摘要:
In one embodiment, a method for specifying addressability in a memory-mapped device is disclosed. A data access primitive is used to model addressablity for the memory-mapped device. Addressability comprises an address matching function, a lane matching function and one or more bus connections. A first starting address for the memory-mapped device is specified. A first set of addressing matching function, lane matching function and one or more bus connections for the memory-mapped device is generated using the data access primitive and the first starting address.
摘要:
A method for asserting an address alignment of an address for a memory-mapped device in a logic design is disclosed. An align primitive comprising an alignment size port, an input address port and an output address port is used. The alignment size port has data indicating a desired address boundary. The input address port is used for an address to be verified against the desired address boundary. The output address port is used to provide an address that is on the desired address boundary. The address to be verified against the desired address boundary is provided at the output address port when that address meets the desired address boundary. Another method for specifying an offset address for a memory-mapped device in a logic design is disclosed. An offset primitive is used to assert an address for the memory-mapped device. The offset primitive comprises an incoming address port, an outgoing address port and an offset value port. The offset value port has a data value indicating a desired address offset. The incoming address port has a base address to calculate an offset address. The outgoing address port has the offset address. The offset value is a multiple of a transaction size at the memory-mapped device.
摘要:
A bidirectional bus structure includes a first multiplexer path propagating signals in a first direction and a second multiplexer path propagating signals in a second direction. For one embodiment, the bus structure further includes a circuit for selectively combining the signals on the first and second paths and selectively propagating the signal on one of the first and second paths. For another embodiment, the bus structure further includes a logic gate for combining the signals on the first and second paths and a circuit for selectively propagating the signal on one of the first path, the second path, and an output signal of the logic gate. For both embodiments, the present invention allows multiple signals to use the bus without contention, thereby providing an extremely flexible interconnect routing resource. This bidirectional bus can selectively drive signals onto the general interconnect as well as onto a system bus in a configurable system on a chip.
摘要:
The configurable processor system includes a processor, an internal system bus, and a programmable logic all interconnected via the internal system bus, on a single integrated circuit.
摘要:
Interface logic is disclosed. The interface logic comprises a first address decoder, a first set of mode logic coupled to the address decoder and a first selector coupled to the first set of mode logic. The interface logic is adaptable to connect the programmable logic to the system interconnect via one of a plurality of access modes supported by the system interconnect.
摘要:
A transceiver with non-deterministic delay characteristics is analyzed and adjusted to provide a transceiver with deterministic delay characteristics. The transceiver may be implemented with a variety of device types to support high bandwidth operation over a wide range of frequencies. Deterministic behavior allows use of the transceiver in source synchronous interfaces. The transceiver may also be dynamically analyzed and adjusted during operation as operation frequency changes.
摘要:
A method of configuring a logic block of a programmable logic device (PLD) during physical implementation of a circuit design, wherein ports of the logic block are selectively registered, can include identifying the logic block of the PLD, wherein the logic block is located on a critical path. For each of a plurality of selectively registerable portions of the logic block, the method can include computing input slacks and output slacks based upon potential register usage within the logic block. The method further can include determining register usage for the logic block by maximizing a function which depends upon a measure of worst case slack for pipeline stages.
摘要:
A computer-implemented method of implementing a circuit design within a target integrated circuit (IC) can include, during technology mapping of the circuit design, determining a plurality of implementations of at least one sub-circuit of the circuit design and placing the circuit design on the target IC using a primary implementation of the plurality of implementations of the sub-circuit. The primary implementation of the sub-circuit can be selectively replaced with an alternate implementation of the sub-circuit selected from the plurality of implementations of the sub-circuit. The placed circuit design, including either the primary implementation or the alternate implementation of the sub-circuit, can be output.
摘要:
Methods of directly targeting specified routing resources in a PLD, e.g., routing resources that need to be tested. Test designs are produced that implement observable nets using the targeted routing resources. A PLD router is used to route from a target routing resource backwards through the routing fabric of the PLD to the source of an observable net. The net is identified based on the source, and loads of the net are identified as router load targets. The router is then used to route from the target routing resource forwards to one of the loads on the net. This process can be repeated for a list of target routing resources to provide a test design that tests as many of the targeted routing resources as possible. Additional test designs can be created to test remaining target routing resources. In other embodiments, the router routes first forwards, then backwards.
摘要:
Associative management of distributed multimedia assets and associated resources using multi-domain agent-based communication between heterogeneous peers is achieved using an Asset/Resource Management (ARM) platform architecture that has an ARM Framework that is used by Asset Management Agents. The ARM Framework includes an ARM Infrastructure which is a system of protocols and libraries from which communities of agents that are grouped in logical Agent Domains are built. The agents communicate via the KQML language embedded within TCP/IP messages, advertise their capabilities and cooperate together to perform meaningful work. An XML-based language is used to embed “content” within the KQML language, providing a self-describing data representation using various character sets. The ARM Framework includes system agents including in each Agent Domain a Resolver for keeping track of asset logical locations, an Agent Name Server (ANS) for keeping track of security access to the assets, and the “Yellow Pages” containing the Advertised System Knowledge (ASK) agent for keeping track of the capabilities of the Asset Management Agents.