-
公开(公告)号:US20240168807A1
公开(公告)日:2024-05-23
申请号:US18056949
申请日:2022-11-18
申请人: Intel Corporation
发明人: Jorge Eduardo Parra Osorio , Guei-Yuan Lueh , Maxim Kazakov , Fangwen Fu , Supratim Pal , Kaiyu Chen
CPC分类号: G06F9/5027 , G06F9/48 , G06F9/522 , G06F15/8046
摘要: An apparatus to facilitate cross-thread register sharing for matrix multiplication compute is disclosed. The apparatus includes matrix acceleration hardware comprising a plurality of data processing units, wherein the respective plurality of data processing units are to: receive a decoded instruction for a first thread having a first register space, wherein the decoded instruction is for a matrix multiplication operation and comprises an indication to utilize a second register space of a second thread for an operand of the decoded instruction for the first thread; access the second register space of the second thread to obtain data for the operand of the decoded instruction; and perform the matrix multiplication operation for the first thread using the data for the operand from the second register space of the second thread.
-
公开(公告)号:US20240134527A1
公开(公告)日:2024-04-25
申请号:US17971290
申请日:2022-10-20
申请人: Intel Corporation
发明人: Joydeep Ray , Michael Apodaca , Yoav Harel , Guei-Yuan Lueh , John A. Wiegert
CPC分类号: G06F3/061 , G06F3/0655 , G06F3/0679 , G06T1/60
摘要: Embodiments described herein provide a technique to enable access to entries in a surface state or sampler state using 64-bit virtual addresses. One embodiment provides a graphics core that includes memory access circuitry configured to facilitate access to the memory by functional units of the graphics core. The memory access circuitry is configured to receive a message to access an entry in a surface state or a sampler state associated with a parallel processing operation. The message specifies a base address for a surface state entry or sampler state entry. The circuitry can add the base address and the offset to determine a 64-bit virtual address for the entry in the surface state entry or the sampler state and submit a memory access request to the memory to access the entry of the surface state or sampler state.
-
公开(公告)号:US11762696B2
公开(公告)日:2023-09-19
申请号:US17520583
申请日:2021-11-05
申请人: Intel Corporation
发明人: Abhishek R Appu , Altug Koker , Balaji Vembu , Joydeep Ray , Kamal Sinha , Prasoonkumar Surti , Kiran C. Veernapu , Subramaniam Maiyuran , Sanjeev S. Jahagirdar , Eric J. Asperheim , Guei-Yuan Lueh , David Puffer , Wenyin Fu , Nikos Kaburlasos , Bhushan M. Borole , Josh B. Mastronarde , Linda L. Hurd , Travis T. Schluessler , Tomasz Janczak , Abhishek Venkatesh , Kai Xiao , Slawomir Grajewski
CPC分类号: G06F9/5016 , G06F1/329 , G06F9/4893 , G06F9/5044 , G06T1/20 , G06T1/60 , G06T15/005 , G06T2200/28 , Y02D10/00
摘要: In an example, an apparatus comprises a plurality of execution units comprising at least a first type of execution unit and a second type of execution unit and logic, at least partially including hardware logic, to analyze a workload and assign the workload to one of the first type of execution unit or the second type of execution unit. Other embodiments are also disclosed and claimed.
-
公开(公告)号:US11748302B2
公开(公告)日:2023-09-05
申请号:US17561427
申请日:2021-12-23
申请人: Intel Corporation
发明人: Altug Koker , Prasoonkumar Surti , David Puffer , Subramaniam Maiyuran , Guei-Yuan Lueh , Abhishek R. Appu , Joydeep Ray , Balaji Vembu , Tomer Bar-On , Andrew T. Lauritzen , Hugues Labbe , John G. Gierach , Gabor Liktor
IPC分类号: G06F16/13 , G06F9/38 , G06F9/30 , G06F16/11 , G06F16/172 , G06F9/46 , G06F12/1036 , G06F12/1045 , G06F12/0831
CPC分类号: G06F16/13 , G06F9/30 , G06F9/38 , G06F9/3836 , G06F9/461 , G06F16/113 , G06F16/172 , G06F12/0831 , G06F12/1036 , G06F12/1045 , G06F2201/84
摘要: In an example, an apparatus comprises a plurality of execution units, and a first memory communicatively couple to the plurality of execution units, wherein the first shared memory is shared by the plurality of execution units and a copy engine to copy context state data from at least a first of the plurality of execution units to the first shared memory. Other embodiments are also disclosed and claimed.
-
5.
公开(公告)号:US20230195685A1
公开(公告)日:2023-06-22
申请号:US18170900
申请日:2023-02-17
申请人: Intel Corporation
发明人: Subramaniam Maiyuran , Shubra Marwaha , Ashutosh Garg , Supratim Pal , Jorge Parra , Chandra Gurram , Varghese George , Darin Starkey , Guei-Yuan Lueh
IPC分类号: G06F15/78 , G06F9/30 , G06F12/128 , G06F17/16 , G06F12/0811 , G06F12/02 , G06F12/0866 , G06F7/544 , G06F9/50 , G06F17/18 , G06F9/38 , G06F12/0891 , G06F12/06 , G06F12/0888 , G06F12/0802 , G06T1/60 , G06F12/0871 , G06T1/20 , H03M7/46 , G06F12/0875 , G06F12/0862 , G06F15/80 , G06F12/0897 , G06F12/0893 , G06F12/0804 , G06F12/0882 , G06F7/575 , G06F12/1009 , G06F12/0895 , G06F7/58 , G06T15/06 , G06N3/08
CPC分类号: G06F15/7839 , G06F9/30043 , G06F12/128 , G06F17/16 , G06F12/0811 , G06F12/0238 , G06F12/0866 , G06F9/30014 , G06F7/5443 , G06F9/5077 , G06F12/0246 , G06F17/18 , G06F9/3887 , G06F12/0891 , G06F12/0607 , G06F12/0888 , G06F12/0802 , G06T1/60 , G06F9/30079 , G06F12/0871 , G06F9/30036 , G06T1/20 , H03M7/46 , G06F12/0215 , G06F12/0875 , G06F12/0862 , G06F15/8046 , G06F9/30047 , G06F9/30065 , G06F12/0897 , G06F9/5011 , G06F12/0893 , G06F12/0804 , G06F12/0882 , G06F9/3001 , G06F7/575 , G06F12/1009 , G06F9/3004 , G06F12/0895 , G06F7/588 , G06F2212/401 , G06F2212/1044 , G06F9/3867 , G06F9/3818 , G06F9/3802 , G06F2212/455 , G06F2212/1021 , G06F2212/60 , G06F2212/1008 , G06T15/06 , G06N3/08 , G06F2212/302
摘要: Described herein is a graphics processing unit (GPU) configured to receive an instruction having multiple operands, where the instruction is a single instruction multiple data (SIMD) instruction configured to use a bfloat16 (BF16) number format and the BF16 number format is a sixteen-bit floating point format having an eight-bit exponent. The GPU can process the instruction using the multiple operands, where to process the instruction includes to perform a multiply operation, perform an addition to a result of the multiply operation, and apply a rectified linear unit function to a result of the addition.
-
公开(公告)号:US11640297B2
公开(公告)日:2023-05-02
申请号:US17304153
申请日:2021-06-15
申请人: Intel Corporation
发明人: Subramaniam Maiyuran , Guei-Yuan Lueh , Supratim Pal , Ashutosh Garg , Chandra S. Gurram , Jorge E. Parra , Junjie Gu , Konrad Trifunovic , Hong Bin Liao , Mike B. MacPherson , Shubh B. Shah , Shubra Marwaha , Stephen Junkins , Timothy R. Bauer , Varghese George , Weiyu Chen
摘要: Embodiments described herein provided for an instruction and associated logic to enable GPGPU program code to access special purpose hardware logic to accelerate dot product operations. One embodiment provides for a graphics processing unit comprising a fetch unit to fetch an instruction for execution and a decode unit to decode the instruction into a decoded instruction. The decoded instruction is a matrix instruction to cause the graphics processing unit to perform a parallel dot product operation. The GPGPU also includes systolic dot product circuitry to execute the decoded instruction across one or more SIMD lanes using multiple systolic layers, wherein to execute the decoded instruction, a dot product computed at a first systolic layer is to be output to a second systolic layer, wherein each systolic layer includes one or more sets of interconnected multipliers and adders, each set of multipliers and adders to generate a dot product.
-
公开(公告)号:US11314515B2
公开(公告)日:2022-04-26
申请号:US16724831
申请日:2019-12-23
申请人: Intel Corporation
发明人: Supratim Pal , Sasikanth Avancha , Ishwar Bhati , Wei-Yu Chen , Dipankar Das , Ashutosh Garg , Chandra S. Gurram , Junjie Gu , Guei-Yuan Lueh , Subramaniam Maiyuran , Jorge E. Parra , Sudarshan Srinivasan , Varghese George
摘要: Embodiments described herein provide for an instruction and associated logic to enable a vector multiply add instructions with automatic zero skipping for sparse input. One embodiment provides for a general-purpose graphics processor comprising logic to perform operations comprising fetching a hardware macro instruction having a predicate mask, a repeat count, and a set of initial operands, where the initial operands include a destination operand and multiple source operands. The hardware macro instruction is configured to perform one or more multiply/add operations on input data associated with a set of matrices.
-
公开(公告)号:US11042370B2
公开(公告)日:2021-06-22
申请号:US15957728
申请日:2018-04-19
申请人: Intel Corporation
发明人: Subramaniam Maiyuran , Guei-Yuan Lueh , Supratim Pal , Ashutosh Garg , Chandra S. Gurram , Jorge E. Parra , Junjie Gu , Konrad Trifunovic , Hong Bin Liao , Mike B. Macpherson , Shubh B. Shah , Shubra Marwaha , Stephen Junkins , Timothy R. Bauer , Varghese George , Weiyu Chen
摘要: Embodiments described herein provided for an instruction and associated logic to enable GPGPU program code to access special purpose hardware logic to accelerate dot product operations. One embodiment provides for a graphics processing unit comprising a fetch unit to fetch an instruction for execution and a decode unit to decode the instruction into a decoded instruction. The decoded instruction is a matrix instruction to cause the graphics processing unit to perform a parallel dot product operation. The GPGPU also includes a systolic dot product unit to execute the decoded instruction across one or more SIMD lanes using multiple systolic layers, wherein to execute the decoded instruction, a dot product computed at a first systolic layer is to be output to a second systolic layer, wherein each systolic layer includes one or more sets of interconnected multipliers and adders, each set of multipliers and adders to generate a dot product.
-
公开(公告)号:US20210150663A1
公开(公告)日:2021-05-20
申请号:US17095590
申请日:2020-11-11
申请人: Intel Corporation
发明人: Subramaniam Maiyuran , Durgaprasad Bilagi , Joydeep Ray , Scott Janus , Sanjeev Jahagirdar , Brent Insko , Lidong Xu , Abhishek R. Appu , James Holland , Vasanth Ranganathan , Nikos Kaburlasos , Altug Koker , Xinmin Tian , Guei-Yuan Lueh , Changliang Wang
IPC分类号: G06T1/60 , G06T1/20 , G06N5/04 , G06F12/0802
摘要: Embodiments described herein are generally directed to improvements relating to power, latency, bandwidth and/or performance issues relating to GPU processing/caching. According to one embodiment, a system includes a producer intellectual property (IP) (e.g., a media IP), a compute core (e.g., a GPU or an AI-specific core of the GPU), a streaming buffer logically interposed between the producer IP and the compute core. The producer IP is operable to consume data from memory and output results to the streaming buffer. The compute core is operable to perform AI inference processing based on data consumed from the streaming buffer and output AI inference processing results to the memory.
-
公开(公告)号:US20210125581A1
公开(公告)日:2021-04-29
申请号:US17062871
申请日:2020-10-05
申请人: Intel Corporation
发明人: Joydeep Ray , Altug Koker , Balaji Vembu , Murali Ramadoss , Guei-Yuan Lueh , James A. Valerio , Prasoonkumar Surti , Abhishek R. Appu , Vasanth Ranganathan , Kalyan K. Bhiravabhatla , Arthur D. Hunter, JR. , Wei-Yu Chen , Subramaniam M. Maiyuran
IPC分类号: G09G5/36 , G06F12/0875 , G06F9/46 , G09G5/00
摘要: A mechanism is described for facilitating using of a shared local memory for register spilling/filling relating to graphics processors at computing devices. A method of embodiments, as described herein, includes reserving one or more spaces of a shared local memory (SLM) to perform one or more of spilling and filling relating to registers associated with a graphics processor of a computing device.
-
-
-
-
-
-
-
-
-