Chapter 2 Process, Thread and Scheduling Soloris IPC

1
Chapter 2 Process, Thread and
Scheduling Soloris IPC
2
Outline

• Generic System V IPC Support
• Shared Memory
• System V Semaphores
• System V Message Queues
• POSIX IPC
• Signal
• Door

3
Generic System V IPC Support

• Module Creation
• The System V IPC kernel modules are implemented
  as dynamically loadable modules
• Each facility has a corresponding loadable module
  in the /kernel/sys directory(shmsys, semsys, and
  msgsys).
• When an IPC resource is initially created, a
  positive integer, known as an identifier, is
  assigned to identify the IPC object.

4
Generic System V IPC Support

• Resource Maps
• Two of the three IPC facilities use a low-level
  kernel memory allocation scheme known as resource
  maps.
• message queues
• Semaphores
• Resource maps are a means by which small units of
  kernel memory can be allocated and freed from a
  larger pool of kernel pages that have been
  preallocated.
• The amount of space allocated for resource maps
  for the IPC facilities is determined by kernel
  tunable parameters, one parameter each for
  message queues and semaphores.

5
Outline

• Generic System V IPC Support
• Shared Memory
• System V Semaphores
• System V Message Queues
• POSIX IPC
• Signal
• Door

6
Shared Memory

• What is shared Memory
• Shared memory provides an extremely efficient
  means of sharing data between multiple processes
  on a Solaris system
• The data need not actually be moved from one
  processs address space to another
• The sharing of the same physical memory (RAM)
  pages by multiple processes
• Each process has mappings to the same physical
  pages and can access the memory through pointer
  dereferencing in code

7
Shared Memory APIs

• Some implementation APIs
• shmat(2)
• shmget(2)
• shmdt(2)
• shmctl(2)

8
Shared Memory APIs
9
namespace(shmid)

• shmid identifier
• A shared memory identifier, shmid, is initialized
  and maintained by the operating system whenever a
  shmget(2) system call is executed successfully
• The shmid identifies a shared segment that has
  two components
• The actual shared RAM pages
• A data structure that maintains information about
  the shared segment, the shmid_ds data structure

10
Intimate Shared Memory (ISM)

• ISM shared segment
• Intimate shared memory (ISM) is an optimization
  introduced first in Solaris 2.2
• Allows for the sharing of the translation tables
  involved in the virtual -to-physical address
  translation for shared memory pages
• As opposed to just sharing the actual physical
  memory pages
• With many processes attaching to shared memory,
  this scheme creates a lot of redundant mappings
  to the same physical pages that the kernel must
  maintain.

11
ISM versus Non-ISM
The difference
12
ISM useful feature

• ISM provides useful feature
• Translation table sharing
• Small TLB Page Size
• Locked pages
• Default shared memory mode for Oracle RDBMS

System V Semaphores
13
Outline

• Generic System V IPC Support
• Shared Memory
• System V Semaphores
• System V Message Queues
• POSIX IPC
• Signal
• Door

14
System V Semaphores

• What is semaphore
• A semaphore, as defined in the dictionary, is a
  mechanical signalling device or a means of doing
  visual signalling
• The analogy typically used is the railroad
  mechanism of signalling trains, where mechanical
  arms would swing down to block a train from a
  section of track that another train was currently
  using. When the track was free, the arm would
  swing up, and the waiting train could then
  proceed.

15
System V Semaphores

• Two semaphore operations
• wait and signal (which correlate nicely to the
  railroad example). The operations were referred
  to as P and V operations.
• The P operation was the wait, which decremented
  the value of the semaphore if it was greater than
  zero
• The V operation was the signal, which incremented
  the semaphore value.

16
System V Semaphores

• Some features about Solaris semaphore
• The semaphore implementation in Solaris (System V
  semaphores) allows for semaphore sets.
• The actual value of a semaphore can never be
  negative.
• A kernel mutex lock is created for each semaphore
  set.
• The creation of a semaphore set by an application
  requires a call to semget(2).

17
Outline

• Generic System V IPC Support
• Shared Memory
• System V Semaphores
• System V Message Queues
• POSIX IPC
• Signal
• Door

18
System V Message Queues

• Something about System V Message Queues
• Message queues provide a means for processes to
  send and receive messages of various size in an
  asynchronous fashion on a Solaris system.
• Sent messages are placed at the back of the
  queue, and messages are received from the front
  of the queue thus the queue is implemented as a
  FIFO (First In, First Out).

19
System V Message Queues

• Kernel Resources for Message Queues
• The number of resources that the kernel allocates
  for message queues is tunable.
• Values for various message queue tunable
  parameters can be increased from their default
  values so more resources are made available for
  systems running applications that make heavy use
  of message queues.

20
System V Message Queues
msgque points to the beginning of the kernel
space allocated to hold all the system msqid_ds
structures.
The beginning of the map structures used for
maintaining resource allocation maps
21
Outline

• Generic System V IPC Support
• Shared Memory
• System V Semaphores
• System V Message Queues
• POSIX IPC
• Signal
• Door

22
OVERVIEW OF POSIX IPC

• The evolution of the POSIX standard and
  associated APIs resulted in a set of
  industry-standard interfaces that provide the
  same types of facilities as the System V IPC set
• shared memory
• semaphores
• message queues

23
POSIX APIs for the three IPC facilities
24
POSIX Shared Memory

• Something about POSIX shared memory
• The POSIX shared memory interfaces provide an API
  for support of the POSIX IPC name abstraction.
• The interfaces, shm_open(3R) and shm_unlink(3R),
  do not allocate or map memory into a calling
  processs address space.
• The programmer using POSIX shared memory must
  create the address space mapping with an explicit
  call to mmap(2)

25
POSIX Semaphores

• The POSIX specification provides for two types of
  semaphores
• POSIX named semaphores follow the POSIX IPC name
  convention discussed earlier and are created with
  the sem_open(3R) call
• POSIX unnamed semaphores, which do not have a
  name in the file system space and are memory
  based

26
POSIX Named Semaphores
sad_next provides the pointer for support of a
singly linked list.
The linked list exists within the processs
address space, not in the kernel.
Semheadp points to the first semaddr structure on
the list
27
POSIX Message Queues

• POSIX message queues are constructed on a linked
  list built by the internal libposix4 library
  code.
• The essential interfaces for using message queues
  are mq_open(3R) which opens, or creates and
  opens, a queue, making it available to the
  calling process

28
POSIX Message Queue Structures
total size in bytes
And so on
maximum size of each message
maximum number of messages allowed on the queue
29
Outline

• Generic System V IPC Support
• Shared Memory
• System V Semaphores
• System V Message Queues
• POSIX IPC
• Signal
• Door

30
Introduction to Solaris signals

• What is signal
• Signals are a means by which a process or thread
  can be notified of a particular event. Signals
  are often compared with hardware interrupts, when
  a hardware subsystem, such as a disk I/O
  interface (e.g., a SCSI host adapter), generates
  an interrupt to a processor as a result of an I/O
  being completed.
• The signal facility provides a means interrupt a
  process or thread within a process as a result of
  a specific event

31
Introduction to Solaris signals

• Exit
• Terminate the process
• Core
• Create a core image of the process and terminate
• Stop
• Suspend process execution (typically, job control
  or debug)
• Ignore
• Discard the signal and take no action, even if
  the signal is blocked

32
Introduction to Solaris signals

• Signal representation
• Each signal has a unique signal number, we use a
  structure member of sufficient width, such that
  we can represent every signal by simply setting
  the bit that responds to the signals number of
  the signal we wish to post.

33
Signal Implementation
Figure illustrates all the structures, data
types, and linked lists required to support
signals in Solaris. There are, of course,
differences between a multithreadedprocess and a
nonthreaded process.
34
Signal Implementation
High-Level Signal Flow
35
Synchronous signals

• Synchronous signals occur as a direct result of
  the executing instruction stream, where an
  unrecoverble error requires an immediate
  termination of the process.
• Synchronous signals are sometimes referred to as
  traps

36
Asynchronous signals

• Asynchronous signals are as the term implies
  external (and in some cases unrelated) to the
  current execution context.
• Asynchronous signals are also referred to as
  interrupts.

37
Outline

• Generic System V IPC Support
• Shared Memory
• System V Semaphores
• System V Message Queues
• POSIX IPC
• Signal
• Door

38
Solaris Doors

• Doors function
• Doors provide a facility for processes to issue
  procedure calls to functions in other
• Using the APIs, a process can become a door
  server, exporting a function through a door it
  creates by using the door_create(3X) interface
• Other processes can then invoke the procedure by
  issuing a door_call(3X), specifying the correct
  door descriptor

39
How doors provide an IPC
1
2
40
Doors Implementation

• Doors are implemented in the kernel as a
  pseudofile system, doorfs, which is loaded from
  the /kernel/sys directory during boot
• Within a process, a door is referenced through
  its door descriptor, which is similar in form a
  function to a file descriptor, and, in fact, the
  allocation of a door descriptor in a process uses
  an available file descriptor slot

41
Solaris Doors Structures

42
Reference

• Jim Mauro, Richard McDougall, Solaris
  Internals-Core Kernel Components, Sun
  Microsystems Press, 2000
• Max Bruning, Threading Model In Solaris,
  Training lectures,2005
• Solaris internals and performance management,
  Richard McDougall, 2002

43
End