The VERIFY macro turns on various paranoid consistency checks, but
the complete functionality should still be tested without it.
This also adds a couple of static test points for extremely small
split inputs/outputs. The existing bounds vectors already check
extremely large outputs.
This allows us to shift by 256+128 = 384 bits, which is a multiple of the limb size of
the scalar representation. This also happens to be the most precision possible for g2
that still fits into a 256-bit value.
a45c1fa63cb3020225d72049ef9c1cf300014795 Rename testrand functions to have test in name (Pieter Wuille)
Pull request description:
Suggested here: https://github.com/bitcoin-core/secp256k1/pull/808#discussion_r488871913
ACKs for top commit:
real-or-random:
ACK a45c1fa63cb3020225d72049ef9c1cf300014795 diff looks good
elichai:
utACK a45c1fa63cb3020225d72049ef9c1cf300014795
Tree-SHA512: a15c29b88877e0f1a099acab90cbfa1e70420527e07348a69c8a5b539319a3131b771b86852e772a669a1eb3475d508d0f7e10f37eec363dc6640d4eaf967536
4eecb4d6ef6d4f18be8870a5929feb1dae376d15 travis: VALGRIND->RUN_VALGRIND to avoid confusion with WITH_VALGRIND (Jonas Nick)
66a765c7752b76d99be02d0f84dc05105bf4e70d travis: Explicitly set --with-valgrind (Jonas Nick)
Pull request description:
Also remove CPPFLAGS=-DVALGRIND because that's redundant with when
configured with --enable-valgrind.
ACKs for top commit:
real-or-random:
ACK 4eecb4d6ef6d4f18be8870a5929feb1dae376d15 diff and travis output look good
sipa:
utACK 4eecb4d6ef6d4f18be8870a5929feb1dae376d15
elichai:
ACK 4eecb4d6ef6d4f18be8870a5929feb1dae376d15
Tree-SHA512: c22d79fccaa926a074272b63a61f052f4bec3b1e5a871e3f08a4f6c19046da575779126a7008eb8a7513e70997b32d1dc6565dfb7aa41c57c0b6ef15ebbc8303
c0041b5cfca5efb160aa9a5616350069c89a8c29 Add static assertion that uint32_t is unsigned int or wider (Tim Ruffing)
Pull request description:
Solves one item in #792 .
ACKs for top commit:
sipa:
utACK c0041b5cfca5efb160aa9a5616350069c89a8c29
elichai:
ACK c0041b5cfca5efb160aa9a5616350069c89a8c29
Tree-SHA512: 9f700e89be39e15983260da94642593d16b9c437171e10377837ac73731ca7ba5dd7e328b3d93d0a24d143fb9e73abd11c578f6b58e2f94c82b783e977173b0c
8b7dcdd955a4f57174f478e36bdae5b84784fb9c Add exhaustive test for extrakeys and schnorrsig (Pieter Wuille)
08d7d89299a6492bf9388b4662b709d268c8ea29 Make pubkey parsing test whether points are in the correct subgroup (Pieter Wuille)
87af00b511f2938b6b4799f94d446a005730515e Abstract out challenge computation in schnorrsig (Pieter Wuille)
63e1b2aa7d396209aa5e26aa540d9593ede312a6 Disable output buffering in tests_exhaustive.c (Pieter Wuille)
39f67dd072fc44c7c0d27b95610ba8912de56db5 Support splitting exhaustive tests across cores (Pieter Wuille)
e99b26fcd54cb4096515ba80cf0f79d147b2683c Give exhaustive_tests count and seed cmdline inputs (Pieter Wuille)
49e6630bca5f6628bd1fd92d70d465273d4d873f refactor: move RNG seeding to testrand (Pieter Wuille)
b110c106fa9704e30f6b0c2ffa6a2697031e89a8 Change exhaustive test groups so they have a point with X=1 (Pieter Wuille)
cec7b18a34e68adb04f31a71a2eb4c5fc97674ce Select exhaustive lambda in function of order (Pieter Wuille)
78f6cdfaae9866694dcb0eee966332688753a8c3 Make the curve B constant a secp256k1_fe (Pieter Wuille)
d7f39ae4b67ea1ac6f085e6262a5f53afc0c5a25 Delete gej_is_valid_var: unused outside tests (Pieter Wuille)
8bcd78cd791fd9209d72d6bce455c8d3cf2c0249 Make secp256k1_scalar_b32 detect overflow in scalar_low (Pieter Wuille)
c498366e5b2d9c60e2e677949cf7373dbe877515 Move exhaustive tests for recovery to module (Pieter Wuille)
be317915436909573733afe3972a9abdee9357f7 Make group order purely compile-time in exhaustive tests (Pieter Wuille)
Pull request description:
A few miscellaneous improvements:
* Just use EXHAUSTIVE_TEST_ORDER as order everywhere, rather than a variable
* Move exhaustive tests for recovery module to the recovery module directory
* Make `secp256k1_scalar_set_b32` detect overflow correctly for scalar_low (a comment in the recovery exhaustive test indicated why this was the case, but this looks incorrect).
* Change the small test groups so that they include a point with X coordinate 1.
* Initialize the RNG seed, allowing configurating from the cmdline, and report it.
* Permit changing the number of iterations (re-randomizing for each).
* Support splitting the work across cores from the cmdline.
And a big one:
* Add exhaustive tests for schnorrsig module (and limited ones for extrakeys).
ACKs for top commit:
real-or-random:
ACK 8b7dcdd955a4f57174f478e36bdae5b84784fb9c
jonasnick:
ACK 8b7dcdd955a4f57174f478e36bdae5b84784fb9c
Tree-SHA512: 18d7f362402085238faaced164c0ca34079717a477001fc0b13448b3529ea2ad705793a13b7a36f34bf12e9231fee11070f88cc51bfc2a83ca82aa13f7aaae71
37dba329c6cb0f7a4228a11dc26aa3a342a3a5d0 Remove unnecessary sign variable from wnaf_const (Jonas Nick)
6bb0b77e158fc2f9e56e4b65b08bcb660d4c588b Fix test_constant_wnaf for -1 and add a test for it. (Jonas Nick)
Pull request description:
ACKs for top commit:
jonasnick:
ACK 37dba329c6cb0f7a4228a11dc26aa3a342a3a5d0
Tree-SHA512: 4a529579f04dfa8d5abded16cbc0ad2747ccdbef41501c984c348ffd154afdd28333d739db60c869410211187850e3c05dfe91dfb184ad8dca8d5c59b3a158ed
412bf874d09517b559eba4f7addb4c181cc2780b configure: Allow specifying --with[out]-valgrind explicitly (Luke Dashjr)
Pull request description:
ACKs for top commit:
sipa:
ACK 412bf874d09517b559eba4f7addb4c181cc2780b. Tested by running configure on a system with and without valgrind, and with no argument, with `--with-valgrind`, and with `--without-valgrind`.
real-or-random:
ACK 412bf874d09517b559eba4f7addb4c181cc2780b
jonasnick:
ACK 412bf874d09517b559eba4f7addb4c181cc2780b
Tree-SHA512: 92417609751e5af813faff1661055cd37f3d00dbcf109a8f14f8ba59d9f3d620c9c6b67d2b1629b6ab75e2afcd47d2b3898a0427931567fb505bc92fa5ee3532
This enables testing overflow is correctly encoded in the recid, and
likely triggers more edge cases.
Also introduce a Sage script to generate the parameters.
34debf7a6d36bbd9a52e68e079ddfc446faf5bef Modify .travis.yml to explictly pass no in env vars instead of setting to nothing (Elichai Turkel)
ef37761feed0172baa03dd94c842f1547bdf3016 Change travis.sh to check if variables are equal to yes instead of not-empty. Before this, setting `VALGRIND=wat` was considered as true, and to make it evaluate as false you had to unset the variable `VALGRIND=` but not it checks if `VALGRIND=yes` and if it's not `yes` then it's evaluated to false (Elichai Turkel)
Pull request description:
ACKs for top commit:
real-or-random:
ACK 34debf7a6d36bbd9a52e68e079ddfc446faf5bef
jonasnick:
ACK 34debf7a6d36bbd9a52e68e079ddfc446faf5bef
Tree-SHA512: 91becfbc9cb7587ee55b2bceb604ea0aed8860990d63a5f414b11db92180c090ea8bcc048c2fb67a094e892138e3be46f00562bf78b7c3369232457289cde447
5738e8622d8ba02caa984425c23c072a3f14352c tests: Initialize random group elements fully (Tim Ruffing)
Pull request description:
Also fix add a missing comment.
ACKs for top commit:
sipa:
utACK 5738e8622d8ba02caa984425c23c072a3f14352c
Tree-SHA512: c7723e225434e7044379f307b2977a3a5251080793bd87b377a2bbf1d18b39ca05f6fb3b427acec32c3b34f4de678fe7087a2dcca4b5f03ec1fc680a88d82b9a
