Patents by Inventor Sagi Gurfinkel
Sagi Gurfinkel has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Reduction of silicon area for ethernet PFC protocol implementation in queue based network processors
Patent number: 9912604Abstract: In a pipelined network processor, a first stage in the pipeline is responsive to receipt of a pause indication from a third stage. The pause indication is associated with one of a plurality of ports and priority classes of frames advancing through the pipeline. The first stage asserts a hold indication to a second stage in response to the pause indication. The second stage is responsive to the hold indication by marking frames associated with the one of a plurality of ports and priority classes as they arrive from the pipeline at the second stage by returning them to the first stage interface instead of transmitting them to the third stage. The marked frames are stored in memory external to the network processor.Type: GrantFiled: July 29, 2015Date of Patent: March 6, 2018Assignee: NXP USA, Inc.Inventors: Roman Nos, Noam Efrati, Sagi Gurfinkel -
Patent number: 9904313Abstract: In a processing system, a method includes selecting, at a timer management component of a processor, a timer ring of a set of timer rings for a requested timer based on a time unit granularity associated with the requested timer, wherein each timer ring of the set has a different time unit granularity. The method further includes instantiating the requested timer in a selected entry of the selected timer ring. Instantiating the requested timer may include fixedly maintaining a record for the requested timer in the selected entry of the selected timer ring for the entire time span of the requested timer.Type: GrantFiled: July 13, 2015Date of Patent: February 27, 2018Assignee: NXP USA, Inc.Inventors: Ron-Michael Bar, Eran Glickman, Sagi Gurfinkel
-
Reduction Of Silicon Area for Ethernet PFC Protocol Implementation In Queue Based Network Processors
Publication number: 20170034069Abstract: In a pipelined network processor, a first stage in the pipeline is responsive to receipt of a pause indication from a third stage. The pause indication is associated with one of a plurality of ports and priority classes of frames advancing through the pipeline. The first stage asserts a hold indication to a second stage in response to the pause indication. The second stage is responsive to the hold indication by marking frames associated with the one of a plurality of ports and priority classes as they arrive from the pipeline at the second stage by returning them to the first stage interface instead of transmitting them to the third stage. The marked frames are stored in memory external to the network processor.Type: ApplicationFiled: July 29, 2015Publication date: February 2, 2017Inventors: Roman Nos, Noam Efrati, Sagi Gurfinkel -
Publication number: 20170017260Abstract: In a processing system, a method includes selecting, at a timer management component of a processor, a timer ring of a set of timer rings for a requested timer based on a time unit granularity associated with the requested timer, wherein each timer ring of the set has a different time unit granularity. The method further includes instantiating the requested timer in a selected entry of the selected timer ring. Instantiating the requested timer may include fixedly maintaining a record for the requested timer in the selected entry of the selected timer ring for the entire time span of the requested timer.Type: ApplicationFiled: July 13, 2015Publication date: January 19, 2017Inventors: Ron-Michael Bar, Eran Glickman, Sagi Gurfinkel
-
Patent number: 8572296Abstract: A method for arbitrating between direct memory access task requests, the method includes receiving multiple DMA task requests; the method is characterized by selecting a DMA task request out of the multiple DMA task requests in response to timing deadlines associated with the DMA tasks. A device that includes an interface, that is adapted to receive DMA task requests; the device is characterized by including an arbiter that is adapted to select a DMA task request out of the multiple DMA task requests in response to timing deadlines associated with the DMA tasks.Type: GrantFiled: June 30, 2005Date of Patent: October 29, 2013Assignee: Freescale Semiconductor, Inc.Inventors: Uri Shasha, Sagi Gurfinkel, Gilad Hassid, Eran Kahn, Yehuda Shvager
-
Patent number: 8001430Abstract: A method for controlling an execution of a first DMA task, the method includes comprises monitoring an execution of the first DMA task, the method characterized by including defining a first DMA task execution interval and a first DMA task execution sub-interval; and performing a first possible timing violation responsive operation if the first DMA task was not completed during the first DMA task execution sub-interval. A device having a first DMA task controlling capabilities, the device includes a memory unit; characterized by including a DMA controller that is adapted to monitor an execution of the first DMA task that involves an access to the memory unit, and to perform a first possible timing violation responsive operation if the first DMA task was not completed during a first DMA task execution sub-interval.Type: GrantFiled: June 30, 2005Date of Patent: August 16, 2011Assignee: Freescale Semiconductor, Inc.Inventors: Uri Shasha, Sagi Gurfinkel, Gilad Hassid, Eran Kahn
-
Patent number: 7930444Abstract: A method for controlling multiple DMA tasks, the method includes receiving multiple DMA task requests; the method is characterized by defining multiple buffer descriptors for each of a plurality of DMA channel; wherein at least two buffer descriptors comprise timing information that controls an execution of cyclic time based DMA tasks; selecting a DMA task request out of the multiple DMA task requests; executing a DMA task or a DMA task iteration and updating the buffer descriptor associated with the selected DMA task request to reflect the execution; and jumping to the stage of selecting.Type: GrantFiled: June 30, 2005Date of Patent: April 19, 2011Assignee: Freescale Semiconductor, Inc.Inventors: Uri Shasha, Sagi Gurfinkel, Gilad Hassid, Eran Kahn
-
Publication number: 20090216917Abstract: A method for arbitrating between direct memory access task requests, the method includes receiving multiple DMA task requests; the method is characterized by selecting a DMA task request out of the multiple DMA task requests in response to timing deadlines associated with the DMA tasks. A device that includes an interface, that is adapted to receive DMA task requests; the device is characterized by including an arbiter that is adapted to select a DMA task request out of the multiple DMA task requests in response to timing deadlines associated with the DMA tasks.Type: ApplicationFiled: June 30, 2005Publication date: August 27, 2009Applicant: Freescale Semiconductor, IncInventors: Uri Shasha, Sagi Gurfinkel, Gilad Hassid, Eran Kahn, Yehuda Shvager
-
Publication number: 20090144589Abstract: A method for controlling an execution of a first DMA task, the method includes comprises monitoring an execution of the first DMA task, the method characterized by including defining a first DMA task execution interval and a first DMA task execution sub-interval; and performing a first possible timing violation responsive operation if the first DMA task was not completed during the first DMA task execution sub-interval. A device having a first DMA task controlling capabilities, the device includes a memory unit; characterized by including a DMA controller that is adapted to monitor an execution of the first DMA task that involves an access to the memory unit, and to perform a first possible timing violation responsive operation if the first DMA task was not completed during a first DMA task execution sub-interval.Type: ApplicationFiled: June 30, 2005Publication date: June 4, 2009Applicant: FREESCALE SEMICONDUCTOR, INC.Inventors: Uri Shasha, Sagi Gurfinkel, Gilad Hassid, Eran Kahn
-
Publication number: 20090125647Abstract: A method for executing a DMA task, the method includes receiving a request to execute a DMA task; the method characterized by including: defining inter-buffer jumping points at substantially an end of one or more dimensions of each multidimensional buffer out of a plurality of multidimensional buffers; and executing multiple DMA sub-tasks, wherein the executing includes jumping between buffers at the inter-buffer jumping points. A device hat includes at least one memory unit and a DMA controller adapted to access the memory unit; the device is characterized by being adapted to implement multidimensional buffers within the at least one memory unit; wherein the DMA controller is adapted to execute multiple DMA sub-tasks, wherein the execution comprises jumping between buffers at inter-buffer jumping points; and wherein the inter-buffer jumping points are defined at substantially an end of one or more dimensions of each multidimensional buffer out of a plurality of multidimensional buffers.Type: ApplicationFiled: June 30, 2005Publication date: May 14, 2009Applicant: CITIBANK, N.A.Inventors: Uri Shasha, Sagi Gurfinkel, Gilad Hassid, Eran Kahn