Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

Is there a special way to map an address to SDR SDRAM?

Status
Not open for further replies.
M

matrixofdynamism

Advanced Member level 2
Joined
Apr 17, 2011
Messages
593
Helped
24
Reputation
48
Reaction score
23
Trophy points
1,298
Activity points
7,681
Is there special way to map address to SDR SDRAM?

SDR SDRAM like other DRAMs contains banks, rows, columns. Thus, to get maximum efficiency, I assume that there would be some non-trivial way to translate memory address to bank, row and column address. Is this true? If so, then how is it usually done?
 

Re: Is there special way to map address to SDR SDRAM?

You read the data sheet.
There will be a number of address pins each for row, bank and column. So you know how many of each by doing 2^(npins)
 

Re: Is there special way to map address to SDR SDRAM?

The optimal scheme is almost trivial. You get fastest access when reading or writing full pages, thus the lower address bits are usually mapped to column address lines.
 
Re: Is there special way to map address to SDR SDRAM?

normally, the addresses are mapped either (bank, row, column) or (row, bank, column). You can choose a different scheme if you want though. placing bank at higher address bits has a little advantage if you have a small number of independent applications that are allocated a large memory space. placing bank at lower address bits allows a larger sequential region to have only column accesses once all banks are open.
 

Re: Is there special way to map address to SDR SDRAM?

To access SDRAM in an efficient manner, it is better to follow linear addressing scheme as told in #4 that is combining row address, column address and bank address into a single vector and then increment this single vector to access different memory locations.

But as switching from one row to another row is slow, therefore keeping the row address in MSB is recommended.

Therefore I think (row,bank,column) will be efficient in terms of time.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top