public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Dimitrije Milosevic <dimitrije.milosevic@syrmia.com>
To: gcc-patches@gcc.gnu.org
Cc: djordje.todorovic@syrmia.com
Subject: [PATCH 0/2] ivopts: Fix candidate selection for architectures with limited addressing modes.
Date: Fri, 21 Oct 2022 15:52:01 +0200	[thread overview]
Message-ID: <20221021135203.626255-1-dimitrije.milosevic@syrmia.com> (raw)

Architectures like Mips are very limited when it comes to addressing modes. Therefore, the expected
behavior would be that, for the BASE + OFFSET addressing mode, complexity is lower, while, for more
complex addressing modes (e.g. BASE + INDEX << SCALE), which are not supported, complexity is
higher. Currently, the complexity calculation algorithm bails out if BASE + INDEX addressing mode
is not supported by the target architecture, resuling in 0-complexities for all candidates, which
leads to non-optimal candidate selection, especially in scenarios where there are multiple nested
loops.

Additionally, when bumping up the register pressure cost, the number of invariants should also be
considered, in addition to the number of candidates.

Dimitrije Milosevic (2):
  ivopts: Revert computation of address cost complexity.
  ivopts: Consider number of invariants when calculating register pressure.

 gcc/tree-ssa-address.cc     |   2 +-
 gcc/tree-ssa-address.h      |   2 +
 gcc/tree-ssa-loop-ivopts.cc | 220 +++++++++++++++++++++++++++++++++++++++++---
 3 files changed, 210 insertions(+), 14 deletions(-)
---
2.25.1



             reply	other threads:[~2022-10-21 13:54 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-10-21 13:52 Dimitrije Milosevic [this message]
2022-10-21 13:52 ` [PATCH 1/2] ivopts: Revert computation of address cost complexity Dimitrije Milosevic
2022-10-25 11:08   ` Richard Biener
2022-10-25 13:00     ` Dimitrije Milosevic
2022-10-27 23:02   ` Jeff Law
2022-10-28  6:43     ` Dimitrije Milosevic
2022-10-28  7:00       ` Richard Biener
2022-10-28 13:39         ` Dimitrije Milosevic
2022-11-01 18:46         ` Jeff Law
2022-11-02  8:40           ` Dimitrije Milosevic
2022-11-07 13:35             ` Richard Biener
2022-12-15 15:26               ` Dimitrije Milosevic
2022-12-16  9:58                 ` Richard Biener
2022-12-16 11:37                   ` Dimitrije Milosevic
2022-12-16 11:58                     ` Richard Biener
2022-10-21 13:52 ` [PATCH 2/2] ivopts: Consider number of invariants when calculating register pressure Dimitrije Milosevic
2022-10-25 11:07   ` Richard Biener
2022-10-25 13:00     ` Dimitrije Milosevic
2022-10-28  7:38       ` Richard Biener
2022-10-28 13:39         ` Dimitrije Milosevic
2022-11-07 12:56           ` Richard Biener

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=20221021135203.626255-1-dimitrije.milosevic@syrmia.com \
    --to=dimitrije.milosevic@syrmia.com \
    --cc=djordje.todorovic@syrmia.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).