* Re: [Kernel-janitors] [PATHC] fix ide.c warning
2004-04-14 2:17 [Kernel-janitors] [PATHC] fix ide.c warning Don Koch
@ 2004-04-14 18:02 ` Luiz Fernando N. Capitulino
2004-04-14 21:10 ` Michael Still
` (4 subsequent siblings)
5 siblings, 0 replies; 7+ messages in thread
From: Luiz Fernando N. Capitulino @ 2004-04-14 18:02 UTC (permalink / raw)
To: kernel-janitors
[-- Attachment #1: Type: text/plain, Size: 926 bytes --]
Hi Don,
Em Tue, 13 Apr 2004 22:17:53 -0400
Don Koch <aardvark@krl.com> escreveu:
| This fixes the "implicit declaration of function `pnpide_init'" warning
| in ide.c. Based on 2.6.5-mm5.
|
| ===== drivers/ide/ide.c 1.132 vs edited =====
| --- 1.132/drivers/ide/ide.c Sat Apr 10 18:22:11 2004
| +++ edited/drivers/ide/ide.c Tue Apr 13 20:59:52 2004
| @@ -2245,7 +2245,10 @@
| return 1;
| }
| #if defined(CONFIG_BLK_DEV_IDEPNP) && defined(CONFIG_PNP) && defined(MODULE)
| - pnpide_init(0);
| + {
| + extern void pnpide_init(int enable);
| + pnpide_init(0);
| + }
| #endif /* CONFIG_BLK_DEV_IDEPNP */
| #ifdef CONFIG_PROC_FS
| ide_remove_proc_entries(drive->proc, DRIVER(drive)->proc);
what about to declare pnpide_init() in include/linux/ide.h (like ide_end_request),
and just call it ? It is a bad idea ?
--
Luiz Fernando N. Capitulino
<lcapitulino@prefeitura.sp.gov.br>
<http://www.telecentros.sp.gov.br>
[-- Attachment #2: Type: text/plain, Size: 167 bytes --]
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.osdl.org
http://lists.osdl.org/mailman/listinfo/kernel-janitors
^ permalink raw reply [flat|nested] 7+ messages in thread* Re: [Kernel-janitors] [PATHC] fix ide.c warning
2004-04-14 2:17 [Kernel-janitors] [PATHC] fix ide.c warning Don Koch
2004-04-14 18:02 ` Luiz Fernando N. Capitulino
@ 2004-04-14 21:10 ` Michael Still
2004-04-14 21:17 ` Don Koch
` (3 subsequent siblings)
5 siblings, 0 replies; 7+ messages in thread
From: Michael Still @ 2004-04-14 21:10 UTC (permalink / raw)
To: kernel-janitors
[-- Attachment #1: Type: text/plain, Size: 1428 bytes --]
Luiz Fernando N. Capitulino wrote:
> Hi Don,
>
> Em Tue, 13 Apr 2004 22:17:53 -0400
> Don Koch <aardvark@krl.com> escreveu:
>
> | This fixes the "implicit declaration of function `pnpide_init'" warning
> | in ide.c. Based on 2.6.5-mm5.
> |
> | ===== drivers/ide/ide.c 1.132 vs edited =====
> | --- 1.132/drivers/ide/ide.c Sat Apr 10 18:22:11 2004
> | +++ edited/drivers/ide/ide.c Tue Apr 13 20:59:52 2004
> | @@ -2245,7 +2245,10 @@
> | return 1;
> | }
> | #if defined(CONFIG_BLK_DEV_IDEPNP) && defined(CONFIG_PNP) && defined(MODULE)
> | - pnpide_init(0);
> | + {
> | + extern void pnpide_init(int enable);
> | + pnpide_init(0);
> | + }
> | #endif /* CONFIG_BLK_DEV_IDEPNP */
> | #ifdef CONFIG_PROC_FS
> | ide_remove_proc_entries(drive->proc, DRIVER(drive)->proc);
>
> what about to declare pnpide_init() in include/linux/ide.h (like ide_end_request),
> and just call it ? It is a bad idea ?
I would think it's much better form to either:
- put the prototype in a header
- repeat the prototype at the top of this file (there are risks here)
With the first one being much prefered in my book. It's just not
idiomatic in C to do it this way. You also don't need the extern:
void pnpide_init(int enable);
should be sufficient.
Additionally, putting that block of code in braces is evil.
Cheers,
Mikal
PS: I didn't look at the definition of pnpide_init, and assume it's not
static or something.
[-- Attachment #2: Type: text/plain, Size: 167 bytes --]
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.osdl.org
http://lists.osdl.org/mailman/listinfo/kernel-janitors
^ permalink raw reply [flat|nested] 7+ messages in thread* Re: [Kernel-janitors] [PATHC] fix ide.c warning
2004-04-14 2:17 [Kernel-janitors] [PATHC] fix ide.c warning Don Koch
2004-04-14 18:02 ` Luiz Fernando N. Capitulino
2004-04-14 21:10 ` Michael Still
@ 2004-04-14 21:17 ` Don Koch
2004-04-14 21:33 ` Luiz Fernando N. Capitulino
` (2 subsequent siblings)
5 siblings, 0 replies; 7+ messages in thread
From: Don Koch @ 2004-04-14 21:17 UTC (permalink / raw)
To: kernel-janitors
[-- Attachment #1: Type: text/plain, Size: 1836 bytes --]
mikal@stillhq.com said:
> Luiz Fernando N. Capitulino wrote:
> > Hi Don,
> >
> > Em Tue, 13 Apr 2004 22:17:53 -0400
> > Don Koch <aardvark@krl.com> escreveu:
> >
> > | This fixes the "implicit declaration of function `pnpide_init'" warning
> > | in ide.c. Based on 2.6.5-mm5.
> > |
> > | ===== drivers/ide/ide.c 1.132 vs edited =====
> > | --- 1.132/drivers/ide/ide.c Sat Apr 10 18:22:11 2004
> > | +++ edited/drivers/ide/ide.c Tue Apr 13 20:59:52 2004
> > | @@ -2245,7 +2245,10 @@
> > | return 1;
> > | }
> > | #if defined(CONFIG_BLK_DEV_IDEPNP) && defined(CONFIG_PNP) && defined(MODULE)
> > | - pnpide_init(0);
> > | + {
> > | + extern void pnpide_init(int enable);
> > | + pnpide_init(0);
> > | + }
> > | #endif /* CONFIG_BLK_DEV_IDEPNP */
> > | #ifdef CONFIG_PROC_FS
> > | ide_remove_proc_entries(drive->proc, DRIVER(drive)->proc);
> >
> > what about to declare pnpide_init() in include/linux/ide.h (like ide_end_request),
> > and just call it ? It is a bad idea ?
>
> I would think it's much better form to either:
>
> - put the prototype in a header
> - repeat the prototype at the top of this file (there are risks here)
>
> With the first one being much prefered in my book. It's just not
> idiomatic in C to do it this way. You also don't need the extern:
>
> void pnpide_init(int enable);
>
> should be sufficient.
>
> Additionally, putting that block of code in braces is evil.
>
> Cheers,
> Mikal
>
> PS: I didn't look at the definition of pnpide_init, and assume it's not
> static or something.
>
The patch echos other calls to pnpide_init() in the same file.
The evil was done before I got there. (I agree, it's ugly.)
I'll poke around and see if I can find a more appropriate way
to handle all of the declarations and submit a follow-up patch.
--
Don Koch
http://www.krl.com/
[-- Attachment #2: Type: text/plain, Size: 167 bytes --]
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.osdl.org
http://lists.osdl.org/mailman/listinfo/kernel-janitors
^ permalink raw reply [flat|nested] 7+ messages in thread* Re: [Kernel-janitors] [PATHC] fix ide.c warning
2004-04-14 2:17 [Kernel-janitors] [PATHC] fix ide.c warning Don Koch
` (2 preceding siblings ...)
2004-04-14 21:17 ` Don Koch
@ 2004-04-14 21:33 ` Luiz Fernando N. Capitulino
2004-04-14 21:39 ` Luiz Fernando N. Capitulino
2004-04-14 22:58 ` Michael Still
5 siblings, 0 replies; 7+ messages in thread
From: Luiz Fernando N. Capitulino @ 2004-04-14 21:33 UTC (permalink / raw)
To: kernel-janitors
[-- Attachment #1: Type: text/plain, Size: 1810 bytes --]
Hi Michael,
Em Thu, 15 Apr 2004 07:10:02 +1000
Michael Still <mikal@stillhq.com> escreveu:
| > | This fixes the "implicit declaration of function `pnpide_init'" warning
| > | in ide.c. Based on 2.6.5-mm5.
| > |
| > | ===== drivers/ide/ide.c 1.132 vs edited =====
| > | --- 1.132/drivers/ide/ide.c Sat Apr 10 18:22:11 2004
| > | +++ edited/drivers/ide/ide.c Tue Apr 13 20:59:52 2004
| > | @@ -2245,7 +2245,10 @@
| > | return 1;
| > | }
| > | #if defined(CONFIG_BLK_DEV_IDEPNP) && defined(CONFIG_PNP) && defined(MODULE)
| > | - pnpide_init(0);
| > | + {
| > | + extern void pnpide_init(int enable);
| > | + pnpide_init(0);
| > | + }
| > | #endif /* CONFIG_BLK_DEV_IDEPNP */
| > | #ifdef CONFIG_PROC_FS
| > | ide_remove_proc_entries(drive->proc, DRIVER(drive)->proc);
| >
| > what about to declare pnpide_init() in include/linux/ide.h (like ide_end_request),
| > and just call it ? It is a bad idea ?
|
| I would think it's much better form to either:
|
| - put the prototype in a header
| - repeat the prototype at the top of this file (there are risks here)
I don't think is a good way to create a new .h file only to put
one prototype there. include/linux/ide.h already does that for another ide
functions, so it can be used (well, was what a quick look show me).
| With the first one being much prefered in my book. It's just not
| idiomatic in C to do it this way. You also don't need the extern:
|
| void pnpide_init(int enable);
|
| should be sufficient.
IMHO, the extern declaration makes it easy to understand.
| Additionally, putting that block of code in braces is evil.
Yes, is what I'm trying to avoid (but I already added evil code
in the kernel :-)).
--
Luiz Fernando N. Capitulino
<lcapitulino@prefeitura.sp.gov.br>
<http://www.telecentros.sp.gov.br>
[-- Attachment #2: Type: text/plain, Size: 167 bytes --]
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.osdl.org
http://lists.osdl.org/mailman/listinfo/kernel-janitors
^ permalink raw reply [flat|nested] 7+ messages in thread* Re: [Kernel-janitors] [PATHC] fix ide.c warning
2004-04-14 2:17 [Kernel-janitors] [PATHC] fix ide.c warning Don Koch
` (3 preceding siblings ...)
2004-04-14 21:33 ` Luiz Fernando N. Capitulino
@ 2004-04-14 21:39 ` Luiz Fernando N. Capitulino
2004-04-14 22:58 ` Michael Still
5 siblings, 0 replies; 7+ messages in thread
From: Luiz Fernando N. Capitulino @ 2004-04-14 21:39 UTC (permalink / raw)
To: kernel-janitors
[-- Attachment #1: Type: text/plain, Size: 2062 bytes --]
Em Wed, 14 Apr 2004 17:17:45 -0400
Don Koch <aardvark@krl.com> escreveu:
| > > Em Tue, 13 Apr 2004 22:17:53 -0400
| > > Don Koch <aardvark@krl.com> escreveu:
| > >
| > > | This fixes the "implicit declaration of function `pnpide_init'" warning
| > > | in ide.c. Based on 2.6.5-mm5.
| > > |
| > > | ===== drivers/ide/ide.c 1.132 vs edited =====
| > > | --- 1.132/drivers/ide/ide.c Sat Apr 10 18:22:11 2004
| > > | +++ edited/drivers/ide/ide.c Tue Apr 13 20:59:52 2004
| > > | @@ -2245,7 +2245,10 @@
| > > | return 1;
| > > | }
| > > | #if defined(CONFIG_BLK_DEV_IDEPNP) && defined(CONFIG_PNP) && defined(MODULE)
| > > | - pnpide_init(0);
| > > | + {
| > > | + extern void pnpide_init(int enable);
| > > | + pnpide_init(0);
| > > | + }
| > > | #endif /* CONFIG_BLK_DEV_IDEPNP */
| > > | #ifdef CONFIG_PROC_FS
| > > | ide_remove_proc_entries(drive->proc, DRIVER(drive)->proc);
| > >
| > > what about to declare pnpide_init() in include/linux/ide.h (like ide_end_request),
| > > and just call it ? It is a bad idea ?
| >
| > I would think it's much better form to either:
| >
| > - put the prototype in a header
| > - repeat the prototype at the top of this file (there are risks here)
| >
| > With the first one being much prefered in my book. It's just not
| > idiomatic in C to do it this way. You also don't need the extern:
| >
| > void pnpide_init(int enable);
| >
| > should be sufficient.
| >
| > Additionally, putting that block of code in braces is evil.
| >
| > Cheers,
| > Mikal
| >
| > PS: I didn't look at the definition of pnpide_init, and assume it's not
| > static or something.
| >
|
| The patch echos other calls to pnpide_init() in the same file.
| The evil was done before I got there. (I agree, it's ugly.)
| I'll poke around and see if I can find a more appropriate way
| to handle all of the declarations and submit a follow-up patch.
Will be very good if we remove the warning with a good way.
--
Luiz Fernando N. Capitulino
<lcapitulino@prefeitura.sp.gov.br>
<http://www.telecentros.sp.gov.br>
[-- Attachment #2: Type: text/plain, Size: 167 bytes --]
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.osdl.org
http://lists.osdl.org/mailman/listinfo/kernel-janitors
^ permalink raw reply [flat|nested] 7+ messages in thread* Re: [Kernel-janitors] [PATHC] fix ide.c warning
2004-04-14 2:17 [Kernel-janitors] [PATHC] fix ide.c warning Don Koch
` (4 preceding siblings ...)
2004-04-14 21:39 ` Luiz Fernando N. Capitulino
@ 2004-04-14 22:58 ` Michael Still
5 siblings, 0 replies; 7+ messages in thread
From: Michael Still @ 2004-04-14 22:58 UTC (permalink / raw)
To: kernel-janitors
[-- Attachment #1: Type: text/plain, Size: 760 bytes --]
Luiz Fernando N. Capitulino wrote:
> | I would think it's much better form to either:
> |
> | - put the prototype in a header
> | - repeat the prototype at the top of this file (there are risks here)
>
> I don't think is a good way to create a new .h file only to put
> one prototype there. include/linux/ide.h already does that for another ide
> functions, so it can be used (well, was what a quick look show me).
Hey, I don't really mind which header it goes in. If there is an
existing suitable header, then that sounds like a good path.
Cheers,
Mikal
--
Michael Still (mikal@stillhq.com) | "All my life I've had one dream,
http://www.stillhq.com | to achieve my many goals"
UTC + 11 | -- Homer Simpson
[-- Attachment #2: Type: text/plain, Size: 167 bytes --]
_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.osdl.org
http://lists.osdl.org/mailman/listinfo/kernel-janitors
^ permalink raw reply [flat|nested] 7+ messages in thread