Sharing of a logic operator having a work register
A circuit for calculating a discriminating function with successive iterations and with a work register on data divided into blocks, including: a single operator in wired logic for executing the function; a plurality of work registers sharing the operator; and an element for selecting one of the work registers to be associated with the operator.
1. A circuit for calculating a discriminating function by performing successive iterations on data divided into a plurality of data blocks using a work register, the circuit comprising:
a single operator in wired logic for executing the discriminating function;
a plurality of work registers sharing the operator, each work register from the plurality of work registers storing a current state of the operator and a rank of a corresponding iteration; and
a selection element for selecting one of the work registers to be associated with the operator.
2. The circuit of claim 1 , wherein the discriminating function is a Hash function.
3. The circuit of claim 1 , wherein a multiplexer forming the selection element is controlled by a priority decoder associated with an integrated processor containing said calculation circuit.
4. The circuit of claim 3 , wherein inputs/outputs of the plurality of work registers are connected to a plurality of inputs of the multiplexer.
5. The circuit of claim 4 , wherein an input/output of the multiplexer is connected to an input and an output of the operator.
6. The circuit of claim 1 , wherein each work register from the plurality of work registers is assigned to an application from a plurality of applications.
7. The circuit of claim 6 , wherein the plurality of applications use the discriminating function.
8. The circuit of claim 1 , wherein the operator executing the discriminating function performs acts comprising:
executing a first iteration of the discriminating function on a first data block from the plurality of data blocks;
determining whether an end of the data is reached;
if it is determined than the end of the data is not reached:
incrementing the rank of the corresponding iteration; and
executing a second iteration of the function on a second data block from the plurality of data blocks; and
if it is determined than the end of the data is reached, providing the current state of the operator as an output.
9. A method of calculating a discriminating function by performing successive iterations on data divided into a plurality of data blocks, the method comprising:
sharing an operator for executing the discriminating function between a plurality of work registers;
selecting a work register from the plurality of work registers to be associated with the operator; and
after each execution of the discriminating function by the operator on a data block from the plurality of data blocks, storing, in the work register, a current state of the operator and a rank of a current iteration of the execution of the discriminating function.
10. The method of claim 9 , wherein each work register from the plurality of work registers is assigned to an application from a plurality of applications.
11. The method of claim 10 , wherein the plurality of applications use the discriminating function.
12. The method of claim 9 , wherein the discriminating function is a hash function.
13. The method of claim 9 , wherein a priority decoder associated with an integrated processor containing a calculation circuit performing the method controls a multiplexer that performs the selection of the work register.
14. The method of claim 13 , wherein inputs/outputs of the plurality of work registers are connected to a plurality of inputs of the multiplexer.
15. The method of claim 13 , wherein an input/output of the multiplexer is connected to an input and an output of the operator.
16. The method of claim 9 , wherein the operator executing the discriminating function performs acts comprising:
executing a first iteration of the discriminating function on a first data block from the plurality of data blocks;
determining whether an end of the data is reached;
if it is determined than the end of the data is not reached:
incrementing the rank of the corresponding iteration; and
executing a second iteration of the function on a second data block from the plurality of data blocks; and
if it is determined than the end of the data is reached, providing the current state of the operator as an output.