Abstract: Techniques for handling uncorrectable errors occurring during memory accesses reduce the likelihood of mis-correction of errors due to the presence of noise. When an uncorrectable memory error is detected in response to an access to a memory device, a memory controller managing the interface to the memory halts issuing of access requests to the memory device until a predetermined time period has elapsed. In-flight memory requests are marked for retry, and responses to pending request are flushed. A calibration command may be issued after the predetermined time period has elapsed. After the predetermined time period has elapsed and any calibration performed, the requests marked for retry are issued to the memory device.

Abstract: A method and structures are provided for implementing decoupling capacitors within a DRAM TSV stack. A DRAM is formed with a plurality of TSVs extending completely through the substrate and filled with a conducting material. A layer of glass is grown on both the top and bottom of the DRAM providing an insulator. A layer of metal is grown on each glass layer providing a conductor. The metal and glass layers are etched through to TSVs with a gap provided around the perimeter of via pads. A respective solder ball is formed on the TSVs to connect to another DRAM chip in the DRAM TSV stack. The metal layers are connected to at least one TSV by one respective solder ball and are connected to a voltage source and a dielectric is inserted between the metal layers in the DRAM TSV stack to complete the decoupling capacitor.

Abstract: According to one embodiment of the present invention, a method for bank sparing in a 3D memory device that includes detecting, by a memory controller, a first error in the 3D memory device and detecting a second error in a first element in a first rank of the 3D memory device, wherein the first element in the first rank has an associated first chip select. The method also includes sending a command to the 3D memory device to set mode registers in a master logic portion of the 3D memory device that enable a second element to receive communications directed to the first element and wherein the second element is in a second rank of the 3D memory device, wherein the first element and second element are each either a bank or a bank group that comprise a plurality of chips.

Abstract: According to one embodiment of the present invention, a method for operating a three dimensional (“3D”) memory device includes detecting, by a memory controller, a first error on the 3D memory device and detecting, by the memory controller, a second error in a first chip in a first rank of the 3D memory device, wherein the first chip has an associated first chip select. The method also includes powering up a second chip in a second rank, sending a command from the memory controller to the 3D memory device to replace the first chip in the first chip select with the second chip and correcting the first error using an error control code.

Abstract: Isolation of faulty links in a transmission medium including a method that includes receiving an atomic data unit via a multi-link transmission medium that has a plurality of transmission links. An error condition is detected and it is determined that the error condition is isolated to a single transmission link. It is determined if the single transmission link has been isolated previously as a failing transmission link a specified number of times within an interval specified by a timer. If the single transmission link has been isolated previously as a failing transmission link a specified number of times within an interval specified by a timer then: identifying the single transmission link as a faulty transmission link; resetting the timer; and outputting an identifier of the single transmission link.

Abstract: A method includes directing an access of a memory location of a memory device to an error correction code (ECC) decoder in response to receiving a test activation request indicating the memory location. The method also includes writing a test pattern to the memory location and reading a value from the memory location. The method further includes determining whether a fault is detected at the memory location based on a comparison of the test pattern and the value.

Abstract: An approach for saving power in a memory subsystem that uses memory access idle timer to enable low power mode and memory scrub operation within computing system has been provided. The computing system determines that a memory subsystem is switched out of low power operation mode due to a memory scrub operation. In addition, the computing system bypasses the low power operation mode of an idle timer of the memory subsystem such that the memory subsystem is returned to the low power operation mode upon completion of the memory scrub operation. The computing system further sets a scrub flag of the memory subsystem to a high state, and clears the scrub flag to a low state to track if the idle timer should be bypassed.

Abstract: An approach for saving power in a memory subsystem that uses memory access idle timer to enable low power mode and memory scrub operation within computing system has been provided. The computing system determines that a memory subsystem is switched out of low power operation mode due to a memory scrub operation. In addition, the computing system bypasses the low power operation mode of an idle timer of the memory subsystem such that the memory subsystem is returned to the low power operation mode upon completion of the memory scrub operation. The computing system further sets a scrub flag of the memory subsystem to a high state, and clears the scrub flag to a low state to track if the idle timer should be bypassed.

Abstract: This disclosure includes a method for securing a memory of an electronic system that includes initializing the memory, creating a security key, transmitting the security key to memory, storing the security key in the memory, transmitting the current security key and a a new security key to the memory by the memory controller. If the current security key transmitted is the same as the security key stored in memory, then access to the memory is enabled and the current security key in the memory is replaced with the new security key. If the current security key transmitted is not the same as the security key stored in the memory, then access to the memory is disabled.

Abstract: A method includes directing an access of a memory location of a memory device to an error correction code (ECC) decoder in response to receiving a test activation request indicating the memory location. The method also includes writing a test pattern to the memory location and reading a value from the memory location. The method further includes determining whether a fault is detected at the memory location based on a comparison of the test pattern and the value.

Abstract: A method includes reading, at a memory controller, data from a first dynamic random-access memory (DRAM) die layer of a DRAM stack. The method also includes writing the data to a second DRAM die layer of the DRAM stack. The method further includes sending a request to a test engine to test the first DRAM die layer after writing the data to the second DRAM die layer.

Abstract: Mechanisms are provided for efficient storage of meta-bits within a system memory. The mechanisms combine an L/G bit and an SUE bit to form meta-bits. The mechanisms then determine the local/global state of a cache line on the first cycle of data. The mechanisms forward the data to the requesting cache, and the requesting cache may reissue the request globally based on the local/global state of the cache line. The mechanisms then determine the special uncorrectable error state of the cache line on the second or subsequent cycle of data. The mechanisms perform error processing regardless of whether the request was reissued globally.

Abstract: Mechanisms are provided for efficient storage of meta-bits within a system memory. The mechanisms combine an L/G bit and an SUE bit to form meta-bits. The mechanisms then determine the local/global state of a cache line on the first cycle of data. The mechanisms forward the data to the requesting cache, and the requesting cache may reissue the request globally based on the local/global state of the cache line. The mechanisms then determine the special uncorrectable error state of the cache line on the second or subsequent cycle of data. The mechanisms perform error processing regardless of whether the request was reissued globally.

Abstract: A method includes modifying, at a bit error injection circuit, a multiplier value by a first value according to an occurrence of a first event. The method also includes, in response to a determination that the modified multiplier value matches a first threshold, modifying, at the bit error injection circuit, the offset value according to an occurrence of a second event. The method further includes, in response to a determination that the modified offset value matches a second threshold, asserting, at the bit error injection circuit, an error injection signal. The method further includes asserting a first error pattern to be transmitted via a bus lane based on the error injection signal.

Abstract: A method provides processor initialization in different platform environments via a single code set. The method includes: in response to detecting a power-on operation of the processor, a microcontroller retrieving hardware procedures (HWP) framework code from a storage and triggering execution of the HWP framework code on the processor. The execution of the HWP framework code generates a HWP framework that comprises a plurality of application programming interfaces (APIs) which govern how all communication processes involving hardware procedures can be accomplished. The method further includes performing one or more initialization procedures by communicating one or more attribute data via the HWP framework to configure the processor for operation within a specific platform environment in which the processor is to be operated. The HWP framework includes standard interfaces and enables direct updates to hardware procedures without requiring a new flash code or a firmware patch.

Abstract: A system and computer program product provide processor initialization in different platform environments via a single code set. The system includes: in response to detecting a power-on operation of the processor, a microcontroller retrieving hardware procedures (HWP) framework code from a storage and triggering execution of the HWP framework code on the processor. The execution of the HWP framework code generates a HWP framework that comprises a plurality of application programming interfaces (APIs) which govern how all communication processes involving hardware procedures can be accomplished. The system further includes the microcontroller performing one or more initialization procedures by communicating one or more attribute data via the HWP framework to configure the processor for operation within a specific platform environment in which the processor is to be operated.

Abstract: A method and structures are provided for implementing decoupling capacitors within a DRAM TSV stack. A DRAM is formed with a plurality of TSVs extending completely through the substrate and filled with a conducting material. A layer of glass is grown on both the top and bottom of the DRAM providing an insulator. A layer of metal is grown on each glass layer providing a conductor. The metal and glass layers are etched through to TSVs with a gap provided around the perimeter of via pads. A respective solder ball is formed on the TSVs to connect to another DRAM chip in the DRAM TSV stack. The metal layers are connected to at least one TSV by one respective solder ball and are connected to a voltage source and a dielectric is inserted between the metal layers in the DRAM TSV stack to complete the decoupling capacitor.

Abstract: A method and structures are provided for implementing decoupling capacitors within a DRAM TSV stack. A DRAM is formed with a plurality of TSVs extending completely through the substrate and filled with a conducting material. A layer of glass is grown on both the top and bottom of the DRAM providing an insulator. A layer of metal is grown on each glass layer providing a conductor. The metal and glass layers are etched through to TSVs with a gap provided around the perimeter of via pads. A respective solder ball is formed on the TSVs to connect to another DRAM chip in the DRAM TSV stack. The metal layers are connected to at least one TSV by one respective solder ball and are connected to a voltage source and a dielectric is inserted between the metal layers in the DRAM TSV stack to complete the decoupling capacitor.

Abstract: A method and apparatus for controlling marking store updates in a central electronic complex with a plurality of core processors and eDRAM cache and interconnect bus to a service processor for loading memory controller firmware to dual-channel DDR3 memory controllers with an internal marking store. Loaded firmware of the memory controllers is responsible for tracking of ECC errors using a ECC decoder control whereby said marking store is written by a slow ECC decoder, and read by a fast ECC decoder for every read operation of said memory controllers to provide a blocking mechanism for notifying marking store firmware when the marking store has been updated and which guarantees that marking store firmware cannot write to the marking store until the marking store firmware has seen updates without causing the marking store hardware to time out.

Abstract: According to one embodiment of the present invention, a method for operating a three dimensional (“3D”) memory device includes detecting, by a memory controller, a first error on the 3D memory device and detecting, by the memory controller, a second error in a first chip in a first rank of the 3D memory device, wherein the first chip has an associated first chip select. The method also includes powering up a second chip in a second rank, sending a command from the memory controller to the 3D memory device to replace the first chip in the first chip select with the second chip and correcting the first error using an error control code.