Re: [RFC 0/6] uprobes: kill uprobes_srcu/uprobe_srcu_id

From: Peter Zijlstra
Date: Mon Apr 16 2012 - 06:17:35 EST

On Mon, 2012-04-16 at 01:44 +0200, Oleg Nesterov wrote:

> No. Please note that if is_swbp_at_addr_fast() sets is_swbp == 0 we
> restart this insn.

Ah, see I was missing something.. Hmm ok, let me think about this a
little more though.. but at least I think I'm now (finally!) seeing what
you propose.

> And. I have another reason for down_write() in register/unregister.
> I am still not sure this is possible (I had no time to try to
> implement), but it seems to me we can kill the uprobe counter in
> mm_struct.

You mean by making register/unregister down_write, you're exclusive with
munmap() and thus we can rely on is_swbp_at_addr_fast() to inspect the
address to see if there's a breakpoint or not and avoid the rest of the
work that way?
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at