Defrag Tools #186 - Time Travel Debugging - Advanced

Defrag Tools - Channel 9

Content provided by Defrag Tools - Channel 9. All podcast content including episodes, graphics, and podcast descriptions are uploaded and provided directly by Defrag Tools - Channel 9 or their podcast platform partner. If you believe someone is using your copyrighted work without your permission, you can follow the process outlined here https://player.fm/legal.

7y ago 27:45

MP4•Episode home

Archived series ("Inactive feed" status)

When? This feed was archived on February 06, 2022 05:29 (2+ y ago). Last successful fetch was on March 12, 2020 12:40 (4+ y ago)

Why? Inactive feed status. Our servers were unable to retrieve a valid podcast feed for a sustained period.

What now? You might be able to find a more up-to-date version using the search function. This series will no longer be checked for updates. If you believe this to be in error, please check if the publisher's feed link below is valid and contact support to request the feed be restored or if you have any other concerns about this.

In this episode of Defrag Tools, Andrew Richards is joined by JCAB (Juan Carlos Arevalo Baeza) and Jordi Mola from the Windows Debugger team to demonstrate some more advanced usage of a new feature of WinDbg Preview: Time Travel Debugging (TTD).

Related Links:

WinDbg Preview (download from Microsoft Store)
Time Travel Debugging Overview (Online documentation)
Debugging Tools for Windows Blog
Time Travel Debugging FAQ

Timeline:
[00:00] Introductions
[01:07] Seeing a memory corruption crash in the Chakra Core when running a script. Difficult to debug!
[05:33] Now reproduce the same crash while recording a Time Travel Debugging trace
[07:06] Looking at the TTD trace with unoptimized code
[07:55] Use the !events command to list interesting events and exceptions in the trace and jump to them
[11:43] Found the corrupt memory, step backwards to figure out where it came from.
[13:15] Identifying the memory location containing a bad value with dx command, and setting a data breakpoint (with ba) to see who previously wrote to it.
[17:37] Getting closer. Keep following the trail backwards...
[19:29] Found where the bad value came from!
[21:08] Another use case: Find where a value is bad and track it back from there with a binary search (use !tt with a percentage value to jump to locations in the trace)
[22:09] Second demo: Looking at the same crash but with optimized production code.
[25:09] Exceptions will be hit when running the trace either forward or backward.
[26:54] To give feedback on WinDbg Preview, use the Feedback Hub.

25 episodes

#Tech #montserrat #Video