-
Notifications
You must be signed in to change notification settings - Fork 273
For Java skip the check for dereference of a deallocated or dead object #89
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
For Java skip the check for dereference of a deallocated or dead object #89
Conversation
This will enable recovering the full path to a source file where necessary.
Used for coverage output and dead-code analysis
Catch decl in the try/catch is not used. Making string constants to be ID_constant expressions with string_typet Typecheking function call. Added "nan" symbol. Assigning type to lhs after unknown function call. White spaces, positioning { in new line. Cleaning up the code. Correct includes. Renaming jsil types. Explicit constructors. Inlining expressions if used only once. Asserting that expression does not have type set just yet. More efficient implementation for is_subtype. Introduced jsil_incompatible_types to hide low level implementation. Fixed bug in is_subtype. Typechecking throw and return labels. Typechecking bitwise operators. Avoiding duplicate type-checking by marking symbols as type-checked. Add empty decl symbol. Special case for index expression. Clean up. Converting all procedure declarations to symbols before typechecking.
Supports, e.g., x86_64-apple-darwin14-llvm-gcc-4.2 when invoked as x86_64-apple-darwin14-llvm-goto-gcc-4.2 cr https://cr.amazon.com/r/4975025/
Windows vs. Unix need distinct treatment for executables; cleanup of duplicate declaration (using include instead) and formatting.
Do not generate necessarily dead instructions
a5bc493 introduced cases where information in the frame may become inconsistent with the actual renaming being used, as shown in the regression test.
Follow-up to 3f01cce to support void foo() __attribute__((constructor)); in addition to void __attribute__((constructor)) foo();
Fixes if_expr1 regresion test.
Fix java bytecode translation of control-flow joins
Support for postfixed __attribute__((constructor))
Object descriptor should skip typecasts
Fix consistency error in SSA level 1 renaming
…uffix, if unambiguous
… are missing (pointer related assertions missing from the goto program)
…umented with pointer checks
…id/cbmc into java_pointer_checks
I'm closing this pull request as it got too messy and difficult to review (I didn't rebase until too late). I'll make a cleaner version. |
@cristina-david note that you are always free to push --force a new series into the same branch to do such cleanup without being force to closing and creating a new pull request! |
@tautschnig noted, thanks. Being in doubt yesterday, I relied on the git "manual" https://xkcd.com/1597/ :) (courtesy of @martin-cs ) |
…aint_summaries_PR Improving HTML dump of taint summaries.
This is when running cbmc with the "--pointer-check" flag on Java programs.