Re: [RFC, ARM] later split of symbol_refs

[Ramana Radhakrishnan <ramana dot radhakrishnan at linaro dot org>]

On 27 June 2012 15:58, Dmitry Melnik <dm@ispras.ru> wrote:
> Hi,
>
> We'd like to note about CodeSourcery's patch for ARM backend, from which GCC
> mainline can gain 4% on SPEC2K INT:
> http://cgit.openembedded.org/openembedded/plain/recipes/gcc/gcc-4.5/linaro/gcc-4.5-linaro-r99369.patch
> (also the patch is attached).
>
> Originally, we noticed that GNU Go works 6% faster on cortex-a8 with
> -fno-gcse. ?After profiling we found that this is most likely caused by
> cache misses when accessing global variables. ?GCC generates ldr
> instructions for them, while this can be avoided by emitting movt/movw pair
> for such cases. RTL expressions for these instructions is high_ and lo_sum.
> ?Currently, symbol_ref expands as high_ and lo_sum but then cprop1 decides
> that this is redundant and merges them into one load insn.
>
> The problem was also found by Linaro community:
> https://bugs.launchpad.net/gcc-linaro/+bug/886124 .

The reason IIRC this isn't in our later releases is that it wasn't
thought beneficial enough to upstream. Now you've got some evidence to
the contrary.

> Also there is a patch from codesourcery (attached), which was ported to
> linaro gcc 4.5, but is missing in later linaro releases.
> This patch makes split of symbol_refs at the later stage (after cprop),
> instead of generating movt/movw at expand.

I must admit that I had been suggesting to Zhenqiang about turning
this off by tightening the movsi_insn predicates rather than adding a
split, but given that it appears to produce enough benefit in this
case I don't have any reasons to object ...

However it's interesting that this doesn't seem to help vpr ....


Ramana