mirrors/git-annex
2
0
Fork 0
Code Issues Pull requests Projects Releases Packages 1 Wiki Activity Actions

Added a comment

Browse source
This commit is contained in:
http://id.pvgoran.name/ 2020-03-03 03:16:25 +00:00 • committed by admin
parent d9b890ba40
commit 2a5b640ade
1 changed files with 12 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

12
doc/bugs/__34__Illegal_instruction__34___when_running_certain_tests_on_AMD_Phenom_II/comment_3_28ede555e6a372ca0278148016cae4b0._comment Normal file
View file

@ -0,0 +1,12 @@
[[!comment format=mdwn
username="http://id.pvgoran.name/"
nickname="pvgoran"
avatar="http://cdn.libravatar.org/avatar/e32a61d9c49989ae31d7a30d6af27f5c73d8d46ba03c840a612791fe5c820b87"
subject="comment 3"
date="2020-03-03T03:16:25Z"
content="""
After rebuilding `cryptonite` without AES-NI support as suggested in https://github.com/haskell-crypto/cryptonite/issues/260#issuecomment-484185981 (by passing `--constraint=\"cryptonite -support_aesni\"` to `cabal install --dependencies-only` and to `cabal build`), all of QuickCheck tests pass on the machine in question. (Many other tests fail, but it's probably because I run tests from a not-yet-installed `git-annex` binary.)
So `cryptonite` apparently uses these AES-NI instructions (if configured to support them) without runtime checks (or at least without working runtime checks). Which constitutes a problem for any binary distributions that include `cryptonite` code: either they won't work on older CPUs, or they will not work as fast as they could on newer CPUs.
"""]]