[PATCH] x86_64: Use peephole2 to eliminate redundant moves.

Roger Sayle roger@nextmovesoftware.com
Wed Aug 12 07:55:17 GMT 2020

Hi Uros,

Many thanks for the review, and your help/explanation around constant
materialization on x86_64/i386.

Your suggestion to try x86_64_general_operand as a predicate was awesome!
Not only does this improvement survive "make bootstrap" and "make -k check"
on x86_64-pc-linux-gnu, and fix the previous test case, but this now triggers
15070 times during stage2 and stage3, and an additional 8926 times
during make check.  Many thanks for the impressive improvement.

Here's the version as committed.

2020-08-12  Roger Sayle  <roger@nextmovesoftware.com>
            Uroš Bizjak  <ubizjak@gmail.com>

	* config/i386/i386.md (peephole2): Reduce unnecessary
	register shuffling produced by register allocation.


-----Original Message-----
From: Uros Bizjak <ubizjak@gmail.com> 
Sent: 11 August 2020 11:14
To: Roger Sayle <roger@nextmovesoftware.com>
Cc: GCC Patches <gcc-patches@gcc.gnu.org>
Subject: Re: [PATCH] x86_64: Use peephole2 to eliminate redundant moves.

On Tue, Aug 11, 2020 at 9:34 AM Roger Sayle <roger@nextmovesoftware.com> wrote:
> The recent fix for mul_widen_cost revealed an interesting quirk of 
> ira/reload register allocation on x86_64.  As shown in 
> https://gcc.gnu.org/pipermail/gcc-patches/2020-August/551648.html
> for gcc.target/i386/pr71321.c we generate the following code that 
> performs unnecessary register shuffling.
>         movl    $-51, %edx
>         movl    %edx, %eax
>         mulb    %dil
> which is caused by reload generating the following instructions 
> (notice the set of the first register is dead in the 2nd insn):
> (insn 7 4 36 2 (set (reg:QI 1 dx [94])
>         (const_int -51 [0xffffffffffffffcd])) {*movqi_internal}
>      (expr_list:REG_EQUIV (const_int -51 [0xffffffffffffffcd])
>         (nil)))
> (insn 36 7 8 2 (set (reg:QI 0 ax [93])
>         (reg:QI 1 dx [94])) {*movqi_internal}
>      (expr_list:REG_DEAD (reg:QI 1 dx [94])
>         (nil)))
> Various discussions in bugzilla seem to point to reload preferring not 
> to load constants directly into CLASS_LIKELY_SPILLED_P registers.

This can extend the lifetime of a register over the instruction that needs one of the CLASS_LIKELY_SPILLED_P registers. Various MUL, DIV and even shift insns were able to choke the allocator for x86 targets, so this is a small price to pay to avoid regalloc failure.

> Whatever the cause, one solution (workaround), that doesn't involve 
> rewriting a register allocator, is to use peephole2 to spot this 
> weirdness and eliminate it.  In fact, this use case is (probably) the 
> reason peephole optimizers were originally developed, but it's a 
> little disappointing this application of them is still required today.  
> On a positive note, this clean-up is cheap, as we're already 
> traversing the instruction stream with liveness (REG_DEAD notes) 
> already calculated.
> With this peephole2 the above three instructions (from pr71321.c) are 
> replaced with:
>         movl    $-51, %eax
>         mulb    %dil
> This patch has been tested on x86_64-pc-linux-gnu with "make bootstrap"
> and "make -k check" with no new failures.  This peephole triggers
> 1435 during stage2 and stage3 of a bootstrap, and a further 1274 times 
> during "make check".  The most common case is DX_REG->AX_REG (as 
> above) which occurs 421 times.  I've restricted this pattern to 
> immediate constant loads into general operand registers, which fixes 
> this particular problem, but broader predicates may help similar cases.
> Ok for mainline?
> 2020-08-11  Roger Sayle  <roger@nextmovesoftware.com>
>         * config/i386/i386.md (peephole2): Reduce unnecessary
>         register shuffling produced by register allocation.

LGTM, but I wonder if the allocator is also too conservative with memory operands. Perhaps x86_64_general_operand can be used here.

> Thanks in advance,
> Roger
> --
> Roger Sayle
> NextMove Software
> Cambridge, UK
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: patchr3.txt
URL: <https://gcc.gnu.org/pipermail/gcc-patches/attachments/20200812/b5aa4d91/attachment.txt>

More information about the Gcc-patches mailing list