Leveraging code churn analytics to optimize sanitizer performance
Systems and methods sanitize computer code. In particular, fragile portions of computer code are identified based on instances of bug/defect-related churn data associated with the computer code. A control flow graph representative of the computer code may be generated, the control flow graph including nodes and edges. Nodes whose source location falls within the reported fragile sections are identified, and may be flagged as being susceptible. Thereafter, a sanitizer is run on the flagged nodes.
1 . A method comprising:
identifying instances of bug/defect-related churn data associated with a source code file;
identifying a number of bug/defect-related churn data instances associated with each source code line of the source code file;
clustering data within a data structure representing the source code line with data representing consecutive source code line of the source code file to form a clustered data representing source code lines based on similarity of numbers of bug/defect-related churn data instances associated with the consecutive source code line;
ranking the clusters of data:
wherein the ranking of the clusters is at least based on the numbers of bug/defect-related churn data instances;
selecting highest ranked clusters of data as suspect source code ranges;
generating a control flow graph comprising nodes and edges;
flagging as susceptible, those nodes associated with a source code location within the suspect source code ranges; and
running a sanitizer on the nodes flagged as being susceptible.
2 . The method of claim 1 , wherein generating the control flow graph comprising nodes and edges is part of a normal compilation process.
3 . The method of claim 1 , wherein flagging as susceptible, those nodes associated with the source code location within the suspect source code ranges comprises marking control flow passing through the flagged nodes as susceptible code flow and enabled for sanitizer instrumentation.
4 . The method of claim 3 , wherein instrumenting only the nodes flagged as being susceptible comprises adding sanitizer instrumentation into an executable code.
5 . The method of claim 4 , wherein instrumenting on only the nodes flagged as being susceptible comprises running the executable code with the sanitizer instrumentation enabled on only the control flow marked as susceptible code flow.
6 . A method comprising:
identifying records of bug/defect-related commits associated with a source code file;
identifying a number of bug/defect-related commit records associated with each source code line of the source code file;
clustering data within a data structure representing the source code line with data representing consecutive source code line of the source code file to form a clustered data representing source code lines based on similarity of numbers of bug/defect-related commit records associated with the consecutive source code line;
ranking the clusters of data:
wherein the ranking of the clusters is at least based on the numbers of bug/defect-related commit records;
selecting highest ranked clusters of data as suspect source code ranges;
generating a control flow graph comprising nodes and edges;
flagging as susceptible, those nodes associated with a source code location within the suspect source code ranges; and
instrumenting on the nodes flagged as being susceptible.
7 . The method of claim 6 , wherein generating the control flow graph comprising nodes and edges is part of a normal compilation process.
8 . The method of claim 6 , wherein flagging as susceptible, those nodes associated with the source code location within the suspect source code ranges comprises marking control flow passing through the flagged nodes as susceptible code flow and enabled for sanitizer instrumentation.
9 . The method of claim 8 , wherein instrumenting only the nodes flagged as being susceptible comprises building the sanitizer into an executable code.
10 . The method of claim 9 , wherein instrumenting on only the nodes flagged as being susceptible comprises running the executable code with the sanitizer instrumentation enabled on only the control flow marked as susceptible code flow.
11 . A computer system comprising:
a processor; and
a memory including instructions that, when executed by the processor, cause the processor to:
identify instances of bug/defect-related churn data associated with a source code file;
identify a number of bug/defect-related churn data instances associated with each source code line of the source code file;
cluster data within a data structure representing the source code line with data representing consecutive source code line of the source code file to form a clustered data representing source code lines based on similarity of numbers of bug/defect-related churn data instances associated with the consecutive source code line;
rank the clusters of data:
wherein the ranking of the clusters is at least based on the numbers of bug/defect-related churn data instances;
select highest ranked clusters of data as suspect source code ranges;
generate a control flow graph comprising nodes and edges;
flag as susceptible, those nodes associated with a source code location within the suspect source code ranges; and
run a sanitizer on the nodes flagged as being susceptible.
12 . The computer system of claim 11 , wherein generating the control flow graph comprising nodes and edges is part of a normal compilation process.
13 . The computer system of claim 11 , wherein flagging as susceptible, those nodes associated with the source code location within the suspect source code ranges comprises marking control flow passing through the flagged nodes as susceptible code flow and enabled for sanitizer instrumentation.
14 . The computer system of claim 13 , wherein instrumenting only the nodes flagged as being susceptible comprises adding sanitizer instrumentation into an executable code.
15 . The method of claim 14 , wherein instrumenting on only the nodes flagged as being susceptible comprises running the executable code with the sanitizer instrumentation enabled on only the control flow marked as susceptible code flow.