-
For some input numbers, `BN_mod_mul_montgomery()` produces incorrect results on SPARC sun4v CPUs. Specifically, for `1DEA64498CAE6197AB5283F8A8A05F343D0CCA7F5F05EDF3C9892F80851E0E40A39BD8855B4824C5BCC…
-
What's the logic for converting bignums to fixnums if possible in `>integer`? It would seem better to "always convert" inside the `bignum.cpp`, returning a `fixnum` if possible, or a `bignum` if not.…
-
Bignums are bigger than 64-bit numbers. Since all of our coercions are going into numbers that are max of 64-bit I don't see how this can work correctly.
-
I came into a situation where I want to run coqtop.byte in ocamldebug to analyze some runtime behavior. From those plugins which I use from the Coq Platform, the Bignums syntax plugin is the only one …
-
I am definitely splitting hairs here, but technically, the arithmetic functions do not use bignums (as the nums are converted into `long`s for the operations and then back into nums), which slightly d…
-
*_Reported by: pnkfelix on Mon Mar 5 13:41:41 2007 *_
An email from Aziz.
(We should check if these results are consistent across platforms, or if its only a problem on IasnLarceny...)
```
On Mar 5…
-
Using fixed-precision integers in the runtime is a massive footgun, and requires audits for overflows and other bad behavior. Using arbitrary-precision integers avoids these problems. While there is…
-
At present, no least upper bound (supremum) is defined by any of the #time* feature designators; rather, they say "all values", which effectively translates to a requirement for arbitrary bignums. It …
-
This is mostly to help us give help to users. We need to be able to say "the bignums library" or "the crayons library" and have the user be able to find it. And probably the list should be in alphab…
-
in rsa_modExp in rsa.c, it seems there is a check for integer overflow (checking for negative numbers) that leads to an exit. It would be great to at least print out a warning when that happens. Would…