Abstract:
PROBLEM TO BE SOLVED: To provide a technology for distributing a message via a point-to-point network. SOLUTION: In embodiment of this invention, a message is distributed between electronic components inside the point-to-point mutual connection. This system is constructed of a first bus agent, which is arranged on a first bus for transmitting a broadcast message to one or more bus agents arranged on one or more buses except the first bus, and a second bus agent arranged on a second bus for receiving the broadcast message and notifying receipt of the broadcast message to the first bus agent. COPYRIGHT: (C)2006,JPO&NCIPI
Abstract:
PROBLEM TO BE SOLVED: To realize transfer of a transaction without dead lock through a system present for a number of system components of different protocols. SOLUTION: In one embodiment of the present invention, a method may include a step for separating an input transaction to an agent of a coherent system into at least first, second and third channels on the basis of a type of the input transaction. The input transaction may be sent by an equal apparatus connected to the coherent system. The transaction is separated on the basis of the type, so that dead locks may be avoided. COPYRIGHT: (C)2006,JPO&NCIPI
Abstract:
Vorrichtung, umfassend:ein Mehrkernprozessor einschließlich einer Vielzahl von Kernen, ein gemeinsam benutzter Cache-Speicher, ein integriertes Ein-/Ausgabe-(IIO)-Modul, um eine Schnittstelle zwischen dem Mehrkernprozessor und wenigstens einem IO-Gerät herzustellen, das mit dem Mehrkernprozessor gekoppelt ist, und einen Caching-Agenten, wobei der Caching-Agent Kohärenz-Logik einschließt, um Cache-Kohärenz-Operationen für die Vielzahl von Kernen und das IIO-Modul auszuführen und kohärenten Zugriff in Bezug auf die Kerntransaktionen als auch die IO-Gerät-Transaktionen zu ermöglichen, wobei das IIO-Modul eine erste Warteschlange für gepostete Anforderungen und eine zweite Warteschlange für nicht gepostete Anforderungen aufweist und wobei das IIO-Modul eine PCI Express-Schnittstelle aufweist, wobei das IIO-Modul ferner eingerichtet ist, mittels der Anforderungen direkt auf den Cache-Speicher zuzugreifen,wobei der Caching-Agent derart eingerichtet ist, dass bei einem Cache-Speicher-Zugriff eine gepostete Anforderung einer nicht geposteten Anforderung vorausgeht.
Abstract:
Memory read and write requests are received. The read is received in accordance with a communication protocol that has a transction ordering rule in which a memory read cannot pass a memory write. The memory read and write requests are forwarded to the first device in accordance with another communication protocol that has a transaction ordering rule in which a memory read may pass a memory write. The forwarded memory read request is allowed to pass the forwarded memory write request whenever a relaxed ordering flag in the received read request is asserted. Other embodiments are also described and claimed.
Abstract:
Bei einer Ausführungsform schließt die vorliegende Erfindung einen Mehrkernprozessor ein, der eine Vielzahl von Kernen aufweist, einen gemeinsam benutzten Cache-Speicher, ein integriertes Ein-/Ausgabe-(IIO)-Modul, um eine Schnittstelle zwischen dem Mehrkernprozessor und wenigstens einem IO-Gerät herzustellen, das mit dem Mehrkernprozessor gekoppelt ist, und einen Caching-Agenten, um Cache-Kohärenz-Operationen für die Vielzahl von Kernen und das IIO-Modul auszuführen.
Abstract:
The ability to configure an integrated device with a decoder in a processor or network component according to PCI or PCI Express interconnects.
Abstract:
The ability to configure an integrated device with a decoder in a processor or network component according to PCI or PCI Express interconnects.
Abstract:
A POWER FAILURE MODE FOR A MEMORY CONTROLLER, SUCH AS A MEMORY CONTROLLER (200) USED IN AN INPUT/OUTPUT PROCESSOR, WHICH, WHEN THE MEMORY CONTROLLER HAS SYSTEM POWER, REFRESHES A MEMORY UNIT, SUCH AS AN SDRAM MEMORY UNIT (210), AS REQUIRED TO MAINTAIN THE MEMORY IMAGE. IN ONE EMBODIMENT, WHEN A POWER FAILURE OCCURS, THE MEMORY CONTROLLER (200) ISSUES A SELF-REFRESH COMMAND TO THE MEMORY, WHICH HAS BATTERY-BACKUP POWER. A PCI RESET SIGNAL MAY BE USED TO DETERMINE WHEN A POWER FAILURE HAS OCCURRED. THE SELF-REFRESH COMMAND PLACES THE MEMORY IN A SELF-REFRESH MODE, AND A PROGRAMMABLE LOGIC DEVICE MAY BE USED TO ENSURE THAT A CLOCK ENABLE SIGNAL INPUT TO THE MEMORY MAINTAINS THE SELF-REFRESH MODE. WHEN SYSTEM POWER RETURNS, THE MEMORY CONTROLLER (200) RESUMES REFRESHING THE MEMORY (210). MOST ILLUSTRATIVE DRAWING IS (FIGURE 5).
Abstract:
A power failure mode for a memory controller (200), such as a memory controller used in an input/output processor (100), which, when the memory controller has system power, refreshes a memory unit (210), such as an SDRAM memory unit, as required to maintain the memory image. In one embodiment, when a power failure occurs, the memory controller issues a self-refresh command to the memory, which has battery-backup power. A PCI reset signal may be used to determine when a power failure has occured. The self-refresh command places the memory in a self-refresh mode, and a programmable logic device may be used to ensure that a clock enable signal input to the memory maintains the self-refresh mode. When system power returns, the memory controller resumes refreshing the memory.
Abstract:
A shared memory controller is to service load and store operations received, over data links, from a plurality of independent nodes to provide access to a shared memory resource. Each of the plurality of independent nodes is to be permitted to access a respective portion of the shared memory resource. Interconnect protocol data and memory access protocol data are sent on the data links and transitions between the interconnect protocol data and memory access protocol data can be defined and identified.