Prevent out-of-bound memory access. #124
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Before the change, the scan_delimiters function could return max_delims+1. The scan_delimiters function is called by extensions/strikethrough with: 18: inline_parser, sizeof(buffer)-1, '~' where buffer is: 12: char buffer[101]; So scan_delimiters can return delims = 101. A few lines later, it is used to 0-terminate the buffer with: 23: buffer[delims] = 0; so it effectively does buffer[101] = 0, which is an out-of-bound access. A test was added. That's the test that I used inside Google to reproduce the bug, thanks to internal infrastructure (sanitizers) to detect bad memory access. Unfortunately, I couldnt' reproduce the bug with cmark-gfm build suite. I tried `make asan test` but it gave nothing. I'm still including the test for reference.
|
The bug in this PR was found through fuzzing through the Google Autofuzz project. Autofuzz is an internal project, but a similar service is offered for free to open-source software. This is called OSS-Fuzz and described on the Google Opensource blog. |
This is not really a bug, but it's possible to send an input markdown consisting of lots of @ signs, and the recursion will cause memory explosion. The limit depends on the running environment, but there is no reason to accept arbitrarily long sequence of @, so let's just cut off at 1000.
|
The PR is really one commit. It shows as 3 because I got confused when preparing another PR in the same branch, when I now know it should be in a separate branch. |
|
Thank you! I'll squash the commits into one. |
talum
pushed a commit
that referenced
this pull request
Sep 14, 2021
* Prevent out-of-bound memory access. Before the change, the scan_delimiters function could return max_delims+1. The scan_delimiters function is called by extensions/strikethrough with: 18: inline_parser, sizeof(buffer)-1, '~' where buffer is: 12: char buffer[101]; So scan_delimiters can return delims = 101. A few lines later, it is used to 0-terminate the buffer with: 23: buffer[delims] = 0; so it effectively does buffer[101] = 0, which is an out-of-bound access. A test was added. That's the test that I used inside Google to reproduce the bug, thanks to internal infrastructure (sanitizers) to detect bad memory access. Unfortunately, I couldnt' reproduce the bug with cmark-gfm build suite. I tried `make asan test` but it gave nothing. I'm still including the test for reference. * Limit the recursion in autolink extension. This is not really a bug, but it's possible to send an input markdown consisting of lots of @ signs, and the recursion will cause memory explosion. The limit depends on the running environment, but there is no reason to accept arbitrarily long sequence of @, so let's just cut off at 1000. * Reverted change
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Before the change, the scan_delimiters function could return max_delims+1.
The scan_delimiters function is called by extensions/strikethrough with:
where buffer is:
So scan_delimiters can return delims = 101. A few lines later, it is
used to 0-terminate the buffer with:
so it effectively does buffer[101] = 0, which is an out-of-bound access.
A test was added. That's the test that I used inside Google to reproduce
the bug, thanks to internal infrastructure (sanitizers) to detect bad
memory access. Unfortunately, I couldnt' reproduce the bug with
cmark-gfm build suite. I tried
make asan testbut it gave nothing. I'mstill including the test for reference.