This is the mail archive of the
libc-ports@sources.redhat.com
mailing list for the libc-ports project.
Re: [PATCH] ARM: Add pointer guard support.
- From: Will Newton <will dot newton at linaro dot org>
- To: Andrew Haley <aph at redhat dot com>
- Cc: "Carlos O'Donell" <carlos at redhat dot com>, "libc-ports at sourceware dot org" <libc-ports at sourceware dot org>, Patch Tracking <patches at linaro dot org>
- Date: Wed, 25 Sep 2013 20:32:43 +0100
- Subject: Re: [PATCH] ARM: Add pointer guard support.
- Authentication-results: sourceware.org; auth=none
- References: <5242A79D dot 1030709 at linaro dot org> <52430AA4 dot 70703 at redhat dot com> <CANu=DmhCMr-LpHLaDRsLOHDVXsWm-sxHzRTMJxNCJs3Ae0uPZg at mail dot gmail dot com> <5243165F dot 804 at redhat dot com>
On 25 September 2013 17:59, Andrew Haley <aph@redhat.com> wrote:
> On 09/25/2013 05:23 PM, Will Newton wrote:
>> From a back of the envelope calculation the cost of accessing TLS is
>> one cycle faster than accessing a global in best case (e.g.
>> Cortex-A15)
>
> Why is TLS so fast on Cortex-A15?
I don't have exact numbers but the mrc instruction used to load the
thread pointer into a general purpose register has very variable
timings depending on which core you have. A15 is the fastest of the
Cortex-A cores in this regard.
--
Will Newton
Toolchain Working Group, Linaro