On 08/08/2010 07:13 AM, Florian Weimer wrote:
* Vincent Lefevre:
On 2010-08-07 13:38:05 +0200, Florian Weimer wrote:
* John Regehr:
[...]
Let me know if more detail is needed or if it would be better for me to
file all 71 bug reports.
I wonder if we should give up and make -fwrapv the default.
Do you really mean that all these integer overflows should behave
like -fwrapv and none of them can hide a real bug?
There quite a few instances of the x & -x pattern, which would be fine
with -fwrapv.
It's always valid if you know that x is not INT_MIN, which you do in
many cases (for example if x is an offset).
Paolo
