Sun Microsystems, Inc.

Oracle America, Inc.

Howard L Levy

Nadeem Eleyan

Hong Seok Kim

Harsh D Sharma

Jeffrey Y Su

Robert T Golla

Jeffrey S Brooks

Dennis L Wendell

Jama Ismail Barreh

Jin-uk Shin

Nayon Tomsio

Techniques and structures are described which allow the detection of certain dependency conditions, including evil twin conditions, during the execution of computer instructions. Information used to detect dependencies may be stored in a logical map table, which may include a content-addressable memory. The logical map table may maintain a logical register to physical register mapping, including entries dedicated to physical registers available as rename registers. In one embodiment, each entry in the logical map table includes a first value usable to indicate whether only a portion of the physical register is valid and whether the physical register includes the most recent update to the logical register being renamed. Use of this first value may allow precise detection of dependency conditions, including evil twin conditions, upon an instruction reading from at least two portions of a logical register having an entry in the logical map table whose first value is set.

Logical map table for detecting dependency conditions between instructions having varying width operand values

Post-manufacture compensation for a sensing offset can be provided, at least in part, by selectively exposing one of a pair of cross-coupled transistors in a sense amplifier to a bias voltage selected to cause a compensating shift in a characteristic of the exposed transistor. In designs susceptible to post-manufacture data dependent creep in a device characteristic, such exposure may be advantageously provided in situ by causing the sense amplifier to sense values purposefully skewed toward a predominate value selected to cause the compensating shift. In some realizations, an on-chip test block is employed to identify and characterize sensing mismatch. Typically, the techniques described herein may be employed to address sensing offsets that have developed post-manufacture due to a data-dependent effect. However, in some realizations, the techniques described herein may be used to address a sensing offset arising at least in part from other or additional sources.

Test circuit for measuring sense amplifier and memory mismatches

A signal duration measurement system compares a known duration, Tof a test data signal with the duration, Tof a state of a signal under test. In one embodiment, if Tcompares favorably with Tthen the circuit generating the signal under test 'passes.' Otherwise the signal under test 'fails,' and a problem has been identified. Furthermore, in one embodiment, Tcan be selectively adjusted to more accurately measure TIn one embodiment, the test data signal is allowed to travel a signal path, having a known signal propagation delay time, during a single state of the signal under test. The data signal at the beginning of the state, e.g. during the rise of the signal under test, is compared to the data signal captured at the end of the state, e.g. during the fall of the signal under test. If the initial and captured data signals are the same, then the duration of the state of the signal under test is greater than or equal to the signal propagation delay time. The signal propagation time can be adjusted by inserting varying delay elements into the signal path traversed by test data signal. The signal duration measurement system can be fabricated on-chip, thus making its use more practical. The signal duration measurement system is, for example, useful for measuring the state duration of signals such as self-resetting signals, which are difficult to externally measure.

On-chip measurement of signal state duration

It has been discovered that initialization of a memory array can be improved by setting the nodes of the memory array to a predetermined value automatically upon applying power to the integrated circuit. Data input nodes and a memory write enable node are configured to store the predetermined values on the nodes of the memory array in response to successive enablement of word lines corresponding to the nodes of the memory array and automatic reset of the word lines. Circuitry included for initializing control and data signals of the memory array are effectively disabled upon termination of the initialization. Inclusion of circuitry that initiates and terminates the initialization obviates an additional input/output pin for this purpose.

On-chip power-on voltage initialization

A Negative Bias Temperature Instability (NBTI) tolerant sense amplifier is provided. The sense amplifier includes an input stage having a pair of balanced isolation devices. Each of the balanced isolation devices has an input connected to receive a separate one of a pair of differential input signals. Each of the balanced isolation devices also has a gate that is connected to receive a common bias voltage. The sense amplifier further includes a sense stage connected to the input stage. The sense stage is configured to receive and amplify a higher signal to be provided by the pair of balanced isolation devices. The sense amplifier is also equipped to operate a low voltage levels.

Sense amplifier and method for making the same

Signal state durations, such as the pulse-width, of on-chip signals are often critical to the successful operation of an integrated circuit. The signal state durations measured by on-chip technology provide signal state duration information to an on-chip signal state duration control system. The signal state duration control system uses the information to adjust the signal state duration of an on-chip signal. In one embodiment, the signal state duration of the on-chip signal is the pulse width of the on-chip signal. The signal duration measurement and adjustment system is, for example, useful for measuring the state duration of signals such as self-resetting signals, which are difficult to externally measure and adjust signal state durations using on-chip technology.

On-chip signal state duration measurement and adjustment

A circuit has been developed that reduces the effective strength of a keeper circuit during an interval in which at least one path of an evaluation circuit is sensitive to a keeper device. The keeper circuit includes a keeper gating device coupled to a keeper device that is responsive to a keeper control. The keeper device is sized to overcome leakage current in the evaluation circuit. In some configurations, the keeper circuit includes a weak keeper device that is minimally sized to overcome noise while the keeper device is effectively disabled. In some configurations, the reduction in effective strength of the keeper circuit occurs before arrival of the fastest signal coupled to a sensitive output path of the evaluation circuit and the effective strength is restored after arrival of the slowest signal coupled to the sensitive output path of the evaluation circuit.

Pulsed dynamic keeper gating

Post-manufacture compensation for a sensing offset can be provided, at least in part, by selectively exposing one of a pair of cross-coupled transistors in a sense amplifier to a bias voltage selected to cause a compensating shift in a characteristic of the exposed transistor. Such exposure may be advantageously provided in situ by causing the sense amplifier to sense values purposefully skewed toward a predominate value selected to cause the compensating shift. In some realizations, purposefully skewed values (e.g., value and value_1) are introduced directly into the sense amplifier. In some realizations, an on-chip test block is employed to identify and characterize sensing mismatch.

Measuring and correcting sense amplifier and memory mismatches using NBTI

The locked frequency of a PLL is used to latch a test signal through various latching devices (flip-flops or the like). Various different delays are selectively applied to the test signal to provide a delayed test signal and the delayed test signal is measured to determine whether the delay in the test signal matches the jitter in the locked frequency of the PLL. When the delay in the test signal matches the jitter in the locked frequency of the PLL, the respective delay of the test-signal is used to determine the effective locked frequency of the PLL.

On-chip PLL locked frequency determination method and system

Post-manufacture variation of timing may be employed to address data-dependent degradation or creep in device characteristics affecting a differential circuit. One particular example of such data-dependent degradation or creep involves Negative Bias Temperature Instability (NBTI). In certain memory circuit configurations, NBTI can cause threshold voltage (V ) of PMOS devices to increase by an amount that depends on the historical amount of voltage bias that has been applied across gate and source/drain nodes. In the case of many sense amplifier designs, a predominant value read out using the sense amp may tend to disparately affect one device (or set of devices) as compared with an opposing device (or set of devices). In other words, if the same data value is read over and over again, then one of two opposing PMOS devices of a typical sense amp will accumulate an NBTI-related V shift, while the opposing PMOS device will accumulate little or no shift. The accumulated mismatch tends to cause an increase in the sense amp fail-point.

Variable delay compensation for data-dependent mismatch in characteristic of opposing devices of a sense amplifier

According to one embodiment, a memory is disclosed. The memory includes a differential sense amplifier that receives a data input and a complementary data input; and a switching mechanism, coupled to the amplifier, that switches the data input and the complementary data input to minimize a negative bias temperature instability (NBTI) effect on the amplifier.

Growing community of inventors

Cedar Park, TX, United States of America

Howard L Levy