CPSC 3220 - DAY 11
OCTOBER 5, 2017

================================================================================

Chapter 6
---------

Multi-object Programs

Synchronization Performance

	-A program with lots of concurrent threads can still have poor 
	performance on a multiprocessor

		-Overhead of creating threads, if not needed
		-Lock contention: only one thread at a time can hold a given 
		lock.
		-Shared data protected by a lock may ping back and forth 	
		between cores
		-False sharing: communication between cores even for data is 
		not shared.


Multiprocessor Cache Coherence

	Scenario:
		-Thread A modifies data inside a critical section and releases
		a lock.

		-Thread B acquires lock and reads data.

	Easy if all accesses go to main memory.
		-Ex. Thread A changes main mem, thread B reads it.


Write Back Cache Coherence

	Cache coherence = system behaves as if there is one copy of the data
		-If the data is only being read, any number of caches can
		have a copy.

		-If data is being modified, at most one cached copy

	On write: (get ownership)
		-Invalidate all cached copies, before doing write
		-Modified data stays in cache ("write back")

	On read:
		-Fetch value from owner or from memory.

Directory Based Cache coherence

	How do we know which cores have a location cached?

		-Hardware keeps track of all cached copies
		-On a read miss, if held exclusive, fetch latest copy and 
		invalidate that copy.
		-On a write miss, invalidate all copies

	Read-modify-write instructions
		-Fetch cache entry exlusive, prevent any other cache from
		reading the data until instruction completes.