CactiParser Class Reference

Class that provides the interface between ZigZag and CACTI. More...

Public Member Functions
def	__init__ (self)
bool	item_exists (self, int size, int r_bw, int r_port, int w_port, int rw_port, int bank, float technology, str mem_pool_path=MEM_POOL_PATH)
	This function checks whether the provided memory configuration was already used in the past. More...
None	create_item (self, str mem_type, int size, int r_bw, int r_port, int w_port, int rw_port, int bank, float technology=0.022, str mem_pool_path=MEM_POOL_PATH, str cacti_top_path=CACTI_TOP_PATH)
	This function simulates a new item by calling CACTI7 based on the provided parameters. More...
tuple[float, float, float]	get_item (self, *str mem_name, str mem_type, int size, int r_bw, int r_port, int w_port, int rw_port, int bank, float technology=0.022, str mem_pool_path=MEM_POOL_PATH, str cacti_top_path=CACTI_TOP_PATH)
	This functions checks first if the memory with the provided parameters was already simulated once. More...

Static Public Attributes
	cacti_path = os.path.dirname(os.path.realpath(__file__))
	Path of current directory. More...
string	MEM_POOL_PATH = f"{cacti_path}/cacti_master/example_mem_pool.yaml"
	Path to cached cacti simulated memories. More...
string	CACTI_TOP_PATH = f"{cacti_path}/cacti_master/cacti_top.py"
	Path to cacti python script to extract costs. More...

Detailed Description

Class that provides the interface between ZigZag and CACTI.

Constructor & Destructor Documentation

__init__()

def __init__

(

self

)

Member Function Documentation

create_item()

None create_item	(		self,
		str	mem_type,
		int	size,
		int	r_bw,
		int	r_port,
		int	w_port,
		int	rw_port,
		int	bank,
		float	technology = 0.022,
		str	mem_pool_path = MEM_POOL_PATH,
		str	cacti_top_path = CACTI_TOP_PATH
	)

This function simulates a new item by calling CACTI7 based on the provided parameters.

Parameters

mem_pool_path	Path to cached cacti simulated memories
cacti_top_path	Path to cacti python script to extract costs

Here is the caller graph for this function:

get_item()

tuple[float, float, float] get_item	(		self,
		*str	mem_name,
		str	mem_type,
		int	size,
		int	r_bw,
		int	r_port,
		int	w_port,
		int	rw_port,
		int	bank,
		float	technology = 0.022,
		str	mem_pool_path = MEM_POOL_PATH,
		str	cacti_top_path = CACTI_TOP_PATH
	)

This functions checks first if the memory with the provided parameters was already simulated once.

In case it hasn't been simulated, then it will create a new memory item based on the provided parameters.

Parameters

mem_pool_path	Path to cached cacti simulated memories
cacti_top_path	Path to cacti python script to extract costs

Here is the call graph for this function:

item_exists()

bool item_exists	(		self,
		int	size,
		int	r_bw,
		int	r_port,
		int	w_port,
		int	rw_port,
		int	bank,
		float	technology,
		str	mem_pool_path = MEM_POOL_PATH
	)

This function checks whether the provided memory configuration was already used in the past.

Parameters

mem_pool_path

Path to cached cacti simulated memories

Returns

Return wether the requested memory item has been simulated before.

Here is the caller graph for this function:

Member Data Documentation

cacti_path

cacti_path = os.path.dirname(os.path.realpath(__file__))

static

Path of current directory.

CACTI_TOP_PATH

string CACTI_TOP_PATH = f"{cacti_path}/cacti_master/cacti_top.py"

static

Path to cacti python script to extract costs.

MEM_POOL_PATH

string MEM_POOL_PATH = f"{cacti_path}/cacti_master/example_mem_pool.yaml"

static

Path to cached cacti simulated memories.

---

The documentation for this class was generated from the following file:

• /home/runner/work/zigzag/zigzag/zigzag/cacti/cacti_parser.py