| Electronic Components Datasheet Search |
|
MPC8540 Datasheet(PDF) 400 Page - Freescale Semiconductor, Inc |
|
|
|||||||||||||||||||||||||||||
MPC8540 Datasheet(HTML) 400 Page - Freescale Semiconductor, Inc |
|
400 / 1302 page ![]() MPC8540 PowerQUICC III Integrated Host Processor Reference Manual, Rev. 1 9-46 Freescale Semiconductor DDR Memory Controller When the DDR memory controller operates in open page mode, it retains the currently active SDRAM page by not issuing a precharge command. The page remains opens until one of the following conditions occurs: • Refresh interval is met • The user-programmable DDR_SDRAM_INTERVAL[BSTOPRE] value is exceeded. • There is a logical bank row collision with another transaction that must be issued. Page mode can dramatically reduce access latencies for page hits. Depending on the memory system design and timing parameters, using page mode can save two to three clock cycles for subsequent burst accesses that hit in an active page. Also, better performance can be obtained by using more banks, especially in systems which use many different channels. Page mode is disabled by clearing DDR_SDRAM_INTERVAL[BSTOPRE] and CS_n_CONFIG[AP_n_EN] 9.5.11 Error Checking and Correcting (ECC) The DDR memory controller supports error checking and correcting (ECC) for the data path between the core master and system memory. The memory detects all double-bit errors, detects all multi-bit errors within a nibble, and corrects all single-bit errors. Other errors may be detected, but are not guaranteed to be corrected or detected. Multiple-bit errors are always reported when error reporting is enabled. When a single-bit error occurs, the single-bit error counter register is incremented, and its value compared to the single-bit error trigger register. An error is reported when these values are equal. The single-bit error registers can be programmed such that minor memory faults are corrected and ignored, but a catastrophic memory failure generates an interrupt. For writes that are smaller than 64 bits, the DDR memory controller performs a double-word read from system memory of the address for the write (checking for errors), and merges the write data with the data read from memory. Then, a new ECC code is generated for the merged double word. The data and ECC code is then written to memory. If a multi-bit error is detected on the read, the transaction completes the read-modify-write to keep the DDR memory controller from hanging. This read-modify-write operation is performed as an atomic transaction in the DDR controller. The write command is then issued 3–5 memory clocks after the completion of the read, depending on various system parameters. However, the corrupt data is masked on the write, so the original contents in SDRAM remain unchanged. The syndrome encodings for the ECC code are shown in Table 9-32 and Table 9-33. Table 9-32. DDR SDRAM ECC Syndrome Encoding Data Bit Syndrome Bit Data Bit Syndrome Bit 012 3456 7 0123 4567 0 •• • 32 •• • 1 •• • 33 •• • 2 •• • 34 •• • |
|
Link URL |
| Does ALLDATASHEET help your business so far? [ DONATE ] |
About Alldatasheet | Advertisement | Contact us | Privacy Policy | Link to Datasheet | Link Exchange | Manufacturer List All Rights Reserved©Alldatasheet.com |
| Russian : Alldatasheetru.com | Korean : Alldatasheet.co.kr | Spanish : Alldatasheet.es | French : Alldatasheet.fr | Italian : Alldatasheetit.com Portuguese : Alldatasheetpt.com | Polish : Alldatasheet.pl | Vietnamese : Alldatasheet.vn Indian : Alldatasheet.in | Mexican : Alldatasheet.com.mx | British : Alldatasheet.co.uk | New Zealand : Alldatasheet.co.nz |
|
Family Site : ic2ic.com |
icmetro.com |