Session2-2

Stacks….
 ‫ﺍﺻﻞ ﺷﻤﺎﺭﻩ ‪:3‬‬

‫ﻣﻮﺍﺭﺩ ﻣﺸﺘﺮﻙ ﺭﺍ ﺗﺴﺮﻳﻊ ﻛﻨﻴﺪ‪.‬‬

‫ﻋﻤﻠﻮﻧﺪﻫﺎﻱ ﺛﺎﺑﺖ ﻣﺮﺗﺒﺎ ﻣﻮﺭﺩ ﺍﺳﺘﻔﺎﺩﻩ ﻗﺮﺍﺭ ﻣﻲ ﮔﻴﺮﻧﺪ ﻭ ﺩﺳﺘﻮﺭﺍﻟﻌﻤﻞ‬

‫ﻫﺎﻱ ﺣﺴﺎﺑﻲ ﺷﺎﻣﻞ ﺁﻧﻬﺎ ﺳﺮﻳﻌﺘﺮ ﺍﺯ ﺩﺳﺘﻮﺭﺍﺕ ﺧﺎﻟﻲ ﻛﻪ ﻣﻲ ﺧﻮﺍﻫﻨﺪ‬
‫ﻋﻤﻠﻮﻧﺪﻫﺎ ﺭﺍ ﺍﺯ ﺣﺎﻓﻈﻪ ﺑﺮﺩﺍﺭﻧﺪ ﺍﺟﺮﺍ ﻣﻲ ﺷﻮﺩ‪.‬‬

‫‪Advanced Computer Architecture & Design‬‬

‫‪2‬‬
‫‪Science & Research Branch‬‬
 ‫ﻣﺸﮑﻼﺕ ﺩﺭ ﻣﻌﻤﺎﺭی ﺣﺎﻓﻈﻪ ﺑﻪ ﺣﺎﻓﻈﻪ‬
‫• ﺣﺎﻓﻈﻪ ﺍﺻﻠﯽ ﺧﻴﻠﯽ ﮐﻨﺪﺗﺮ ﺍﺯ ﻣﺪﺍﺭﺍﺕ ﻣﺤﺎﺳﺒﺎﺗﯽ ﺍﺳﺖ‬
‫‪ -‬ﺧﺎﻧﻪ ﻫﺎی ﺯﻳﺎﺩی ﺑﺮﺍی ﻣﺸﺨﺺ ﻧﻤﻮﺩﻥ ﺁﺩﺭﺳﻬﺎی ﺣﺎﻓﻈﻪ‬
‫ﮔﺮﻓﺘﻪ ﻣﯽ ﺷﻮﻧﺪ‪.‬‬
‫• ﻣﻌﻤﻮﻻ ﻧﺘﺎﻳﺞ ﻳﮏ ﻳﺎ ﺩﻭ ﺩﺳﺘﻮﺭﺍﻟﻌﻤﻞ ﺑﻌﺪ ﻣﻮﺭﺩ ﺍﺳﺘﻔﺎﺩﻩ ﻗﺮﺍﺭ ﻣﯽ‬
‫ﮔﻴﺮﻧﺪ‪.‬‬

‫ﺑﻪ ﺧﺎﻃﺮ ﺩﺍﺷﺘﻪ ﺑﺎﺷﻴﺪ‪ :‬ﺑﺨﺸﻬﺎی ﺍﺷﺘﺮﺍﮐﯽ ﺭﺍ ﺳﺮﻳﻌﺘﺮ ﻧﻤﺎﺋﻴﺪ!‬

‫ﺭﺍﻩ ﺣﻞ‪ :‬ﻧﺘﺎﻳﺞ ﻣﻴﺎﻧﯽ ﻳﺎ ﻣﻮﻗﺘﯽ ﺭﺍ ﺩﺭ ﺣﺎﻓﻈﻪ ﻫﺎی ﺳﺮﻳﻊ ﻭ ﻧﺰﺩﻳﮏ ﺑﻪ ﻭﺍﺣﺪ‬
‫ﻣﺤﺎﺳﺒﻪ ﺫﺧﻴﺮﻩ ﻧﻤﺎﺋﻴﺪ‪.‬‬
‫‪Advanced Computer Architecture & Design‬‬
‫‪3‬‬
‫‪Science & Research Branch‬‬
 ‫• ﻋﻤﻠﻴﺎﺕ ﺭﻳﺎﺿﻲ ﻣﻌﻤﻮﻻً ﺑﻪ ﺻﻮﺭﺕ ﮔﺎﻡ ﺑﻪ ﮔﺎﻡ ﺍﻧﺠﺎﻡ ﻣﻲﺷﻮﻧﺪ‪ ،‬ﻭ ﺍﺯ ﻧﺘﺎﻳﺞ ﻫﺮ‬
‫ﻣﺮﺣﻠﻪ ﺩﺭ ﻣﺮﺣﻠﻪ ﻱ ﺑﻌﺪ ﺑﻪ ﺻﻮﺭﺕ ﻭﺭﻭﺩﻱ ﺍﺳﺘﻔﺎﺩﻩ ﻣﻲﺷﻮﺩ‪.‬‬

‫• ﺩﺭ ﺭﺍﻳﺎﻧﻪﻫﺎﻱ ﺍﺑﺘﺪﺍﻳﻲ ﻧﺘﺎﻳﺞ ﻫﺮ ﻣﺮﺣﻠﻪ ﺩﺭﻭﻥ ﺣﺎﻓﻈﻪﻱ ﻣﻐﻨﺎﻃﻴﺴﻲ ﺫﺧﻴﺮﻩ‬

‫ﻣﻲﺷﺪ‪ .‬ﻋﻤﻠﻴﺎﺕ ﻧﻮﺷﺘﻦ ﺩﺭﻭﻥ ﺣﺎﻓﻈﻪﻱ ﻣﻐﻨﺎﻃﻴﺴﻲ ﻣﺪﺕ ﺯﻣﺎﻥ ﺑﺴﻴﺎﺭ ﺯﻳﺎﺩﻱ‬
‫ﻃﻮﻝ ﻣﻲﻛﺸﻴﺪ ﻭ ﺑﺎﻋﺚ ﻛﻨﺪ ﺷﺪﻥ ﺳﻴﺴﺘﻢ ﻣﻲﺷﺪ‪.‬‬

‫• ﺍﺯ ﻃﺮﻑ ﺩﻳﮕﺮ ﺩﺭ ﻋﻤﻠﻴﺎﺕ ﺑﻌﺪﻱ ﻣﻌﻤﻮﻻً ﻧﻴﺎﺯ ﺑﻮﺩ ﻛﻪ ﻫﻤﺎﻥ ﺩﺍﺩﻩﺍﻱ ﻛﻪ ﺩﺭ‬
‫ﻣﺮﺣﻠﻪﻱ ﻗﺒﻠﻲ ﺩﺭﻭﻥ ﺣﺎﻓﻈﻪ ﻣﻐﻨﺎﻃﻴﺴﻲ ﻧﻮﺷﺘﻪ ﺷﺪﻩ ﺑﻮﺩ ﺍﺯ ﺭﻭﻱ ﻫﻤﺎﻥ‬
‫ﺣﺎﻓﻈﻪ ﺧﻮﺍﻧﺪﻩ ﺷﻮﺩ‪ ،‬ﺍﺯ ﺁﻥ ﺟﺎ ﻛﻪ ﻋﻤﻠﻴﺎﺕ ﺧﻮﺍﻧﺪﻥ ﺍﺯ ﺣﺎﻓﻈﻪ ﻣﻐﻨﺎﻃﻴﺴﻲ ﻧﻴﺰ‬
‫ﻋﻤﻠﻴﺎﺗﻲ ﻛﻨﺪ ﺑﻮﺩﻧﺪ ﺑﺎﻋﺚ ﻣﻲﺷﺪ ﻛﻪ ﻛﻨﺪﻱ ﺩﺭ ﻛﺎﺭ ﺩﻭ ﭼﻨﺪﺍﻥ ﺷﻮﺩ‪.‬‬

‫‪Advanced Computer Architecture & Design‬‬

‫‪4‬‬
‫‪Science & Research Branch‬‬
 ‫ﺍﻧﺒﺎﺷﺘﮕﺮﻫﺎ ﺑﺎ ﻓﺮﺍﻫﻢ ﺳﺎﺧﺘﻦ ﻳﻚ ﺣﺎﻓﻈﻪﻱِ ﭼﺮﻛﻨﻮﻳﺲ ﻛﻪ ﺍﻳﻦ‬
‫ﻗﺎﺑﻠﻴﺖ ﺭﺍ ﺩﺍﺭﺩ ﻛﻪ ﻧﺘﺎﻳﺞ ﻫﺮ ﻣﺮﺣﻠﻪ ﺭﺍ ﺑﺪﻭﻥ ﻛﻨﺪ ﻛﺮﺩﻥ ﺳﻴﺴﺘﻢ ﺑﻪ‬
‫ﻣﺮﺣﻠﻪﻱ ﺑﻌﺪ ﺍﻧﺘﻘﺎﻝ ﺩﻫﺪ‪ ،‬ﺑﺎﺯﺩﻫﻲ ﺭﺍ ﺑﻪ ﺷﺪﺕ ﺑﻬﺒﻮﺩ ﻣﻲ ﺑﺨﺸﻨﺪ‪.‬‬

‫ﺑﺎ ﺍﻳﻦ ﻛﺎﺭ ﺩﺭ ﻫﺮ ﻣﺮﺣﻠﻪ ﻳﻚ ﻋﻤﻠﻴﺎﺕ ﺧﻮﺍﻧﺪﻥ ﻭ ﻧﻮﺷﺘﻦ ﺣﺎﻓﻈﻪﻱ‬

‫ﻣﻐﻨﺎﻃﻴﺴﻲ ﺻﺮﻓﻪ ﺟﻮﻳﻲ ﻣﻲﺷﻮﺩ ﻛﻪ ﺁﻥ ﻋﻤﻠﻴﺎﺕ ﺩﻩﻫﺎ ﻭ ﺻﺪﻫﺎ ﺑﺮﺍﺑﺮ‬
‫ﻛﻨﺪﺗﺮ ﺍﺯ ﻋﻤﻠﻴﺎﺕ ﺭﻳﺎﺿﻲ ﻫﺴﺘﻨﺪ‪.‬‬

‫‪Advanced Computer Architecture & Design‬‬

‫‪5‬‬
‫‪Science & Research Branch‬‬
 ‫‪Accumulator‬‬
‫ﺩﺭ ﭘﺮﺩﺍﺯﻧﺪﻩ ﻣﺮﮐﺰی ﻳﮏ ﮐﺎﻣﭙﻴﻮﺗﺮ‪ ،‬ﺍﻧﺒﺎﺷﺘﮕﺮ ﺭﺟﻴﺴﺘﺮی ﺍﺳﺖ ﮐﻪ ﺩﺭ ﺁﻥ ﺩﺍﺩﻩﻫﺎی ﻣﻴﺎﻧﯽ ﮐﻪ ﺗﻮﺳﻂ ﻭﺍﺣﺪ‬ ‫•‬

‫ﻣﺤﺎﺳﺒﺎﺕ ﻭ ﻣﻨﻄﻖ ﺗﻮﻟﻴﺪ ﻣﯽﺷﻮﻧﺪ‪ ،‬ﺫﺧﻴﺮﻩ ﻣﯽﺷﻮﺩ‪.‬‬

‫ﺩﺳﺘﺮﺳﯽ ﺑﻪ ﺣﺎﻓﻈﻪ ﺍﺻﻠﯽ ﺑﺴﻴﺎﺭ ﮐﻨﺪﺗﺮ ﺍﺯ ﺩﺳﺘﺮﺳﯽ ﺑﻪ ﺭﺟﻴﺴﺘﺮ ﺍﺳﺖ ﻭ ﺍﻳﻦ ﺑﻪ ﻋﻠﺖ ﻓﻨﺎﻭﺭی ﺍﺳﺘﻔﺎﺩﻩ ﺷﺪﻩ ﺩﺭ‬ ‫•‬

‫ﺳﺎﺧﺖ ﺁﻥ ﻫﺎﺳﺖ‪ ،‬ﺣﺎﻓﻈﻪ ﺍﺻﻠﯽ ﮔﻨﺠﺎﻳﺶ ﺑﻴﺸﺘﺮی ﺩﺍﺭﺩ ﻭ ﺩﺭ ﻋﻮﺽ ﮐﻨﺪﺗﺮ ﺍﺯ ﺭﺟﻴﺴﺘﺮ ﻫﺎﺳﺖ‪.‬‬

‫ﮐﺎﻣﭙﻴﻮﺗﺮﻫﺎی ﺍﺑﺘﺪﺍﻳﯽ ﺑﻪ ﺩﻭ ﺩﺳﺘﻪ ﺗﻘﺴﻴﻢ ﻣﯽ ﺷﺪﻧﺪ‪ ،‬ﺁﻥﻫﺎﻳﯽ ﮐﻪ ﺭﺟﻴﺴﺘﺮ ﺍﻧﺒﺎﺷﺘﮕﺮ ﺩﺍﺷﺘﻨﺪ ﻭ ﺁﻥﻫﺎﻳﯽ ﮐﻪ ﭼﻨﻴﻦ‬ ‫•‬

‫ﻗﺎﺑﻠﻴﺘﯽ ﻧﺪﺍﺷﺘﻨﺪ‪.‬‬

‫ﺩﺭ ﺻﻮﺭﺗﯽ ﮐﻪ ﺭﺟﻴﺴﺘﺮ ﺑﺮﺍی ﺫﺧﻴﺮﻩ ﺳﺎﺯی ﺩﺍﺩﻩﻫﺎی ﻣﻴﺎﻧﯽ ﻭﺟﻮﺩ ﻧﺪﺍﺷﺘﻪ ﺑﺎﺷﺪ‪ ،‬ﻻﺯﻡ ﺧﻮﺍﻫﺪ ﺑﻮﺩ ﮐﻪ ﺍﻳﻦ‬ ‫•‬

‫ﺍﻃﻼﻋﺎﺕ ﺑﻌﺪ ﺍﺯ ﻫﺮ ﻣﺤﺎﺳﺒﻪ ﺩﺭﻭﻥ ﺣﺎﻓﻈﻪ ﺍﺻﻠﯽ ﻧﻮﺷﺘﻪ ﺷﻮﺩ‪ ،‬ﻭ ﺍﺣﺘﻤﺎﻝ ﺩﺍﺭﺩ ﮐﻪ ﻫﻤﺎﻥ ﺩﺍﺩﻩ ﺩﻭﺑﺎﺭﻩ ﺍﺯ ﺣﺎﻓﻈﻪ‬

‫ﺍﺻﻠﯽ ﺧﻮﺍﻧﺪﻩ ﺷﻮﺩ ﻭ ﺩﺭ ﻋﻤﻠﻴﺎﺕ ﺑﻌﺪی ﻣﻮﺭﺩ ﺍﺳﺘﻔﺎﺩﻩ ﻗﺮﺍﺭ ﮔﻴﺮﺩ‪.‬‬

‫‪Advanced Computer Architecture & Design‬‬

‫‪6‬‬
‫‪Science & Research Branch‬‬
 ‫ﻛﺎﻣﭙﻴﻮﺗﺮﻫﺎﻱ ﺍﻣﺮﻭﺯﻱ ﻣﻌﻤﻮﻻً ﺷﺎﻣﻞ ﭼﻨﺪ ﺭﺟﻴﺴﺘﺮ ﻣﺨﺘﻠﻒ ﻭ‬
‫ﭼﻨﺪ ﻣﻨﻈﻮﺭﻩ ﻫﺴﺘﻨﺪ ﻛﻪ ﻣﻲ ﺗﻮﺍﻧﻨﺪ ﻧﻘﺶ ﺭﺟﻴﺴﺘﺮ ﺍﻧﺒﺎﺷﺘﮕﺮ ﺭﺍ ﻫﻢ‬
‫ﺍﻳﻔﺎ ﻛﻨﻨﺪ‪ ،‬ﺍﺯ ﺍﻳﻦ ﺭﻭ ﻭﺍژﻩ "ﺍﻧﺒﺎﺷﺘﮕﺮ" ﺍﻣﺮﻭﺯﻩ ﺑﻪ ﻣﺎﻧﻨﺪ ﮔﺬﺷﺘﻪ ﺭﺍﻳﺞ‬
‫ﻧﻴﺴﺖ‪.‬‬
‫ﺑﺎ ﺍﻳﻦ ﺣﺎﻝ ﺍﻣﺮﻭﺯﻩ ﻧﻴﺰ ﭘﺮﺩﺍﺯﻧﺪﻩﻫﺎﻱ ﺧﺎﺹ ﻣﻨﻈﻮﺭﻩﺍﻱ ﻭﺟﻮﺩ ﺩﺍﺭﻧﺪ‬
‫ﻛﻪ ﺑﺮﺍﻱ ﺭﻋﺎﻳﺖ ﺳﺎﺩﮔﻲ ﺩﺭ ﻃﺮﺍﺣﻴﺸﺎﻥ ﺍﺯ ﺗﻨﻬﺎ ﻳﻚ ﺭﺟﻴﺴﺘﺮ‬
‫ﺍﻧﺒﺎﺷﺘﮕﺮ ﺑﺮﺍﻱ ﻛﺎﺭﻫﺎﻳﺸﺎﻥ ﺍﺳﺘﻔﺎﺩﻩ ﻣﻲ ﻛﻨﻨﺪ‬

‫‪Advanced Computer Architecture & Design‬‬

‫‪7‬‬
‫‪Science & Research Branch‬‬
 ‫ﻣﺎﺷﻴﻨﻬﺎی ﻣﺒﺘﻨﯽ ﺑﺮ ﺍﻧﺒﺎﺷﺘﮕﺮ‬
‫‪Accumulator‬‬
‫‪ ‬ﻳﮏ ﻣﺎﺷﻴﻦ ﺍﻧﺒﺎﺷﺘﮕﺮ‪ ،‬ﻳﮏ ﺑﺎﻓﺮ ﭘﺮﺳﺮﻋﺖ ﻭﺍﺣﺪ ﺭﺍ ﻧﺰﺩﻳﮏ ﻭﺍﺣﺪ ﻣﺤﺎﺳﺒﻪ ﻣﻨﻄﻖ‬
‫ﻧﮕﻬﺪﺍﺭی ﻣﯽ ﮐﻨﺪ‪.‬‬
‫‪ ‬ﻣﺎﻧﻨﺪ ﻳﮏ ﻣﺠﻤﻮﻋﻪ ﺍﺯ ‪D latch‬ﻫﺎ ﻳﺎ ﻓﻠﻴﭗ ﻓﻼﭘﻬﺎ ‪ ،‬ﻫﺮ ﮐﺪﺍﻡ ﺑﺎ ﻳﮏ ﺑﻴﺖ ﺩﺍﺩﻩ‬
‫‪ ‬ﻭﻳﮋﮔﯽ ﮐﻪ ﺭﺟﻴﺴﺘﺮ ﺍﻧﺒﺎﺷﺘﮕﺮ ﺭﺍ ﺑﺎ ﺩﻳﮕﺮ ﺍﻧﻮﺍﻉ ﺭﺟﻴﺴﺘﺮﻫﺎ ﻣﺘﻤﺎﻳﺰ ﻣﯽﮐﻨﺪ )ﺍﮔﺮ‬
‫ﻣﻌﻤﺎﺭی ﺩﺍﺭﺍی ﺭﺟﻴﺴﺘﺮ ﺍﻧﺒﺎﺷﺘﮕﺮ ﺑﺎﺷﺪ(‪ ،‬ﺍﻳﻦ ﺍﺳﺖ ﮐﻪ ﺍﻳﻦ ﺭﺟﻴﺴﺘﺮ ﺑﻪ‬
‫ﻃﻮﺭ ﺿﻤﻨﯽ ﺑﻪ ﻋﻨﻮﺍﻥ ﻋﻤﻠﻮﻧﺪ ﺩﺭ ﺩﺳﺘﻮﺭﻫﺎی ﺣﺴﺎﺑﯽ ﺍﺳﺘﻔﺎﺩﻩ ﻣﯽﺷﻮﺩ‪.‬‬
‫ﺩﺭ ﺳﺎﺩﻩ ﺗﺮﻳﻦ ﺣﺎﻟﺖ‪ ،‬ﻓﻘﻂ ﻳﮏ ﻋﻤﻠﻮﻧﺪ ﻣﯽ ﺗﻮﺍﻧﺪ ﻣﺸﺨﺺ ﮔﺮﺩﺩ; ﺍﻧﺒﺎﺷﺘﮕﺮ ﺑﻪ ﺻﻮﺭﺕ‬
‫ﻣﺠﺎﺯی ﺑﻪ ﻣﻔﻬﻮﻡ “‪ ”OP Operand‬ﻣﯽ ﺑﺎﺷﺪ ﻳﻌﻨﯽ‪:‬‬
‫‪acc. = acc. OP operand‬‬

‫‪Example:‬‬
‫‪# Load B into acc.‬‬ ‫]‪LOAD M[48‬‬
‫)‪# Add C to acc. (now has B+C‬‬ ‫]‪ADD M[76‬‬
‫‪# Write acc. To A‬‬ ‫]‪STORE M[100‬‬
‫‪8‬‬
 ‫ﺿﻌﻒ ﻣﺎﺷﻴﻨﻬﺎی ﻣﺒﺘﻨﯽ ﺑﺮ ﺍﻧﺒﺎﺷﺘﮕﺮ‬
‫• ﻫﻨﻮﺯ ﻧﻴﺎﺯﻣﻨﺪ ﺫﺧﻴﺮﻩ ﺳﺎﺯی ﺑﺴﻴﺎﺭی ﻣﻘﺎﺩﻳﺮ ﻣﻮﻗﺘﯽ ﻭ ﻣﻴﺎﻧﯽ‬
‫ﺩﺭ ﺣﺎﻓﻈﻪ ﻣﯽ ﺑﺎﺷﻴﻢ‬

‫• ﺩﺭ ﻭﺍﻗﻊ ﺍﻧﺒﺎﺷﺘﮕﺮ ﻓﻘﻂ ﺑﺮﺍی ﻳﮏ ﺗﺮﺗﻴﺐ ﺍﺯ ﻣﺤﺎﺳﺒﺎﺕ ﮐﻪ ﺩﺭ‬

‫ﺁﻥ ﻧﺘﻴﺠﻪ ﻳﮑﯽ‪ ،‬ﻭﺭﻭﺩی ﺑﺮﺍی ﺑﻌﺪی ﺍﺳﺖ‪ ،‬ﻣﻔﻴﺪ ﻣﯽ ﺑﺎﺷﺪ‪.‬‬

‫‪Advanced Computer Architecture & Design‬‬

‫‪10‬‬
‫‪Science & Research Branch‬‬
‫ﭘﻴﺸﻨﻬﺎﺩﺍﺕ ﺑﺮﺍی ﻣﺎﺷﻴﻨﻬﺎی ﻣﺒﺘﻨﯽ ﺑﺮ ﺍﻧﺒﺎﺷﺘﮕﺮ‬

‫• ﺍﮔﺮ ﻣﻨﺎﺑﻊ ﺳﺨﺖ ﺍﻓﺰﺍﺭی ﺑﻴﺸﺘﺮی ﺩﺭ ﺩﺳﺘﺮﺱ ﺍﺳﺖ‪ ،‬ﻣﮑﺎﻧﻬﺎی‬

‫ﺫﺧﻴﺮﻩ ﺳﺎﺯی ﺳﺮﻳﻊ ﺭﺍ ﺩﺭ ﮐﻨﺎﺭ ﺍﻧﺒﺎﺷﺘﮕﺮ ﻗﺮﺍﺭ ﺩﻫﻴﺪ‪:‬‬
‫• ﻣﺎﺷﻴﻨﻬﺎی ﻣﺒﺘﻨﯽ ﺑﺮ ﭘﺸﺘﻪ‬
‫• ﻣﺎﺷﻴﻨﻬﺎی ﻣﺒﺘﻨﯽ ﺑﺮ ﺭﺟﻴﺴﺘﺮ‬
‫‪ -‬ﺧﺎﺹ ﻣﻨﻈﻮﺭﻩ‬
‫‪ -‬ﻋﺎﻡ ﻣﻨﻈﻮﺭﻩ‬

‫‪Advanced Computer Architecture & Design‬‬

‫‪11‬‬
‫‪Science & Research Branch‬‬
 ‫ﻣﺎﺷﻴﻨﻬﺎی ﻣﺒﺘﻨﯽ ﺑﺮ ﭘﺸﺘﻪ‬

‫ﺍﻳﺪﻩ‪ :‬ﻳﮏ ﺳﺘﻮﻥ ﺍﺯ ﻣﮑﺎﻧﻬﺎی ﺫﺧﻴﺮﻩ ﺳﺎﺯی ﺳﺮﻳﻊ ﺩﺍﺭﺍی ﻳﮏ ﺑﺎﻻ)‪ (top‬ﻭ ﻳﮏ ﭘﺎﻳﻴﻦ )‪(bottom‬‬

‫ﻳﮏ ﺩﺳﺘﻮﺭﺍﻟﻌﻤﻞ ﻓﻘﻂ ﺍﺯ ﻣﻘﺪﺍﺭ ﺑﺎﻻی ﭘﺸﺘﻪ)‪ (top‬ﻗﺎﺑﻠﻴﺖ‬

‫‪Address‬‬ ‫‪Contents‬‬ ‫ﺑﺮﺩﺍﺷﺖ ﺩﺍﺭﺩ‪ ،‬ﻳﺎ ﺷﺎﻳﺪ ﺩﻭ ﻳﺎ ﺳﻪ ﻣﻘﺪﺍﺭ ﺑﻪ ﻋﻨﻮﺍﻥ ﺑﺎﻻی‬
‫‪top‬‬ ‫‪14‬‬ ‫ﭘﺸﺘﻪ ﺩﺭ ﻧﻈﺮ ﮔﺮﻓﺘﻪ ﺷﻮﻧﺪ‪.‬‬
‫‪2nd from top‬‬ ‫‪3‬‬
‫‪3rd from top‬‬ ‫‪99‬‬
‫ﻣﯽ ﺗﻮﺍﻧﻴﻢ ﻣﻘﺎﺩﻳﺮ ﺟﺪﻳﺪ ﺭﺍ ﺩﺭ ﺑﺎﻻی ﭘﺸﺘﻪ ﻗﺮﺍﺭ ﺩﻫﻴﻢ‬
‫‪.‬‬ ‫‪.‬‬
‫‪.‬‬
‫‪.‬‬
‫‪.‬‬
‫‪.‬‬
‫)“‪ (”push‬ﻳﺎ ﺍﺯ ﺑﺎﻻی ﭘﺸﺘﻪ ﺑﺮﺩﺍﺭﻳﻢ )“‪(”pop‬‬
‫‪bottom‬‬ ‫‪0‬‬
‫ﻧﻤﯽ ﺗﻮﺍﻧﻴﻢ ﺑﻪ ﻣﮑﺎﻧﻬﺎﻳﯽ ﺩﺭ ﺯﻳﺮ ﭘﺸﺘﻪ ﺩﺳﺘﺮﺳﯽ ﺩﺍﺷﺘﻪ ﺑﺎﺷﻴﻢ‬
‫ﻣﮕﺮ ﺍﻳﻨﮑﻪ ﻫﺮ ﭼﻴﺰ ﺑﺎﻻی ﺁﻥ ﺭﺍ ﺧﺎﺭﺝ ﮐﻨﻴﻢ‪.‬‬

‫‪Advanced Computer Architecture & Design‬‬

‫‪12‬‬
‫‪Science & Research Branch‬‬
 ‫ﻣﻌﻤﺎﺭی ﻣﺠﻤﻮﻋﻪ ﺩﺳﺘﻮﺭﺍﻟﻌﻤﻞ ﻣﺎﺷﻴﻨﻬﺎی ﻣﺒﺘﻨﯽ ﺑﺮ‬
‫ﭘﺸﺘﻪ‬
‫ﺍﻋﻤﺎﻝ ﺍﺻﻠﯽ ﺷﺎﻣﻞ‪:‬‬

‫ﺑﺎﺭﮔﺬﺍﺭی‪ :‬ﺑﺮﺩﺍﺷﺖ ﻣﻘﺪﺍﺭ ﺍﺯ ﺣﺎﻓﻈﻪ ﻭ ﻗﺮﺍﺭ ﺩﺍﺩﻥ ﺁﻥ ﺑﺮ ﺭﻭی ﭘﺸﺘﻪ‬

‫ﺫﺧﻴﺮﻩ ﺳﺎﺯی‪ :‬ﺑﺮﺩﺍﺷﺖ ﻣﻘﺪﺍﺭ ﺍﺯ ﭘﺸﺘﻪ ﻭ ﺫﺧﻴﺮﻩ ﺁﻥ ﺩﺭ ﺣﺎﻓﻈﻪ‬

‫ﺣﺴﺎﺑﯽ‪ :‬ﺧﺎﺭﺝ ﮐﺮﺩﻥ ﻳﮏ ﻳﺎ ﺩﻭ ﻣﻘﺪﺍﺭ ﺍﺯ ﭘﺸﺘﻪ؛ ﻗﺮﺍﺭ ﺩﺍﺩﻥ ﻧﺘﻴﺠﻪ ﺭﻭی ﭘﺸﺘﻪ‬

‫ﺩﻭﻧﺴﺨﻪ ﺍی )‪ :(Dup‬ﮔﺮﻓﺘﻦ ﻣﻘﺪﺍﺭ ﺍﺯ ﺑﺎﻻی ﭘﺸﺘﻪ ﺑﺪﻭﻥ ﺣﺬﻑ ﺁﻥ؛ ﻗﺮﺍﺭ ﺩﺍﺩﻥ ﻳﮏ ﮐﭙﯽ ﺟﺪﻳﺪ ﺍﺯ ﺁﻥ ﺩﺭ‬
‫ﺭﻭی ﭘﺸﺘﻪ‬

‫‪Advanced Computer Architecture & Design‬‬

‫‪13‬‬
‫‪Science & Research Branch‬‬
‫ﺭﺍ ﺍﻧﺠﺎﻡ ﻣﯽ‬ A=(B+C)*(D+E) ‫ﻣﺎﺷﻴﻦ ﻣﺒﺘﻨﯽ ﺑﺮ ﭘﺸﺘﻪ ﮐﻪ‬
‫ﺩﻫﺪ‬

XXX (stack top at start) (D+E)

XXX ADD

(D)
XXX LOAD M[20] (B)
(D+E)
XXX LOAD M[48]
(E)
(D) (continued next slide)
XXX LOAD M[32]

Advanced Computer\ Architecture & Design 14

Science & Research Branch
 ‫ﻣﺎﺷﻴﻦ ﻣﺒﺘﻨﯽ ﺑﺮ ﭘﺸﺘﻪ‬

(C) ((B+C)*(D+E))
(B) XXX MULT
(D+E)
XXX LOAD M[76]

XXX STORE M[100]

(B+C)
(D+E)
XXX ADD

.‫ﺩﻗﺖ ﮐﻨﻴﺪ ﮐﻪ ﺍﮐﻨﻮﻥ ﭘﺸﺘﻪ ﻣﺸﺎﺑﻪ ﺯﻣﺎﻧﯽ ﺍﺳﺖ ﮐﻪ ﺷﺮﻭﻉ ﻧﻤﻮﺩﻳﻢ‬

Advanced Computer Architecture & Design
15
Science & Research Branch
‫ﮐﺎﺭﺑﺮﺩ ﻣﺎﺷﻴﻨﻬﺎی ﻣﺒﺘﻨﯽ ﺑﺮﭘﺸﺘﻪ‬

‫• ﺑﺴﻴﺎﺭی ﺍﺯ ﮐﺎﻣﭙﻴﻮﺗﺮﻫﺎی ﺍﻭﻟﻴﻪ‬

‫• ﻭﺍﺣﺪ ﻣﻤﻴﺰ ﺷﻨﺎﻭﺭ ﺩﺭ ‪) 8086‬ﺗﻘﺮﻳﺒﺎ(‬

‫• ﻣﺎﺷﻴﻦ ﻣﺠﺎﺯی ﺟﺎﻭﺍ‬

‫‪Advanced Computer Architecture & Design‬‬

‫‪16‬‬
‫‪Science & Research Branch‬‬
 ‫ﻣﺎﺷﻴﻨﻬﺎی ﻣﺒﺘﻨﯽ ﺑﺮ ﺭﺟﻴﺴﺘﺮ‬
‫ﺍﻳﺪﻩ‪ :‬ﺑﻪ ﮐﺎﺭ ﺑﺮﺩﻥ ﻣﮑﺎﻧﻬﺎی ﺫﺧﻴﺮﻩ ﺳﺎﺯی ﺯﻳﺎﺩ ﺭﺟﻴﺴﺘﺮﻫﺎ ﻧﺰﺩﻳﮏ ﺍﻧﺒﺎﺷﺘﮕﺮ‬

‫ﺭﺟﻴﺴﺘﺮ ﻫﺎ ﻧﺎﻡ ‪ /‬ﺷﻤﺎﺭﻩ ﻫﺎی ﻣﺸﺨﺼﯽ ﺩﺍﺭﻧﺪ ﮐﻪ ﻣﯽ ﺗﻮﺍﻧﻨﺪ ﺑﻪ ﺟﺎی ﺣﺎﻓﻈﻪ ﺍﺳﺘﻔﺎﺩﻩ ﺷﻮﻧﺪ‬ ‫•‬

‫ﺩﺳﺘﻴﺎﺑﯽ ﻧﺴﺒﺖ ﺑﻪ ﺣﺎﻓﻈﻪ ﺍﺻﻠﯽ ﺧﻴﻠﯽ ﺳﺮﻳﻌﺘﺮ ﺍﺳﺖ‬ ‫•‬

‫ﺭﺟﻴﺴﺘﺮ ﻫﺎ ﻧﺴﺒﺖ ﺑﻪ ﻣﮑﺎﻥ ﻫﺎی ﺣﺎﻓﻈﻪ ﺧﻴﻠﯽ ﻧﺰﺩﻳﮏ ﺗﺮﻧﺪ‬ ‫•‬

‫‪ 32 MIPS ‬ﺗﺎ ﺛﺒﺎﺕ ‪ 32‬ﺑﻴﺘﯽ ﺩﺍﺭﺩ‬

‫ﻭﺗﻌﺪﺍﺩ ﺑﻴﺘﻬﺎی ﮐﻤﺘﺮ ﺑﺮﺍی ﻧﺎﻡ ﮔﺬﺍﺭی ﺁﻧﻬﺎ‬ ‫ﺁﺩﺭﺱ ﻫﺎی ﮐﻮﭼﮑﺘﺮ‬ ‫‪ ‬ﺭﺟﻴﺴﺘﺮ ﻫﺎی ﮐﻤﺘﺮ‬

‫‪ ‬ﻣﻨﺎﺑﻊ ﺍﺳﺘﻔﺎﺩﻩ ﮐﻤﺘﺮ ﺍﺳﺖ ﻭ ﺑﺎﻳﺪ ﺑﺎ ﺩﻗﺖ ﺍﺯ ﺁﻧﻬﺎ ﺍﺳﺘﻔﺎﺩﻩ ﺷﻮﺩ‬

‫‪Advanced Computer Architecture & Design‬‬

‫‪17‬‬
‫‪Science & Research Branch‬‬
 ‫ﺭﺟﻴﺴﺘﺮﻫﺎی ﻋﺎﻡ ﻣﻨﻈﻮﺭﻩ‪/‬ﺧﺎﺹ ﻣﻨﻈﻮﺭﻩ‬
‫• ﻳﮏ ﺭﺟﻴﺴﺘﺮ ﺧﺎﺹ ﻣﻨﻈﻮﺭﻩ ﺑﺮﺍی ﺍﻫﺪﺍﻑ ﻣﺸﺨﺺ ﺍﺳﺘﻔﺎﺩﻩ ﻣﯽ ﺷﻮﺩ‬
‫ﻭ ﻣﻤﮑﻦ ﺍﺳﺖ ﻋﻤﻠﻴﺎﺗﯽ ﺭﺍ ﮐﻪ ﺍﺳﺘﻔﺎﺩﻩ ﻣﯽ ﺷﻮﺩ ﺭﺍ ﻣﺤﺪﻭﺩﮐﻨﺪ‬
‫‪‬ﻃﺮﺍﺣﯽ ﺁﺳﺎﻧﺘﺮ ﺳﺨﺖ ﺍﻓﺰﺍﺭ‪ :‬ﺛﺒﺎﺕ ﺭﺍ ﺩﺭ ﺟﺎﻳﯽ ﮐﻪ ﺩﻗﻴﻘﺎ ﻧﻴﺎﺯﺍﺳﺖ‬
‫ﻗﺮﺍﺭ ﺑﺪﻩ‬
‫• ﻳﮏ ﺭﺟﻴﺴﺘﺮﻋﺎﻡ ﻣﻨﻈﻮﺭﻩ ﻣﯽ ﺗﻮﺍﻧﺪ ﺩﺭﺑﻴﺸﺘﺮ ﻣﺴﻴﺮﻫﺎی ﻋﻤﻠﻴﺎﺗﯽ‬
‫ﺍﺳﺘﻔﺎﺩﻩ ﺷﻮﺩ ‪.‬‬
‫‪ ‬ﺑﻨﺎﺑﺮﺍﻳﻦ ﻣﺴﻴﺮﻳﺎﺑﯽ ﺑﺮ ﺍﺳﺎﺱ ﮐﺎﺭﺑﺮﺩ ﻣﯽ ﺗﻮﺍﻧﺪ ﻣﺸﮑﻞ ﺑﺎﺷﺪ‬

‫‪Advanced Computer Architecture & Design‬‬

‫‪18‬‬
‫‪Science & Research Branch‬‬
 ‫ﻣﻌﻤﺎﺭی‬MIPS‫ﺑﻴﺎﻥ ﺷﺪﻩ‬
‫ﻋﻤﻠﻮﻧﺪﻫﺎی‬MIPS

Name Example Comments

$s0-$s7, $t0-$t9, Fast locations for data. In MIPS, data must be in registers to perform
32 registers $zero, $a0-$a3, $v0- arithmetic. MIPS register $zero always equals 0. $gp (28) is the global
$v1, $gp, $fp, $sp, $ra pointer, $sp(29) is the stack pointer, $fp (30) is the frame pointer, and $ra
(31) is the return address.
Memory [0], Accessed only by data transfer instructions. MIPS uses byte addresses, so
30
2 memory words Memory [4],…, sequential words differ by 4. Memory holds data structures, such as arrays,
Memory[42949672920 and spilled register, such as those saved on procedure calls.

MIPS ‫ﺯﺑﺎﻥ ﺍﺳﻤﺒﻠﯽ‬

Advanced Computer Architecture & Design

19
Science & Research Branch
 ‫ﻧﻤﺎﻳﺶ ﺩﺳﺘﻮﺭﺍﻟﻌﻤﻞ ﻫﺎ ﺩﺭ ﮐﺎﻣﭙﻴﻮﺗﺮ‬
‫‪ ‬ﺩﺭ ﺯﺑﺎﻥ ﺍﺳﻤﺒﻠﯽ ‪: MIPS‬‬
‫ﺭﺟﻴﺴﺘﺮﻫﺎی ‪ $s7 - $s0‬ﺑﻪ ﺭﺟﻴﺴﺘﺮﻫﺎی ‪ 16‬ﺗﺎ ‪23‬‬
‫ﻭ‬
‫ﺭﺟﻴﺴﺘﺮﻫﺎی ‪ $t7 - $t0‬ﺑﻪ ﺭﺟﻴﺴﺘﺮﻫﺎی ‪ 8‬ﺗﺎ ‪ 15‬ﻧﮕﺎﺷﺖ ﻣﯽ‬
‫ﺷﻮﻧﺪ‪.‬‬
‫‪ ‬ﻫﺮﮐﺪﺍﻡ ﺍﺯ ﺑﺨﺶ ﻫﺎی ﺩﺳﺘﻮﺭﺍﻟﻌﻤﻞ ﻳﮏ ﻣﻴﺪﺍﻥ ﻳﺎ ‪field‬‬
‫ﻧﺎﻣﻴﺪﻩ ﻣﯽ ﺷﻮﺩ‪.‬‬

‫‪Advanced Computer Architecture & Design‬‬

‫‪20‬‬
‫‪Science & Research Branch‬‬
‫‪ ‬ﺑﺮﺍی ﺗﻤﺎﻳﺰ ﺑﺎ ﺯﺑﺎﻥ ﺍﺳﻤﺒﻠﺮ‪،‬ﻧﻮﻉ ﻋﺪﺩی ﺩﺳﺘﻮﺭﺍﻟﻌﻤﻞ ﺭﺍ ﺯﺑﺎﻥ‬
‫ﻣﺎﺷﻴﻦ ﻭ ﺗﺮﺗﻴﺒﯽ ﺍﺯ ﺩﺳﺘﻮﺭﺍﻟﻌﻤﻞ ﻫﺎ ﺭﺍ ﮐﺪ ﻣﺎﺷﻴﻦ ﻣﯽ ﻧﺎﻣﻴﻢ‪.‬‬

‫‪ ‬ﺍﻳﻦ ﺗﺮﮐﻴﺐ ﺩﺳﺘﻮﺭﺍﻟﻌﻤﻞ ‪ ،‬ﻗﺎﻟﺐ ﺩﺳﺘﻮﺭﺍﻟﻌﻤﻞ ﻧﺎﻡ ﺩﺍﺭﺩ‪.‬‬

‫‪ ‬ﺍﺻﻞ ﺷﻤﺎﺭﻩ ‪:4‬‬

‫ﻃﺮﺍﺣﯽ ﺧﻮﺏ ﺑﻪ ﺍﻧﻄﺒﺎﻕ ﻭ ﺳﺎﺯﺵ ﻧﻴﺎﺯ ﺩﺍﺭﺩ‪.‬‬

‫‪Advanced Computer Architecture & Design‬‬

‫‪21‬‬
‫‪Science & Research Branch‬‬
Advanced Computer Architecture & Design
22
Science & Research Branch
Advanced Computer Architecture23
& Design
Science & Research Branch