Skip to content
This repository has been archived by the owner on Sep 20, 2023. It is now read-only.

cryptonite-0.25 test suite fails with "too many pending signals" #260

Closed
peti opened this issue Dec 9, 2018 · 7 comments
Closed

cryptonite-0.25 test suite fails with "too many pending signals" #260

peti opened this issue Dec 9, 2018 · 7 comments

Comments

@peti
Copy link

peti commented Dec 9, 2018

Citing from https://hydra.nixos.org/build/85709602:

Running 1 test suites...
Test suite test-cryptonite: RUNNING...
cryptonite
  number
    num-bits:                            OK
      +++ OK, passed 100 tests.
    num-bits2:                           OK
      +++ OK, passed 100 tests.
    generate-param:                      OK (0.01s)
      +++ OK, passed 100 tests.
    generate-param2:                     OK
      +++ OK, passed 100 tests.
    generate-param-odd:                  OK
      +++ OK, passed 100 tests.
    generate-range:                      OK
      +++ OK, passed 100 tests.
    generate-prime:                      OK (0.67s)
      +++ OK, passed 100 tests.
    generate-safe-prime:                 OK (10.58s)
      +++ OK, passed 100 tests.
    marshalling:                         OK
      +++ OK, passed 100 tests.
    marshalling-kat-to-bytearray
      0:                                 OK
      1:                                 OK
      2:                                 OK
      3:                                 OK
      4:                                 OK
    marshalling-kat-to-integer
      0:                                 OK
      1:                                 OK
      2:                                 OK
      3:                                 OK
      4:                                 OK
  number.F2m
    addF2m
      commutative:                       OK
        +++ OK, passed 100 tests.
      associative:                       OK
        +++ OK, passed 100 tests.
      0 is neutral:                      OK
        +++ OK, passed 100 tests.
      nullable:                          OK
        +++ OK, passed 100 tests.
      works per bit:                     OK
        +++ OK, passed 100 tests.
    modF2m
      idempotent:                        OK
        +++ OK, passed 100 tests.
      upper bound:                       OK
        +++ OK, passed 100 tests.
      reach upper:                       OK
        +++ OK, passed 100 tests.
      lower bound:                       OK
        +++ OK, passed 100 tests.
      reach lower:                       OK
        +++ OK, passed 100 tests.
      additive:                          OK
        +++ OK, passed 100 tests.
    mulF2m
      commutative:                       OK
        +++ OK, passed 100 tests.
      associative:                       OK
        +++ OK, passed 100 tests.
      1 is neutral:                      OK
        +++ OK, passed 100 tests.
      0 is annihilator:                  OK
        +++ OK, passed 100 tests.
      distributive:                      OK
        +++ OK, passed 100 tests.
    squareF2m
      sqr(a) == a * a:                   OK
        +++ OK, passed 100 tests.
    invF2m
      1 / a * a == 1:                    OK
        +++ OK, passed 100 tests.
      1 / a == a (mod a^2-1):            OK
        +++ OK, passed 100 tests.
    divF2m
      1 / a == inv a:                    OK
        +++ OK, passed 100 tests.
      a / b == a * inv b:                OK
        +++ OK, passed 100 tests.
      a * b / b == a:                    OK
        +++ OK, passed 100 tests.
  hash
    KATs
      MD2
        1:                               OK
        2:                               OK
        3:                               OK
      MD4
        1:                               OK
        2:                               OK
        3:                               OK
      MD5
        1:                               OK
        2:                               OK
        3:                               OK
      SHA1
        1:                               OK
        2:                               OK
        3:                               OK
      SHA224
        1:                               OK
        2:                               OK
        3:                               OK
      SHA256
        1:                               OK
        2:                               OK
        3:                               OK
      SHA384
        1:                               OK
        2:                               OK
        3:                               OK
      SHA512
        1:                               OK
        2:                               OK
        3:                               OK
      SHA512/224
        1:                               OK
        2:                               OK
        3:                               OK
      SHA512/256
        1:                               OK
        2:                               OK
        3:                               OK
      RIPEMD160
        1:                               OK
        2:                               OK
        3:                               OK
      Tiger
        1:                               OK
        2:                               OK
        3:                               OK
      Skein256-256
        1:                               OK
        2:                               OK
        3:                               OK
      Skein512-384
        1:                               OK
        2:                               OK
        3:                               OK
      Skein512-512
        1:                               OK
        2:                               OK
        3:                               OK
      Whirlpool
        1:                               OK
        2:                               OK
        3:                               OK
      Keccak-224
        1:                               OK
        2:                               OK
        3:                               OK
      Keccak-256
        1:                               OK
        2:                               OK
        3:                               OK
      Keccak-384
        1:                               OK
        2:                               OK
        3:                               OK
      Keccak-512
        1:                               OK
        2:                               OK
        3:                               OK
      SHA3-224
        1:                               OK
        2:                               OK
        3:                               OK
      SHA3-256
        1:                               OK
        2:                               OK
        3:                               OK
      SHA3-384
        1:                               OK
        2:                               OK
        3:                               OK
      SHA3-512
        1:                               OK
        2:                               OK
        3:                               OK
      Blake2b-160
        1:                               test-cryptonite: too many pending signals
Test suite test-cryptonite: FAIL
@crockeea
Copy link

crockeea commented Feb 6, 2019

Tests pass for me on Windows 10 and Ubuntu 18.04.

@ahibal
Copy link

ahibal commented Feb 15, 2019

I have EXACTLY the same output when I run:
nix-env -f https://github.com/obsidiansystems/obelisk/archive/master.tar.gz -iA command

system:
debian 9
Kernel Linux 4.9.0-8-amd64 x86_64
2.9 GiB memory
AMD Athlon(tm) 64 X2 Dual Core Processor 6000+ × 2

"too many pending signals" is it because I have low memory?

@ryantrinkle
Copy link

I'm seeing these tests take a very long time (more than an hour), but I'm not sure whether that's related.

@ocheron
Copy link
Contributor

ocheron commented Apr 17, 2019

Can those having BLAKE2 failure try removing flag support_aesni?
I suspect SSSE3 is not supported on old AMD processors.

@pvgoran
Copy link

pvgoran commented Mar 3, 2020

I have this exact problem with the current master. When I rebuild and test cryptonite with --constraint="cryptonite -support_aesni", the problem goes away.

@pvgoran
Copy link

pvgoran commented Mar 3, 2020

I re-posted this as #314, with more detailed description of the problem.

@ocheron
Copy link
Contributor

ocheron commented Aug 11, 2020

Fixed in cryptonite-0.27 and now in Stackage Nightly.
There should not be error at runtime anymore with default flags, the cost is C compiler support being made more difficult for old versions.

@ocheron ocheron closed this as completed Aug 11, 2020
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

No branches or pull requests

6 participants