From: Alexander Monakov <amonakov@ispras.ru>
To: Peter Bergner <bergner@vnet.ibm.com>
Cc: GCC <gcc@gcc.gnu.org>, Vladimir N Makarov <vmakarov@redhat.com>,
Pat Haugen <pthaugen@linux.vnet.ibm.com>
Subject: Re: Why does IRA force all pseudos live across a setjmp call to be spilled?
Date: Fri, 02 Mar 2018 19:55:00 -0000 [thread overview]
Message-ID: <alpine.LNX.2.20.13.1803022252110.3749@monopod.intra.ispras.ru> (raw)
In-Reply-To: <141cfa78-f202-029a-e530-24e657692bff@vnet.ibm.com>
On Fri, 2 Mar 2018, Peter Bergner wrote:
> But currently ira-lives.c:process_bb_node_lives() has:
>
> /* Don't allocate allocnos that cross setjmps or any
> call, if this function receives a nonlocal
> goto. */
> if (cfun->has_nonlocal_label
> || find_reg_note (insn, REG_SETJMP,
> NULL_RTX) != NULL_RTX)
> {
> SET_HARD_REG_SET (OBJECT_CONFLICT_HARD_REGS (obj));
> SET_HARD_REG_SET (OBJECT_TOTAL_CONFLICT_HARD_REGS (obj));
> }
>
> ...which forces us to spill everything live across the setjmp by forcing
> the pseudos to interfere all hardregs. That can't be good for performance.
> What am I missing?
FWIW there's a similar issue with exceptions where IRA chooses memory
for pseudos inside the loop even though the throwing call is outside:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82242#c3
Alexander
next prev parent reply other threads:[~2018-03-02 19:55 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-03-02 19:45 Peter Bergner
2018-03-02 19:55 ` Alexander Monakov [this message]
2018-03-02 21:26 ` Jeff Law
2018-03-03 3:28 ` Peter Bergner
2018-03-03 16:29 ` Jeff Law
2018-03-03 23:47 ` Peter Bergner
2018-03-04 0:30 ` Peter Bergner
2018-03-04 8:57 ` Richard Biener
2018-03-04 9:11 ` Andreas Schwab
2018-03-05 2:12 ` Jeff Law
2018-03-04 13:57 ` Eric Botcazou
2018-03-04 16:40 ` Peter Bergner
2018-03-04 17:39 ` Andreas Schwab
2018-03-05 0:16 ` Segher Boessenkool
2018-03-05 7:01 ` Eric Botcazou
2018-03-05 7:09 ` Eric Botcazou
2018-03-05 15:33 ` Segher Boessenkool
2018-03-05 16:18 ` Eric Botcazou
2018-03-06 16:20 ` Peter Bergner
2018-03-03 3:21 ` Vladimir Makarov
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.2.20.13.1803022252110.3749@monopod.intra.ispras.ru \
--to=amonakov@ispras.ru \
--cc=bergner@vnet.ibm.com \
--cc=gcc@gcc.gnu.org \
--cc=pthaugen@linux.vnet.ibm.com \
--cc=vmakarov@redhat.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).