-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Hi there,
Am Samstag, den 05.11.2016, 23:47 +0100 schrieb Stefan Pöschel: > To be honest: I have no knowledge in terms of ABI and which changes > affect it. But I have a strong feeling that both of my proposals > break it :-/ probably. As a rule of thumb, if a change in a library requires depending binaries to get rebuilt in order for the change to take effect, then it's an ABI break. Sure, I want to avoid that. > The first occasion also works without the cast as far as I can see. Strictly speaking, setting an unsigned variable to a negative value should be undefined behaviour, but I am pretty sure that the compiler does an implicit cast and simply sets the variable value to 0xff. > The second cast should IMO be annoted with a respective comment, to > describe the issue. Probably. I'll forward this issue upstream and see what he has to say about it. Thanks! - Fabian -----BEGIN PGP SIGNATURE----- iQIcBAEBCAAGBQJYH01bAAoJEMvqjpcMzVnf6x0QAKsIKRTvCkvrFcSkA9QEG0GW r+JpIBZR765BsXBsP8lJ3JSHLa3QeMhU7J/vaEF9K4amGjfCpkJjn6V7t8uH6ksL MH+BVczvVWNP31PQ09tHavulTmbBN1GDeqWlqaaspciyPFlEI61kWpjA11YoZrYV ssDYUbdCU4vOtN7ADjgBhw3UI/IfyquW3jVSv00sTIU5omw6n+fiXJN7IE6di3vR CJbf4E4j3QtIQUpTa4R2sPFyJ/vuwB9Vbom/w3GkmAWer/8hiNdBvAPeoBYUyPp6 PCRpbcC5fXUaDfpJb+eqxEk6COTk3qiDLF2vtu8wukPhQJuDE0Wz1VikmddnmPJl 8XJPZUJDsb1AzwZz9M8uTEAfNLzPiCKS1MxHrFXacp5EQIzA3APG31mmrPE759gn S0uVnjgJI/7xtsh6TRuI3OE2Z5XajhvyMaK+4MAlVSNuS3KOm29QDbAgbXsWT4FO qwFbOuxwAgmvCuNDGH/z/nd7DBKMEaW8PzrP4z9u08NGMoFj20BcO/FufmsXIsgM qaybbKqaINjbSwb2+MLbsirxTYcjkOxYnRJwxMNvSC4YIaDAHtN7zpcj/LMvWG11 22tHCQl169FV9G9rYB3HrPR0Pd7WbtmExXr9b56XPhsUQpckpqwb7bXZY421b0N+ ajOMxDsr2uEOxUtk+bSN =E/9U -----END PGP SIGNATURE-----
