Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Nov 29, 2016

• update submodules to public repos

  matthewfl

  committed
  c0738be

• add a real README

  matthewfl

  committed
  a6a04da

Commits on Sep 4, 2016

• some changes to timer to try and get it such that performance actually improves

  matthewfl

  committed
  ad2b7b0

Commits on Aug 31, 2016

• minor bisect change to try running some commands on ipython

  matthewfl

  committed
  239df8a

• Merge branch 'merge-ret-issue'

  matthewfl

  committed
  d08d6fa

• handful of minor fixes

  matthewfl

  committed
  f912efd

Commits on Aug 28, 2016

• appears that it might be working

  matthewfl

  committed
  6b1dd19

• taking useful changes from wait-till-ret branch

  matthewfl

  committed
  baffd83

Commits on Aug 27, 2016

• need to track the delta changes in the stack height when performing the resumes

  matthewfl

  committed
  a62fcc0

Commits on Aug 26, 2016

• fix issue with it deleting the call when it was not running the tracer

  matthewfl

  committed
  6ee2f04

• check to see if resuming another trace inappropriate

  matthewfl

  committed
  c11a7aa

Commits on Aug 24, 2016

• add vpbroadcast instruction to udis86, hope it is correct

  matthewfl

  committed
  c18153f

• fix issue with the merging happening at the wrong stack frame by checking the rsp pointer

  Show description for 64e70c8
  matthewfl

  committed
  64e70c8

Commits on Aug 6, 2016

• using a map to track the merge rip instead of embedding it in the assembly

  matthewfl

  committed
  4c25ed5

Commits on Aug 5, 2016

• going to start checking the rip on merge blocks, thinking that this should be traced in an external strcture instead of through embedding the address in assembly code

  matthewfl

  committed
  87dc9cb

Commits on Aug 4, 2016

• need to refactor the method and merge stacks to use custom types since it would make more since to have them contain more information

  matthewfl

  committed
  6baafda

Commits on Aug 3, 2016

• trying to get it to merge on ret from method

  matthewfl

  committed
  e5eadf8

Commits on Aug 2, 2016

• some cleanup and interfaces for limiting what is traced depending on time frames etc

  matthewfl

  committed
  0e5751b

Commits on Aug 1, 2016

• it is working with ipython now

  matthewfl

  committed
  10aa760

• resume tracers from end branchable frame

  matthewfl

  committed
  a7967a8

• branchable frame to prevent traces from escaping the function they started in

  Show description for 0efe508
  matthewfl

  committed
  0efe508

Commits on Jul 30, 2016

• some issue with the forward branch inlining, not able to pin down the problem at the moment but it happens when running ipython and disabling it gets past at least this issue

  matthewfl

  committed
  aebfaf6

Commits on Jul 29, 2016

• cleaned up some management interface stuff

  matthewfl

  committed
  5fbfcb5

• need to refactor the abort system since it is causing too many edge cases

  matthewfl

  committed
  c0341d8

Commits on Jul 28, 2016

• fix bug with sub traces being recreated instead of branched to

  matthewfl

  committed
  cc80611

Commits on Jul 27, 2016

• merge block code

  matthewfl

  committed
  073da57

• fix inlined branches from not be included if followed by another branch

  Show description for 255acbf
  matthewfl

  committed
  255acbf

Commits on Jul 26, 2016

• fix encoding (,%r12,8) operand

  matthewfl

  committed
  25281e7

Commits on Jul 24, 2016

• able to handle generating non general register rip instructions, but still have bug with saving float registers

  matthewfl

  committed
  6e676b6

Commits on Jul 23, 2016

• changing code buffs if the current one runs out of space

  matthewfl

  committed
  ae0dcd1

Commits on Jul 22, 2016

• largely working

  matthewfl

  committed
  a56edc5

• not crashing the python repl

  matthewfl

  committed
  3e78813

Commits on Jul 21, 2016

• inlining small jumps

  matthewfl

  committed
  fc5985f

• add disabling branches, used for yield since their control flow is screwy

  matthewfl

  committed
  f3d7620

• should handle tail calls using jumps and patching address in for branches in traces

  matthewfl

  committed
  fb9276a

Pagination

PreviousNext