Re: [RFC PATCH 3/3] fs: detect that the i_rwsem has already been taken exclusively

From: Eric W. Biederman
Date: Sun Oct 01 2017 - 18:06:41 EST

Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> writes:

> On Sep 30, 2017 18:33, "Eric W. Biederman" <ebiederm@xxxxxxxxxxxx> wrote:.
> That would require a task_work or another kind of work callback so that
> the writes of the xattr are not synchronous with the vfs callback
> correct?
> No, why?
> You should just invalidate the IMA on xattr write or other operations that make the measurement invalid. You only need the inner
> lock.
> Why are you guys making up all these things just to make it complicated?

I am not trying to make things complicated I am just trying to
understand the conversation.

Unless I misread something it was being pointed out there are some vfs
operations today on which ima writes an ima xattr as a side effect. And
those operations hold the i_sem. So perhaps I am misunderstanding
things or writing the ima xattr needs to happen at some point. Which
implies something like queued work.

But perhaps I a misunderstanding the conversation and ima. I frequenly
misunderstand ima.