This is the mail archive of the
binutils@sourceware.org
mailing list for the binutils project.
Re: [Patch, AArch64, BFD, LD][2.25] Cortex-A53 erratum 835769 linker workaround.
- From: Tristan Gingold <gingold at adacore dot com>
- To: Tejas Belagod <tejas dot belagod at arm dot com>
- Cc: binutils at sourceware dot org, Marcus Shawcroft <Marcus dot Shawcroft at arm dot com>, Richard Earnshaw <Richard dot Earnshaw at arm dot com>
- Date: Fri, 24 Oct 2014 12:46:33 +0200
- Subject: Re: [Patch, AArch64, BFD, LD][2.25] Cortex-A53 erratum 835769 linker workaround.
- Authentication-results: sourceware.org; auth=none
- References: <544A2C95 dot 3040901 at arm dot com>
> On 24 Oct 2014, at 12:40, Tejas Belagod <tejas.belagod@arm.com> wrote:
>
> Hi,
>
> Some early revisions of the Cortex-A53 have an erratum (835769) whereby
> it is possible for a 64-bit multiply-accumulate instruction in AArch64 state to
> generate an incorrect result. The details are quite complex and hard to
> determine statically, since branches in the code may exist in some
> circumstances, but all cases end with a memory (load, store, or
> prefetch) instruction followed immediately by the multiply-accumulate
> operation.
>
> We employ a linker patching technique, whereby we move the potentially
> affected multiply-accumulate instruction into a patch region and
> replace the original instruction with a branch to the patch.
>
> This patch achieves the linker patch. Also, a patch to turn this on from the compiler driver is being submitted independently to the gcc lists.
>
> The fix is disabled by default and can be turned on by the
> --fix-cortex-a53-835769 linker option.
>
> Tested on aarch64-none-linux-gnu (and bootstrap with the option) and built and run various large
> benchmarks with it.
>
> Ok for 2.25?
I have missed the initial question, sorry. Yes, ok for backport.