Re: [PATCH] x86: Rename copy_from_user_nmi() to copy_from_user_trace()

[Peter Zijlstra]

On Fri, Feb 28, 2014 at 04:01:39PM -0500, Steven Rostedt wrote:
> OK, I'll go with that. So instead of an _nmi() which it has nothing to
> do with NMIs, 

Well, how is 'safe to use from NMI' nothing to do with NMIs?

> how about making it like spin_lock_irq() and a more
> descriptive name about what it does?
> 
>  copy_from_user_savecr2()

So how expensive are these CR2 reads? Because if they're more expensive
than one or two cycles (normal instructions) I'd really rather not put
them in there.

I seem to have missed what was wrong with Jiri's patch. Are we worried
about the function trace events?

So IF we're going to push them into this side of things; I'd really
rather see the CR2 foo in perf_callchain_user() and make
arch_perf_out_copy_user() a real function which includes the CR2 goo
too.
--
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/