linux-kernel - Re: [RFC][PATCH] tracepoints: Move the work out of line from hotpath sections

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Thu, 9 Aug 2012 22:05:23 -0400
From:	Mathieu Desnoyers <mathieu.desnoyers@...icios.com>
To:	Steven Rostedt <rostedt@...dmis.org>
Cc:	David Daney <ddaney.cavm@...il.com>,
	"H. Peter Anvin" <hpa@...ux.intel.com>,
	LKML <linux-kernel@...r.kernel.org>,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	Ingo Molnar <mingo@...e.hu>,
	Thomas Gleixner <tglx@...utronix.de>,
	Peter Zijlstra <peterz@...radead.org>,
	Masami Hiramatsu <masami.hiramatsu.pt@...achi.com>,
	Avi Kivity <avi@...hat.com>,
	Christoph Hellwig <hch@...radead.org>,
	Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [RFC][PATCH] tracepoints: Move the work out of line from
	hotpath sections

* Steven Rostedt (rostedt@...dmis.org) wrote:
> On Thu, 2012-08-09 at 16:54 -0700, David Daney wrote:
> > On 08/09/2012 04:16 PM, H. Peter Anvin wrote:
> > > On 08/09/2012 03:25 PM, Steven Rostedt wrote:
> > >>>
> > >>> It might be better to improve gcc to move really cold branches out of
> > >>> line (really, really far away), and use the compiler to do this, rather
> > >>> than to use an extra indirection that adds bloat and complexity to the
> > >>> kernel.
> > 
> > Oh, you mean like: -freorder-blocks-and-partition
> 
> Actually, what would be really nice is to place a block in a section of
> your choice. Something like:
> 
> 
> 	if (unlikely(x)) __attribute__((section(".unlikely"))) {
> 		/* code here will be in the ".unlikely" section */
> 	}

In your example, is the attribute attached to the if() or the following
basic block ? Attaching it to the basic block allows a nice level of
genericity:

if (unlikely(x)) __attribute__((section(".unlikely"))) {
        ...
} else __attribute__((section(".likely"))) {
        ...
}

or

switch (x) {
        case 0:
        case 1:
        case 6:
        __attribute__((section(".likely"))) {
                ...
                break;
        }
        default:
        __attribute__((section(".unlikely"))) {
                ...
                break;
        }
}

Thanks,

Mathieu

> 
> -- Steve
> 
> 

-- 
Mathieu Desnoyers
Operating System Efficiency R&D Consultant
EfficiOS Inc.
http://www.efficios.com
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/