Hugetlb mem wipe
Using a custom memory wiping program from initramfs is implemented in
the feature/hugetlb_mem_wipe
branch.
It makes the wipe much faster and better looking, but it’s not as efficient as Tails’ current parallel sdmem approach when using a PAE kernel (which usually results in 0 occurrences in my tests, if not it’s just a few hundred occurences), which arguably is what most users will use. For the non-PAE kernel I believe this branch is better, though.
Next thing to do: fine tune the algorithm parameters and/or memory settings to be as efficient on PAE than current implementation.
Test results
Both tests performed in the same 8 GiB-RAM VM:
- With PAE-kernel: 137K occurences =~ 2.1 MiB of unwiped memory.
- With non-PAE-kernel: 155K occurences =~ 2.4 MiB of unwiped memory.
Feature Branch: feature/hugetlb_mem_wipe
Related issues
- Related to #5762 (closed)
- Related to #5456 (closed)
- Blocks #6006 (closed)
- Blocks #5658 (closed)
Original created by @tails on 5342 (Redmine)