Re: [PATCH] mm: Enable MAP_UNINITIALIZED for archs with mmu

[Balbir Singh]

On Thu, Jan 19, 2012 at 12:21 AM, Arun Sharma <asharma@xxxxxx> wrote:
>
> This enables malloc optimizations where we might
> madvise(..,MADV_DONTNEED) a page only to fault it
> back at a different virtual address.
>
> To ensure that we don't leak sensitive data to
> unprivileged processes, we enable this optimization
> only for pages that are reused within a memory
> cgroup.
>

So the assumption is that only apps that have access to each others
VMA's will run in this cgroup?

> The idea is to make this opt-in both at the mmap()
> level and cgroup level so the default behavior is
> unchanged after the patch.
>

Sorry, I am not convinced we need to do this

1. I know that zeroing out memory is expensive, but building a
potential loop hole is not a good idea
2. How do we ensure that tasks in a cgroup should be allowed to reuse
memory uninitialized, how does the cgroup admin know what she is
getting into?

So I am going to NACK this.

Balbir
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/