Buffer overflow fixes for radiotap. #322
Conversation
current_flags_ starts from 0.
Reasons stated in #320 (comment).
|
Sorry it took me a while to merge this. Thanks for the PR! |
|
I just tried to build v4.2 in a x86 QEMU machine and the test for radiotap failed. It was the same failure in the same environment that let me notice the bug and prepare the PR. Actually, I'm quite surprised to see this because all the tests had passed before I submitted this PR months ago. I think I must have made some mistakes when testing the patch, but anyway, it's clear that the bug hasn't been completely fixed yet. I'm still investigating this and will try to prepare another patch to fix the failure I'm experiencing now. Update: It seems to be a different issue. I just found out that though radiotap tests are failing, they are different test cases than the original issue. |
|
@mfontanini It turns out that the failures are caused by float precision problems. I also fixed some other bugs in the Radiotap Ritter class, and the PR is ready: #333 |
Previous discussions: #320