公开(公告)号：US10067911B2

公开(公告)日：2018-09-04

申请号：US15219672

申请日：2016-07-26

申请人： Advanced Micro Devices, Inc.

发明人： Amir Gholaminejad ,  Bragadeesh Natarajan

IPC分类号： G06F17/16 ,  G06F7/78 ,  G06F13/16 ,  G06F13/42 ,  G06F7/58

摘要： Systems, apparatuses, and methods for performing in-place matrix transpose operations are disclosed. Operations for transposing tiles of a matrix are scheduled in an order determined by moving diagonally through tiles of the matrix. When a diagonal line hits a boundary, then a tile on a new diagonal line of the matrix is selected and operations are scheduled for transposing this tile. Only tiles within a triangular region of the matrix are scheduled for being transposed. This allows memory access operations to be performed in parallel, expediting the matrix transpose operation compared to linear tile indexing.

公开(公告)号：US10067709B2

公开(公告)日：2018-09-04

申请号：US15269289

申请日：2016-09-19

申请人： Advanced Micro Devices, Inc.

发明人： Leonardo Piga ,  Mauricio Breternitz

IPC分类号： G06F12/00 ,  G06F3/06 ,  G06F17/30 ,  G06F13/00 ,  G06F13/28

摘要： Systems, apparatuses, and methods for accelerating page migration using a two-level bloom filter are disclosed. In one embodiment, a system includes a GPU and a CPU and a multi-level memory hierarchy. When a memory request misses in a first memory, the GPU is configured to check a first level of a two-level bloom filter to determine if a page targeted by the memory request is located in a second memory. If the first level of the two-level bloom filter indicates that the page is not in the second memory, then the GPU generates a page fault and sends the memory request to a third memory. If the first level of the two-level bloom filter indicates that the page is in the second memory, then the GPU sends the memory request to the CPU.

公开(公告)号：US10558606B1

公开(公告)日：2020-02-11

申请号：US16118172

申请日：2018-08-30

申请人： Advanced Micro Devices, Inc.

发明人： Shomit N. Das ,  Matthew Tomei ,  Shrikanth Ganapathy ,  John Kalamatianos

IPC分类号： G06F13/42 ,  G06F1/3296 ,  H03M13/05

摘要： Systems, apparatuses, and methods for reliably transmitting data over voltage scaled links are disclosed. A computing system includes at least first and second devices connected via a link. In one implementation, if a data block can be compressed to less than or equal to half the original size of the data block, then the data block is compressed and sent on the link in a single clock cycle rather than two clock cycles. If the data block cannot be compressed to half the original size, but if the data block can be compressed enough to include error correction code (ECC) bits without exceeding the original size, then ECC bits are added to the compressed block which is sent on the link at a reduced voltage. The ECC bits help to correct for any errors that are generated as a result of operating the link at the reduced voltage.

公开(公告)号：US10545743B2

公开(公告)日：2020-01-28

申请号：US15706297

申请日：2017-09-15

申请人： Oracle International Corporation

发明人： Petr Andreevich Kudriavtsev ,  Vladimir Viktorovich Voskresenskii

IPC分类号： G06F8/51

摘要： Systems, apparatuses, and methods for converting computer program source code from a first high level language to a functionally equivalent second high level language different from the first high level language. Source code in a first high level language that includes implicitly created temporary objects is analyzed. Source code in the second high level programming language is generated that is functionally equivalent to the source code in the first high level programming language. The second high level language does not support temporary objects. For temporary objects identified in the first high level source language, source code is generated to explicitly create and track a corresponding object within a similar scope. In addition, source code is generated to destroy the explicitly created object at a time that generally corresponds to destruction of the temporary object in the original source language.

公开(公告)号：US10540316B2

公开(公告)日：2020-01-21

申请号：US15856799

申请日：2017-12-28

申请人： Advanced Micro Devices, Inc.

发明人： Vydhyanathan Kalyanasundharam ,  Eric Christopher Morton ,  Chen-Ping Yang ,  Amit P. Apte ,  Elizabeth M. Cooper

IPC分类号： G06F13/00 ,  G06F13/42 ,  G06F12/0806 ,  G06F12/0808 ,  G06F13/364 ,  G06F12/0811

摘要： Systems, apparatuses, and methods for implementing a cancel and replay mechanism for ordered requests are disclosed. A system includes at least an ordering master, a memory controller, a coherent slave coupled to the memory controller, and an interconnect fabric coupled to the ordering master and the coherent slave. The ordering master generates a write request which is forwarded to the coherent slave on the path to memory. The coherent slave sends invalidating probes to all processing nodes and then sends an indication that the write request is globally visible to the ordering master when all cached copies of the data targeted by the write request have been invalidated. In response to receiving the globally visible indication, the ordering master starts a timer. If the timer expires before all older requests have become globally visible, then the write request is cancelled and replayed to ensure forward progress in the fabric and avoid a potential deadlock scenario.

公开(公告)号：US10474468B2

公开(公告)日：2019-11-12

申请号：US15439540

申请日：2017-02-22

申请人： Advanced Micro Devices, Inc.

发明人： Michael J. Mantor ,  Brian D. Emberling ,  Mark Fowler ,  Mark M. Leather

IPC分类号： G06F9/38 ,  G06T1/20 ,  G06F9/30

摘要： Systems, apparatuses, and methods for processing variable wavefront sizes on a processor are disclosed. In one embodiment, a processor includes at least a scheduler, cache, and multiple execution units. When operating in a first mode, the processor executes the same instruction on multiple portions of a wavefront before proceeding to the next instruction of the shader program. When operating in a second mode, the processor executes a set of instructions on a first portion of a wavefront. In the second mode, when the processor finishes executing the set of instructions on the first portion of the wavefront, the processor executes the set of instructions on a second portion of the wavefront, and so on until all portions of the wavefront have been processed. The processor determines the operating mode based on one or more conditions.

公开(公告)号：US10440765B2

公开(公告)日：2019-10-08

申请号：US14862735

申请日：2015-09-23

申请人： Apple Inc.

发明人： Farouk Belghoul ,  Tarik Tabet ,  Dawei Zhang

IPC分类号： H04W76/15 ,  H04W28/08 ,  H04W28/02 ,  H04W36/28 ,  H04W36/14 ,  H04W84/12 ,  H04W88/06 ,  H04W76/16

摘要： This disclosure relates to aggregation of radio resources provided according to multiple radio interfaces. According to some embodiments, a base station may establish a radio bearer with a wireless user equipment (UE) device. The radio bearer may initially utilize a first radio interface between the base station and the UE. The base station may receive an indication to aggregate radio resources of an access point that utilizes a second radio interface for the UE. Based on the indication to aggregate radio resources for the UE, the base station may redirect at least a portion of data of the radio bearer by way of the access point to be exchanged with the UE using the second radio interface.

公开(公告)号：US10440593B2

公开(公告)日：2019-10-08

申请号：US16414377

申请日：2019-05-16

申请人： Apple Inc.

发明人： Francisco J. Gonzalez ,  Rafael L. Rivera-Barreto ,  Rohan C. Malthankar ,  Sarma V. Vangala ,  Tarik Tabet

IPC分类号： H04W24/02 ,  H04W76/10 ,  H04W8/04 ,  H04W28/18 ,  H04L29/08 ,  H04B1/3827 ,  H04W74/00 ,  H04W88/06

摘要： System and methods for providing special radio provisions to link budget limited devices. Before establishing a cellular connection with a cellular network, a user equipment device (UE), such as a link budget limited UE, may negotiate the special radio provisions with the network by presenting requested profile information to the network, and receiving back approved profile information from the network. The UE may perform these communications either directly, via a non-cellular communication connection with the network, or indirectly, via a companion or proxy device. When the UE initiates a cellular connection with the network, both entities may use information specified in the approved profile information, which may improve efficiency of the connection. In some scenarios, the approved profile information may identify a class of devices to which the UE belongs. The approved profile information may also be used in establishing connections with other devices of the same class.

公开(公告)号：US10437595B1

公开(公告)日：2019-10-08

申请号：US15070435

申请日：2016-03-15

申请人： Apple Inc.

发明人： Pradeep Kanapathipillai ,  Stephan G. Meier ,  Gerard R. Williams, III ,  Mridul Agarwal ,  Kulin N. Kothari

IPC分类号： G06F9/38 ,  G06F9/30

摘要： Systems, apparatuses, and methods for optimizing a load-store dependency predictor (LSDP). When a younger load instruction is issued before an older store instruction and the younger load is dependent on the older store, the LSDP is trained on this ordering violation. A replay/flush indicator is stored in a corresponding entry in the LSDP to indicate whether the ordering violation resulted in a flush or replay. On subsequent executions, a dependency may be enforced for the load-store pair if a confidence counter is above a threshold, with the threshold varying based on the status of the replay/flush indicator. If a given load matches on multiple entries in the LSDP, and if at least one of the entries has a flush indicator, then the given load may be marked as a multimatch case and forced to wait to issue until all older stores have issued.

公开(公告)号：US10433211B2

公开(公告)日：2019-10-01

申请号：US15593702

申请日：2017-05-12

申请人： Apple Inc.

发明人： Farouk Belghoul ,  Lydi Smaini ,  Tarik Tabet ,  Syed Aon Mujtaba

IPC分类号： H04W28/08 ,  H04W16/14 ,  H04W72/04

摘要： A wireless communication device (UE) may monitor grants received from respective cells associated with one or more first (e.g. licensed) frequency band(s) and one or more second (e.g. unlicensed) frequency band(s). Based on the received grants the UE may determine if an imbalance exists in the radio resource allocation, whereby the UE communicates more than intended in the second frequency band(s). The imbalance may be specified with respect to the ratio or portion of the communications that are conducted in the second frequency band(s) with respect to all wireless communications of the UE. If the UE detects an imbalance, it may transmit a report to the network (e.g. to the base station serving the UE) indicating unfavorable conditions for the UE to be operating in the second frequency band(s). In response, the network may disable the respective cell(s) associated with the second frequency band(s) and/or it may disable carrier aggregation for the UE.