Continuing to test a hardware component after occurrence of a checkstop
A computer-implemented method, system, and computer program product for testing hardware in an environment involving a checkstop. Errors are injected into a model of a design under test (DUT), which includes a first hardware component interfaced with a second hardware component, at various states causing an occurrence of a checkstop in the first hardware component. A simulation of the model of the DUT is then performed with the injected errors for a number of cycles after the checkstop based on a quiesce time where the injected errors are suppressed. Furthermore, a post-processing script is executed to test a model of the DUT which ignores the errors of the first hardware component that were identified after the checkstop in the simulation of the model of the DUT. Any non-ignored errors from the execution of the post-processing script are detected and analyzed to determine if the second hardware component failed the test.
1 . A computer-implemented method for testing hardware in an environment involving a checkstop, the method comprising:
injecting errors into a model of a design under test comprising a first hardware component and a second hardware component at various states causing an occurrence of a checkstop in said first hardware component, wherein said first hardware component is interfaced with said second hardware
performing a simulation of said model of said design under test with said injected errors for a number of cycles after said checkstop based on a quiesce time where said injected errors are suppressed;
executing a post-processing script to test said model of said design under test which ignores errors of said first hardware component that were identified after said checkstop in said simulation of said model of said design under test;
analyzing a non-ignored error from said execution of said post-processing script; and
outputting an indication that said second hardware component failed said test after said occurrence of said checkstop in response to said analyzed error causing said model of said design under test to fail said test.
2 . The method as recited in claim 1 , wherein said injected errors are injected using dynamic-link library injection to run code within another process.
3 . The method as recited in claim 1 , wherein said injected errors are injected at random times to cover a state space.
4 . The method as recited in claim 1 further comprising:
outputting an indication that said second hardware component did not fail said test after said occurrence of said checkstop in response to said analyzed error not causing said model of said design under test to fail said test.
5 . The method as recited in claim 1 further comprising:
stopping a clock signal from being applied to said first hardware component after said occurrence of said checkstop.
6 . The method as recited in claim 5 further comprising:
driving random inputs on an interface from said second hardware component to said first hardware component to confirm that said first hardware component has been checkstopped.
7 . The method as recited in claim 1 , wherein said first hardware component corresponds to a coherence fabric and said second hardware component corresponds to a level 3 cache.
8 . A computer program product for testing hardware in an environment involving a checkstop, the computer program product comprising one or more computer readable storage mediums having program code embodied therewith, the program code comprising programming instructions for:
injecting errors into a model of a design under test comprising a first hardware component and a second hardware component at various states causing an occurrence of a checkstop in said first hardware component, wherein said first hardware component is interfaced with said second hardware
performing a simulation of said model of said design under test with said injected errors for a number of cycles after said checkstop based on a quiesce time where said injected errors are suppressed;
executing a post-processing script to test said model of said design under test which ignores errors of said first hardware component that were identified after said checkstop in said simulation of said model of said design under test;
analyzing a non-ignored error from said execution of said post-processing script; and
outputting an indication that said second hardware component failed said test after said occurrence of said checkstop in response to said analyzed error causing said model of said design under test to fail said test.
9 . The computer program product as recited in claim 8 , wherein said injected errors are injected using dynamic-link library injection to run code within another process.
10 . The computer program product as recited in claim 8 , wherein said injected errors are injected at random times to cover a state space.
11 . The computer program product as recited in claim 8 , wherein the program code further comprises the programming instructions for:
outputting an indication that said second hardware component did not fail said test after said occurrence of said checkstop in response to said analyzed error not causing said model of said design under test to fail said test.
12 . The computer program product as recited in claim 8 , wherein the program code further comprises the programming instructions for:
stopping a clock signal from being applied to said first hardware component after said occurrence of said checkstop.
13 . The computer program product as recited in claim 12 , wherein the program code further comprises the programming instructions for:
driving random inputs on an interface from said second hardware component to said first hardware component to confirm that said first hardware component has been checkstopped.
14 . The computer program product as recited in claim 8 , wherein said first hardware component corresponds to a coherence fabric and said second hardware component corresponds to a level 3 cache.
15 . A system, comprising:
a memory for storing a computer program for testing hardware in an environment involving a checkstop; and
a processor connected to said memory, wherein said processor is configured to execute program instructions of the computer program comprising:
injecting errors into a model of a design under test comprising a first hardware component and a second hardware component at various states causing an occurrence of a checkstop in said first hardware component, wherein said first hardware component is interfaced with said second hardware component;
performing a simulation of said model of said design under test with said injected errors for a number of cycles after said checkstop based on a quiesce time where said injected errors are suppressed;
executing a post-processing script to test said model of said design under test which ignores errors of said first hardware component that were identified after said checkstop in said simulation of said model of said design under test;
analyzing a non-ignored error from said execution of said post-processing script; and
outputting an indication that said second hardware component failed said test after said occurrence of said checkstop in response to said analyzed error causing said model of said design under test to fail said test.
16 . The system as recited in claim 15 , wherein said injected errors are injected using dynamic-link library injection to run code within another process.
17 . The system as recited in claim 15 , wherein said injected errors are injected at random times to cover a state space.
18 . The system as recited in claim 15 , wherein the program instructions of the computer program further comprise:
outputting an indication that said second hardware component did not fail said test after said occurrence of said checkstop in response to said analyzed error not causing said model of said design under test to fail said test.
19 . The system as recited in claim 15 , wherein the program instructions of the computer program further comprise:
stopping a clock signal from being applied to said first hardware component after said occurrence of said checkstop.
20 . The system as recited in claim 19 , wherein the program instructions of the computer program further comprise:
driving random inputs on an interface from said second hardware component to said first hardware component to confirm that said first hardware component has been checkstopped.