Xilinx, Inc.

Cypress Semiconductor Corporation

Suresh Parameswaran

Thinh Tran

Boon Yong Ang

Joseph Tzou

Yuqing Gong

Henley Liu

Myongseob Kim

Cheang-Whang Chang

Gamal Refai-Ahmed

Suresh Ramalingam

Morgan Andrew Whately

Ankur Jain

Hoa Lap Do

Toshiyuki Hisamura

Sarayanan Balakrishnan

Scott McCann

In one example, a method includes providing a first side of a semiconductor substrate with a plurality of transistors, etching a second side of the substrate, opposite the first side, with a pattern of trenches, the trenches having a pre-defined depth and width, and providing the etched semiconductor substrate in a package. In one example, the predefined depth and width of the trenches is such so as to increase the surface area of the second side of the substrate by at least 20 percent. In one example, the method also includes providing a layer of a thermal interface material (TIM) on the second side of the substrate, including to fill at least a portion of the trenches.

Thermal enablement of dies with impurity gettering

A test vehicle, along with methods for fabricating and using a test vehicle, are disclosed herein. In one example, a test vehicle is provided that includes a substrate, at least a first passive die mounted on the substrate, and at least a first test die mounted on the substrate. The first test die includes test circuitry configured to test continuity through solder interconnects formed between the substrate and the first passive die.

Test vehicle for package testing

A thermal management system includes an integrated circuit (IC). The IC includes a plurality of digitally addressable sectors. Each sector includes an on-die sensing element. The on-die sensing element includes an on-die temperature sensor configured to measure a sector temperature and provide an analog signal associated with the sector temperature; and an on-die digitizer configured to generate a digital sensed temperature signal based on the analog signal. The IC further includes a first output configured to output a plurality of digital sensed temperature signals from the plurality of sectors.

Systems and methods for on-die heat generation and temperature sensing

The disclosure provides simple, low-cost but accurate systems and related methods for on-die temperature sensing typically using calibration and without the need for precision voltage references. In some implementations, the system utilizes two user selectable temperature sensing elements and two user selectable DACs to provide a digital code for the sensed temperature. In some implementations, the two sensing elements can be used to calibrate against each other. For example, calibration can be useful to account for silicon local/global variation. Typically, one of the temperature sensors is diode-based, while the other is resistor-based. However, those of skill in the art will recognize that, in accordance with the disclosure, more than two sensors can be provided that can be calibrated against one another.

On-die temperature sensing and digitization system

A circuit for testing bond connections between a first die and a second die is described. The circuit comprises a defect monitoring circuit implemented on the first die, which is configured as a test die; and a plurality of bond connections between the first die and the second die; wherein the defect monitoring circuit is configured to detect a defect in a bond connection of the plurality of bond connections between the first die and the second die. A method of testing bond connections between a first die and a second die is also described.

Circuit for and method of testing bond connections between a first die and a second die

An integrated circuit (IC) comprises routing circuitry including a plurality of signal line segments in routing layers of the IC, and a plurality of micro-bump contacts coupled to the routing circuitry. The IC includes a plurality of test circuits coupled to respective subsets of the plurality of signal line segments. Each test circuit is configured to connect micro-bump contacts in the respective subset to form first and second sets of daisy chains. Each test circuit is configured to test the first and second sets of daisy chains for open circuits and test for short circuits between the first and second sets of daisy chains. Each test circuit is configured to determine the locations of detected open circuits and determine the locations of detected short circuits.

Method of testing a semiconductor structure

An integrated circuit may include at least a first replica driver stage coupled between a reference impedance input and a first power supply node and having a first programmable driver impedance set by a first driver configuration value in the same manner as a first output driver section of the integrated circuit. At least a first replica input termination stage may be coupled between the reference impedance input and the first power supply node and having a first programmable termination impedance set by a first termination configuration value in the same manner as a first input termination section of the integrated circuit. An impedance programming circuit may generate at least the first driver configuration value and the first termination configuration value in response to a potential at the reference node.

Circuits and methods for programming integrated circuit input and output impedances

An improved circuit and method for programmable cascading of impedance matching in a multi-chip configuration are disclosed. Handshaking is implemented in cascaded chips by defining a master-slave configuration, and impedance is evaluated in cascaded chips in a non-overlapping manner. The circuit includes a plurality of chips arranged in a cascading configuration. A cascade output pin of a chip is coupled to a cascade input pin of a cascaded chip to enable handshaking between the plurality of chips. The plurality of chips are coupled to a common precision resistor via a common impedance line to enable each chip to calibrate impedance of the chip. Each of the plurality of chips includes a control circuit. Each control circuit includes a state machine circuit. The control circuit is configured to control a non-overlapping clock cycle of each chip during which the impedance of the chip is evaluated.

Circuit and method for cascading programmable impedance matching in a multi-chip system

Embodiments of the invention relate to the testing and reduction of read disturb failures in a memory, e.g., an array of SRAM cells. A read disturb test mode may be added during wafer sort to identify any marginal memory cells that may fail read disturb, thus minimizing yield loss. The read disturb test mode may include first writing data to the memory. After a predetermined time period, the read disturb test mode reads data from the same memory, and compares the read data with the data previously written to the memory. A repair signal may be generated, when the read data is different from the data previously written to the memory. Additionally, a system may be implemented to reduce read disturb failures in the memory. The system may include a match logic circuit and a data selecting circuit. When a match condition is satisfied, data is read from a register that stores the previous written data, instead of from the memory. The match logic circuit may be selectively enabled or disabled.

Memory having read disturb test mode

Data paths (and) can be configured to accommodate two or four burst data sequences, with a data value being input/output each half clock cycle. A data sequence can be a fixed order or user-defined order depending upon a selected option. A data input path () can reduce power consumption with an enable signal (dinen) timed to activate after data input lines have settled values. A data output path () can access output data in a parallel fashion for subsequent output according to a burst sequence. Cycle latencies for such output data can include one clock cycle latency or one and a half-clock cycles. A data output path () can also accommodate various clocking modes, including: single clocking with a delay locked loop (DLL) type circuit enabled, single clocking with a delay locked loop (DLL) type circuit disabled, and double clocking, with a phase difference between an input clock and output clock of up to 180&#xb0;.

Configurable data path architecture and clocking scheme

Synchronous SRAM may conform to Std. Sync or early-write at an external interface whilst providing late-write internally.

Single late-write for standard synchronous SRAMs

A memory interface system and method are provided for transferring data between a memory controller and an array of storage elements. The storage elements are preferably SRAM elements, and the memory interface is preferably one having separate address bus paths and separate data bus paths. One address bus path is reserved for receiving read addresses and the other address bus path is reserved for receiving write addresses. One of the data bus paths is reserved for receiving read data from the array, and the other data bus path is reserved for receiving data written to the array. While bifurcating the address and data bus paths within the interface is transparent to the memory controller, the separate paths afford addressing phases of a read and write address operation to be partially overlapped, as well as the data transfer phases. This will essentially reduce the cycle time between a read and write memory access, and proves useful when maximizing the data throughput across the data bus when implementing double data rate (QDR) mechanisms.

Growing community of inventors

Fremont, CA, United States of America

Suresh Parameswaran