public inbox for gcc-prs@sourceware.org
help / color / mirror / Atom feed
* middle-end/7048: GCC fails to emit addr_diff_vec table after casesi in -O2.
@ 2002-06-16 13:24 Matt Thomas
0 siblings, 0 replies; 3+ messages in thread
From: Matt Thomas @ 2002-06-16 13:24 UTC (permalink / raw)
To: gcc-gnats
>Number: 7048
>Category: middle-end
>Synopsis: GCC fails to emit addr_diff_vec table after casesi in -O2.
>Confidential: no
>Severity: serious
>Priority: medium
>Responsible: unassigned
>State: open
>Class: wrong-code
>Submitter-Id: net
>Arrival-Date: Sun Jun 16 13:16:00 PDT 2002
>Closed-Date:
>Last-Modified:
>Originator: Matt Thomas
>Release: 3.2 20020615 (experimental)
>Organization:
3am Software Foundry
>Environment:
System: NetBSD pilot.local 1.5ZC NetBSD 1.5ZC (PILOT) #59: Sat May 18 12:43:46 PDT 2002 matt@pilot.local:/export/i386/kobj/PILOT i386
host: i386-unknown-netbsdelf1.5ZC
build: i386-unknown-netbsdelf1.5ZC
target: vax-dec-netbsdelf
configured with: /export/toolchain/gcc/configure --enable-static --disable-shared --target=vax-netbsdelf : (reconfigured) : (reconfigured)
>Description:
When compiling C source with a switch statement, depending on the
complexity of the switch cases, the addr_diff_vec table is sometimes
omitted when compiling -O2.
>How-To-Repeat:
Compile the following with -O2 and examine the resultant .s file;
notice the missing addr_diff_vec table:
__asm__(".section " ".ident" " ; .asciz \"" "$NetBSD: sysconf.c,v 1.14 2002/01/31 00:32:47 kleink Exp $" "\" ; .text");
typedef long long int __int64_t;
typedef __int64_t quad_t;
typedef quad_t rlim_t;
typedef unsigned int u_int;
typedef unsigned long int size_t;
int _sysctl (int *, u_int, void *, size_t *, const void *, size_t);
struct rlimit {
rlim_t rlim_cur;
rlim_t rlim_max;
};
int getrlimit (int, struct rlimit *);
int *__errno (void);
long
__sysconf(name)
int name;
{
struct rlimit rl;
size_t len;
int mib[2], value;
len = sizeof(value);
switch (name) {
case 1:
mib[0] = 1;
mib[1] = 8;
break;
case 2:
return (getrlimit(7, &rl) ? -1 : (long)rl.rlim_cur);
case 3:
return (100);
case 6:
mib[0] = 1;
mib[1] = 19;
goto yesno;
case 4:
mib[0] = 1;
mib[1] = 18;
break;
case 5:
return (getrlimit(8, &rl) ? -1 : (long)rl.rlim_cur);
case 26:
mib[0] = 8;
mib[1] = 19;
break;
case 27:
mib[0] = 8;
mib[1] = 20;
break;
case 7:
mib[0] = 1;
mib[1] = 20;
goto yesno;
case 8:
mib[0] = 1;
mib[1] = 17;
break;
case 28:
mib[0] = 6;
mib[1] = 7;
break;
case 29:
mib[0] = 1;
mib[1] = 32;
goto yesno;
case 31:
mib[0] = 1;
mib[1] = 37;
goto yesno;
case 33:
mib[0] = 1;
mib[1] = 40;
goto yesno;
case 34:
mib[0] = 1;
mib[1] = 41;
goto yesno;
case 35:
mib[0] = 1;
mib[1] = 42;
goto yesno;
case 36:
mib[0] = 1;
mib[1] = 43;
goto yesno;
case 38:
mib[0] = 1;
mib[1] = 60;
goto yesno;
case 37:
mib[0] = 1;
mib[1] = 44;
break;
case 9:
mib[0] = 8;
mib[1] = 2;
break;
case 10:
mib[0] = 8;
mib[1] = 3;
break;
case 11:
mib[0] = 8;
mib[1] = 4;
break;
case 12:
mib[0] = 8;
mib[1] = 5;
break;
case 13:
mib[0] = 8;
mib[1] = 6;
break;
case 14:
mib[0] = 8;
mib[1] = 7;
break;
case 15:
mib[0] = 8;
mib[1] = 8;
break;
case 16:
mib[0] = 8;
mib[1] = 9;
break;
case 17:
mib[0] = 8;
mib[1] = 10;
break;
case 18:
mib[0] = 8;
mib[1] = 11;
goto yesno;
case 19:
mib[0] = 8;
mib[1] = 12;
goto yesno;
case 20:
mib[0] = 8;
mib[1] = 13;
goto yesno;
case 21:
mib[0] = 8;
mib[1] = 14;
goto yesno;
case 22:
mib[0] = 8;
mib[1] = 15;
goto yesno;
case 23:
mib[0] = 8;
mib[1] = 16;
goto yesno;
case 24:
mib[0] = 8;
mib[1] = 17;
goto yesno;
case 25:
mib[0] = 8;
mib[1] = 18;
goto yesno;
case 32:
mib[0] = 1;
mib[1] = 38;
break;
case 30:
mib[0] = 1;
mib[1] = 35;
yesno: if (_sysctl(mib, 2, &value, &len, 0, 0) == -1)
return (-1);
if (value == 0)
return (-1);
return (value);
break;
default:
(*__errno()) = 22;
return (-1);
}
return (_sysctl(mib, 2, &value, &len, 0, 0) == -1 ? -1 : value);
}
>Fix:
>Release-Note:
>Audit-Trail:
>Unformatted:
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: middle-end/7048: GCC fails to emit addr_diff_vec table after casesi in -O2.
@ 2003-01-14 23:03 bangerth
0 siblings, 0 replies; 3+ messages in thread
From: bangerth @ 2003-01-14 23:03 UTC (permalink / raw)
To: gcc-bugs, gcc-prs, matt, rth
Synopsis: GCC fails to emit addr_diff_vec table after casesi in -O2.
State-Changed-From-To: feedback->closed
State-Changed-By: bangerth
State-Changed-When: Tue Jan 14 15:03:29 2003
State-Changed-Why:
No feedback. Richard's patch has been applied.
http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=7048
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: middle-end/7048: GCC fails to emit addr_diff_vec table after casesi in -O2.
@ 2002-06-19 11:41 rth
0 siblings, 0 replies; 3+ messages in thread
From: rth @ 2002-06-19 11:41 UTC (permalink / raw)
To: gcc-bugs, gcc-prs, matt, nobody, rth
Synopsis: GCC fails to emit addr_diff_vec table after casesi in -O2.
Responsible-Changed-From-To: unassigned->rth
Responsible-Changed-By: rth
Responsible-Changed-When: Wed Jun 19 10:16:11 2002
Responsible-Changed-Why:
.
State-Changed-From-To: open->feedback
State-Changed-By: rth
State-Changed-When: Wed Jun 19 10:16:11 2002
State-Changed-Why:
http://gcc.gnu.org/ml/gcc-patches/2002-06/msg01591.html
http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=7048
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2003-01-14 23:03 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2002-06-16 13:24 middle-end/7048: GCC fails to emit addr_diff_vec table after casesi in -O2 Matt Thomas
2002-06-19 11:41 rth
2003-01-14 23:03 bangerth
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).