des wrote _at_ Thu, 11 Dec 2003 14:25:08 +0100: > andy_at_splashground.de writes: > > And in this case, as far as i can see, the code does violate the > > strict aliasing rules. > > Yes, and you are welcome to send a patch, but there was none in the > PR. PRs consisting of nothing but a compiler error message are worse > than useless. I see. It just wasn't clear to me from your comment in the PR wether you considered the code to be a problem at all. Do you prefer the union version, then please see Andreas Jäger's patch, or my local version with more varibles and casts, it's rather untested, or just -fno-strict-aliasing for the offending code parts or a combination of all of these? AndyReceived on Thu Dec 11 2003 - 05:38:14 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:33 UTC