768 for 32-bit
 110           systems and up to 262,144 for 64-bit systems.
 111
 112           Select a specific number if testing RCU itself.
 113           Take the default if unsure.
 114
 115 config RCU_FANOUT_EXACT
 116         bool "Disable tree-based hierarchical RCU auto-balancing"
 117         depends on TREE_RCU
 118         default n
 119         help
 120           This option forces use of the exact RCU_FANOUT value specified,
 121           regardless of imbalances in the hierarchy.  This is useful for
 122           testing RCU itself, and might one day be useful on systems with
 123           strong NUMA behavior.
 124
 125           Without RCU_FANOUT_EXACT, the code will balance the hierarchy.
 126
 127           Say n if unsure.
 128
 129