| commit | c88aec662b639caebd9ce41d31e087757a7e2400 | |
| author | Dan Carpenter <dan.carpenter@oracle.com> | |
| Thu, 5 Oct 2017 10:32:10 +0000 (5 13:32 +0300) | ||
| committer | Dan Carpenter <dan.carpenter@oracle.com> | |
| Thu, 5 Oct 2017 10:32:10 +0000 (5 13:32 +0300) | ||
| tree | 25a8933329c6516349fd79248a1a42a940cda369 | treesnapshot (tar.gz zip) |
| parent | 410e8a89da7d5b76e0c2d98a7b74cca7be34cef0 | commitdiff |
comparison: fix a bug with unsigned/signed impossible returned comparisons
The problem is that we're saying that "foo is UNSIGNED_LT bar" and then
the db doesn't record signed/unsigned information. Then there is probably
another bug that filter_comparison doesn't handle signed/unsigned correct?
I'm not positive if this a bug or a feature...
Anyway, I feel like this patch is probably the correct thing. We can
normally assume that the comparisons and the DB comparisons are going to
use the same signedness...
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
The problem is that we're saying that "foo is UNSIGNED_LT bar" and then
the db doesn't record signed/unsigned information. Then there is probably
another bug that filter_comparison doesn't handle signed/unsigned correct?
I'm not positive if this a bug or a feature...
Anyway, I feel like this patch is probably the correct thing. We can
normally assume that the comparisons and the DB comparisons are going to
use the same signedness...
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
| smatch_comparison.c | diffblobblamehistory |