[Python-Dev] New re failures on Windows

Tim Peters tim.one@comcast.net
Mon, 21 Apr 2003 23:33:05 -0400

[Martin v. L=F6wis]
>> Instead of trying various compilers hoping that the problem goes a=
>> I recommend that you try to narrow down the test case that fails.

[Andrew MacIntyre]
> I never had any hope the problem would "go away".  I've been trying=
> quantify the extent of the problem, by finding out which compilers
> exhibit the failure with what optimisation settings, so that the
> autoconf configurations generated don't result in interpreters that
> blow up unexpectedly.

Narrowing it down to the specific C code that's at fault is still the=
hope.  There are two reasons for that:

1. It's very easy to write ill-defined code in C, and for all we know
   now some part of _sre is depending on undefined, or implementation
   defined (but apparently likely), behavior.

2. If that's not the problem, optimization bugs are usually easy to
   sidestep via minor code changes.  You have to know which code is
   getting screwed first, though.

> ...
> I have a patch to configure.in which I'll upload to SF shortly whic=
> lowers the optimisation for FreeBSD.  Not my preferred outcome, but=
> I'm able to offer in my current circumstances.

Narrowing it down is indeed A Project.