[nvptx] Re: Mostly rewrite genrecog

[Thomas Schwinge]

Hi!

On Mon, 27 Apr 2015 11:20:30 +0100, Richard Sandiford 
<richard.sandif...@arm.com> wrote:
> I think it's been the case for a while that parallel builds of GCC tend
> to serialise around the compilation of insn-recog.c, especially with
> higher --enable-checking settings.  This patch tries to speed that
> up by replacing most of genrecog with a new algorithm.  [...]

> Here's a comparison of the number of lines of code in insn-recog.c
> before and after the patch on one target per config/ CPU:
> 
> aarch64-linux-gnueabi                           115526    38169 :   33.04%
> alpha-linux-gnu                                  24479    10740 :   43.87%
> arm-linux-gnueabi                               169208    67759 :   40.04%
> avr-rtems                                        55647    22127 :   39.76%
> bfin-elf                                         13928     6498 :   46.65%
> c6x-elf                                          29928    13324 :   44.52%
> [...]

Nice work!  For a nvptx-none build of r222446 and r222860, respectively:

    $ wc -l {prev/,}build-gcc/gcc/insn-recog.c | head -n -1
      5042 prev/build-gcc/gcc/insn-recog.c
      2570 build-gcc/gcc/insn-recog.c


I'm mostly illiterate regarding GCC's machine description files, and
genrecog, but noticed one thing: with the genrecog rewrite, I get the
following diff compared to a previous build log:

     build/genrecog [...]/source-gcc/gcc/common.md 
[...]/source-gcc/gcc/config/nvptx/nvptx.md \
               insn-conditions.md > tmp-recog.c
    -[...]/source-gcc/gcc/config/nvptx/nvptx.md:1206: warning: operand 0 
missing mode?
    -[...]/source-gcc/gcc/config/nvptx/nvptx.md:1206: warning: operand 1 
missing mode?
    +Statistics for recog:
    +  Number of decisions:    799
    +  longest path:            28 (code:    208)
    +  longest backtrack:        2 (code:    136)
    +Statistics for split_insns:
    +  Number of decisions:      0
    +  longest path:             0 (code:     -1)
    +  longest backtrack:        0 (code:     -1)
    +Statistics for peephole2_insns:
    +  Number of decisions:      0
    +  longest path:             0 (code:     -1)
    +  longest backtrack:        0 (code:     -1)
    +Shared 655 out of 1350 states by creating 153 new states, saving 502

gcc/config/nvptx/nvptx.md:

    1206 (define_insn "allocate_stack"
    1207   [(set (match_operand 0 "nvptx_register_operand" "=R")
    1208         (unspec [(match_operand 1 "nvptx_register_operand" "R")]
    1209                   UNSPEC_ALLOCA))]
    1210   ""
    1211   "%.\\tcall (%0), %%alloca, (%1);")

Are these two (former) warnings a) something that should still be
reported by genrecog, and b) something that should be addressed (Bernd)?


Grüße,
 Thomas

signature.asc
Description: PGP signature