摘要:
To enable devices to detect L3 roaming users and to take appropriate forwarding actions, L3 knowledge is introduced inside a bridge in a non-intrusive way. In particular, as a client moves from a subnet associated with a first network element to a subnet associated with a second network element, a determination is made regarding whether the client is roaming. This is done by evaluating a source IP address within a L3 packet header within a first frame received at the second network element. If, as a result of the evaluating step, it is determined that the client is roaming, an L2 bridge forwarding table in the second network element is configured to include a source MAC address of the client together with information identifying at least a destination interface for use in directing client data traffic back towards the subnet associated with the first network element. The first frame is then forwarded. In one embodiment, the traffic is directed back towards the subnet associated with the first network element via a GRE encapsulation tunnel, although any convenient tunneling mechanism can be used. According to another feature, given information cached at the foreign access point is used to enable the roaming client to continue to seamlessly receive inbound traffic prior to or during the configuration of the L2 bridge forwarding table (i.e., before any outbound traffic is actually sent from the client).
摘要:
A mass storage system made of flash electrically erasable and programmable read only memory (“EEPROM”) cells organized into blocks, the blocks in turn being grouped into memory banks, is managed to even out the numbers of erase and rewrite cycles experienced by the memory banks in order to extend the service lifetime of the memory system. Since this type of memory cell becomes unusable after a finite number of erase and rewrite cycles, although in the tens of thousands of cycles, uneven use of the memory banks is avoided so that the entire memory does not become inoperative because one of its banks has reached its end of life while others of the banks are little used. Relative use of the memory banks is monitored and, in response to detection of uneven use, have their physical addresses periodically swapped for each other in order to even out their use over the lifetime of the memory.
摘要:
To reduce FIFO access cycles across a system bus in a multi-processor system in which two processors communicate across a system bus through a FIFO, two separate FIFO descriptors are provided. The first descriptor is maintained by the processor located on-board with the FIFO, and the second descriptor is maintained by an off-board processor which communicates with the FIFO across the bus. When one processor performs a FIFO operation, the processor updates the other processor's descriptor via a memory access across the bus. Additionally, one module passes credits to the other to indicate that the latter has permission to perform a plurality of FIFO operations consecutively. In one embodiment a special non-valid data value is used to indicate an empty FIFO position.
摘要:
In a sector in a flash memory array PAGE ERASE and MULTIPLE PAGE ERASE modes of operation are provided. In the PAGE ERASE and MULTIPLE PAGE ERASE modes of operation, a preferred tunneling potential of approximately −10 Volts is applied to the gates of the flash memory cells on the row or rows being selected for erasure, and the bitlines connected to the drains of the flash memory cells are driven to a preferred voltage of approximately 6.5 Volts. To reduce the unintended erasure of memory cells in rows other than the selected row or rows, a preferred bias voltage of approximately 1 to 2 Volts is applied to the gates of all the flash memory cells in the rows other than the selected row or rows.
摘要:
Microprocessor main programs and their interrupt handling routines are written in a high level programming language such as C. Each is compiled separately, and each is compiled invoking a compiler option which commands the compiler to not use a given set of registers in the compiled code. Post-processing is then performed on the compiled interrupt code to replace accesses to a first set of registers with accesses to the given set of registers. The result is that while both the main program and the interrupt handler were written in C, the compiled code for each employs different registers. This allows context switching from the main program to the interrupt handler and back again with almost none of the overhead traditionally associated with context switching register save and restore operations during exception handling.
摘要:
A flash memory array architecture comprising a flash memory array, first and second memory buffer, and I/O interface circuit which has several operating modes which permit data to be read from the flash memory array, several operating modes which permit data to be programmed into the flash memory array, and a mode for rewriting the data in the flash memory array.In the four read modes, one of the pages stored in the flash memory array is read, the data stored in either of first or second memory buffers is read, the data in one of the pages of data stored in the flash memory array is read and then written into either of first or second memory buffers, the data in one of the pages of data stored in the flash memory array is read and then compared to the data read from either of first or second memory buffers. In the four write modes, data from an input stream is written into a selected first or second memory buffer, one of the pages of data stored in the flash memory array is erased, and then in the same cycle, data in either of first or second memory buffers is written into the erased page in the flash memory array, data in either of first or second memory buffers is written into a previously erased page in the flash memory array, and data from an input stream is written into the selected first or second memory buffer, one of the pages of data stored in the flash memory array is erased, and then in the same cycle, data in either of first or second memory buffers is written into the erased page in the flash memory array. In the auto page rewrite mode, the data in one of the pages of data stored in the flash memory array is read and then written into either of first or second memory buffers. The data stored in the page of the flash memory array just read is erased, and then in the same cycle, the page of data stored in the selected first or second memory buffer is written into the erased page in the flash memory array.
摘要:
A look-up table circuit implemented with MOS transistors that uses combinational logic to generate signals that enable the transistors. A circuit using 16 inputs and 4 select lines is disclosed. Two of the select lines are used as inputs to combinational logic including four NOR gates to generate enable signals for transistors in a third stage of the circuit. This produces a reduction in the propagation delay of a signal from the input to the output of the look-up table circuit.
摘要:
A negative voltage decoder applies a negative voltage to the sense line of a selected row of a memory array but not to sense lines of unselected rows. The negative voltage decoder includes a negative voltage source, an array of P-channel transistors, and a negative voltage address signal generator. P-channel transistors in the array have gates coupled to address lines, so that address signals on the address lines turn on the P-channel transistors and connect only the selected sense line to the negative voltage source. A negative voltage charge pumps in the negative voltage address signal generator generates address signals lower than the negative voltage source. In one embodiment, the transistor array has rows of P-channel transistors which fit the pitch of the memory array and individual P-channel transistors which are stacked laterally away from the memory array, and each row of P-channel transistors couples through a set of individual transistor to a set of sense lines. When a positive voltage decoder applies a positive voltage to the sense lines, the negative voltage address signal generator provides a high voltage to shut off the transistors directly coupled to the sense lines. An isolation circuit isolates the positive voltage row decoder from the negative voltage applied by the negative voltage decoder.
摘要:
A look-up table circuit implemented with MOS transistors that uses combinational logic to generate signals that enable the transistors. A circuit using 16 inputs and 4 select lines is disclosed. Two of the select lines are used as inputs to combinational logic including four NOR gates to generate enable signals for transistors in a third stage of the circuit. This produces a reduction in the propagation delay of a signal from the input to the output of the look-up table circuit.
摘要:
A charge pump for providing programming voltages to the word lines of a semiconductor memory array is disclosed. The charge pump, configured as a combination of enhancement and native MOS transistors, prevents DC current from flowing from the source of the programming voltage to ground through unselected word lines, and thereby permits the design of semiconductor programmable memory arrays having on-chip programming voltage generation, allowing for design of semiconductor programmable memory arrays which operate from a single voltage power supply.