public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug target/19493] New: [4.0 regression] TImode no longer supported on powerpc
@ 2005-01-17 23:18 schwab at suse dot de
2005-01-17 23:25 ` [Bug target/19493] " pinskia at gcc dot gnu dot org
` (4 more replies)
0 siblings, 5 replies; 6+ messages in thread
From: schwab at suse dot de @ 2005-01-17 23:18 UTC (permalink / raw)
To: gcc-bugs
$ cat timode.c
typedef unsigned foo __attribute__ ((mode (TI)));
$ gcc-4.0 -S timode.c
timode.c:1: error: unable to emulate ‘TI’
--
Summary: [4.0 regression] TImode no longer supported on powerpc
Product: gcc
Version: 4.0.0
Status: UNCONFIRMED
Severity: normal
Priority: P2
Component: target
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: schwab at suse dot de
CC: gcc-bugs at gcc dot gnu dot org
GCC target triplet: powerpc-linux
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19493
^ permalink raw reply [flat|nested] 6+ messages in thread* [Bug target/19493] [4.0 regression] TImode no longer supported on powerpc
2005-01-17 23:18 [Bug target/19493] New: [4.0 regression] TImode no longer supported on powerpc schwab at suse dot de
@ 2005-01-17 23:25 ` pinskia at gcc dot gnu dot org
2005-01-17 23:34 ` pinskia at gcc dot gnu dot org
` (3 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2005-01-17 23:25 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From pinskia at gcc dot gnu dot org 2005-01-17 23:25 -------
Actually this is not a regression TImode was never really supported on ppc.
If you did the following on a 3.4 compiler it would ICE:
void main (void)
{
typedef unsigned __uint128_t __attribute__ (( __mode__ (__TI__)));
__uint128_t x,a,w;
x=a*w+a-w/a;
}
So we just reject the code before we ICE which is better than ICEing.
Really any place which used TI mode should be changed because TI mode was never supported on ppc-
* (now ppc64 is a different story).
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19493
^ permalink raw reply [flat|nested] 6+ messages in thread* [Bug target/19493] [4.0 regression] TImode no longer supported on powerpc
2005-01-17 23:18 [Bug target/19493] New: [4.0 regression] TImode no longer supported on powerpc schwab at suse dot de
2005-01-17 23:25 ` [Bug target/19493] " pinskia at gcc dot gnu dot org
@ 2005-01-17 23:34 ` pinskia at gcc dot gnu dot org
2005-01-17 23:37 ` [Bug target/19493] [3.4/4.0 " schwab at suse dot de
` (2 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2005-01-17 23:34 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From pinskia at gcc dot gnu dot org 2005-01-17 23:34 -------
TI mode is not supported in 32bit.
If this was accepted before 4.0.0, well that was a bug as shown by my example about.
--
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |RESOLVED
Resolution| |INVALID
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19493
^ permalink raw reply [flat|nested] 6+ messages in thread* [Bug target/19493] [3.4/4.0 regression] TImode no longer supported on powerpc
2005-01-17 23:18 [Bug target/19493] New: [4.0 regression] TImode no longer supported on powerpc schwab at suse dot de
2005-01-17 23:25 ` [Bug target/19493] " pinskia at gcc dot gnu dot org
2005-01-17 23:34 ` pinskia at gcc dot gnu dot org
@ 2005-01-17 23:37 ` schwab at suse dot de
2005-01-17 23:56 ` [Bug target/19493] TImode is not " pinskia at gcc dot gnu dot org
2005-01-18 0:03 ` dje at gcc dot gnu dot org
4 siblings, 0 replies; 6+ messages in thread
From: schwab at suse dot de @ 2005-01-17 23:37 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From schwab at suse dot de 2005-01-17 23:37 -------
Doesn't ICE with gcc-3.3 (but TImode arithmetic is incomplete here too).
Glibc assumes that __uint128_t is defined for gcc >= 3.1.
--
What |Removed |Added
----------------------------------------------------------------------------
Status|RESOLVED |UNCONFIRMED
Resolution|INVALID |
Summary|[4.0 regression] TImode no |[3.4/4.0 regression] TImode
|longer supported on powerpc |no longer supported on
| |powerpc
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19493
^ permalink raw reply [flat|nested] 6+ messages in thread* [Bug target/19493] TImode is not supported on powerpc
2005-01-17 23:18 [Bug target/19493] New: [4.0 regression] TImode no longer supported on powerpc schwab at suse dot de
` (2 preceding siblings ...)
2005-01-17 23:37 ` [Bug target/19493] [3.4/4.0 " schwab at suse dot de
@ 2005-01-17 23:56 ` pinskia at gcc dot gnu dot org
2005-01-18 0:03 ` dje at gcc dot gnu dot org
4 siblings, 0 replies; 6+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2005-01-17 23:56 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From pinskia at gcc dot gnu dot org 2005-01-17 23:56 -------
(In reply to comment #3)
> Doesn't ICE with gcc-3.3 (but TImode arithmetic is incomplete here too).
> Glibc assumes that __uint128_t is defined for gcc >= 3.1.
Well if TImode is incomplete why support it?
See PR 17279 which is why __uint128_t does not get defined.
I would take Alan's advice in
<http://gcc.gnu.org/ml/gcc/2005-01/msg00513.html>.
See RTH's patch here which basically made TImode not avaiable in 32bit mdoe:
http://gcc.gnu.org/ml/gcc-patches/2004-08/msg02114.html
If anything should be the case, we should not create another regression, PR 17279 was a regression in
3.4.0 from 3.3.3 on powerpc-darwin because HWI was 32bit and changed to 64bit for 3.4.0 but since
TImode is rejected we don't get an ICE either.
--
What |Removed |Added
----------------------------------------------------------------------------
Severity|normal |minor
Summary|[3.4/4.0 regression] TImode |TImode is not supported on
|no longer supported on |powerpc
|powerpc |
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19493
^ permalink raw reply [flat|nested] 6+ messages in thread* [Bug target/19493] TImode is not supported on powerpc
2005-01-17 23:18 [Bug target/19493] New: [4.0 regression] TImode no longer supported on powerpc schwab at suse dot de
` (3 preceding siblings ...)
2005-01-17 23:56 ` [Bug target/19493] TImode is not " pinskia at gcc dot gnu dot org
@ 2005-01-18 0:03 ` dje at gcc dot gnu dot org
4 siblings, 0 replies; 6+ messages in thread
From: dje at gcc dot gnu dot org @ 2005-01-18 0:03 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From dje at gcc dot gnu dot org 2005-01-18 00:03 -------
alanm's comment and recommendation are correct. The use of TImode in
PowerPC-specific portion of glibc is incorrect.
--
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |RESOLVED
Resolution| |WONTFIX
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19493
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2005-01-18 0:03 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2005-01-17 23:18 [Bug target/19493] New: [4.0 regression] TImode no longer supported on powerpc schwab at suse dot de
2005-01-17 23:25 ` [Bug target/19493] " pinskia at gcc dot gnu dot org
2005-01-17 23:34 ` pinskia at gcc dot gnu dot org
2005-01-17 23:37 ` [Bug target/19493] [3.4/4.0 " schwab at suse dot de
2005-01-17 23:56 ` [Bug target/19493] TImode is not " pinskia at gcc dot gnu dot org
2005-01-18 0:03 ` dje at gcc dot gnu dot org
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).