public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Alexander Monakov <amonakov@ispras.ru>
To: Sandra Loosemore <sandra@codesourcery.com>
Cc: gcc-patches@gcc.gnu.org, Nathan Sidwell <nathan@acm.org>
Subject: Re: [PATCH 3/8] nvptx -muniform-simt
Date: Wed, 22 Jun 2016 18:39:00 -0000	[thread overview]
Message-ID: <alpine.LNX.2.20.1606222138030.6439@monopod.intra.ispras.ru> (raw)
In-Reply-To: <alpine.LNX.2.20.1606131910560.28650@monopod.intra.ispras.ru>

Ping.

On Mon, 13 Jun 2016, Alexander Monakov wrote:
> On Sun, 12 Jun 2016, Sandra Loosemore wrote:
> > On 06/09/2016 10:53 AM, Alexander Monakov wrote:
> > > +@item -muniform-simt
> > > +@opindex muniform-simt
> > > +Generate code that allows to keep all lanes in each warp active, even when
> > 
> > Allows *what* to keep?  E.g. what is doing the keeping here?  If it is the
> > generated code itself, please rephrase as
> > 
> > Generate code that keeps....
> 
> Let me try to expand and rephrase what I meant:
> 
> Allows the compiler to emit code that, at run time, may have all lanes active,
> particularly in those regions of the program where observable effects from
> execution must happen as if one lane is active (outside of SIMD loops).
> 
> But nevertheless generated code can run just like conventionally generated
> code does: with each lane being active/inactive independently, and side
> effects happening from each active lane (inside of SIMD loops).
> 
> Whether it actually runs in the former (let's call it "uniform") or the latter
> ("conventional") way is switchable at run time. The compiler itself is
> responsible for emitting mode changes at SIMD region boundaries.
> 
> Does this help? Below I went with your suggestion, but changed "keeps" to "may
> keep" because that's generally true only outside of SIMD regions.
> 
> > > +observable effects from execution should appear as if only one lane was
> > 
> > s/was/is/
> > 
> > > +active. This is achieved by instrumenting syscalls and atomic instructions
> > > in
> > > +a lightweight way that allows to switch behavior at runtime. This code
> > 
> > Same issue here....  allows *what* to switch behavior?  (And how would you
> > select which run-time behavior you want?)
> 
> Sorry. This gives compiler itself a way to emit code that will switch behavior
> of the subsequently running code.
> 
> > Also, in the snippet above where it is used as a noun, please
> > s/runtime/run time/
> 
> Thanks. Does the following look better?
> 
> @item -muniform-simt
> @opindex muniform-simt
> Generate code that may keep all lanes in each warp active, even when
> observable effects from execution must appear as if only one lane is active.
> This is achieved by instrumenting syscalls and atomic instructions in a
> lightweight way, allowing the compiler to emit code that can switch at run
> time between this and conventional execution modes. This code generation
> variant is used for OpenMP offloading, but the option is exposed on its own
> for the purpose of testing the compiler; to generate code suitable for linking
> into programs using OpenMP offloading, use option @option{-mgomp}.
> 
> Alexander

  reply	other threads:[~2016-06-22 18:39 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-06-09 16:54 [PATCH 0/8] NVPTX offloading to NVPTX: backend patches Alexander Monakov
2016-06-09 16:54 ` [PATCH 7/8] nvptx: new insns for OpenMP SIMD-on-SIMT Alexander Monakov
2016-06-09 16:54 ` [PATCH 5/8] nvptx mkoffload: pass -mgomp for OpenMP offloading Alexander Monakov
2016-06-09 16:54 ` [PATCH 1/8] nvptx -msoft-stack Alexander Monakov
2016-06-09 16:54 ` [PATCH 4/8] nvptx -mgomp Alexander Monakov
2016-06-13  3:57   ` Sandra Loosemore
2016-06-09 16:54 ` [PATCH 6/8] new target hook: TARGET_SIMT_VF Alexander Monakov
2016-06-09 16:54 ` [PATCH 3/8] nvptx -muniform-simt Alexander Monakov
2016-06-13  3:55   ` Sandra Loosemore
2016-06-13 17:37     ` Alexander Monakov
2016-06-22 18:39       ` Alexander Monakov [this message]
2016-06-24 20:41         ` Sandra Loosemore
2016-06-09 16:54 ` [PATCH 2/8] nvptx: implement predicated instructions Alexander Monakov
2016-06-09 16:54 ` [PATCH 8/8] nvptx: handle OpenMP "omp target entrypoint" Alexander Monakov
2016-06-09 17:00 ` [PATCH 0/8] NVPTX offloading to NVPTX: backend patches Jakub Jelinek
2016-10-14 16:40 Alexander Monakov
2016-10-14 16:40 ` [PATCH 3/8] nvptx -muniform-simt Alexander Monakov

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=alpine.LNX.2.20.1606222138030.6439@monopod.intra.ispras.ru \
    --to=amonakov@ispras.ru \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=nathan@acm.org \
    --cc=sandra@codesourcery.com \
    /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).