OpenCL autotune for actual salt and/or hash count
Nobody has claimed this yet.
- Dominant language
- C
- Stars
- 13.6k
- Forks
- 2.6k
- PR merge metrics
- No merged PRs in 30d
Description
As discussed in comments on #3779, we might want to autotune for the actual salt count. This could affect the ratio of calls to set_key, crypt_all, and cmp_all, maybe in some cases resulting in different LWS/GWS being optimal. We should use the salt count remaining for cracking (with john.pot entries removal already processed).
Further, with unsalted hashes we might want to autotune for the actual hash count, perhaps using the actual hashes for the comparisons. This affects whether cmp_all or a get_hash*() (and which one) is used. With formats supporting on-device comparison, this affects performance of such comparison.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Read the comments on #3779 first, then trace the OpenCL paths involving set_key, crypt_all, cmp_all, and get_hash*. Check how john.pot removals expose the remaining salt or hash counts. Done means autotuning accounts for those actual counts when selecting LWS/GWS and the comparison strategy.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- c
- Domain
- performance
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 25/100