Make curve25519_basepoint_spot_check() faster

This function is currently around 9% of our startup time, and it does a bunch of curve25519 operations. We could make it a bunch faster, since our curve25519_basepoint function has not been observed to fail in the wild.

changed milestone to %Tor: 0.3.5.x-final

added 035-backport actualpoints::0 component::core tor/tor milestone::Tor: 0.3.5.x-final owner::nickm parent::28481 performance priority::medium resolution::implemented reviewer::asn severity::normal sponsor::8-can startup status::closed type::defect version::tor 0.3.4.9 labels

Trac:
Keywords: android startup performance controller deleted, startup performance added

Trac:
Status: new to accepted
Owner: N/A to nickm
Keywords: startup performance deleted, startup performance 035-backport added

See branch ticket28835_035 in my public repository, with PR at https://github.com/torproject/tor/pull/585

This change is so simple that I think it would be safe to backport to Tor 0.3.5.

Trac:
Status: accepted to needs_review

Very simple, lgtm;

I'm not familiar enough with this though to judge how much that is a "bad" idea vs "good" idea. So I'll rely on you for the reasoning behind this choice. Apart from that, trivial change.

Trac:
Reviewer: N/A to asn

LGTM. Maybe we can also consider moving this check on the unittests.

Trac:
Status: needs_review to merge_ready

Merged to 0.3.5 and forward; I think we already do more spot-checking than this in the unit tests.

Trac:
Resolution: N/A to implemented
Status: merge_ready to closed

Trac:
Actualpoints: N/A to 0

Merged to 0.3.5.

Trac:
Milestone: Tor: 0.4.0.x-final to Tor: 0.3.5.x-final

closed

mentioned in issue #28851 (moved)

moved to tpo/core/tor#28838 (closed)

mentioned in issue tpo/core/tor#28851 (closed)