| commit | a93547affa0657a746c2771819c2a49c293ced2f | |
| author | Dan Carpenter <dan.carpenter@oracle.com> | |
| Wed, 19 Apr 2017 10:28:15 +0000 (19 13:28 +0300) | ||
| committer | Dan Carpenter <dan.carpenter@oracle.com> | |
| Wed, 19 Apr 2017 10:28:15 +0000 (19 13:28 +0300) | ||
| tree | 8c30eef6fa6837c77f3a40ee69c43f09d1712d4b | treesnapshot (tar.gz zip) |
| parent | fd90ac01996c608043f4cdb1f2be85ee7cfb9509 | commitdiff |
flow, db: fix how inline function assignments are handled
The problem here is that big_expression_stack is really unreliable and we
should probably just get rid of it. I think it was putting the inline
statements on the stack so it was returning false instead of true. This
is used in smatch_function_hooks and it means that we parse the call twice
once as a function call and once as a function assignment call. It
affected some new code I'm working on where it said that we're calling a
function twice resulting in a double free.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
The problem here is that big_expression_stack is really unreliable and we
should probably just get rid of it. I think it was putting the inline
statements on the stack so it was returning false instead of true. This
is used in smatch_function_hooks and it means that we parse the call twice
once as a function call and once as a function assignment call. It
affected some new code I'm working on where it said that we're calling a
function twice resulting in a double free.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
| smatch_flow.c | diffblobblamehistory |