public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: "Yangfei (Felix)" <felix.yang@huawei.com>
To: "augustine.sterling@gmail.com" <augustine.sterling@gmail.com>
Cc: "gcc-patches@gcc.gnu.org" <gcc-patches@gcc.gnu.org>,
	       Felix Yang	<fei.yang0953@gmail.com>
Subject: Re: [PATCH] Add zero-overhead looping for xtensa backend
Date: Fri, 24 Oct 2014 06:43:00 -0000	[thread overview]
Message-ID: <DA41BE1DDCA941489001C7FBD7A8820E5554BB72@szxema507-mbx.china.huawei.com> (raw)
In-Reply-To: <CAGSvup9yzkQCV_TaByLfvFLmqjVhy3+yhjUBDvAG3WJmnMuA4Q@mail.gmail.com>

1. The original xtensa port never generates "loop" instruction at all.
2. A port doesn't need to implement hwloop_pattern_reg hook if it has no zero-cost loop instruction.

Is that clear?

> 
> I mean without your patch at all.
> 
> On Thu, Oct 23, 2014 at 11:30 PM, Yangfei (Felix) <felix.yang@huawei.com>
> wrote:
> >>
> >> On Thu, Oct 23, 2014 at 9:12 PM, Yangfei (Felix) <felix.yang@huawei.com>
> wrote:
> >> >> Here the key point is we need a general purpose register for the "loop"
> >> >> instruction.
> >>
> >> So the question to ask here is, "How does this work today, without
> >> loop instructions?" Somehow--even when it has been spilled--a branch
> >> instruction can test the trip count. There should be no difference.
> >>
> >> >> And we cannot use zero-cost looping in this situation.
> >> >> And that's why I spilt the zero_cost_loop_end into a normal test and
> branch.
> >>
> >> > Also note that the hwloop_pattern_reg interface also expects a
> >> > general
> >> purpose register in the doloop_end pattern.
> >>
> >> If there were no loop instruction, how would this work?
> >
> >
> > Just take a look at my patch. I handle this in the new define_split:
> >
> > +(define_split
> > +  [(set (pc)
> > +        (if_then_else (ne (match_operand:SI 0 "nonimmediate_operand"
> "")
> > +                          (const_int 1))
> > +                      (label_ref (match_operand 1 "" ""))
> > +                      (pc)))
> > +   (set (match_operand:SI 2 "nonimmediate_operand" "")
> > +        (plus:SI (match_dup 0)
> > +                 (const_int -1)))
> > +   (unspec [(const_int 0)] UNSPEC_LSETUP_END)
> > +   (clobber (match_scratch 3))]
> > +  "TARGET_LOOPS && optimize && reload_completed"
> > +  [(const_int 0)]
> > +{
> > +  if (!REG_P (operands[0]))
> > +    {
> > +      rtx test;
> > +
> > +      /* Fallback into a normal conditional branch insn.  */
> > +      emit_move_insn (operands[3], operands[0]);
> > +      emit_insn (gen_addsi3 (operands[3], operands[3], constm1_rtx));
> > +      emit_move_insn (operands[0], operands[3]);
> > +      test = gen_rtx_NE (VOIDmode, operands[3], const0_rtx);
> > +      emit_jump_insn (gen_cbranchsi4 (test, operands[3],
> > +                                      const0_rtx, operands[1]));
> > +    }
> > +  else
> > +    {
> > +      emit_jump_insn (gen_loop_end (operands[0], operands[1],
> operands[2]));
> > +    }
> > +
> > +  DONE;
> > +})

  reply	other threads:[~2014-10-24  6:40 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-01-08 16:27 Felix Yang
2014-01-08 16:49 ` Sterling Augustine
2014-01-09 15:08   ` Felix Yang
2014-01-09 23:51     ` Felix Yang
2014-01-10  3:49       ` Yangfei (Felix)
2014-01-13 17:24         ` Sterling Augustine
2014-10-09 11:04           ` Felix Yang
2014-10-10 14:01             ` Felix Yang
2014-10-11  9:32               ` [PING] [PATCH, xtensa] " Yangfei (Felix)
2014-10-13 16:09               ` [PATCH] " augustine.sterling
2014-10-13 16:30                 ` Felix Yang
2014-10-14 15:43                   ` Felix Yang
2014-10-15 19:51                     ` augustine.sterling
2014-10-16  4:52                       ` Yangfei (Felix)
2014-10-21 14:57                         ` augustine.sterling
2014-10-22  2:20                           ` Yangfei (Felix)
2014-10-22  5:20                           ` Yangfei (Felix)
2014-10-23 17:51                             ` augustine.sterling
2014-10-24  1:49                               ` Yangfei (Felix)
2014-10-24  4:32                               ` Yangfei (Felix)
2014-10-24  6:28                                 ` augustine.sterling
2014-10-24  6:33                                   ` Yangfei (Felix)
2014-10-24  6:40                                     ` augustine.sterling
2014-10-24  6:43                                       ` Yangfei (Felix) [this message]
2014-10-24  6:49                                         ` augustine.sterling
2014-10-24  6:53                                           ` Yangfei (Felix)
2014-10-24  7:15                                             ` Andrew Pinski
2014-10-28 12:24                                               ` Yangfei (Felix)
2014-10-30 22:02                                                 ` augustine.sterling

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=DA41BE1DDCA941489001C7FBD7A8820E5554BB72@szxema507-mbx.china.huawei.com \
    --to=felix.yang@huawei.com \
    --cc=augustine.sterling@gmail.com \
    --cc=fei.yang0953@gmail.com \
    --cc=gcc-patches@gcc.gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).