Re: [pacman-dev] [PATCH] mbscasecmp fails in 1% of time.

2016-06-12 Thread Allan McRae
On 11/06/16 17:26, Tobias Stoeckmann wrote: > On Fri, Jun 10, 2016 at 07:10:56PM -0400, Andrew Gregory wrote: >> Good catch, but the solution needs work. > > You were able to grasp the idea of this non-issue. > > Now you can adjust your code the way you want. > Good old passive aggression... D

Re: [pacman-dev] [PATCH] mbscasecmp fails in 1% of time.

2016-06-11 Thread Tobias Stoeckmann
On Fri, Jun 10, 2016 at 07:10:56PM -0400, Andrew Gregory wrote: > Good catch, but the solution needs work. You were able to grasp the idea of this non-issue. Now you can adjust your code the way you want.

Re: [pacman-dev] [PATCH] mbscasecmp fails in 1% of time.

2016-06-10 Thread Andrew Gregory
On 06/10/16 at 06:40pm, Tobias Stoeckmann wrote: > The width of wchar_t is allowed to be of the same width as long, > according to standards. The return type of mbscasecmp is int though. > > On amd64 with a 32 bit int, this means that mbscasecmp can return > zero (indicating that strings are equal

[pacman-dev] [PATCH] mbscasecmp fails in 1% of time.

2016-06-10 Thread Tobias Stoeckmann
The width of wchar_t is allowed to be of the same width as long, according to standards. The return type of mbscasecmp is int though. On amd64 with a 32 bit int, this means that mbscasecmp can return zero (indicating that strings are equal) even though the input strings differ. No known system ex