Re: middle-end/10011: Broken __builtin_constant_p

Re: middle-end/10011: Broken __builtin_constant_p

[bangerth]

Synopsis: Broken __builtin_constant_p

State-Changed-From-To: open->feedback
State-Changed-By: bangerth
State-Changed-When: Sat Mar 15 04:03:42 2003
State-Changed-Why:
    Andreas, I cannot reproduce this with a snapshot of
    yesterday on x86 linux. I am surprised that this should
    be target dependent, so can I ask you to re-check on PPC
    whether you still see the problem?
    
    Thanks
      Wolfgang

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=10011

Re: middle-end/10011: Broken __builtin_constant_p

[Andreas Schwab]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 964 bytes --]

The following reply was made to PR middle-end/10011; it has been noted by GNATS.

From: Andreas Schwab <schwab@suse.de>
To: bangerth@dealii.org
Cc: gcc-bugs@gcc.gnu.org, gcc-prs@gcc.gnu.org, gcc-gnats@gcc.gnu.org
Subject: Re: middle-end/10011: Broken __builtin_constant_p
Date: Sat, 15 Mar 2003 16:44:52 +0100

 bangerth@dealii.org writes:
 
 |>     Andreas, I cannot reproduce this with a snapshot of
 |>     yesterday on x86 linux. I am surprised that this should
 |>     be target dependent, so can I ask you to re-check on PPC
 |>     whether you still see the problem?
 
 I can reproduce it on ia64 as of today.  Also checked with an m68k cross
 compiler as of yesterday, and it was ok.  I'll try to check other targets.
 
 Andreas.
 
 -- 
 Andreas Schwab, SuSE Labs, schwab@suse.de
 SuSE Linux AG, Deutschherrnstr. 15-19, D-90429 Nürnberg
 Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
 "And now for something completely different."

middle-end/10011: Broken __builtin_constant_p

[schwab]

>Number:         10011
>Category:       middle-end
>Synopsis:       Broken __builtin_constant_p
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          wrong-code
>Submitter-Id:   net
>Arrival-Date:   Sun Mar 09 20:46:01 UTC 2003
>Closed-Date:
>Last-Modified:
>Originator:     Andreas Schwab
>Release:        3.4 20030306 (experimental)
>Organization:
>Environment:
System: Linux whitebox 2.4.20-ben7 #1 Thu Feb 27 00:38:54 CET 2003 ppc unknown unknown GNU/Linux
Architecture: ppc

host: powerpc-unknown-linux-gnu
build: powerpc-unknown-linux-gnu
target: powerpc-unknown-linux-gnu
configured with: ../gcc/configure --prefix=/usr/local/powerpc-linux --host=powerpc-linux --with-__cxa_atexit --with-as=/usr/local/powerpc-linux/bin/as --with-ld=/usr/local/powerpc-linux/bin/ld --enable-languages=c,c++ --enable-shared
>Description:
__builtin_constant_p expands to non-zero although the argument is not constant.
>How-To-Repeat:
$ cat builtin_constant_p.c
extern void __bad_udelay(void);
static void usb_hub_port_wait_reset(unsigned int delay)
{
	int delay_time;
	for (delay_time = 0; delay_time < 500; delay_time += delay) {
		if (__builtin_constant_p(delay))
			__bad_udelay();
	}
}
$ gcc -S -O2 builtin_constant_p.c
$ grep __bad_udelay builtin_constant_p.s
	bl __bad_udelay
>Fix:

>Release-Note:
>Audit-Trail:
>Unformatted: