Read only memory (ROM) with redundancy
A read only memory (ROM) with redundancy and methods of use are provided. The ROM with redundancy includes a programmable array coupled to a repair circuit having one or more redundant repairs. The one or more redundant repairs include a word address match logic block, a data I/O address, and a tri-state buffer. The word address match logic block is provided to the tri-state buffer as a control input and the data I/O address is provided to the tri-state buffer as an input. An output of the tri-state buffer of each redundant repair is provided as a first input to one or more logic devices. One or more data outputs of a ROM bit cell array is provided as a second input to a respective one of the one or more logic devices.
1. A method, comprising:
performing a read-only memory (ROM) built-in-self-test (BIST) to determine a bit failure;
identifying a row having the bit failure;
comparing columns of the row having the bit failure with expected results to determine the location of the bit failure;
performing a soft repair to repair the bit failure in a ROM;
determining whether the soft repair corrected the bit failure using a functional BIST; and
performing a hard repair to store a corrected data value in a hardware component when the soft repair corrected the bit failure,
wherein the identifying the row having the bit failure comprises:
predetermining an expected multiple input signature register (MISR) value for a first half of rows of a ROM bit cell array;
comparing the first half of the rows with the expected MISR value;
determining whether the bit failure is in the first half of the rows or a second half of the rows;
splitting the determined half of the rows having the bit failure into a first half and a second half; and
iteratively comparing the determined half of the rows to the expected MISR value and determining whether the bit failure is in the first half or the second half of the determined half of the rows.
2. The method of claim 1 , further comprising:
reading a content of columns in the failing row into a tester; and
comparing the contents to expected content of the columns of the row having the bit failure to determine an address of the bit failure.
3. The method of claim 1 , wherein a failed bit is an incorrect value, a stuck-at-0 bit, or a stuck-at-1 bit.
4. A method, comprising:
performing a read-only memory (ROM) built-in-self-test of an error correction code (ECC) word to determine a bit failure;
conducting an ECC-based diagnostic to determine whether there are multiple bit failures;
identifying an address of the bit failure;
performing a soft repair to repair the bit failure;
performing a hard repair to store a corrected data value of the bit failure in a hardware component;
serializing a plurality of words to form the ECC word by inserting “m” number of ECC bits according to a hamming code logic;
storing the address of the bit failure in a serial bit position counter;
re-running the serialized ECC word through the control logic; and
terminating the control logic when a value of a bit counter is greater than the address of the bit failure,
wherein the “m” number of ECC bits is given by a data width<=(2^(m−1)−m),
wherein the identifying the address of the bit failure comprises:
providing the ECC word to a control logic, wherein the control logic:
inserts the ECC bits into the ECC word;
computes a parity for the ECC bits;
outputs parity bits; and
comparing the parity bits with an expected output value, wherein parity bits with a different value than the expected output value form a parity word which indicates the address of the bit failure.
5. The method of claim 4 , wherein:
the expected output value is all zeros; and
when the parity bits match the expected output value, a bit repair is unnecessary.
6. The method of claim 4 , further comprising translating a location of the bit failure of the ECC word to a repair location word address, a repair location column address, and a bit data value.
7. A structure, comprising:
a read-only memory (ROM) bit cell array;
an error correction code (ECC) serialized word embedded in the ROM bit cell array;
a repair circuit coupled to a programmable array and one or more outputs of the ROM bit cell array;
a ROM built-in-self-test (RBIST) multiple input signature register (MISR) and a hamming code logic circuit coupled to an output of the repair circuit; wherein:
the ROM bit cell array is expanded to include ECC bits in the serialized word;
the hamming code logic circuit identifies a location of a bit failure; and
the repair circuit comprises one or more redundant repairs having a word address match logic block, a data I/O address, and at least one tri-state buffer;
the word address match logic block is provided to the at least one tri-state buffer as a control input;
the data I/O address is provided to the at least one tri-state buffer as an input;
an output of the at least one tri-state buffer of each redundant repair is provided as a first input to one or more logic devices; and
one or more data outputs of a ROM bit cell array is provided as a second input to a respective one of the one or more logic devices.
8. The structure of claim 7 , wherein:
the data I/O address is provided as the input to a first tri-state buffer;
a data value is provided as an input to a second tri-state buffer;
the word address match logic block is provided to the first tri-state buffer and the second tri-state buffer as the control input;
an output of the first tri-state buffer is provided to the one or more logic devices as a select signal; and
an output of the second tri-state buffer is provided as the first input to one or more logic devices.