On Sun, 20 Feb 2011 14:51:12 -0500, Nico Williams <n...@cryptonector.com> wrote:
> On Sun, Feb 20, 2011 at 6:24 AM, Samuel Adam <a...@certifound.com> wrote: [big snip] >> Any numerics experts (which I am not) or fp-software gurus care to chime >> in? *If* I am correct in my inductive hypothesis that the SQLite core >> may >> in rare circumstances trip SIGFPE, I have some suggestions as to [snip] > > Why don’t you try it? It’s not a question which can be answered by a trivial type-numbers-into-the-shell test. I can immediately think of at least two-and-a-half distinct practical reasons why not; a real numerics expert with extensive fp programming experience could easily infer what I am talking about, then add a few more I’ve probably never heard of. One reason is that different compilers, compiler switches, processor targets, processor version “errata” (bugs), and combinations thereof botch up floating point calculations in different ways—meaning that your ./sqlite3 and my sqlite3.exe may give subtly different results in some tests; the other reasons, which are more obvious, I will leave as an exercise to the reader. I know enough to understand the problem area conceptually, and to ask some precise questions for addressing the practical impact thereof. In other words, with apologies to Einstein, I am sufficiently knowledgeable to understand that I know almost nothing. Thus do I ask, if any Real Mathematicians or fp-calc specialists hit this thread, please do feel free to strut your stuff. As I said: *If* SQLite can in rare circumstances cause a signal to be raised, the result of which is officially “implementation-defined” in C, that really ought be investigated and documented. (And if SQLite can *guarantee* no SIGFPE *ever*—under any possible inputs, period, bar none—then I would be pleasantly quite surprised, and that should also be documented.) Very truly, Samuel Adam ◊ <http://certifound.com/> 763 Montgomery Road ◊ Hillsborough, NJ 08844-1304 ◊ United States Legal advice from a non-lawyer: “If you are sued, don’t do what the Supreme Court of New Jersey, its agents, and its officers did.” http://www.youtube.com/watch?v=iT2hEwBfU1g _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
