sanchan wrote: > missing -fno-strict-aliasing wherever there are bad programming > practices and so on;
IIUC, GCC's current -Wstrict-aliasing implementation has a number of flaws. According to a recent thread[i] on the GCC ml.. > The current implementation of -Wstrict-aliasing looks at a single expression > at once. Although it does OK given the limited scope, it has several > drawbacks. > First, it produces false positives, i.e., it warns when it should not. > For instance, it warns about pointer conversions even when the pointers > are never dereferenced. > Second, it has many false negatives, i.e., it does not warn when it should. > It does not warn when an address is not taken, for instance: > float *f = ... > int *i = (int*)f; > Third, it does not display the name or type of the aliased objects, which > makes the warnings hard to understand in templates or macros. Until that improves it might be a good idea to take strict aliasing warnings with a grain of salt. [i] http://permalink.gmane.org/gmane.comp.gcc.patches/131449 -- by design, by neglect dirtyepic gentoo org for a fact or just for effect 9B81 6C9F E791 83BB 3AB3 5B2D E625 A073 8379 37E8 (0x837937E8)
signature.asc
Description: OpenPGP digital signature
