Submitted by: emmanuel.azencot@bull.net
Reviewed by: steve

Prevent infinite loop in BN_GF2m_mod_inv().

Exact improvement coefficients vary from one benchmark and platform to
another, e.g. it performs 70%-33% better on ARM, hereafter less for
longer keys, and 100%-90% better on x86_64.

of algorithm test program.

(The CHANGES entry covers the change from PR #2111 as well, submitted by
Martin Olsson.)

Submitted by: Neel Mehta

Submitted by: Huang Ying
Reviewed by: Douglas Stebila

the BN library more generally useful.

Submitted by: Douglas Stebila

Submitted by: David Hartman <david_hartman@symantec.com>

keep disclaiming narrower than 32-bit support.

("perl util/ck_errf.pl */*.c */*/*.c" still reports many more.)

ec_curve.c (unify comments, etc).

Submitted by: Nils Larsch
Reviewed by: Bodo Moeller, Geoff Thorpe

Remove certain redundant BN_zero() initialisations, because BN_CTX_get(),
BN_init(), [etc] already initialise to zero.

Correct error checking in bn_sqr.c, and be less wishy-wash about how/why
the result's 'top' value is set (note also, 'max' is always > 0 at this
point).

automatically to accomodate the value, some compilers fail to do so. Most
notably 0x0123456789ABCDEF should come out as long long in 32-bit context,
but HP compiler truncates it to 32-bit value. Which in turn breaks GF(2^m)
arithmetics in hpux-parisc2-cc build. Therefore this fix...

white-space.

optimizations.

Submitted by: Nils Larsch

is itself experimental, and in addition may cause execution to break on
existing openssl "bugs" that previously were harmless or at least
invisible.

Submitted by: Bodo Moeller

Submitted by: Nils Larsch

Submitted by: Nils Larsch, Douglas Stebila

and it does not yet work correctly)

Submitted by: Douglas Stebila

The 'OPENSSL_NO_SUN_DIV' default is still subject to change,
so I didn't bother to finish the CHANGES entry yet.

Submitted by: Douglas Stebila <douglas.stebila@sun.com>, Sheueling Chang <sheueling.chang@sun.com>
(CHANGES entry by Bodo Moeller)