Industry accepted computer processing standards use the concept of virtual slots, wherein multiple modules, cards or boards are interconnected. Virtual slot techniques permit the use of self stacking modules, cards or boards over a processor bus to allow multiple cards on a system without the burden of backplanes. The processor is logically connected to the stacked cards through the bus, and a connection to the bus is shared by the stacked cards.
Processor resources, such as chip select signals, interrupts, common clocks, etc., are allocated across the stacked cards. Manual settings may be used to associate the cards to the resources. This allows similar or even dissimilar hardware units to be cascaded or stacked with different jumper/switch settings or loaded with configured software, so that resources may be distributed without clash. Such jumper/switch settings, software or other manual settings may require system knowledge to set or configure properly. Lack of such knowledge, or mistakes in making the settings can lead to performance and resource clash issues. Some system flexibility, such as the order cards are stacked, may also be sacrificed with this conventional approach.
In the following description, reference is made to the accompanying drawings that form a part hereof, and in which is shown by way of illustration specific embodiments which may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the invention, and it is to be understood that other embodiments may be utilized and that structural, logical and electrical changes may be made without departing from the scope of the present invention. The following description of example embodiments is, therefore, not to be taken in a limited sense, and the scope of the present invention is defined by the appended claims.
The functions or algorithms described herein are implemented in software or a combination of software and human implemented procedures in one embodiment. The software may consist of computer executable instructions stored on computer readable media such as memory or other type of storage devices. The term “computer readable media” is also used to represent any means by which the computer readable instructions may be received by the computer, such as by different forms of wireless transmissions. Further, such functions correspond to modules, which are software, hardware, firmware or any combination thereof. Multiple functions are performed in one or more modules as desired, and the embodiments described are merely examples. The software may be executed on a digital signal processor, ASIC, microprocessor, or other type of processor operating on a computer system, such as a personal computer, server or other computer system.
Stacking of electronics modules, boards or cards, hereinafter referred to as cards is described. In one embodiment, a baseboard has electronic communication with a plurality of daughterboards. The daughterboards may be of the same type (homogeneous) or different types (heterogeneous). The electronic communication is channeled through a route running across the daughterboards.
In a further embodiment, each card in a stack is connected logically and can detect its relative position in the stack on initial power on. Relative address information is available through out operation and can be used dynamically if desired. The relative address information may be used to obtain or associate an appropriate resource pool. In one embodiment, a card, such as a bottom most or top most card is used as a reference and the rest of the cards in the stack derive a relative address with respect to such reference card. A few lines are dedicated between neighboring cards through which the cards can share their relative address information with succeeding cards and automate resource allocation based on the address information.
In one embodiment, the addresses generated are card level addresses, which are then used to enable resource allocation to the cards. In further embodiments, the card level addresses may be used for resource negotiation as well, depending on the complexity of the system. Stacked, or cascaded cards are then self aware of their position in a virtual slot. In addition, information regarding all the cards in a stack can be provided to a processor or other logic device to assist in resource allocation or association. In some embodiments, relative address generation of stacked cards occurs on power-on, and need not involve firmware overhead for derivation of the addresses. Cards may be stacked in any desired order without the need for manual settings. Daughter cards may also use a similar address derivation algorithm.
In one embodiment, the top most stacked card 130 provides a termination of the physical transmission line. In the case of cascaded modules, the termination is provided on the outermost module. In one embodiment, a dedicated add on termination card may be used as the last card in the stack. Address generation techniques may be implemented even with out termination)
Each card in the stack passes through the bus to cards further up the stack. In addition to the bus, a set of address pins 210 are provided that couple adjacent cards in the stack, as shown in
In one embodiment the PLD or other suitable logic, increments the address by adding binary “0001” or otherwise changes the address, and passes that changed address on to the next card. In further embodiments, the address may be changed by any increment/decrement operations of any step or even by a number derived by functions using multiplication/division or further mathematical or table look-up operations. A top most or bottom most card in the stack starts the process by passing a beginning address, which may be “0000” or “0001” or other default value as desired. The default value may be generated by many different means. One example is the use of pull-up or pull-down resistors, or combinations of such resistors connected to input address lines. This default address may be overridden for all cards except the reference card by a neighboring card.
In one embodiment, each device stores the address passed to it, or in the case of the reference card, storing the originally generated address. The address may be stored or may be dynamically used in further embodiments. Such address becomes a virtual address of the card in the stack. In further embodiments, the incremented address may be used, so long as each card obtains a unique address relative to the other cards in the stack. By using a known algorithm for changing the address, each card will also know its relative position within the stack. Further, the host 110 will also know how many cards are in the stack by the last address passed by the lowest card to the bus. The PLD or other logic may then obtain or associate resources for each of the cards, such as chip select, IRQ, clock pin, etc. In further embodiments, other logic in the cards, or host, or anywhere desired may allocate the resources to each of the cards as a function of the virtual address of each card and available system resources. The PLD or other logic may act as a multiplexer/demultiplexer for the hardware resources, which may be board centric.
In one embodiment, the PLD or other logic may detect the top most card, address “0000”, or other desired address, in the stack and automatically enable the termination of the card. A state of the input address lines will be decided based on pull ups/pull downs in the case of reference card. In the case of an intermediate card the state will be driven by its neighboring card which can override the default state set by the pull ups/pull downs. The method of assigning resources may operate without regard to which card is the top most card, also referred to as being in the last virtual slot. This eliminates the need for a separate termination board or manually enabling the termination.
In such embodiment, all cards may be provided with bus termination circuitry which is enabled or disabled according to the card's position in the stack. The PLD or other logic may be used to automatically enable the termination on the last card where the relative address is detected as 0×0000 or another desired address. Other cards in the stack then deactivate or do not enable termination. Controlling the termination may be done through FET or other switch control circuitry or through tri state buffers in the PLD. Other methods may also be used to control the termination.
Resources may then be assigned to cards in the stack as a function of the assigned addresses as indicated at 330. In one embodiment, at least some of the resources are selected from the group consisting of memory, communication channels, clock, interrupts, chip selects, resets, control lines, power controls, analog inputs, digital inputs, power tracks and general purpose I/Os or any other type of electrical/electronic resources. The first address, that of the reference card may be predetermined by pull down or pull up resistors or other means as desired.
In one embodiment, the address is changed by adding 1. The address may be four bit address that is a logical address that identifies the relative position of the card.
In further embodiments, a terminating resistance or impedance may be enabled on the top card when the card is detected as the top card as indicated at 335. The method may be executed when power is applied to the cards or as otherwise desired, such as when a new card is added to the stack of cards. In one embodiment, the cards may be added when power is on, and is referred to as hot plugging.
The Abstract is provided to comply with 37 C.F.R. §1.72(b) to allow the reader to quickly ascertain the nature and gist of the technical disclosure. The Abstract is submitted with the understanding that it will not be used to interpret or limit the scope or meaning of the claims.
Number | Name | Date | Kind |
---|---|---|---|
4773868 | Heinecke | Sep 1988 | A |
5201038 | Fielder | Apr 1993 | A |
5437019 | Brockmann | Jul 1995 | A |
5530887 | Harper et al. | Jun 1996 | A |
5564055 | Asnaashari et al. | Oct 1996 | A |
5604871 | Pecone | Feb 1997 | A |
5948078 | Joffe | Sep 1999 | A |
5963464 | Dell et al. | Oct 1999 | A |
6496940 | Horst et al. | Dec 2002 | B1 |
6502161 | Perego et al. | Dec 2002 | B1 |
6567870 | Thomas et al. | May 2003 | B1 |
6650332 | Doyle et al. | Nov 2003 | B2 |
6683372 | Wong et al. | Jan 2004 | B1 |
6799234 | Moon et al. | Sep 2004 | B1 |
6839393 | Sidiropoulos | Jan 2005 | B1 |
6864569 | Evans | Mar 2005 | B2 |
6882649 | Gura et al. | Apr 2005 | B1 |
6963941 | Lee | Nov 2005 | B1 |
7000052 | Moon et al. | Feb 2006 | B2 |
7215561 | Park et al. | May 2007 | B2 |
7269042 | Kinsley et al. | Sep 2007 | B2 |
7286384 | Wendt et al. | Oct 2007 | B2 |
7298668 | Hoppe et al. | Nov 2007 | B2 |
7337256 | Korowitz et al. | Feb 2008 | B2 |
7493439 | Gower et al. | Feb 2009 | B2 |
7606959 | Yao et al. | Oct 2009 | B2 |
20020034068 | Weber et al. | Mar 2002 | A1 |
20020129186 | Emerson et al. | Sep 2002 | A1 |
20030049948 | Kim et al. | Mar 2003 | A1 |
20040255178 | Hung | Dec 2004 | A1 |
20050160216 | Norman et al. | Jul 2005 | A1 |
20050172085 | Klingman | Aug 2005 | A1 |
20050177671 | Klingman | Aug 2005 | A1 |
20050228931 | Broyles | Oct 2005 | A1 |
20050268070 | Baxter | Dec 2005 | A1 |
20050271072 | Anderson et al. | Dec 2005 | A1 |
20050283743 | Mulholland et al. | Dec 2005 | A1 |
20060288132 | McCall et al. | Dec 2006 | A1 |
Number | Date | Country |
---|---|---|
849684 | Jun 1998 | EP |
1187033 | Mar 2002 | EP |
918556 | Jan 2005 | EP |
WO-0115499 | Mar 2001 | WO |
WO-2005069151 | Jul 2005 | WO |
Number | Date | Country | |
---|---|---|---|
20080077718 A1 | Mar 2008 | US |