* RE: Warning elimination patch in sched.c
@ 1998-03-10 6:46 Bjørn Wennberg
0 siblings, 0 replies; 6+ messages in thread
From: Bjørn Wennberg @ 1998-03-10 6:46 UTC (permalink / raw)
To: 'Kaveh R. Ghazi', egcs
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1120 bytes --]
I did include the appropriate include-files to get the correct
definition of qsort. That
is why I find the warning messages strange.
bjornw>
> -----Original Message-----
> From: Kaveh R. Ghazi [SMTP:ghazi@caip.rutgers.edu]
> Sent: 10. mars 1998 14:42
> To: Bjorn.Wennberg@lcc.no; egcs@cygnus.com;
> schmid@ltoi.iap.physik.tu-darmstadt.de
> Subject: RE: Warning elimination patch in sched.c
>
> > From: Bjørn Wennberg <Bjorn.Wennberg@lcc.no>
> >
> > This might be the correct solution to the problem but I'm curios to
> why
> > the warnings are reported from gcc.
> > If I create a small file containing:
> > -----
> > static int function(int *, int *);
> > int main() {
> > qsort(0, 0, 0, function);
> > }
> > ------
> > Then I do not get a warning message from gcc. So why do we get a
> warning
> > message when compiling haifa-sched.c (or sched.c)?
> > bjornw>
>
> Your program must include stdlib.h to prototype qsort's argument
> list, or the -W -Wall flags won't know anything's amiss.
> --
> Kaveh R. Ghazi Project Manager / Custom
> Development
> ghazi@caip.rutgers.edu ICon CMT Corp.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: Warning elimination patch in sched.c
1998-03-09 12:52 Peter Schmid
@ 1998-03-12 14:50 ` Jim Wilson
0 siblings, 0 replies; 6+ messages in thread
From: Jim Wilson @ 1998-03-12 14:50 UTC (permalink / raw)
To: Peter Schmid; +Cc: egcs
While compiling gcc/sched.c I get the warning:
sched.c: In function `schedule_block':
sched.c:3138: warning: passing arg 4 of `qsort' from incompatible pointer type
sched.c:3233: warning: passing arg 4 of `qsort' from incompatible pointer type
A patch has been installed to fix this. It is very similar to your patch,
except that it uses GENERIC_PTR instead of void *, as some older K&R C
compiler doesn't handle void *.
Jim
^ permalink raw reply [flat|nested] 6+ messages in thread
* RE: Warning elimination patch in sched.c
@ 1998-03-10 10:28 Kaveh R. Ghazi
0 siblings, 0 replies; 6+ messages in thread
From: Kaveh R. Ghazi @ 1998-03-10 10:28 UTC (permalink / raw)
To: Bjorn.Wennberg, egcs
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1086 bytes --]
> > From: Bjxrn Wennberg <Bjorn.Wennberg@lcc.no>
> >
> > This might be the correct solution to the problem but I'm curios to why
> > the warnings are reported from gcc.
> > If I create a small file containing:
> > -----
> > static int function(int *, int *);
> > int main() {
> > qsort(0, 0, 0, function);
> > }
> > ------
> From: Bjørn Wennberg <Bjorn.Wennberg@lcc.no>
>
> I did include the appropriate include-files to get the correct
> definition of qsort. That
> is why I find the warning messages strange.
>
> bjornw>
Hmm, your posting of the code fragment didn't indicate stdlib.h
was included. Its hard to figure out what's going on with incomplete
information.
Perhaps if you posted the output from "gcc -v", the exact source
code, and the exact flags you passed to gcc to compile the program, it
would help. Also run the compilation with -save-temps and send the .i
file. This will let me see the prototype for qsort on your platform.
--Kaveh
--
Kaveh R. Ghazi Project Manager / Custom Development
ghazi@caip.rutgers.edu ICon CMT Corp.
^ permalink raw reply [flat|nested] 6+ messages in thread
* RE: Warning elimination patch in sched.c
@ 1998-03-10 6:46 Kaveh R. Ghazi
0 siblings, 0 replies; 6+ messages in thread
From: Kaveh R. Ghazi @ 1998-03-10 6:46 UTC (permalink / raw)
To: Bjorn.Wennberg, egcs, schmid
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 679 bytes --]
> From: Bjørn Wennberg <Bjorn.Wennberg@lcc.no>
>
> This might be the correct solution to the problem but I'm curios to why
> the warnings are reported from gcc.
> If I create a small file containing:
> -----
> static int function(int *, int *);
> int main() {
> qsort(0, 0, 0, function);
> }
> ------
> Then I do not get a warning message from gcc. So why do we get a warning
> message when compiling haifa-sched.c (or sched.c)?
> bjornw>
Your program must include stdlib.h to prototype qsort's argument
list, or the -W -Wall flags won't know anything's amiss.
--
Kaveh R. Ghazi Project Manager / Custom Development
ghazi@caip.rutgers.edu ICon CMT Corp.
^ permalink raw reply [flat|nested] 6+ messages in thread
* RE: Warning elimination patch in sched.c
@ 1998-03-10 2:14 Bjørn Wennberg
0 siblings, 0 replies; 6+ messages in thread
From: Bjørn Wennberg @ 1998-03-10 2:14 UTC (permalink / raw)
To: 'Peter Schmid', egcs
This might be the correct solution to the problem but I'm curios to why
the warnings are reported from gcc.
If I create a small file containing:
-----
static int function(int *, int *);
int main() {
qsort(0, 0, 0, function);
}
------
Then I do not get a warning message from gcc. So why do we get a warning
message when compiling haifa-sched.c (or sched.c)?
bjornw>
> -----Original Message-----
> From: Peter Schmid [SMTP:schmid@ltoi.iap.physik.tu-darmstadt.de]
> Sent: 9. mars 1998 20:27
> To: egcs@cygnus.com
> Subject: Warning elimination patch in sched.c
>
> While compiling gcc/sched.c I get the warning:
> sched.c: In function `schedule_block':
> sched.c:3138: warning: passing arg 4 of `qsort' from incompatible
> pointer type
> sched.c:3233: warning: passing arg 4 of `qsort' from incompatible
> pointer type
>
> The following patch eliminates the warning.
>
> Perhaps something like:
> #ifndef __cplusplus
> #ifndef __STDC__
> #define const
> #endif
> #endif
> should also be included, as in c-parse.c.
>
> Peter Schmid
>
> Mon Mar 9 19:34:17 1998 Peter Schmid
> <schmid@ltoi.iap.physik.tu-darmstadt.de>
>
> * sched.c: change rank_for_schedule prototype to
> (const void*,const void*) to avoid passing arg 4 of `qsort'
> incompatible
> pointer type warning
>
> *** gcc/sched.c~ Fri Feb 20 14:24:50 1998
> --- gcc/sched.c Mon Mar 9 20:22:55 1998
> *************** static void sched_analyze_2 PROTO((rtx,
> *** 322,328 ****
> static void sched_analyze_insn PROTO((rtx, rtx, rtx));
> static int sched_analyze PROTO((rtx, rtx));
> static void sched_note_set PROTO((int, rtx, int));
> ! static int rank_for_schedule PROTO((rtx *, rtx *));
> static void swap_sort PROTO((rtx *, int));
> static void queue_insn PROTO((rtx, int));
> static int birthing_insn_p PROTO((rtx));
> --- 322,328 ----
> static void sched_analyze_insn PROTO((rtx, rtx, rtx));
> static int sched_analyze PROTO((rtx, rtx));
> static void sched_note_set PROTO((int, rtx, int));
> ! static int rank_for_schedule PROTO((const void *,
> const void *));
> static void swap_sort PROTO((rtx *, int));
> static void queue_insn PROTO((rtx, int));
> static int birthing_insn_p PROTO((rtx));
> *************** sched_note_set (b, x, death)
> *** 1859,1867 ****
> unstable. */
>
> static int
> ! rank_for_schedule (x, y)
> ! rtx *x, *y;
> {
> rtx tmp = *y;
> rtx tmp2 = *x;
> rtx link;
> --- 1859,1869 ----
> unstable. */
>
> static int
> ! rank_for_schedule (xv, yv)
> ! const void *xv, *yv;
> {
> + rtx *x = (rtx *) xv;
> + rtx *y = (rtx *) yv;
> rtx tmp = *y;
> rtx tmp2 = *x;
> rtx link;
^ permalink raw reply [flat|nested] 6+ messages in thread
* Warning elimination patch in sched.c
@ 1998-03-09 12:52 Peter Schmid
1998-03-12 14:50 ` Jim Wilson
0 siblings, 1 reply; 6+ messages in thread
From: Peter Schmid @ 1998-03-09 12:52 UTC (permalink / raw)
To: egcs
While compiling gcc/sched.c I get the warning:
sched.c: In function `schedule_block':
sched.c:3138: warning: passing arg 4 of `qsort' from incompatible pointer type
sched.c:3233: warning: passing arg 4 of `qsort' from incompatible pointer type
The following patch eliminates the warning.
Perhaps something like:
#ifndef __cplusplus
#ifndef __STDC__
#define const
#endif
#endif
should also be included, as in c-parse.c.
Peter Schmid
Mon Mar 9 19:34:17 1998 Peter Schmid <schmid@ltoi.iap.physik.tu-darmstadt.de>
* sched.c: change rank_for_schedule prototype to
(const void*,const void*) to avoid passing arg 4 of `qsort' incompatible
pointer type warning
*** gcc/sched.c~ Fri Feb 20 14:24:50 1998
--- gcc/sched.c Mon Mar 9 20:22:55 1998
*************** static void sched_analyze_2 PROTO((rtx,
*** 322,328 ****
static void sched_analyze_insn PROTO((rtx, rtx, rtx));
static int sched_analyze PROTO((rtx, rtx));
static void sched_note_set PROTO((int, rtx, int));
! static int rank_for_schedule PROTO((rtx *, rtx *));
static void swap_sort PROTO((rtx *, int));
static void queue_insn PROTO((rtx, int));
static int birthing_insn_p PROTO((rtx));
--- 322,328 ----
static void sched_analyze_insn PROTO((rtx, rtx, rtx));
static int sched_analyze PROTO((rtx, rtx));
static void sched_note_set PROTO((int, rtx, int));
! static int rank_for_schedule PROTO((const void *, const void *));
static void swap_sort PROTO((rtx *, int));
static void queue_insn PROTO((rtx, int));
static int birthing_insn_p PROTO((rtx));
*************** sched_note_set (b, x, death)
*** 1859,1867 ****
unstable. */
static int
! rank_for_schedule (x, y)
! rtx *x, *y;
{
rtx tmp = *y;
rtx tmp2 = *x;
rtx link;
--- 1859,1869 ----
unstable. */
static int
! rank_for_schedule (xv, yv)
! const void *xv, *yv;
{
+ rtx *x = (rtx *) xv;
+ rtx *y = (rtx *) yv;
rtx tmp = *y;
rtx tmp2 = *x;
rtx link;
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~1998-03-12 14:50 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
1998-03-10 6:46 Warning elimination patch in sched.c Bjørn Wennberg
-- strict thread matches above, loose matches on Subject: below --
1998-03-10 10:28 Kaveh R. Ghazi
1998-03-10 6:46 Kaveh R. Ghazi
1998-03-10 2:14 Bjørn Wennberg
1998-03-09 12:52 Peter Schmid
1998-03-12 14:50 ` Jim Wilson
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).