US20010027505A1 - Method and apparatus for multiple tier intelligent bus arbitration on a PCI to PCI bridge - Google Patents

Method and apparatus for multiple tier intelligent bus arbitration on a PCI to PCI bridge Download PDF

Info

Publication number
US20010027505A1
US20010027505A1 US09/796,016 US79601601A US2001027505A1 US 20010027505 A1 US20010027505 A1 US 20010027505A1 US 79601601 A US79601601 A US 79601601A US 2001027505 A1 US2001027505 A1 US 2001027505A1
Authority
US
United States
Prior art keywords
pci
tier
bus
bridge
devices
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
US09/796,016
Other versions
US6393508B2 (en
Inventor
David Rekeita
Chen Ding
Krunali Patel
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Individual
Original Assignee
Individual
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Priority to US09/796,016 priority Critical patent/US6393508B2/en
Publication of US20010027505A1 publication Critical patent/US20010027505A1/en
Application granted granted Critical
Publication of US6393508B2 publication Critical patent/US6393508B2/en
Anticipated expiration legal-status Critical
Expired - Lifetime legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F13/38Information transfer, e.g. on bus
    • G06F13/40Bus structure
    • G06F13/4004Coupling between buses
    • G06F13/4027Coupling between buses using bus bridges
    • G06F13/4031Coupling between buses using bus bridges with arbitration

Definitions

  • the present invention relates in general to the field of PCI to PCI bridge devices and more specifically to a method and apparatus for intelligent bus arbitration in a PCI to PCI bridge.
  • a peripheral component interconnect (PCI) bridge provides a connection path between two independent PCI buses.
  • the primary function of a PCI to PCI bridge is to allow transactions to occur between a device on one PCI bus and a device on the other PCI bus.
  • System and option card designers can use multiple PCI to PCI bridges to create an hierarchy of PCI buses. This allows system and option card designers to overcome electrical loading limits.
  • the PCI device that initiates the transaction is called the master and the other PCI device is called the target. If the master and target are on different PCI buses, the bus that the master resides on is the initiating bus. The bus that the target resides on is the target bus.
  • a PCI to PCI bridge has two PCI interfaces, each connected to a PCI bus.
  • the PCI interface of the PCI to PCI bridge that is connected to the PCI bus that is closest to the CPU is the primary interface.
  • the PCI interface of the PCI to PCI bridge that is connected to the PCI bus that is farthest from the CPU is the secondary interface.
  • the PCI bus that is connected to the primary interface of the PCI to PCI bridge is called the primary PCI bus.
  • the PCI bus that is connected to the secondary interface of the PCI to PCI bridge is called the secondary bus.
  • a PCI to PCI bridge acts essentially as an intermediary between devices located on the secondary bus and devices that are located on the primary bus.
  • the two interfaces of the PCI to PCI bridge bus are capable of both master and target operations.
  • the PCI to PCI bridge acts as a target on the initiating bus on behalf of the target that actually resides on the target bus.
  • the PCI to PCI bridge functions as a master on the target bus on behalf of the master that actually resides on the initiating bus.
  • the PCI to PCI-bridge appears as one device where it actually represents several PCI devices that are located on the secondary bus.
  • One such arbitration scheme is a two tier arbitration scheme.
  • the devices located on the secondary bus are categorized into a high tier and a low tier.
  • Each device in the high tier is given the opportunity to access the secondary bus.
  • a single device in the low tier is given the same opportunity.
  • each device in the high tier is given another opportunity to access the bus before another device in the low tier is given an opportunity to access the secondary bus. This process is repeated such that devices in the high tier are given more opportunities to access the secondary bus than devices in the low tier.
  • a method and apparatus for multiple tier intelligent bus arbitration on a PCI to PCI bridge is provided that substantially eliminates or reduces problems associated with previously developed PCI to PCI bridge arbitration schemes.
  • the method of the present invention includes maintaining a first tier and a second tier of devices that have access to the secondary bus that the PCI to PCI bridge services.
  • Each device that has access to the secondary PCI bus is categorized into either the first tier or the second tier.
  • the devices in the first tier are provided more frequent opportunities to gain access to the secondary PCI bus than devices in the low tier.
  • a pending transaction is recognized when an initiating device that has been categorized into the second tier accesses the secondary PCI bus and attempts a transaction that crosses the PCI to PCI bridge to the primary PCI bus.
  • the PCI to PCI bridge is unable to complete the transaction on the primary PCI bus. Therefore, the PCI to PCI bridge is unable to provide access to any other device on the secondary bus until the pending transaction completes.
  • the device that initiated the pending transaction is categorized into the first tier until the pending transaction is completed.
  • An apparatus of the present invention includes a PCI to PCI bridge arbiter operable to execute the two tier intelligent bus arbitration scheme as described in the above method.
  • a further apparatus of the present invention includes a PCI to PCI bridge with an internal arbiter operable to execute the two tier intelligent bus arbitration scheme as described in the above method.
  • a technical advantage of the present invention is that the pending transaction will be completed faster because the initiating device is placed into the first tier of high priority devices such that it has more opportunities to access the secondary bus.
  • FIG. 1 is a block diagram of typical PCI to PCI bridge applications
  • FIG. 2 is a functional block diagram of the typical operation of a PCI to PCI bridge
  • FIG. 3 is a functional block diagram of one embodiment of a PCI to PCI bridge that contains a secondary PCI bus internal arbiter according to the present invention
  • FIG. 4 is a diagram of one embodiment of a bridge arbitration register of a PCI to PCI bridge with a two tier internal arbitration mechanism according to the present invention
  • FIG. 5 is a flow chart of one embodiment of a method of a two tier intelligent bus arbitration scheme according to the present invention.
  • FIGS. 6A, 6B, 6 C, and 6 D are diagrams showing one embodiment of the operation of a two tier intelligent bus arbitration scheme.
  • FIG. 7 is a diagram of one embodiment of the activity of a PCI to PCI bridge containing a two tier arbitration scheme with and without an intelligent bus arbitration scheme according to the present invention.
  • FIG. 1 is a block diagram of typical PCI to PCI bridge applications.
  • a CPU 10 connects to memory 18 via host bus 14 .
  • a host bridge 22 connects host bus 14 to a PCI bus 26 .
  • PCI bus 26 connects PCI devices 30 to host bridge 22 .
  • PCI bus 26 also connects to a PCI to PCI bridge 38 .
  • PCI to PCI bridge 38 in turn connects to PCI option cards 44 via PCI bus 42 .
  • a second PCI to PCI bridge 39 operates via PCI bus 48 to connect PCI devices 32 to PCI bus 42 .
  • the bus that connects to a PCI to PCI bridge and is closest to the CPU is the primary bus and connects to the primary interface of the PCI to PCI bridge. Therefore, for PCI to PCI bridge 38 , interface 36 is the primary interface and PCI bus 26 is the primary bus.
  • interface 40 is the secondary interface and PCI bus 42 is the secondary bus.
  • PCI bus 42 is the secondary bus.
  • interface 37 is the primary interface and PCI bus 42 is the primary bus
  • interface 41 is the secondary interface and PCI bus 48 is the secondary bus.
  • PCI bus 26 is PCI BUS 0.
  • PCI-PCI BRIDGE 1. This corresponds to PCI to PCI bridge 38 .
  • the PCI bus that acts as the secondary bus to a PCI to PCI bridge is given the number of that PCI to PCI bridge. Therefore, PCI bus 42 is called “PCI BUS 1” because it is the secondary bus to PCI-PCI BRIDGE 1 (PCI to PCI bridge 38 ).
  • PCI BUS 1 because it is the secondary bus to PCI-PCI BRIDGE 1 (PCI to PCI bridge 38 ).
  • PCI to PCI bridge 39 is named “PCI-PCI BRIDGE 2”
  • PCI bus 48 is “PCI BUS 2.”
  • PCI Local Bus Specification, Revision 2.1 allow only ten PCI device loads per PCI bus. Therefore, a system that requires more than ten PCI device loads requires a PCI to PCI bridge. For example, in FIG. 1, a maximum of 10 PCI devices 30 may connect to PCI bus 26 and host bridge 22 . However, PCI to PCI bridge 38 appears as one device to PCI bus 26 and host bridge 22 . This allows option cards 44 that contain further PCI devices 32 to access PCI bus 26 and host bridge 22 maintaining the specified electrical loading limits.
  • PCI device 32 In operation, if PCI device 32 initiates a transaction to memory 18 , PCI device 32 is the master and memory 18 is the target. To complete the transaction, PCI device 32 acts as a master along PCI bus 48 to secondary interface 41 of PCI to PCI bridge 39 . Then, primary interface 37 of PCI to PCI bridge 39 acts as a master to complete a transaction to secondary interface 40 of PCI to PCI bridge 38 . Next, primary interface 36 of PCI to PCI bridge 38 initiates a transaction as master to secondary interface of host bridge 22 . Host bridge 22 then acts as master for a transaction to memory 18 .
  • FIG. 2 is a functional block diagram of the typical operation of a PCI to PCI bridge indicated generally at 38 .
  • Data moving downstream first encounters a primary interface 36 . This action initiates a primary target interface 60 .
  • PCI to PCI bridge 38 is acting as the target of the transaction that is taking place on the primary bus.
  • a secondary master interface 64 is triggered, in order for PCI to PCI bridge 38 to act as the master of the transaction on the secondary bus.
  • a secondary target interface 68 is initiated.
  • PCI to PCI bridge 38 is acting as the target of this transaction that is taking place on the secondary bus.
  • a primary master interface 72 is initiated.
  • PCI to PCI bridge 38 acts as the master to the transaction on the primary bus.
  • These transactions are affected by configuration registers 76 that determine the characteristics of PCI to PCI bridge 38 .
  • optional data buffers 80 may exist on PCI to PCI bridge 38 to buffer data involved in transactions across PCI to PCI bridge 38 .
  • FIG. 3 is a functional block diagram of one embodiment of a PCI to PCI bridge that contains a secondary PCI bus internal arbiter according to the present invention.
  • PCI to PCI bridge indicated generally at 38 , contains a primary interface 36 and a secondary interface 40 .
  • Primary pinouts 43 comprise the signals that interact with the primary bus on primary interface 36 .
  • Secondary pinouts 45 contain the signals that interface with the secondary bus and secondary interface 40 .
  • Secondary interface 40 also contains pinouts 45 that interact with a secondary PCI bus internal arbiter 100 .
  • PCI to PCI bridge 38 of this embodiment also contains configuration registers 76 and data buffers 80 . In operation, PCI to PCI bridge 38 interprets the signals on primary pinouts 43 and secondary pinouts 45 in order to transmit transactions upstream and downstream.
  • secondary PCI bus internal arbiter 100 determines which device may use PCI to PCI bridge 38 .
  • internal arbiter 100 implements a two tier arbitration scheme, with a high tier and a low tier, as described below.
  • FIG. 4 is a diagram of one embodiment of a bridge arbitration register of a PCI to PCI bridge with a two tier internal arbitration mechanism according to the present invention.
  • the embodiment contemplates a system whereby the PCI to PCI bridge may service up to six PCI devices that are on the secondary bus. Including the PCI to PCI bridge, there are seven devices that may access the secondary bus. Bits 0 - 6 allow the user to categorize each device in one of the two tiers. As FIG. 4 shows, the default for this embodiment places the PCI to PCI bridge in the higher priority tier while all other devices are in the lower priority tier.
  • Such a two tier arbitration scheme presents a problem when a device in the low tier initiates but does not complete a transaction across the PCI to PCI bridge. That is, a master device accessed the secondary bus and indicated to the PCI to PCI bridge that the PCI to PCI bridge needs to access the primary bus and initiate a transaction with a device located on the primary bus. However, the PCI to PCI bridge was unable to complete the transaction on the primary bus. Where the PCI to PCI bridge can only handle one transaction at a time on the primary bus, the incomplete transaction is pending. Therefore, the PCI to PCI bridge informs the master device to retry the attempted transaction.
  • the PCI to PCI bridge since the PCI to PCI bridge has not completed the first transaction, it must tell any other device that attempts a transaction to also retry. However, in a simple two tier arbitration scheme, the initial master device, if it is in the low tier, must wait until all the high priority devices have had an opportunity to access the bus and then each low priority device in sequence has an opportunity to access the secondary bus. Even when the PCI to PCI bridge completes the transaction on the primary bus, it must wait until the master device again has an opportunity to access the secondary bus to complete the transaction. Because the master device is in the low tier, this can result in significant non-utilization of the PCI to PCI bridge.
  • FIG. 5 is a flow chart of one embodiment of a method of a two tier intelligent bus arbitration scheme according to the present invention.
  • This methodology resides inside internal arbiter 100 .
  • step 120 a high and low tier of devices with access to a secondary PCI bus is maintained.
  • step 124 each device with access to the secondary bus is categorized into either the high tier or the low tier.
  • step 126 access to the secondary bus is provided such that the devices in the high tier have more opportunities than devices in the low tier to access the secondary bus.
  • step 128 it is determined that if a device in the low tier attempts but cannot complete a transaction over the PCI to PCI bridge, the device is placed into the high tier.
  • step 130 after the transaction completes, the device is placed back into the low tier.
  • this method determines that when a device in the low tier must have access to the secondary bus such that it can complete a transaction that is pending on the primary bus, that device is placed into the high tier in order to allow the transaction to complete more quickly.
  • FIGS. 6A, 6B, 6 C, and 6 D are diagrams showing one embodiment of the operation of a two tier intelligent bus arbitration scheme.
  • FIG. 6A shows a typical placement of a PCI to PCI bridge.
  • Memory 18 resides on PCI primary bus 26 which connects to primary interface 36 of PCI to PCI bridge 38 .
  • PCI to PCI bridge 38 contains secondary PCI bus internal arbiter 100 .
  • PCI to PCI bridge 38 has a secondary interface 40 which connects to PCI secondary bus 42 .
  • PCI devices 30 reside on PCI secondary bus 42 .
  • PCI devices 30 are labeled 0, 1, 2, 3, 4, and 5. In operation, when PCI device 0 desires to access memory 18 , it can conduct a transaction through PCI to PCI bridge 38 .
  • PCI to PCI bridge 38 It can do so by first initiating a transaction over PCI secondary bus 42 to secondary interface 40 .
  • PCI to PCI bridge 38 is acting as the target and PCI device 0 is acting as the master.
  • PCI to PCI bridge 38 must in turn act as the master to initiate a transaction over primary interface 36 to primary PCI bus 26 in order to gain access to memory 18 .
  • PCI to PCI bridge 38 only conducts one such transaction at a time. If PCI to PCI bridge 38 is conducting this transaction for PCI device 0, then it is unable to transact a similar transaction for any other PCI device. In such a state, PCI to PCI bridge 38 contains a pending transaction.
  • PCI to PCI bridge 38 of the present invention initiates a two tier intelligent bus arbitration scheme in order to expedite the transaction.
  • FIG. 6B is a diagram of an operation of a two tier arbitration scheme as implemented by secondary PCI bus internal arbiter 100 .
  • Each PCI device 0-5 is categorized either into high tier 101 or low tier 102 .
  • Each device 0-5 has an opportunity to access secondary PCI bus 42 in the following manner. First, each device in high tier 101 has an opportunity to access secondary PCI bus 42 . Then, one of the devices in low tier 102 has an opportunity to access secondary PCI bus 42 . Therefore, if PCI devices 0-5 are categorized as in FIG. 6B, then devices 0-5 have opportunities to access secondary PCI bus 42 in the following order: 0, 1, 2, 3, 0, 1, 2, 4, 0, 1, 2, 5.
  • FIG. 6C is a diagram of further operation of a two tier arbitration scheme.
  • FIG. 6C shows a situation whereby PCI devices 0-5 are all initially categorized in low tier 102 . Therefore, devices 0-5 have opportunities to access secondary PCI bus 42 in the following order: 0, 1, 2, 3, 4, 5.
  • FIG. 6D is a diagram of the operation of an embodiment of an intelligent two tier bus arbitration scheme, according to the present invention.
  • internal arbiter 100 has determined that PCI device 0 has initiated a transaction that is still pending. Therefore, PCI device 0 is moved temporarily into high tier 101 .
  • PCI device 0 While PCI device 0 is in high tier 101 , the opportunity for each device to access secondary PCI bus 42 will follow the order: 0, 1, 0, 2, 0, 3, 0, 4, 0, 5.
  • FIGS. 6C and 6D show that the present embodiment provides more of an opportunity to access secondary PCI bus 42 when secondary PCI bus internal arbiter determines that a low tier device has a pending transaction. Therefore, the pending transaction can be completed more quickly.
  • FIG. 7 is a diagram of one embodiment of the activity of a PCI to PCI bridge containing a two tier arbitration scheme with and without an intelligent bus arbitration scheme according to the present invention.
  • the top portion of FIG. 7 shows each device 0-5 is in the low tier and thus each device has an equal opportunity to access the secondary bus.
  • This example contemplates that a task initiated by device 0 begins at time A and completes at time C. However, if all devices remain in the low tier, device 0 will not get another opportunity to access the secondary bus until time G. Therefore, the PCI to PCI bridge cannot complete the transaction and cannot process another transaction until time H.
  • the bottom portion of FIG. 7 shows the implementation of a two tier intelligent bus arbitration scheme according to the present invention.
  • a task initiated by device 0 begins at time A and completes at time C.
  • the arbiter has placed 0 in the high tier, thus device 0 has an opportunity to access the secondary bus more frequently. Therefore, the transaction is completed at time C and the bridge is free to process another transaction at time D.

Landscapes

  • Engineering & Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Bus Control (AREA)

Abstract

The method of the present invention includes maintaining a first tier 101 and a second tier 102 of devices 30 that have access to a secondary bus 42 that a PCI to PCI bridge 38 services. Each device 30 that has access to secondary PCI bus 42 is categorized into either first tier 101 or a second tier 102. The devices 30 in first tier 101 are provided more frequent opportunities to gain access to secondary PCI bus 42 than devices in low tier 102. Next, a pending transaction is recognized when an initiating device 30 that has been categorized into second tier 102 accesses secondary PCI bus 42 and attempts a transaction that crosses PCI to PCI bridge 38 to primary PCI bus 26. However, PCI to PCI bridge 38 is unable to complete the transaction on primary PCI bus 26. Therefore, PCI to PCI bridge 38 is unable to provide access to any other device 30 on secondary bus 42 until the pending transaction completes. Next, device 30 that initiated the pending transaction is categorized into first tier 101 until the pending transaction is completed.

Description

    TECHNICAL FIELD OF THE INVENTION
  • The present invention relates in general to the field of PCI to PCI bridge devices and more specifically to a method and apparatus for intelligent bus arbitration in a PCI to PCI bridge. [0001]
  • BACKGROUND OF THE INVENTION
  • A peripheral component interconnect (PCI) bridge provides a connection path between two independent PCI buses. The primary function of a PCI to PCI bridge is to allow transactions to occur between a device on one PCI bus and a device on the other PCI bus. System and option card designers can use multiple PCI to PCI bridges to create an hierarchy of PCI buses. This allows system and option card designers to overcome electrical loading limits. [0002]
  • In a transaction between two PCI devices, the PCI device that initiates the transaction is called the master and the other PCI device is called the target. If the master and target are on different PCI buses, the bus that the master resides on is the initiating bus. The bus that the target resides on is the target bus. [0003]
  • A PCI to PCI bridge has two PCI interfaces, each connected to a PCI bus. The PCI interface of the PCI to PCI bridge that is connected to the PCI bus that is closest to the CPU is the primary interface. The PCI interface of the PCI to PCI bridge that is connected to the PCI bus that is farthest from the CPU is the secondary interface. Similarly, the PCI bus that is connected to the primary interface of the PCI to PCI bridge is called the primary PCI bus. The PCI bus that is connected to the secondary interface of the PCI to PCI bridge is called the secondary bus. [0004]
  • A PCI to PCI bridge acts essentially as an intermediary between devices located on the secondary bus and devices that are located on the primary bus. The two interfaces of the PCI to PCI bridge bus are capable of both master and target operations. The PCI to PCI bridge acts as a target on the initiating bus on behalf of the target that actually resides on the target bus. Similarly, the PCI to PCI bridge functions as a master on the target bus on behalf of the master that actually resides on the initiating bus. To devices located on the primary bus, the PCI to PCI-bridge appears as one device where it actually represents several PCI devices that are located on the secondary bus. A detailed specification for PCI to PCI bridges is set forth in “PCI to PCI Bridge Architecture Specification”, Revision 1.0, Apr. 5, 1994, PCI Special Interest Group, Hillsboro, Oreg. [0005]
  • Because several PCI devices reside on the secondary bus, it is possible that more than one of these devices will attempt to complete a transaction through the PCI to PCI bridge to a device on the primary bus at the same time. Therefore, it is desirable to have some type of arbitration scheme to decide which device on the secondary bus gets to initiate a transaction. [0006]
  • One such arbitration scheme is a two tier arbitration scheme. In the two tier arbitration scheme, the devices located on the secondary bus are categorized into a high tier and a low tier. Each device in the high tier is given the opportunity to access the secondary bus. Then a single device in the low tier is given the same opportunity. Then each device in the high tier is given another opportunity to access the bus before another device in the low tier is given an opportunity to access the secondary bus. This process is repeated such that devices in the high tier are given more opportunities to access the secondary bus than devices in the low tier. [0007]
  • SUMMARY OF THE INVENTION
  • In accordance with the present invention, a method and apparatus for multiple tier intelligent bus arbitration on a PCI to PCI bridge is provided that substantially eliminates or reduces problems associated with previously developed PCI to PCI bridge arbitration schemes. [0008]
  • The method of the present invention includes maintaining a first tier and a second tier of devices that have access to the secondary bus that the PCI to PCI bridge services. Each device that has access to the secondary PCI bus is categorized into either the first tier or the second tier. The devices in the first tier are provided more frequent opportunities to gain access to the secondary PCI bus than devices in the low tier. Next, a pending transaction is recognized when an initiating device that has been categorized into the second tier accesses the secondary PCI bus and attempts a transaction that crosses the PCI to PCI bridge to the primary PCI bus. However, the PCI to PCI bridge is unable to complete the transaction on the primary PCI bus. Therefore, the PCI to PCI bridge is unable to provide access to any other device on the secondary bus until the pending transaction completes. Next, the device that initiated the pending transaction is categorized into the first tier until the pending transaction is completed. [0009]
  • An apparatus of the present invention includes a PCI to PCI bridge arbiter operable to execute the two tier intelligent bus arbitration scheme as described in the above method. [0010]
  • A further apparatus of the present invention includes a PCI to PCI bridge with an internal arbiter operable to execute the two tier intelligent bus arbitration scheme as described in the above method. [0011]
  • A technical advantage of the present invention is that the pending transaction will be completed faster because the initiating device is placed into the first tier of high priority devices such that it has more opportunities to access the secondary bus. [0012]
  • Additional technical advantages should be readily apparent from the drawings, description, and claims. [0013]
  • BRIEF DESCRIPTION OF THE DRAWINGS
  • A more complete understanding of the present invention and advantages thereof may be acquired by referring to the following description taken in conjunction with the accompanying drawings in which like reference numbers indicate like features and wherein: [0014]
  • FIG. 1 is a block diagram of typical PCI to PCI bridge applications; [0015]
  • FIG. 2 is a functional block diagram of the typical operation of a PCI to PCI bridge; [0016]
  • FIG. 3 is a functional block diagram of one embodiment of a PCI to PCI bridge that contains a secondary PCI bus internal arbiter according to the present invention; [0017]
  • FIG. 4 is a diagram of one embodiment of a bridge arbitration register of a PCI to PCI bridge with a two tier internal arbitration mechanism according to the present invention; [0018]
  • FIG. 5 is a flow chart of one embodiment of a method of a two tier intelligent bus arbitration scheme according to the present invention; [0019]
  • FIGS. 6A, 6B, [0020] 6C, and 6D are diagrams showing one embodiment of the operation of a two tier intelligent bus arbitration scheme; and
  • FIG. 7 is a diagram of one embodiment of the activity of a PCI to PCI bridge containing a two tier arbitration scheme with and without an intelligent bus arbitration scheme according to the present invention. [0021]
  • DETAILED DESCRIPTION OF THE INVENTION
  • FIG. 1 is a block diagram of typical PCI to PCI bridge applications. A [0022] CPU 10 connects to memory 18 via host bus 14. A host bridge 22 connects host bus 14 to a PCI bus 26. PCI bus 26 connects PCI devices 30 to host bridge 22. PCI bus 26 also connects to a PCI to PCI bridge 38. PCI to PCI bridge 38 in turn connects to PCI option cards 44 via PCI bus 42. A second PCI to PCI bridge 39 operates via PCI bus 48 to connect PCI devices 32 to PCI bus 42.
  • The bus that connects to a PCI to PCI bridge and is closest to the CPU is the primary bus and connects to the primary interface of the PCI to PCI bridge. Therefore, for PCI to [0023] PCI bridge 38, interface 36 is the primary interface and PCI bus 26 is the primary bus.
  • Accordingly, [0024] interface 40 is the secondary interface and PCI bus 42 is the secondary bus. For PCI to PCI bridge 39, interface 37 is the primary interface and PCI bus 42 is the primary bus, while interface 41 is the secondary interface and PCI bus 48 is the secondary bus.
  • In a configuration such as that shown in FIG. 1, each PCI to PCI bridge and each PCI bus are numbered for identification. The PCI bus that is the secondary bus for the host bridge is given the name “[0025] PCI BUS 0.” Therefore, PCI bus 26 is PCI BUS 0. The PCI to PCI bridge that uses PCI BUS 0 as its primary interface is given the name “PCI-PCI BRIDGE 1.” This corresponds to PCI to PCI bridge 38. The PCI bus that acts as the secondary bus to a PCI to PCI bridge is given the number of that PCI to PCI bridge. Therefore, PCI bus 42 is called “PCI BUS 1” because it is the secondary bus to PCI-PCI BRIDGE 1 (PCI to PCI bridge 38). Each PCI to PCI bridge and each PCI bus is named in this fashion. Thus, for FIG. 1, PCI to PCI bridge 39 is named “PCI-PCI BRIDGE 2” and PCI bus 48 is “PCI BUS 2.”
  • Electrical loading limits, as specified in the PCI Local Bus Specification, Revision 2.1, allow only ten PCI device loads per PCI bus. Therefore, a system that requires more than ten PCI device loads requires a PCI to PCI bridge. For example, in FIG. 1, a maximum of 10 [0026] PCI devices 30 may connect to PCI bus 26 and host bridge 22. However, PCI to PCI bridge 38 appears as one device to PCI bus 26 and host bridge 22. This allows option cards 44 that contain further PCI devices 32 to access PCI bus 26 and host bridge 22 maintaining the specified electrical loading limits.
  • In operation, if [0027] PCI device 32 initiates a transaction to memory 18, PCI device 32 is the master and memory 18 is the target. To complete the transaction, PCI device 32 acts as a master along PCI bus 48 to secondary interface 41 of PCI to PCI bridge 39. Then, primary interface 37 of PCI to PCI bridge 39 acts as a master to complete a transaction to secondary interface 40 of PCI to PCI bridge 38. Next, primary interface 36 of PCI to PCI bridge 38 initiates a transaction as master to secondary interface of host bridge 22. Host bridge 22 then acts as master for a transaction to memory 18.
  • FIG. 2 is a functional block diagram of the typical operation of a PCI to PCI bridge indicated generally at [0028] 38. Data moving downstream first encounters a primary interface 36. This action initiates a primary target interface 60. PCI to PCI bridge 38 is acting as the target of the transaction that is taking place on the primary bus. Next, a secondary master interface 64 is triggered, in order for PCI to PCI bridge 38 to act as the master of the transaction on the secondary bus. When a transaction moves upstream, a secondary target interface 68 is initiated. PCI to PCI bridge 38 is acting as the target of this transaction that is taking place on the secondary bus. Then a primary master interface 72 is initiated. Here, PCI to PCI bridge 38 acts as the master to the transaction on the primary bus. These transactions are affected by configuration registers 76 that determine the characteristics of PCI to PCI bridge 38. Furthermore, optional data buffers 80 may exist on PCI to PCI bridge 38 to buffer data involved in transactions across PCI to PCI bridge 38.
  • FIG. 3 is a functional block diagram of one embodiment of a PCI to PCI bridge that contains a secondary PCI bus internal arbiter according to the present invention. PCI to PCI bridge, indicated generally at [0029] 38, contains a primary interface 36 and a secondary interface 40. Primary pinouts 43 comprise the signals that interact with the primary bus on primary interface 36. Secondary pinouts 45 contain the signals that interface with the secondary bus and secondary interface 40. Secondary interface 40 also contains pinouts 45 that interact with a secondary PCI bus internal arbiter 100. PCI to PCI bridge 38 of this embodiment also contains configuration registers 76 and data buffers 80. In operation, PCI to PCI bridge 38 interprets the signals on primary pinouts 43 and secondary pinouts 45 in order to transmit transactions upstream and downstream.
  • If more than one device on the secondary bus desires to access [0030] secondary interface 40, secondary PCI bus internal arbiter 100 determines which device may use PCI to PCI bridge 38. In one embodiment, internal arbiter 100 implements a two tier arbitration scheme, with a high tier and a low tier, as described below.
  • FIG. 4 is a diagram of one embodiment of a bridge arbitration register of a PCI to PCI bridge with a two tier internal arbitration mechanism according to the present invention. The embodiment contemplates a system whereby the PCI to PCI bridge may service up to six PCI devices that are on the secondary bus. Including the PCI to PCI bridge, there are seven devices that may access the secondary bus. Bits [0031] 0-6 allow the user to categorize each device in one of the two tiers. As FIG. 4 shows, the default for this embodiment places the PCI to PCI bridge in the higher priority tier while all other devices are in the lower priority tier.
  • Such a two tier arbitration scheme presents a problem when a device in the low tier initiates but does not complete a transaction across the PCI to PCI bridge. That is, a master device accessed the secondary bus and indicated to the PCI to PCI bridge that the PCI to PCI bridge needs to access the primary bus and initiate a transaction with a device located on the primary bus. However, the PCI to PCI bridge was unable to complete the transaction on the primary bus. Where the PCI to PCI bridge can only handle one transaction at a time on the primary bus, the incomplete transaction is pending. Therefore, the PCI to PCI bridge informs the master device to retry the attempted transaction. Furthermore, since the PCI to PCI bridge has not completed the first transaction, it must tell any other device that attempts a transaction to also retry. However, in a simple two tier arbitration scheme, the initial master device, if it is in the low tier, must wait until all the high priority devices have had an opportunity to access the bus and then each low priority device in sequence has an opportunity to access the secondary bus. Even when the PCI to PCI bridge completes the transaction on the primary bus, it must wait until the master device again has an opportunity to access the secondary bus to complete the transaction. Because the master device is in the low tier, this can result in significant non-utilization of the PCI to PCI bridge. [0032]
  • FIG. 5 is a flow chart of one embodiment of a method of a two tier intelligent bus arbitration scheme according to the present invention. This methodology resides inside [0033] internal arbiter 100. In step 120, a high and low tier of devices with access to a secondary PCI bus is maintained. Next, at step 124, each device with access to the secondary bus is categorized into either the high tier or the low tier. Then at step 126, access to the secondary bus is provided such that the devices in the high tier have more opportunities than devices in the low tier to access the secondary bus. Next at step 128, it is determined that if a device in the low tier attempts but cannot complete a transaction over the PCI to PCI bridge, the device is placed into the high tier. At step 130, after the transaction completes, the device is placed back into the low tier. In operation, this method determines that when a device in the low tier must have access to the secondary bus such that it can complete a transaction that is pending on the primary bus, that device is placed into the high tier in order to allow the transaction to complete more quickly.
  • FIGS. 6A, 6B, [0034] 6C, and 6D are diagrams showing one embodiment of the operation of a two tier intelligent bus arbitration scheme. FIG. 6A shows a typical placement of a PCI to PCI bridge. Memory 18 resides on PCI primary bus 26 which connects to primary interface 36 of PCI to PCI bridge 38. PCI to PCI bridge 38 contains secondary PCI bus internal arbiter 100. PCI to PCI bridge 38 has a secondary interface 40 which connects to PCI secondary bus 42. PCI devices 30 reside on PCI secondary bus 42. PCI devices 30 are labeled 0, 1, 2, 3, 4, and 5. In operation, when PCI device 0 desires to access memory 18, it can conduct a transaction through PCI to PCI bridge 38. It can do so by first initiating a transaction over PCI secondary bus 42 to secondary interface 40. In this portion of the transaction, PCI to PCI bridge 38 is acting as the target and PCI device 0 is acting as the master. PCI to PCI bridge 38 must in turn act as the master to initiate a transaction over primary interface 36 to primary PCI bus 26 in order to gain access to memory 18. In this embodiment, PCI to PCI bridge 38 only conducts one such transaction at a time. If PCI to PCI bridge 38 is conducting this transaction for PCI device 0, then it is unable to transact a similar transaction for any other PCI device. In such a state, PCI to PCI bridge 38 contains a pending transaction. If the transaction is still pending at the end of PCI device 0's opportunity to access PCI secondary bus 42, the transaction will remain pending until PCI device 0 has another opportunity to access PCI secondary bus 42. While this transaction is pending, PCI to PCI bridge 38 of the present invention initiates a two tier intelligent bus arbitration scheme in order to expedite the transaction.
  • FIG. 6B is a diagram of an operation of a two tier arbitration scheme as implemented by secondary PCI bus [0035] internal arbiter 100. There is a high tier 101 and a low tier 102. Each PCI device 0-5 is categorized either into high tier 101 or low tier 102. Each device 0-5 has an opportunity to access secondary PCI bus 42 in the following manner. First, each device in high tier 101 has an opportunity to access secondary PCI bus 42. Then, one of the devices in low tier 102 has an opportunity to access secondary PCI bus 42. Therefore, if PCI devices 0-5 are categorized as in FIG. 6B, then devices 0-5 have opportunities to access secondary PCI bus 42 in the following order: 0, 1, 2, 3, 0, 1, 2, 4, 0, 1, 2, 5.
  • FIG. 6C is a diagram of further operation of a two tier arbitration scheme. FIG. 6C shows a situation whereby PCI devices 0-5 are all initially categorized in [0036] low tier 102. Therefore, devices 0-5 have opportunities to access secondary PCI bus 42 in the following order: 0, 1, 2, 3, 4, 5.
  • FIG. 6D is a diagram of the operation of an embodiment of an intelligent two tier bus arbitration scheme, according to the present invention. In FIG. 6D, [0037] internal arbiter 100 has determined that PCI device 0 has initiated a transaction that is still pending. Therefore, PCI device 0 is moved temporarily into high tier 101. While PCI device 0 is in high tier 101, the opportunity for each device to access secondary PCI bus 42 will follow the order: 0, 1, 0, 2, 0, 3, 0, 4, 0, 5. FIGS. 6C and 6D show that the present embodiment provides more of an opportunity to access secondary PCI bus 42 when secondary PCI bus internal arbiter determines that a low tier device has a pending transaction. Therefore, the pending transaction can be completed more quickly.
  • FIG. 7 is a diagram of one embodiment of the activity of a PCI to PCI bridge containing a two tier arbitration scheme with and without an intelligent bus arbitration scheme according to the present invention. The top portion of FIG. 7 shows each device 0-5 is in the low tier and thus each device has an equal opportunity to access the secondary bus. This example contemplates that a task initiated by [0038] device 0 begins at time A and completes at time C. However, if all devices remain in the low tier, device 0 will not get another opportunity to access the secondary bus until time G. Therefore, the PCI to PCI bridge cannot complete the transaction and cannot process another transaction until time H. The bottom portion of FIG. 7 shows the implementation of a two tier intelligent bus arbitration scheme according to the present invention. Once again, a task initiated by device 0 begins at time A and completes at time C. However, the arbiter has placed 0 in the high tier, thus device 0 has an opportunity to access the secondary bus more frequently. Therefore, the transaction is completed at time C and the bridge is free to process another transaction at time D.
  • Although the present invention has been described in detail, it should be understood that various changes, substitutions and alterations can be made hereto without departing from the spirit and scope of the invention as defined by the appended claims. [0039]

Claims (19)

What is claimed is:
1. A method for a multiple tier intelligent bus arbitration scheme for a PCI to PCI bridge, comprising the steps of:
maintaining a first tier and a second tier of devices that have access to a secondary PCI bus;
categorizing each device that has access to the secondary PCI bus into the first tier or the second tier;
providing devices in the first tier more frequent opportunities to gain access to the secondary PCI bus than devices in the second tier;
recognizing a pending transaction when an initiating device in the second tier accesses the secondary PCI bus and attempts a transaction that crosses the PCI to PCI bridge to a primary PCI bus, but the PCI to PCI bridge is unable to complete the transaction on the primary PCI bus, such that the PCI to PCI bridge is unable to provide access to the primary PCI bus to other devices on the secondary bus; and
categorizing the initiating device into the first tier until the pending transaction is completed.
2. The method of
claim 1
, whereby the step of providing devices in the high tier more of an opportunity to gain access to the secondary PCI bus than devices in the low tier is comprised of the steps of:
maintaining a sequential order of the devices categorized in the second tier;
providing an opportunity for each device in the first tier an opportunity to gain access to the secondary PCI bus;
providing an opportunity for a single device in the second tier an opportunity to gain access to the secondary bus;
repeating the above steps to grant each device an opportunity to gain access to the secondary bus, such that the device in the second tier that is granted an opportunity to access the secondary bus is the next device in the sequential order from the previous device in the second tier that was allowed an opportunity to access the secondary bus.
3. The method of
claim 1
further comprising the step of categorizing the PCI to PCI bridge into the first or second tier as a device with access to the secondary PCI bus.
4. The method of
claim 3
further comprising the step of initially placing the PCI to PCI bridge into the first tier.
5. The method of
claim 1
whereby the step of categorizing each device with access to the secondary bus comprises categorizing six PCI devices and the PCI to PCI bridge.
6. The method of
claim 5
further comprising the step of initially placing the PCI to PCI bridge into the first tier and all other devices into the second tier.
7. A PCI to PCI bridge arbiter with a multiple tier intelligent bus arbitration scheme, comprising:
an arbiter associated with a PCI to PCI bridge operable to:
receive from the PCI to PCI bridge information on devices associated with a secondary bus;
maintain a first tier and a second tier of devices that have access to the secondary PCI bus;
categorize each device that has access to the secondary PCI bus into the first tier or the second tier;
provide data to the PCI to PCI bridge such that the PCI to PCI bridge provides devices in the first tier more frequent opportunities to gain access to the secondary PCI bus than devices in the second tier;
recognize a pending transaction when an initiating device in the second tier accesses the secondary PCI bus and attempts a transaction that crosses the PCI to PCI bridge to primary PCI bus, but the PCI to PCI bridge is unable to complete the transaction on the primary PCI bus, such that the PCI to PCI bridge is unable to provide access to the primary PCI bus to other devices on the secondary bus; and
categorize the initiating device into the first tier until the pending transaction is completed.
8. The PCI to PCI bridge arbiter of
claim 7
, whereby the arbiter is further operable to:
maintain a sequential order of the devices categorized in the second tier;
provide data to the PCI to PCI bridge such that the PCI to PCI bridge provides an opportunity for each device in the first tier an opportunity to gain access to the secondary PCI bus;
provide data to the PCI to PCI bridge such that the PCI to PCI bridge provides an opportunity for a single device in the second tier an opportunity to gain access to the secondary bus;
repeat the above steps such that the PCI to PCI bridge grants each device an opportunity to gain access to the secondary bus, such that the device in the second tier that is granted an opportunity to access the secondary bus is the next device in the sequential order from the previous device in the second tier that was allowed an opportunity to access the secondary bus.
9. The PCI to PCI bridge arbiter of
claim 7
whereby the arbiter is further operable to categorize the PCI to PCI bridge into the first or second tier as a device with access to the secondary PCI bus.
10. The PCI to PCI bridge arbiter of
claim 9
whereby the arbiter is further operable to initially place the PCI to PCI bridge into the first tier.
11. The PCI to PCI bridge arbiter of
claim 7
whereby the arbiter is further operable to categorize six PCI devices and the PCI to PCI bridge into the first tier or second tier.
12. The PCI to PCI bridge arbiter of
claim 11
whereby the arbiter is further operable to initially place the PCI to PCI bridge into the first tier and all other devices into the second tier.
13. A PCI to PCI bridge with a multiple tier intelligent bus arbitration scheme, comprising:
a primary bus interface operable to communicate with PCI devices associated with a primary bus;
a secondary bus interface associated with the primary bus interface and operable to communicate with PCI devices associated with a secondary bus;
a secondary PCI bus internal arbiter associated with the secondary bus interface operable to:
maintain a first tier and a second tier of devices that have access to the secondary PCI bus;
categorize each device that has access to the secondary PCI bus into the first tier or the second tier;
provide devices in the first tier more frequent opportunities to gain access to the secondary PCI bus than devices in the second tier;
recognize a pending transaction when an initiating device in the second tier accesses the secondary PCI bus and attempts a transaction that crosses the PCI to PCI bridge to the primary PCI bus, but the PCI to PCI bridge is unable to complete the transaction on the primary PCI bus, such that the PCI to PCI bridge is unable to provide access to the primary PCI bus to other devices on the secondary bus; and
categorize the initiating device into the first tier until the pending transaction is completed.
14. The PCI to PCI bridge of
claim 13
, whereby the secondary PCI bus internal arbiter is further operable to:
maintain a sequential order of the devices categorized in the second tier;
provide an opportunity for each device in the first tier an opportunity to gain access to the secondary PCI bus;
provide an opportunity for a single device in the second tier an opportunity to gain access to the secondary bus;
repeat the above steps to grant each device an opportunity to gain access to the secondary bus, such that the device in the second tier that is granted an opportunity to access the secondary bus is the next device in the sequential order from the previous device in the second tier that was allowed an opportunity to access the secondary bus.
15. The PCI to PCI bridge of
claim 13
whereby the secondary PCI bus internal arbiter is further operable to categorize the PCI to PCI bridge into the first or second tier as a device with access to the secondary PCI bus.
16. The PCI to PCI bridge of
claim 15
whereby the secondary PCI bus internal arbiter is further operable to initially place the PCI to PCI bridge into the first tier.
17. The PCI to PCI bridge of
claim 13
whereby the secondary PCI bus internal arbiter is further operable to categorize six PCI devices and the PCI to PCI bridge into the first tier or second tier.
18. The PCI to PCI bridge of
claim 17
whereby the secondary PCI bus internal arbiter is further operable to initially place the PCI to PCI bridge into the first tier and all other devices into the second tier.
19. The PCI to PCI bridge of
claim 13
wherein the first tier is a high tier and the second tier is a low tier.
US09/796,016 1997-09-30 2001-02-28 Method and apparatus for multiple tier intelligent bus arbitration on a PCI to PCI bridge Expired - Lifetime US6393508B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US09/796,016 US6393508B2 (en) 1997-09-30 2001-02-28 Method and apparatus for multiple tier intelligent bus arbitration on a PCI to PCI bridge

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US6046797P 1997-09-30 1997-09-30
US16297598A 1998-09-29 1998-09-29
US09/796,016 US6393508B2 (en) 1997-09-30 2001-02-28 Method and apparatus for multiple tier intelligent bus arbitration on a PCI to PCI bridge

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
US16297598A Continuation 1997-09-30 1998-09-29

Publications (2)

Publication Number Publication Date
US20010027505A1 true US20010027505A1 (en) 2001-10-04
US6393508B2 US6393508B2 (en) 2002-05-21

Family

ID=26739962

Family Applications (1)

Application Number Title Priority Date Filing Date
US09/796,016 Expired - Lifetime US6393508B2 (en) 1997-09-30 2001-02-28 Method and apparatus for multiple tier intelligent bus arbitration on a PCI to PCI bridge

Country Status (1)

Country Link
US (1) US6393508B2 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060036796A1 (en) * 2004-08-10 2006-02-16 Riley Dwight D Method and apparatus for setting a primary port on a PCI bridge
US20060200607A1 (en) * 2005-03-01 2006-09-07 Subramaniam Ganasan Jaya P Bus access arbitration scheme
WO2013101782A1 (en) * 2011-12-26 2013-07-04 Arteris SAS Transport of pci-ordered traffic over independent networks

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TW463093B (en) * 1999-10-16 2001-11-11 Via Tech Inc Method for arbitrarily adjusting PCI compatible device access sequence
US6883057B2 (en) * 2002-02-15 2005-04-19 International Business Machines Corporation Method and apparatus embedding PCI-to-PCI bridge functions in PCI devices using PCI configuration header type 0
US9135193B2 (en) * 2013-03-25 2015-09-15 Hewlett-Packard Development Company, L.P. Expander interrupt processing

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5297260A (en) * 1986-03-12 1994-03-22 Hitachi, Ltd. Processor having a plurality of CPUS with one CPU being normally connected to common bus
US5276845A (en) * 1988-08-25 1994-01-04 Yamaha Corporation Apparatus with multiple buses for permitting concurrent access to a first memory by a processor while a DMA transfer is occurring between a second memory and a communications buffer
JP2986176B2 (en) * 1990-05-11 1999-12-06 株式会社日立製作所 Bus right control system and bus system
US5265211A (en) * 1992-01-02 1993-11-23 International Business Machines Corporation Arbitration control logic for computer system having dual bus architecture
US5280623A (en) * 1992-03-04 1994-01-18 Sun Microsystems, Inc. Versatile peripheral bus
US5396602A (en) 1993-05-28 1995-03-07 International Business Machines Corp. Arbitration logic for multiple bus computer system
US5450551A (en) 1993-05-28 1995-09-12 International Business Machines Corporation System direct memory access (DMA) support logic for PCI based computer system
US5790870A (en) 1995-12-15 1998-08-04 Compaq Computer Corporation Bus error handler for PERR# and SERR# on dual PCI bus system

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060036796A1 (en) * 2004-08-10 2006-02-16 Riley Dwight D Method and apparatus for setting a primary port on a PCI bridge
US8291145B2 (en) 2004-08-10 2012-10-16 Hewlett-Packard Development Company, L.P. Method and apparatus for setting a primary port on a PCI bridge
US20060200607A1 (en) * 2005-03-01 2006-09-07 Subramaniam Ganasan Jaya P Bus access arbitration scheme
US7249210B2 (en) * 2005-03-01 2007-07-24 Qualcomm Incorporated Bus access arbitration scheme
WO2013101782A1 (en) * 2011-12-26 2013-07-04 Arteris SAS Transport of pci-ordered traffic over independent networks
US8788737B2 (en) 2011-12-26 2014-07-22 Qualcomm Technologies, Inc. Transport of PCI-ordered traffic over independent networks

Also Published As

Publication number Publication date
US6393508B2 (en) 2002-05-21

Similar Documents

Publication Publication Date Title
US7308522B2 (en) Method and apparatus to allow dynamic variation of ordering enforcement between transactions in a strongly ordered computer interconnect
US5625779A (en) Arbitration signaling mechanism to prevent deadlock guarantee access latency, and guarantee acquisition latency for an expansion bridge
US5717876A (en) Method for avoiding livelock on bus bridge receiving multiple requests
US6189063B1 (en) Method and apparatus for intelligent configuration register access on a PCI to PCI bridge
EP0737924A2 (en) Bus arbritation and data transfer
US6070209A (en) Delivering transactions between data buses in a computer system
US5564114A (en) Method and an arrangement for handshaking on a bus to transfer information between devices in a computer system
US6012118A (en) Method and apparatus for performing bus operations in a computer system using deferred replies returned without using the address bus
US6694397B2 (en) Request queuing system for a PCI bridge
EP0475005B1 (en) Device for interconnecting channels
US6397279B1 (en) Smart retry system that reduces wasted bus transactions associated with master retries
US6055598A (en) Arrangement and method for allowing sequence-independent command responses across a computer bus bridge
US6604159B1 (en) Data release to reduce latency in on-chip system bus
US20040267992A1 (en) Look ahead split release for a data bus
US6260093B1 (en) Method and apparatus for arbitrating access to multiple buses in a data processing system
US7096290B2 (en) On-chip high speed data interface
US20030229743A1 (en) Methods and structure for improved fairness bus arbitration
US5778235A (en) Computer system and arbitrator utilizing a bus bridge that avoids livelock
US6393508B2 (en) Method and apparatus for multiple tier intelligent bus arbitration on a PCI to PCI bridge
US6260091B1 (en) Method and apparatus for performing out-of-order bus operations in which an agent only arbitrates for use of a data bus to send data with a deferred reply
US6301627B1 (en) Method/system for identifying delayed predetermined information transfer request as bypassable by subsequently-generated information transfer request using bypass enable bit in bridge translation control entry
US6430641B1 (en) Methods, arbiters, and computer program products that can improve the performance of a pipelined dual bus data processing system
US20030131175A1 (en) Method and apparatus for ensuring multi-threaded transaction ordering in a strongly ordered computer interconnect
US5734846A (en) Method for avoiding livelock on bus bridge
US6807593B1 (en) Enhanced bus architecture for posted read operation between masters and slaves

Legal Events

Date Code Title Description
STCF Information on status: patent grant

Free format text: PATENTED CASE

FPAY Fee payment

Year of fee payment: 4

FPAY Fee payment

Year of fee payment: 8

FPAY Fee payment

Year of fee payment: 12