Hi!
On Jun 22, Georgi Kodinov wrote:
> On 22.06.2007, at 14:03, Sergei Golubchik wrote:
> >On Jun 22, kgeorge@stripped wrote:
> >>ChangeSet@stripped, 2007-06-22 13:49:40+03:00, gkodinov@stripped +1 -0
> >> Bug #27383: Crash in test "mysql_client_test"
> >
> >Hm. Okay, I understand why you removed volatile from integer
> >variables. But are you sure it's safe to have float variable not
> >volatile ?
>
> The problem (as we determined over IRC) is in doublestore().
> floatstore() doesn't use such trickery, so IMHO it's safe to leave
> the existing code that uses the storage buffer to check for rounding
> errors (instead of the stack variable as we do for doubles to avoid
> the pointer aliasing effects introduced by doublestore()).
okay. But it's better to keep the code uniform, use the same solution
for float and double. Otherwise one starts wondering why the code is
different for them.
Ok to push after using the same code for float branch.
Regards / Mit vielen Grüssen,
Sergei
--
__ ___ ___ ____ __
/ |/ /_ __/ __/ __ \/ / Sergei Golubchik <serg@stripped>
/ /|_/ / // /\ \/ /_/ / /__ Principal Software Developer
/_/ /_/\_, /___/\___\_\___/ MySQL GmbH, Radlkoferstr. 2, D-81373 München
<___/ Geschäftsführer: Kaj Arnö - HRB
München 162140
