From: Eric Botcazou <ebotcazou@adacore.com>
To: "H.J. Lu" <hjl.tools@gmail.com>
Cc: Bernd Schmidt <bernds@codesourcery.com>,
Richard Henderson <rth@redhat.com>,
Uros Bizjak <ubizjak@gmail.com>,
gcc-patches@gcc.gnu.org, Jeff Law <law@redhat.com>
Subject: Re: [x32] PATCH: PR middle-end/47725: [x32] error: unable to find a register to spill in class DIREG
Date: Thu, 24 Mar 2011 15:56:00 -0000 [thread overview]
Message-ID: <201103241651.28202.ebotcazou@adacore.com> (raw)
In-Reply-To: <AANLkTinjSkpnPcc59JoDTBm-SXi974s16JXN5rkAcQ+g@mail.gmail.com>
> Pointer is promoted to Pmode from ptr_mode.
Indeed. However the problem is the 2 in assign_parm_setup_reg:
/* Store the parm in a pseudoregister during the function, but we may
need to do it in a wider mode. Using 2 here makes the result
consistent with promote_decl_mode and thus expand_expr_real_1. */
promoted_nominal_mode
= promote_function_mode (data->nominal_type, data->nominal_mode, &unsignedp,
TREE_TYPE (current_function_decl), 2);
which is supposed to match the 2 in promote_decl_mode:
if (TREE_CODE (decl) == RESULT_DECL
|| TREE_CODE (decl) == PARM_DECL)
pmode = promote_function_mode (type, mode, &unsignedp,
TREE_TYPE (current_function_decl), 2);
else
pmode = promote_mode (type, mode, &unsignedp);
but doesn't match the 0 in assign_parm_find_data_types:
promoted_mode = promote_function_mode (passed_type, passed_mode, &unsignedp,
TREE_TYPE (current_function_decl), 0);
so you get the redundant extension in the callee. The solution is to define
the promote_function_mode hook for x86 to something like:
static enum machine_mode
ix86_promote_function_mode (const_tree type,
enum machine_mode mode,
int *punsignedp,
const_tree fntype ATTRIBUTE_UNUSED,
int for_return ATTRIBUTE_UNUSED)
{
if (POINTER_TYPE_P (type))
{
*punsignedp = POINTERS_EXTEND_UNSIGNED;
return Pmode;
}
return mode;
}
--
Eric Botcazou
next prev parent reply other threads:[~2011-03-24 15:56 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-02-14 19:04 H.J. Lu
2011-02-14 19:07 ` Jeff Law
2011-02-14 19:11 ` H.J. Lu
2011-02-14 19:17 ` Jeff Law
2011-02-14 19:39 ` Eric Botcazou
2011-02-14 19:49 ` Bernd Schmidt
2011-02-14 19:51 ` Eric Botcazou
2011-02-15 23:09 ` Bernd Schmidt
2011-03-18 0:30 ` H.J. Lu
2011-03-18 4:01 ` H.J. Lu
2011-03-21 4:15 ` H.J. Lu
2011-03-22 20:10 ` Eric Botcazou
2011-03-23 3:29 ` H.J. Lu
2011-03-23 8:27 ` Eric Botcazou
2011-03-23 10:57 ` H.J. Lu
2011-03-24 15:56 ` Eric Botcazou [this message]
2011-03-29 17:50 ` H.J. Lu
2011-02-14 19:54 ` H.J. Lu
2011-02-15 15:53 ` Bernd Schmidt
2011-02-15 17:02 ` H.J. Lu
2011-02-15 17:49 ` Eric Botcazou
2011-02-15 18:14 ` H.J. Lu
2011-02-15 19:03 ` Eric Botcazou
2011-02-15 21:16 ` Jeff Law
2011-02-15 21:39 ` Bernd Schmidt
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=201103241651.28202.ebotcazou@adacore.com \
--to=ebotcazou@adacore.com \
--cc=bernds@codesourcery.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=hjl.tools@gmail.com \
--cc=law@redhat.com \
--cc=rth@redhat.com \
--cc=ubizjak@gmail.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).