From: Richard Henderson <rth@cygnus.com>
To: Jason McMullan <jmcmullan@linuxcare.com>
Cc: gcc@gcc.gnu.org, gcc-patches@gcc.gnu.org, gcc-bugs@gcc.gnu.org
Subject: Re: [PATCH] __builtin_apply / -fdefer-pop bug (ObjC fix!)
Date: Mon, 31 Jul 2000 09:41:00 -0000 [thread overview]
Message-ID: <20000731094105.A6702@cygnus.com> (raw)
In-Reply-To: <20000726163607.A13929@port.evillabs.net>
On Wed, Jul 26, 2000 at 04:36:07PM -0400, Jason McMullan wrote:
> When compiled with -fdefer-pop (without the following
> patch) GCC neglects to set the stack pointer appropriately
> after the memcpy() call (to move the stack arguments), and
> the called function (that __builtin_apply() was trying to call)
> is sent a bogus stack pointer.
You didn't point at a test case, which would have been helpful.
Nor did you work against current gcc sources, which have changed.
However, the solution seems reasonable. I've applied the following.
r~
2000-07-31 Jason McMullan <jmcmullan@linuxcare.com>
* builtins.c (expand_builtin_apply): Don't defer pop during
argument setup.
Index: builtins.c
===================================================================
RCS file: /cvs/gcc/egcs/gcc/builtins.c,v
retrieving revision 1.51
diff -c -p -d -r1.51 builtins.c
*** builtins.c 2000/07/03 18:28:33 1.51
--- builtins.c 2000/07/31 16:34:43
*************** expand_builtin_apply (function, argument
*** 885,892 ****
/* Perform postincrements before actually calling the function. */
emit_queue ();
! /* Push a new argument block and copy the arguments. */
do_pending_stack_adjust ();
/* Save the stack with nonlocal if available */
#ifdef HAVE_save_stack_nonlocal
--- 885,895 ----
/* Perform postincrements before actually calling the function. */
emit_queue ();
! /* Push a new argument block and copy the arguments. Do not allow
! the (potential) memcpy call below to interfere with our stack
! manipulations. */
do_pending_stack_adjust ();
+ NO_DEFER_POP;
/* Save the stack with nonlocal if available */
#ifdef HAVE_save_stack_nonlocal
*************** expand_builtin_apply (function, argument
*** 1025,1030 ****
--- 1028,1035 ----
else
#endif
emit_stack_restore (SAVE_BLOCK, old_stack_level, NULL_RTX);
+
+ OK_DEFER_POP;
/* Return the address of the result block. */
return copy_addr_to_reg (XEXP (result, 0));
prev parent reply other threads:[~2000-07-31 9:41 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2000-07-26 13:36 Jason McMullan
2000-07-27 13:57 ` Ovidiu Predescu
2000-07-27 15:44 ` Jason McMullan
2000-07-31 9:41 ` Richard Henderson [this message]
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=20000731094105.A6702@cygnus.com \
--to=rth@cygnus.com \
--cc=gcc-bugs@gcc.gnu.org \
--cc=gcc-patches@gcc.gnu.org \
--cc=gcc@gcc.gnu.org \
--cc=jmcmullan@linuxcare.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).