Re: [PATCH 0/8] latched RB-trees and __module_address()

[Peter Zijlstra]

On Wed, Mar 18, 2015 at 07:21:14AM -0700, Christoph Hellwig wrote:
> On Wed, Mar 18, 2015 at 02:36:26PM +0100, Peter Zijlstra wrote:
> > This series is aimed at making __module_address() go fast(er).
> 
> What users do hit this so hard that it matters?  Seems like the jump
> label code uses it directly and throgh __module_text_address, and krobes
> use it throug __module_text_address.  Is it the pageattr code or
> lockdep?

>From 7/8:

"One of the users of this is kernel_text_address() which is employed
in many stack unwinders; which in turn are used by perf-callchain and
ftrace (possibly from NMI context).

So by optimizing __module_address() we optimize many stack unwinders
which are used by both perf and tracing in performance sensitive code."

> Also seems interesting that both __module_address and __module_text_address
> are exported, but don't seem to have modular users.

Yeah and a few others too iirc.
--
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/