From: Bernd Schmidt <bernds@codesourcery.com>
To: Jeff Law <law@redhat.com>
Cc: GCC Patches <gcc-patches@gcc.gnu.org>,
John David Anglin <dave@hiauly1.hia.nrc.ca>
Subject: Re: Emit more REG_EQUIV notes for function args (PR42235)
Date: Wed, 21 Jul 2010 22:53:00 -0000 [thread overview]
Message-ID: <4C477A1A.8040604@codesourcery.com> (raw)
In-Reply-To: <4C44827B.5070305@redhat.com>
[-- Attachment #1: Type: text/plain, Size: 199 bytes --]
Here's a different patch, which I've committed as obvious after
bootstrapping and testing i686 and x64_64-linux. It just restricts the
previous change to situations where it's clearly safe.
Bernd
[-- Attachment #2: parm-equiv-fix.diff --]
[-- Type: text/plain, Size: 1526 bytes --]
Index: ChangeLog
===================================================================
--- ChangeLog (revision 162390)
+++ ChangeLog (working copy)
@@ -18,6 +18,10 @@
an old insn, ignore a use that occurs after store_ruid.
* Makefile.in (postreload.o): Update dependencies.
+ * function.c (record_hard_reg_sets): Restrict the previous change
+ to cases where the incoming nominal mode is the same as the
+ incoming promoted mode and everything happens in MODE_INT.
+
2010-07-21 Jakub Jelinek <jakub@redhat.com>
PR debug/45015
Index: function.c
===================================================================
--- function.c (revision 162372)
+++ function.c (working copy)
@@ -2918,7 +2918,10 @@ assign_parm_setup_reg (struct assign_par
|| promoted_nominal_mode != data->promoted_mode);
moved = false;
- if (need_conversion)
+ if (need_conversion
+ && GET_MODE_CLASS (data->nominal_mode) == MODE_INT
+ && data->nominal_mode == data->passed_mode
+ && data->nominal_mode == GET_MODE (data->entry_parm))
{
/* ENTRY_PARM has been converted to PROMOTED_MODE, its
mode, by the caller. We now have to convert it to
@@ -2979,8 +2982,9 @@ assign_parm_setup_reg (struct assign_par
if (moved)
{
emit_insn (insns);
- equiv_stack_parm = gen_rtx_fmt_e (code, GET_MODE (parmreg),
- equiv_stack_parm);
+ if (equiv_stack_parm != NULL_RTX)
+ equiv_stack_parm = gen_rtx_fmt_e (code, GET_MODE (parmreg),
+ equiv_stack_parm);
}
}
}
next prev parent reply other threads:[~2010-07-21 22:53 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-07-14 11:14 Bernd Schmidt
2010-07-14 18:54 ` Jeff Law
2010-07-14 21:30 ` Bernd Schmidt
2010-07-14 22:28 ` Jeff Law
2010-07-15 17:14 ` Bernd Schmidt
2010-07-15 21:26 ` Jeff Law
2010-07-19 9:55 ` Bernd Schmidt
2010-07-19 16:51 ` Jeff Law
2010-07-19 16:53 ` Bernd Schmidt
2010-07-19 16:59 ` Richard Henderson
2010-07-19 17:09 ` Jeff Law
2010-07-19 19:06 ` John David Anglin
2010-07-21 22:53 ` Bernd Schmidt [this message]
2010-07-14 21:48 ` Bernd Schmidt
2010-07-14 22:22 ` Jeff Law
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=4C477A1A.8040604@codesourcery.com \
--to=bernds@codesourcery.com \
--cc=dave@hiauly1.hia.nrc.ca \
--cc=gcc-patches@gcc.gnu.org \
--cc=law@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).