Jason Hiser (ba0e8cde) at 09 Feb 18:01
Make hangman easier to use up all your turns so afl can very very e...
GitLab Runner (389f5052) at 08 Feb 15:36
remove more cmp-or-move problems
GitLab Runner (b1bc61a1) at 07 Feb 22:02
disable recursive cmp parsing in FII with comments about why.
Jason Hiser (c62d1b72) at 13 Nov 18:48
Typo fixes and clarifications
Jason Hiser (cd859152) at 13 Nov 18:45
cross-arch Zipr stuff.
Jason Hiser (05856182) at 23 Oct 14:52
Jason Hiser (fad142b1) at 23 Oct 14:52
Merge branch 'debugging-output-and-jumptable-detection' into 'master'
... and 36 more commits
This merge request got rather convoluted and I'm aware that this is not best practice. If you want to, I can split it up into multiple smaller ones.
The main part of this merge request is a logic change in the backward search that allows accounting for register renames from bounds check until use as index register. This tackles some of the jump table constructs previously documented as "not-handled"
f8a662f7 increases the maximum size used for clamping jump tables and 00917f1f adds the env variable MAX_JUMP_TABLE_CLAMP_SIZE
to adjust this parameter as needed.
Commits 5f5d3169, d434b06b, 9e3f3356, 995ed420, 01262f75, 5cf6ffa9 and 1e2bddd9 include general fixes. If you want to, I can move them into a different merge request
c0e08b6c implements the functions added to IRDB-SDK in opensrc/irdb-sdk!4
Understood. I'm also ready to merge that, so will merge at the same time.
I'm ready to merge this. It does however require opensrc/irdb-sdk!4 as it includes an addition to the API. Just noting this down here again before it gets lost
This has passed testing and my code review, but is still marked as a draft. Are you (@leon.weiss) ready to merge?
Deep testing passed!
I've missed a change that I had done before. This function did not have all correct cases listed
Leon Weiss (05856182) at 23 Oct 10:50
Add missing cases remove extra registers
Deep regression testing started.
I know this is going to break some stuff, but I do want to keep it. I'll have to test to see how much it actually breaks. No action needed, just want to document what's going on with this.
maybe need to change type here to int32_t to match -1 change above.
uint32_t = -1
again. maybe just "auto" here? missed this in the first review!
I don't think it's important.