WO2007106698A2 - Efficient execution of memory barrier bus commands - Google Patents
Efficient execution of memory barrier bus commands Download PDFInfo
- Publication number
- WO2007106698A2 WO2007106698A2 PCT/US2007/063510 US2007063510W WO2007106698A2 WO 2007106698 A2 WO2007106698 A2 WO 2007106698A2 US 2007063510 W US2007063510 W US 2007063510W WO 2007106698 A2 WO2007106698 A2 WO 2007106698A2
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- memory
- processing system
- master device
- weakly
- barrier
- Prior art date
Links
- 230000015654 memory Effects 0.000 title claims abstract description 275
- 230000004888 barrier function Effects 0.000 title claims abstract description 102
- 238000012545 processing Methods 0.000 claims abstract description 74
- 238000000034 method Methods 0.000 claims abstract description 19
- 230000004044 response Effects 0.000 claims description 3
- 208000033748 Device issues Diseases 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 5
- 238000013461 design Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000011664 signaling Effects 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/14—Handling requests for interconnection or transfer
- G06F13/16—Handling requests for interconnection or transfer for access to memory bus
- G06F13/1605—Handling requests for interconnection or transfer for access to memory bus based on arbitration
- G06F13/161—Handling requests for interconnection or transfer for access to memory bus based on arbitration with latency improvement
- G06F13/1621—Handling requests for interconnection or transfer for access to memory bus based on arbitration with latency improvement by maintaining request order
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/14—Handling requests for interconnection or transfer
- G06F13/16—Handling requests for interconnection or transfer for access to memory bus
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/30—Arrangements for executing machine instructions, e.g. instruction decode
- G06F9/38—Concurrent instruction execution, e.g. pipeline, look ahead
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/52—Program synchronisation; Mutual exclusion, e.g. by means of semaphores
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/52—Program synchronisation; Mutual exclusion, e.g. by means of semaphores
- G06F9/522—Barrier synchronisation
Definitions
- the present disclosure relates generally to processing systems, and more particularly, to techniques for efficiently handling memory barrier bus commands in a processing system.
- the types of devices connected to a bus in a processing system may vary depending on the particular application.
- the bus is configured to support a number of processors, memory devices, and peripherals
- the processors often achieve performance benefits by allowing memory operations to be performed out-of-order.
- a processing system may achieve performance benefits by reordering a sequence of memory operations to allow all operations to the same page in memory to be executed before a new page is opened
- Processing systems that are allowed to reorder memory operations are generally referred to as "weakly- ordered" processing systems.
- the reordering of memory operations may unpredictably affect program behavior. For instance, an application may require a processor to write data to memory before the processor reads from memory. In a weakly-ordered processing system, there is no guarantee that this will occur. This result may be unacceptable. fOO05J Various techniques have been employ eel for executing ordered memory operations in a weakly-ordered processing system. One technique is simply to delay certain memory operations until all memory operations before it are executed. In the previous example, the processor may delay issuing a read request until it receives an indication that guarantees that the data has been written to the memory.
- a common technique in modern day processor architectures is to use a bus command known as a "memory barrier' " when an ordered memory operation is required.
- a "memory barrier ' may be used to ensure that all memory access requests issued by a processor before the memory barrier are executed before all memory access requests issued by the processor after the memory barrier.
- a memory barrier could be sent to the memory by the processor before issuing a read request. This would ensure that the processor writes to memory before it reads from the memory
- the processing system includes memory arid a master device configured to issue memory access requests, including memory barriers, to the memory.
- the processing system also includes a slave device configured to provide the master device access to the memory, the slave device being further configured to produce a signal indicating that an ordering constraint imposed by a memory barrier issued by the master device will be enforced, the signal being produced before the execution of all memory access requests issued by the master device to the memory before the memory banner.
- the processing system includes memory and a master device configured to issue memory access requests, including memory barriers, to the memory
- the processing system also includes a slave device configured to provide the master device access to the memory, the slave device being further configured to acknowledge a memory barrier issued by the master device, the memory barrier being acknowledged before the execution of all memory access requests issued by the master device to the memory before the memory barrier iOOt ⁇ ]
- a method of executing memory barriers in weakly-ordered processing system is disclosed
- the processing system includes a master device, a slave device, and memory The method includes issuing a memory barrier from the master device to the memory, and producing, at the slave device, a signal indicating that an ordering constraint imposed by the memory barrier will be enforced, the signal being produced before the execution of all memory access requests to the memory issued by the master device before the memory barrier.
- FIG. 1 is a conceptual block diagram illustrating an example of a weakly- ordered processing system
- FlG 2 is a functional block diagram illustrating an example of a system bus interconnect in a weakly -ordered processing system
- FlG I is a conceptual block diagram illustrating an example of a weakly- ordered processing system
- the processing system 100 may be a computer, resident in a computer, or any other system capable of processing, retrieving and storing information
- the processing system 100 may be a stand-alone system, or alternat ely, embedded in a device, such as a wireless or wired telephone, a persona!
- the processing system 100 may be implemented as integrated circuit, part of an integrated circuit, or distributed across multiple integrated circuits Alternathelv, the processing system 100 may be implemented with discrete components, or any combination of discrete and integrated circuitn J hose skilled in the art will recognize how best to implement the processing system 100 for each particular application
- the processing system 100 is shown with a bus architecture that connects multiple master devices to multiple sla ⁇ e devices
- a "master device” is any device that can issue a bus request
- a "slave device” is any device that responds to a bus request
- the master devices comprise two processors 102a, i02b
- Each processor 102a, i02b may be any type of processing entity including, by way of example, a microprocessor, a digital signal processor (BS?), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic, discrete gate or transistor logic, or discrete hardware components.
- the processors 102a, 102b are connected to multiple slave devices by a high speed bus. commonly referred to as a system bus 104.
- the slave devices include two memory controllers 106a, 106b, each providing access to an independent memory 108a, 108b, respectively.
- each memory 108a, iO8b may be a small, high speed, volatile memory used to store applications and data required often by the processors 102a, 102b.
- the memories 108a, 108b are commonly implemented with RAM, DRAM, or SDRAM, but in practice may be implemented with any suitable storage medium.
- the memories 108a, 108b are generally referred to collectively as the system memory.
- the bridge 1 10 is used to connect the high speed system bus 104 to a slower, lower-level bus 112.
- the lower-level bus 1 12 may be used by the processors 102a, i02b to access peripheral devices (not shown) and additional memories 108c, I OSd.
- These memories 108c, I O8d may be large, inexpensive storage mediums used to store applications and data required less often by the processors 102a, 102b, such as hard drives. Hash memory, removable disks, or the like.
- one or both memories may be non-volatile memory, such as ROM, PROM EPROKl EHPROM memory, CD- ROM, or any other suitable permanent storage medium,
- the bridge 1 10 which is a slave device on the system bus 104, is the master device on the lower-level bus 1 12 for a number of slave devices.
- the slave devices on the lower-level bus 1 12 include a memory controller 106c for the memory- 108c and a second bridge 114.
- the second bridge 1 14 is used to connect the lower-level bus 1 12 to a second lower -level bus 1 16 that provides access to memory 108d through memory controller iO6d.
- the second bridge 1 14 is the master device and the memory controller 106d is the slave device on the second lower-level bus 1 16
- the processors 102a, 102b may be configured to execute instructions under control of an operating system and/or other software.
- the instructions may reside in the various memories I O8a-lO8d. Data may also be stored in the these memory devices 10Sa- I OSd and retrieved by the processors 102a, 102b to execute certain instructions. The new data resulting from the execution of these instructions may be written back into these memory devices 1 OSa-I OSd.
- any processor 102a, 102b may access the first memory 108a, but only the second processor 102b may the second memory 108b.
- each processor 102a, 102b may access the memories 108c, !08d on the lower-leve! buses 1 12, 1 !6.
- a processor may access memory by issuing a "memory access request.' * A '-memory access request' * may be a write request a read request, or any other bus related request
- a processor may issue a write request to a target memory by placing the appropriate address, control information and the pay load on the transmit channel of the system bus 104.
- a processor may place the address and control information on the transmit channel, in response to the read request, the target memory will send the payload back to the processor on the receive channel of the system bus iO4.
- A write request issued by a processor will be received by a system bus interconnect 1 18.
- the system bus interconnect 1 18 decodes the address to determine where to route the payload and the control information.
- the payload, along with the control information is loaded into a buffer (not shown) in the memory controller for the target memory , hi the weakly-ordered processing system 100 of FlG.
- the memory controller may write the contents of its buffer into the target memory out-of-order if performance benefits can be achieved.
- the memory controller may reorder the sequence of memory access requests in the target memory in order to execute all requests to the same page in memory before opening a new page.
- Write requests to the memories 108c, 108d on the lower-level buses 1 12, 1 16 may be executed in a similar fashion, except that the routing is different.
- the system bus interconnect 1 18 routes the payload, along with the control information, to the bridge 1 10.
- the bridge 1 10 transfers the payload and the control information to the lower-level bus 1 12 for delivery to a lower-level bus interconnect 120.
- the lower-level bus interconnect 120 manages access to the memories 10Sc, 108d between the processors 102a, 102b and other bus mastering devices (not shown) within the processing system 100
- the payload and the control information are ultimately delivered by the iower-ieve! bus interconnect 120 to one of the memory controllers !06c or lO ⁇ d
- Each memory controller 106c and l ⁇ d may also include a buffer (not shown) to allow write operations to performed o ⁇ t-of- order if performance benefits can be achieved.
- a processor can issue a read request by placing the address and control information on the transmit channel of the system bos 104.
- the system bus interconnect 1 18 uses the address to determine where to route the control information.
- a read request to either memory 108a or 108b i.e . system memory
- a read request to the other memories 108c, 10Sd causes the system bus interconnect 1 18 to deliver the control information to the bridge I iO where it is transferred to the lower- level bus 1 12.
- the lower-level bus interconnect 120 delivers the control information to a buffer (not shown) in one of the memory controllers iO6c or lO ⁇ d.
- the control information may be used by the memory controller for the target memory to deliver a payload back to the processor issuing the read request
- the memory controller for the target memory may execute the memory access requests in its buffer out of sequence.
- one or more processors may require that the pay-loads for the read requests be received in the order that they were issued. This may be the case, for example, where the performance requirements of the processing system are relaxed in order to reduce complexity.
- a processor may be dynamically switched in and out of a mode that requires payloads to be received in the order they were issued for certain applications.
- the protocol may include an "out-of-order read attribute " as part of the control information to be transmitted by the processors with each read request The out-of-order read attribute may be asserted to indicate that the corresponding read request can be executed in any order.
- the out-of-order read attribute may be deasserted for each of a series of read requests that must be performed in order, either before, after, or interleaved with other pending read requests
- a read request with a deasserted out-of- order read attribute will be referred to herein as an "in-order required" read request.
- a processor may issue a memory barrier.
- the memory barrier is received by the system bus interconnect 1 I S and routed to each memory in the processing system 100 accessible by the processor issuing the memory barrier, either by way of the transmit channel or by sideband signaling
- a memory barrier issued by the first processor 102a will be routed by the system bus interconnect 1 18 to the first memory 108a in system memory, as well as both memories 108c, 108d on the lower -level buses 1 12, 1 16.
- a memory barrier issued by the second processor 102b on the other hand, will be routed by the system bus interconnect ! I S to every memory 108a-108d shown in the processing system of FlG I .
- a "memory barrier acknowledgement” is a signal from a slave device indicating that it can enforce an ordering constraint imposed by the memory barrier.
- the memory barrier acknowledgement is sent from the slave device to the issuing processor over the receive channel, however, as those skilled in the art will readily appreciate, the memory barrier acknowledgement can be sent using sideband signaling.
- FKJ, 2 is a functional block diagram illustrating an example of the system bus interconnect 1 18
- the manner in which the system bus interconnect 1 18 is actually implemented will depend the specific application and the overall design constraints imposed on the processing system. Those skilled in the art will recognize the interchangeability of various designs, and how best to implement the functionality described herein for each particular application.
- the system bus interconnect 1 18 may be used to receive memory access requests from the processors.
- a memory- access request for a read or write operation includes the address for the target memory.
- the memory access request including the address, the control information, and in the case of a write request, the payioad is loaded into a bus register 202.
- the address and control information are also provided to a decoder 204.
- the decoder 204 is used to determine the target memory for each read and write request in the bus register 202.
- the decoder 204 output is used to control a transmit bus switch 206
- the transmit bus switch 206 is used to direct each memory access request in the bus register 202 to the target slave device.
- a receive bus switch 208 is used to direct the payloads from the various memories to the processors issuing the read requests
- the contiol information may also include an om-of- ⁇ rdcr read attribute which is detectable by the decoder 204 If the out-of-order read attribute is deaserted, the decoder 204 alerts a controller 210 The controller 210 determines whether there aie am pending in-order read requests to a memory other than the target memory, that ucic issued earlier b> the same processor If so, then ihe controller 210 delays the release of the read request from the bus register 202 until the payloads from such pending read requests are received b> the system bus interconnect 1 18 If, the controller 210 determmes that there are no pending in-order read requests issued earlier by the same processor, or that all pending in-order read requests issued earlier by the same processor are to the target memory, then the read request can be immediately released b> the bus register 202 and directed to the target memory through the bus switch 206 hi the lattei case, the memory controller to the target memory can ensure that al! pending
- a memory access request may also be a bus command, such as a mention barrier I he memorj barrier is loaded into the bus register 202 and provided to the decoder 204
- the decoder 204 prompts the controller 210 to send the memor> barrier in the bus register 202 to each memory accessible bv the issuing processor using the bus switch 206
- the memories accessible by each processor may be pre-provisioned in the controller 210
- a slave deuce that controls the sequence of all memors access requests to a memory can send a memory barrier acknowledgement to the processor issuing the memory barrier B ⁇ way of example, the memory controllers 106a, 106b control the sequence of all memon, access requests to their respective memories 108a, 108b Raeh of these memory controllers l ⁇ a, 106b can be configured to respond to a memory barrier by executing all memory access requests in its buffer when the memory barrier is received before executing any memory access requests received after the memory barrier, if the memory controllers 106a, 106b are configured in this fashion, each can send an acknowledgement back to the system bus interconnect 1 1 S without having to wait for the execution of all outstanding memory access requests in Us buffer,
- the protocol may need to account for the possibility ⁇ hat one or more processors may not be able to process an acknowledgement prior to receiving the payloads for ail pending in-order required read requests.
- One possible solution is to transmit an "out-of-order enable" attribute with the memory barrier
- a processor that is not constrained by the order in which a memory barrier acknowledgement is received can assert the out-of-order enable attribute transmitted with the memory barrier, if the out-of-order enable attribute is asserted, then each memory controller 106a. 106b receiving the memory barrier can send an acknowledgement back to the issuing processor as soon as it receives the memory barrier.
- a processor that is unable to process a memory barrier acknowledgement prior to receiving the payloads for all in-order required read requests can deassert the out-of-order enable attribute transmitted with the memory barrier.
- a deasserted out-of-order enable attribute will require each, memory controller 106a, 106b receiving the memory barrier to send the payloads for al! in-order required read requests issued prior to the memory barrier before sending an acknowledgement back to the issuing processor. Either way, the processor will be able to issue a subsequent memory access request without having to wait for the execution of all prior memory' access requests. As a result, the potential for under utilizing the system bus 104 may be reduced, thereby improving performance.
- the bridge 1 10 between the system bus 104 and the lower-level bus 1 12 cannot provide an acknowledgement to a memory- barrier issued by a processor because it cannot guarantee an ordering constraint imposed by the memory barrier. More specifically, the bridge 1 10 cannot guarantee that all memory access requests issued before the memory barrier from another bus mastering device (not shown) in the processing system 100 and received by the lower- level bus interconnect 120 will be executed before a memory access request issued by a processor after such processor issues a memory barrier. Consequently, a memory barrier destined for one or both of the memories IQSc, 108d cannot be acknowledged by the bridge 1 10.
- the memory controller 106c on the lower-level bus 1 S 2 can acknowledge a memory barrier issued by a processor.
- the memory controller 106c is responsible for controlling the sequence of all memory access requests to the memory 108c.
- the memory controller 106c can be configured to respond to a memory barrier by executing all memory access requests in its buffer when the memory barrier is received before executing any memory access requests received after the memory barrier. If the memory controller 106c is configured in this Fashion, it too can send an acknowledgement back to the issuing processor early, The exact timing of the acknowledgement will depend on the state of the out-of-order enable attribute.
- the bridge 1 14 on the second lower-level bus 1 16 may also be configured to acknowledge a memory barrier issued by a processor.
- the bridge 114 between the two lower- level buses 1 12, 1 16 can be configured to guarantee an ordering constraint imposed by a memory barrier
- an acknowledgement can be sent by the bridge 1 14 back to the issuing processor early. Again, the timing of the acknowledgement will depend on the state of the out-of-order enable attribute.
- FIG. 3 is a flow chart illustrating an example of the operation of a slave device that can acknowledge a memory barrier before all pending memory access requests are executed.
- the slave device uses the receive channel to acknowledge memory barriers and is configured to support a protocol that allows one or more processors to issue in-order required read requests.
- the slave device executes all pending memory access requests in its buffer before the memory barrier, but can acknowledge the memory barrier earlier as illustrated more fully in FlG. 3
- the slave device receives a memory barrier
- the slave device determines, in step 304, whether the out-of-order enable attribute is asserted or deasserted. If the out-of-order enable attribute is asserted, then the slave device can send a memory barrier acknowledgement over the receive channel on the next bus transaction in step 310. If, on the other hand, the out-of-order enable attribute is deassetted, then the slave device determines, in step 306, whether any i ⁇ -order required read requests are pending in its buffer. The slave device can send the memory barrier acknowledgement over the receive channel on the next bus transaction, in step 310, if there are no pending i ⁇ -order required read requests. Otherwise, the slave device must first send the pay loads for the in-order read requests to the issuing processor in step 308 before acknowledging the memory barrier in step 310.
Abstract
Description
Claims
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2007800080201A CN101395574B (en) | 2006-03-10 | 2007-03-07 | Weak ordered processing system and method for executing memory barrier therein |
EP07758097A EP1999577B1 (en) | 2006-03-10 | 2007-03-07 | Efficient execution of memory barrier bus commands |
JP2009500556A JP5043925B2 (en) | 2006-03-10 | 2007-03-07 | Efficient execution of memory barrier bus commands |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US78109106P | 2006-03-10 | 2006-03-10 | |
US60/781,091 | 2006-03-10 | ||
US11/397,287 US7917676B2 (en) | 2006-03-10 | 2006-04-04 | Efficient execution of memory barrier bus commands with order constrained memory accesses |
US11/397,287 | 2006-04-04 |
Publications (2)
Publication Number | Publication Date |
---|---|
WO2007106698A2 true WO2007106698A2 (en) | 2007-09-20 |
WO2007106698A3 WO2007106698A3 (en) | 2008-03-06 |
Family
ID=38237497
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/US2007/063510 WO2007106698A2 (en) | 2006-03-10 | 2007-03-07 | Efficient execution of memory barrier bus commands |
Country Status (6)
Country | Link |
---|---|
US (2) | US7917676B2 (en) |
EP (2) | EP1999577B1 (en) |
JP (1) | JP5043925B2 (en) |
KR (1) | KR100987210B1 (en) |
CN (1) | CN101395574B (en) |
WO (1) | WO2007106698A2 (en) |
Families Citing this family (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9026744B2 (en) * | 2005-03-23 | 2015-05-05 | Qualcomm Incorporated | Enforcing strongly-ordered requests in a weakly-ordered processing |
US7917676B2 (en) * | 2006-03-10 | 2011-03-29 | Qualcomm, Incorporated | Efficient execution of memory barrier bus commands with order constrained memory accesses |
US7783817B2 (en) * | 2006-08-31 | 2010-08-24 | Qualcomm Incorporated | Method and apparatus for conditional broadcast of barrier operations |
US8032734B2 (en) * | 2006-09-06 | 2011-10-04 | Mips Technologies, Inc. | Coprocessor load data queue for interfacing an out-of-order execution unit with an in-order coprocessor |
US9946547B2 (en) | 2006-09-29 | 2018-04-17 | Arm Finance Overseas Limited | Load/store unit for a processor, and applications thereof |
US7594079B2 (en) | 2006-09-29 | 2009-09-22 | Mips Technologies, Inc. | Data cache virtual hint way prediction, and applications thereof |
US20080082793A1 (en) * | 2006-09-29 | 2008-04-03 | Mips Technologies, Inc. | Detection and prevention of write-after-write hazards, and applications thereof |
US7984202B2 (en) * | 2007-06-01 | 2011-07-19 | Qualcomm Incorporated | Device directed memory barriers |
JP5239769B2 (en) * | 2008-11-14 | 2013-07-17 | 富士通株式会社 | Request order control system, request order control method, and request order control program |
US9461930B2 (en) | 2009-04-27 | 2016-10-04 | Intel Corporation | Modifying data streams without reordering in a multi-thread, multi-flow network processor |
US8949500B2 (en) * | 2011-08-08 | 2015-02-03 | Lsi Corporation | Non-blocking processor bus bridge for network processors or the like |
US9444757B2 (en) | 2009-04-27 | 2016-09-13 | Intel Corporation | Dynamic configuration of processing modules in a network communications processor architecture |
US8489794B2 (en) * | 2010-03-12 | 2013-07-16 | Lsi Corporation | Processor bus bridge for network processors or the like |
TWI454906B (en) * | 2009-09-24 | 2014-10-01 | Phison Electronics Corp | Data read method, and flash memory controller and storage system using the same |
GB2474446A (en) * | 2009-10-13 | 2011-04-20 | Advanced Risc Mach Ltd | Barrier requests to maintain transaction order in an interconnect with multiple paths |
GB2478795B (en) | 2010-03-19 | 2013-03-13 | Imagination Tech Ltd | Requests and data handling in a bus architecture |
GB2488680B (en) * | 2010-03-19 | 2013-03-13 | Imagination Tech Ltd | Requests and data handling in a bus architecture |
JP2012073851A (en) * | 2010-09-29 | 2012-04-12 | Sony Corp | Bus system and deadlock avoidance circuit thereof |
US9208109B2 (en) * | 2011-06-01 | 2015-12-08 | Altera Corporation | Memory controllers with dynamic port priority assignment capabilities |
US9281036B2 (en) | 2013-01-08 | 2016-03-08 | Qualcomm Incorporated | Memory device having an adaptable number of open rows |
US9594713B2 (en) * | 2014-09-12 | 2017-03-14 | Qualcomm Incorporated | Bridging strongly ordered write transactions to devices in weakly ordered domains, and related apparatuses, methods, and computer-readable media |
US9946492B2 (en) * | 2015-10-30 | 2018-04-17 | Arm Limited | Controlling persistent writes to non-volatile memory based on persist buffer data and a persist barrier within a sequence of program instructions |
US10534540B2 (en) * | 2016-06-06 | 2020-01-14 | Micron Technology, Inc. | Memory protocol |
US20180327255A1 (en) * | 2017-05-15 | 2018-11-15 | Honeywell International Inc. | Systems and methods for multi-sensor integrated sensor devices |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6088771A (en) | 1997-10-24 | 2000-07-11 | Digital Equipment Corporation | Mechanism for reducing latency of memory barrier operations on a multiprocessor system |
Family Cites Families (38)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US750045A (en) * | 1904-01-19 | Vania | ||
JPH04190435A (en) | 1990-11-26 | 1992-07-08 | Hitachi Ltd | Memory access order guarantee method for multiprocessor system |
US5664124A (en) | 1994-11-30 | 1997-09-02 | International Business Machines Corporation | Bridge between two buses of a computer system that latches signals from the bus for use on the bridge and responds according to the bus protocols |
WO1997000480A1 (en) | 1995-06-15 | 1997-01-03 | Intel Corporation | Architecture for an i/o processor that integrates a pci to pci bridge |
US5778438A (en) | 1995-12-06 | 1998-07-07 | Intel Corporation | Method and apparatus for maintaining cache coherency in a computer system with a highly pipelined bus and multiple conflicting snoop requests |
JPH09269935A (en) * | 1996-01-31 | 1997-10-14 | Toshiba Corp | Memory controller and memory control method |
US5893165A (en) * | 1996-07-01 | 1999-04-06 | Sun Microsystems, Inc. | System and method for parallel execution of memory transactions using multiple memory models, including SSO, TSO, PSO and RMO |
US5893185A (en) * | 1996-12-16 | 1999-04-13 | Okrepkie; Joseph | Multipurpose electricians hand tool |
US6047334A (en) | 1997-06-17 | 2000-04-04 | Intel Corporation | System for delaying dequeue of commands received prior to fence command until commands received before fence command are ordered for execution in a fixed sequence |
US6370632B1 (en) | 1997-11-18 | 2002-04-09 | Intrinsity, Inc. | Method and apparatus that enforces a regional memory model in hierarchical memory systems |
JPH11167557A (en) * | 1997-12-02 | 1999-06-22 | Hitachi Ltd | Shared memory access sequence assurance method and multiprocessor system |
US6038646A (en) * | 1998-01-23 | 2000-03-14 | Sun Microsystems, Inc. | Method and apparatus for enforcing ordered execution of reads and writes across a memory interface |
US6247102B1 (en) | 1998-03-25 | 2001-06-12 | Compaq Computer Corporation | Computer system employing memory controller and bridge interface permitting concurrent operation |
US6073210A (en) | 1998-03-31 | 2000-06-06 | Intel Corporation | Synchronization of weakly ordered write combining operations using a fencing mechanism |
US6816934B2 (en) | 2000-12-22 | 2004-11-09 | Hewlett-Packard Development Company, L.P. | Computer system with registered peripheral component interconnect device for processing extended commands and attributes according to a registered peripheral component interconnect protocol |
JP2000181891A (en) * | 1998-12-18 | 2000-06-30 | Hitachi Ltd | Shared memory access sequence assurance system |
US6167492A (en) | 1998-12-23 | 2000-12-26 | Advanced Micro Devices, Inc. | Circuit and method for maintaining order of memory access requests initiated by devices coupled to a multiprocessor system |
US6275913B1 (en) | 1999-10-15 | 2001-08-14 | Micron Technology, Inc. | Method for preserving memory request ordering across multiple memory controllers |
US6708269B1 (en) * | 1999-12-30 | 2004-03-16 | Intel Corporation | Method and apparatus for multi-mode fencing in a microprocessor system |
US6963967B1 (en) * | 2000-06-06 | 2005-11-08 | International Business Machines Corporation | System and method for enabling weak consistent storage advantage to a firmly consistent storage architecture |
US6609192B1 (en) | 2000-06-06 | 2003-08-19 | International Business Machines Corporation | System and method for asynchronously overlapping storage barrier operations with old and new storage operations |
FR2816593B1 (en) | 2000-11-16 | 2003-07-25 | Catherine Dumouchel | TRANSPORT PALLET |
US7398376B2 (en) * | 2001-03-23 | 2008-07-08 | International Business Machines Corporation | Instructions for ordering execution in pipelined processes |
US6996812B2 (en) * | 2001-06-18 | 2006-02-07 | International Business Machines Corporation | Software implementation of synchronous memory barriers |
WO2003016232A1 (en) * | 2001-08-20 | 2003-02-27 | Advanced Minerals Corporation | Improved foam glass product |
ATE335237T1 (en) | 2001-08-24 | 2006-08-15 | Intel Corp | A GENERAL INPUT/OUTPUT ARCHITECTURE, PROTOCOL AND CORRESPONDING PROCEDURES FOR IMPLEMENTING FLOW CONTROL |
US20030131175A1 (en) | 2001-12-24 | 2003-07-10 | Heynemann Tom A. | Method and apparatus for ensuring multi-threaded transaction ordering in a strongly ordered computer interconnect |
US6976115B2 (en) | 2002-03-28 | 2005-12-13 | Intel Corporation | Peer-to-peer bus segment bridging |
US7490218B2 (en) * | 2004-01-22 | 2009-02-10 | University Of Washington | Building a wavecache |
WO2005121948A1 (en) * | 2004-06-02 | 2005-12-22 | Sun Microsystems, Inc. | Method and apparatus for enforcing membar instruction semantics in an execute-ahead processor |
US7644409B2 (en) * | 2004-06-04 | 2010-01-05 | Sun Microsystems, Inc. | Techniques for accessing a shared resource using an improved synchronization mechanism |
US7725618B2 (en) * | 2004-07-29 | 2010-05-25 | International Business Machines Corporation | Memory barriers primitives in an asymmetric heterogeneous multiprocessor environment |
US7454570B2 (en) | 2004-12-07 | 2008-11-18 | International Business Machines Corporation | Efficient memory update process for on-the-fly instruction translation for well behaved applications executing on a weakly-ordered processor |
US9026744B2 (en) | 2005-03-23 | 2015-05-05 | Qualcomm Incorporated | Enforcing strongly-ordered requests in a weakly-ordered processing |
US7500045B2 (en) | 2005-03-23 | 2009-03-03 | Qualcomm Incorporated | Minimizing memory barriers when enforcing strongly-ordered requests in a weakly-ordered processing system |
US7917676B2 (en) * | 2006-03-10 | 2011-03-29 | Qualcomm, Incorporated | Efficient execution of memory barrier bus commands with order constrained memory accesses |
US7783817B2 (en) | 2006-08-31 | 2010-08-24 | Qualcomm Incorporated | Method and apparatus for conditional broadcast of barrier operations |
JP2009269935A (en) | 2008-04-30 | 2009-11-19 | Sumitomo Metal Mining Co Ltd | Silver film having golden metallic luster |
-
2006
- 2006-04-04 US US11/397,287 patent/US7917676B2/en active Active
-
2007
- 2007-03-07 EP EP07758097A patent/EP1999577B1/en active Active
- 2007-03-07 EP EP10194652.3A patent/EP2290528B1/en active Active
- 2007-03-07 KR KR1020087024751A patent/KR100987210B1/en active IP Right Grant
- 2007-03-07 JP JP2009500556A patent/JP5043925B2/en active Active
- 2007-03-07 WO PCT/US2007/063510 patent/WO2007106698A2/en active Application Filing
- 2007-03-07 CN CN2007800080201A patent/CN101395574B/en active Active
-
2009
- 2009-09-17 US US12/561,381 patent/US7921249B2/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6088771A (en) | 1997-10-24 | 2000-07-11 | Digital Equipment Corporation | Mechanism for reducing latency of memory barrier operations on a multiprocessor system |
Also Published As
Publication number | Publication date |
---|---|
EP2290528B1 (en) | 2013-07-24 |
US20100005208A1 (en) | 2010-01-07 |
US7921249B2 (en) | 2011-04-05 |
EP1999577A2 (en) | 2008-12-10 |
EP1999577B1 (en) | 2012-06-13 |
KR20080104364A (en) | 2008-12-02 |
US20070214298A1 (en) | 2007-09-13 |
JP2009529748A (en) | 2009-08-20 |
JP5043925B2 (en) | 2012-10-10 |
EP2290528A2 (en) | 2011-03-02 |
KR100987210B1 (en) | 2010-10-12 |
CN101395574A (en) | 2009-03-25 |
CN101395574B (en) | 2013-01-16 |
WO2007106698A3 (en) | 2008-03-06 |
EP2290528A3 (en) | 2011-03-23 |
US7917676B2 (en) | 2011-03-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP1999577A2 (en) | Efficient execution of memory barrier bus commands | |
US7386636B2 (en) | System and method for communicating command parameters between a processor and a memory flow controller | |
US7930457B2 (en) | Channel mechanisms for communicating with a processor event facility | |
US5805927A (en) | Direct memory access channel architecture and method for reception of network information | |
US6006291A (en) | High-throughput interface between a system memory controller and a peripheral device | |
US7502896B2 (en) | System and method for maintaining the integrity of data transfers in shared memory configurations | |
US7778271B2 (en) | Method for communicating instructions and data between a processor and external devices | |
US10078470B2 (en) | Signal transfer device that maintains order of a read request and write request in posted write memory access | |
JP2012073851A (en) | Bus system and deadlock avoidance circuit thereof | |
EP2788882A1 (en) | Auto-ordering of strongly ordered, device, and exclusive transactions across multiple memory regions | |
US10489315B2 (en) | Dynamic adaptation of direct memory transfer in a data processing system with mismatched data-bus widths | |
US8706925B2 (en) | Accelerating memory operations blocked by ordering requirements and data not yet received | |
US20110264827A1 (en) | Performance by reducing transaction request ordering requirements | |
US6742074B2 (en) | Bus to system memory delayed read processing | |
JP4928683B2 (en) | Data processing device | |
WO2007020274A2 (en) | System and method for communicating instructions and data between a processor and external devices | |
US20150227481A1 (en) | System interconnect and operating method of system interconnect | |
JP2004094819A (en) | Memory access arbiter and memory control device | |
US20210055929A1 (en) | Register File Arbitration | |
JP2003108434A (en) | Memory control device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 07758097 Country of ref document: EP Kind code of ref document: A2 |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1899/MUMNP/2008 Country of ref document: IN |
|
WWE | Wipo information: entry into national phase |
Ref document number: 200780008020.1 Country of ref document: CN |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2009500556 Country of ref document: JP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2007758097 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1020087024751 Country of ref document: KR |