[PATCH] Isapnp warning

[PATCH] Isapnp warning

[Geert Uytterhoeven]

Isapnp: Kill warning if CONFIG_PCI is not set

--- linux-2.4.x/drivers/pnp/isapnp.c	Sat Aug 17 14:10:39 2002
+++ linux-m68k-2.4.x/drivers/pnp/isapnp.c	Fri Jun  6 12:27:42 2003
@@ -510,7 +510,9 @@
                                                int dependent, int size)
 {
 	unsigned char tmp[3];
+#ifdef CONFIG_PCI
 	int i;
+#endif
 	struct isapnp_irq *irq, *ptr;
 
 	isapnp_peek(tmp, size);

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

[PATCH] Isapnp warning

[Geert Uytterhoeven]

Isapnp: Kill warning if CONFIG_PCI is not set, like it's done in 2.6.0.

--- linux-2.4.23-pre1/drivers/pnp/isapnp.c	Sat Aug 17 14:10:39 2002
+++ linux-m68k-2.4.23-pre1/drivers/pnp/isapnp.c	Fri Jun  6 12:27:42 2003
@@ -510,7 +510,6 @@
                                                int dependent, int size)
 {
 	unsigned char tmp[3];
-	int i;
 	struct isapnp_irq *irq, *ptr;
 
 	isapnp_peek(tmp, size);
@@ -538,9 +537,13 @@
 	else
 		(*res)->irq = irq;
 #ifdef CONFIG_PCI
-	for (i=0; i<16; i++)
-		if (irq->map & (1<<i))
-			pcibios_penalize_isa_irq(i);
+	{
+		int i;
+
+		for (i=0; i<16; i++)
+			if (irq->map & (1<<i))
+				pcibios_penalize_isa_irq(i);
+	}
 #endif
 }
 

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Re: [PATCH] Isapnp warning

[Jan-Benedict Glaw]

[-- Attachment #1: Type: text/plain, Size: 1199 bytes --]

On Sun, 2003-06-22 16:07:14 +0200, Daniel Phillips <phillips@arcor.de>
wrote in message <200306221607.15232.phillips@arcor.de>:
> Hi Alan,
> 
> On Saturday 21 June 2003 14:31, Alan Cox wrote:
> > On Sul, 2003-06-15 at 19:36, Geert Uytterhoeven wrote:
> 
> How about:
> 
>    #define if_pci(tokens...) tokens
> 
>    int pnp_add_irq_resource(struct pnp_dev *dev, int depnum, struct pnp_irq *data)
>    {
> 	if_pci(int i);
> 	...
>    }
> 
> Admittedly uglier than just having the warning disabled by default.

Even whilest I don't like defining variables where I need them (at an
opening "{" or like in "for (int i, i < x, i++)" as Linus suggested it),
this is quite ugly, too. "if_pci(int i)" looks linke an uglyfied
function call, and even while being ugly, it should basically "work"
like a function call. Here, it doesn't, so I consider this a Bad Thing.

MfG, JBG

-- 
   Jan-Benedict Glaw       jbglaw@lug-owl.de    . +49-172-7608481
   "Eine Freie Meinung in  einem Freien Kopf    | Gegen Zensur | Gegen Krieg
    fuer einen Freien Staat voll Freier Bürger" | im Internet! |   im Irak!
      ret = do_actions((curr | FREE_SPEECH) & ~(IRAQ_WAR_2 | DRM | TCPA));

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [PATCH] Isapnp warning

[Daniel Phillips]

Hi Alan,

On Saturday 21 June 2003 14:31, Alan Cox wrote:
> On Sul, 2003-06-15 at 19:36, Geert Uytterhoeven wrote:
> > Isapnp: Kill warning if CONFIG_PCI is not set
> >
> > --- linux-2.5.x/drivers/pnp/resource.c	Tue May 27 19:03:04 2003
> > +++ linux-m68k-2.5.x/drivers/pnp/resource.c	Sun Jun  8 13:31:20 2003
> > @@ -97,7 +97,9 @@
> >
> >  int pnp_add_irq_resource(struct pnp_dev *dev, int depnum, struct pnp_irq *data)
> >  {
> > +#ifdef CONFIG_PCI
> >  	int i;
> > +#endif
>
> This is far uglier than te warning

How about:

   #define if_pci(tokens...) tokens

   int pnp_add_irq_resource(struct pnp_dev *dev, int depnum, struct pnp_irq *data)
   {
	if_pci(int i);
	...
   }

Admittedly uglier than just having the warning disabled by default.

Regards,

Daniel

Re: [PATCH] Isapnp warning

[Daniel Phillips]

On Sunday 22 June 2003 10:32, Geert Uytterhoeven wrote:
> On Sun, 22 Jun 2003, Paul Mackerras wrote:
> > Andrew Morton writes:
> > > Compared to 2.95.3, gcc-3.3 takes 1.5x as long to compile, and produces
> > > a kernel which is 200k larger.
> >
> > I just tried it on PPC.  Compared to 2.95.4, gcc-3.3 took 36% longer
> > to compile and produced a kernel which was 120k smaller.
>
> I have the same experience w.r.t. kernel size on MIPS, using 2.95.4 from
> Debian woody on the target, and 3.2.2 for cross-compiling.
>
> Perhaps the code increase is for CISC only?

In another branch of this thread Andrew said it's due to extra alignment 
padding (presumably with a view to improving execution speed) and Herbert Xu 
mentioned out-of-line conditional branch compilation, both of which sould 
reasonable to me.  They also sound like they could or should be trivially 
switchable.

Regards,

Daniel

Re: [PATCH] Isapnp warning

[Russell King]

On Sat, Jun 21, 2003 at 10:43:45PM -0300, Arnaldo Carvalho de Melo wrote:
> I don't know, I was just trying to figure out the impact of requiring gcc 3
> to compile the kernel. I never used gcc 2.96 btw.

So, for ARM, we end up with the following gcc versions which appear to
work:

	- gcc 2.95.3 + patch
	- gcc 2.95.4
	- gcc 3.2.2 + patch
	- gcc 3.2.3 + patch
	- gcc 3.3

(Some of the gcc people may beg to differ, but the above list results
from many reports and real life experiences from several people.)

>From what I understand, the gcc people are not all that happy about
the gcc 3.2.x patch, based upon the fact that it has not been applied.
It is also possible that gcc 3.3 happens to work because the bug got
hidden by other changes, or, the real bug did get fixed.  No one seems
to know.  Certainly the bugzilla entry remains open and unresolved.

-- 
Russell King (rmk@arm.linux.org.uk)                The developer of ARM Linux
             http://www.arm.linux.org.uk/personal/aboutme.html

Re: [PATCH] Isapnp warning

[Jörn Engel]

On Sat, 21 June 2003 20:41:02 -0500, Chris Wedgwood wrote:
> On Sat, Jun 21, 2003 at 09:11:01PM -0300, Arnaldo Carvalho de Melo wrote:
> 
> > Humm, I'd love to do that, i.e. to make gcc 3 required, lots of good
> > stuff like this one, anonymous structs, etc, etc, lots of stuff
> > could be done in an easier way, but are we ready to abandon gcc
> > 2.95.*? Can anyone confirm if gcc 2.96 accepts this?
> 
> What *requires* 2.96 still?  Is it a large number of people or obscure
> architecture?

Try these two litter things in Assembler code:

#define LONG_MACRO	\
asm			\
#ifdef something	\
asm			\
#else			\
asm			\
asm			\
#endif			\
asm

Compiles just fine with all 3.x I've tried.

	jump	92f
	...
81:	...
	...
92	...

Again, compiles just fine.

Both are clearly wrong and should not be in the source code.  But
Assembler bugs are subtle, nasty and Code Checkers for Assembler are
rare.  My vote is to keep 2.95+ support.

Jörn

-- 
"Error protection by error detection and correction."
-- from a university class

Re: [PATCH] Isapnp warning

[Geert Uytterhoeven]

On Sun, 22 Jun 2003, Paul Mackerras wrote:
> Andrew Morton writes:
> > Compared to 2.95.3, gcc-3.3 takes 1.5x as long to compile, and produces a
> > kernel which is 200k larger.
> 
> I just tried it on PPC.  Compared to 2.95.4, gcc-3.3 took 36% longer
> to compile and produced a kernel which was 120k smaller.

I have the same experience w.r.t. kernel size on MIPS, using 2.95.4 from Debian
woody on the target, and 3.2.2 for cross-compiling.

Perhaps the code increase is for CISC only?

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Re: [PATCH] Isapnp warning

[Herbert Xu]

Andrew Morton <akpm@digeo.com> wrote:
> 
> Alignment mainly.  It pads stuff all over the place.  A lot of it can be
> defeated - but not all, last time I looked.

And if you're compiling for i386, the main increase comes from
out-of-line branches.
-- 
Debian GNU/Linux 3.0 is out! ( http://www.debian.org/ )
Email:  Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

Re: [PATCH] Isapnp warning

[Paul Mackerras]

Andrew Morton writes:

> Compared to 2.95.3, gcc-3.3 takes 1.5x as long to compile, and produces a
> kernel which is 200k larger.

I just tried it on PPC.  Compared to 2.95.4, gcc-3.3 took 36% longer
to compile and produced a kernel which was 120k smaller.

Paul.

Re: [PATCH] Isapnp warning

[Martin J. Bligh]

--Andrew Morton <akpm@digeo.com> wrote (on Saturday, June 21, 2003 19:17:05 -0700):

> Arnaldo Carvalho de Melo <acme@conectiva.com.br> wrote:
>> 
>> Em Sat, Jun 21, 2003 at 08:41:02PM -0500, Chris Wedgwood escreveu:
>> > On Sat, Jun 21, 2003 at 09:11:01PM -0300, Arnaldo Carvalho de Melo wrote:
>> > 
>> > > Humm, I'd love to do that, i.e. to make gcc 3 required, lots of good
>> > > stuff like this one, anonymous structs, etc, etc, lots of stuff
>> > > could be done in an easier way, but are we ready to abandon gcc
>> > > 2.95.*? Can anyone confirm if gcc 2.96 accepts this?
>> > 
>> > What *requires* 2.96 still?  Is it a large number of people or obscure
>> > architecture?
>> 
>> I don't know, I was just trying to figure out the impact of requiring gcc 3
>> to compile the kernel. I never used gcc 2.96 btw.
>> 
> 
> Compared to 2.95.3, gcc-3.3 takes 1.5x as long to compile, and produces a
> kernel which is 200k larger.
> 
> It is simply worthless.

Agreed. 2.95.4 is also still the default debian compiler. Requiring
3.x seems like a bad plan, until they get it to a point where it's
actually better than 2.95, stable, and widely distributed. It's 
definitely not there yet (seems kind of stable, but not the others).

M.

Re: [PATCH] Isapnp warning

[Andrew Morton]

Chris Wedgwood <cw@f00f.org> wrote:
>
> On Sat, Jun 21, 2003 at 07:17:05PM -0700, Andrew Morton wrote:
> 
> > Compared to 2.95.3, gcc-3.3 takes 1.5x as long to compile, and
> > produces a kernel which is 200k larger.
> 
> Do we know why this is the case?  I assume the fix is far from
> trivial?

Alignment mainly.  It pads stuff all over the place.  A lot of it can be
defeated - but not all, last time I looked.

Re: [PATCH] Isapnp warning

[Chris Wedgwood]

On Sat, Jun 21, 2003 at 07:17:05PM -0700, Andrew Morton wrote:

> Compared to 2.95.3, gcc-3.3 takes 1.5x as long to compile, and
> produces a kernel which is 200k larger.

Do we know why this is the case?  I assume the fix is far from
trivial?

> It is simply worthless.

I stopped using 2.95.x recently because of miscompiles.  Even vendor
compilers seemed to break at times :(

  --cw

Re: [PATCH] Isapnp warning

[Andrew Morton]

Arnaldo Carvalho de Melo <acme@conectiva.com.br> wrote:
>
> Em Sat, Jun 21, 2003 at 08:41:02PM -0500, Chris Wedgwood escreveu:
> > On Sat, Jun 21, 2003 at 09:11:01PM -0300, Arnaldo Carvalho de Melo wrote:
> > 
> > > Humm, I'd love to do that, i.e. to make gcc 3 required, lots of good
> > > stuff like this one, anonymous structs, etc, etc, lots of stuff
> > > could be done in an easier way, but are we ready to abandon gcc
> > > 2.95.*? Can anyone confirm if gcc 2.96 accepts this?
> > 
> > What *requires* 2.96 still?  Is it a large number of people or obscure
> > architecture?
> 
> I don't know, I was just trying to figure out the impact of requiring gcc 3
> to compile the kernel. I never used gcc 2.96 btw.
> 

Compared to 2.95.3, gcc-3.3 takes 1.5x as long to compile, and produces a
kernel which is 200k larger.

It is simply worthless.

Re: [PATCH] Isapnp warning

[Arnaldo Carvalho de Melo]

Em Sat, Jun 21, 2003 at 08:41:02PM -0500, Chris Wedgwood escreveu:
> On Sat, Jun 21, 2003 at 09:11:01PM -0300, Arnaldo Carvalho de Melo wrote:
> 
> > Humm, I'd love to do that, i.e. to make gcc 3 required, lots of good
> > stuff like this one, anonymous structs, etc, etc, lots of stuff
> > could be done in an easier way, but are we ready to abandon gcc
> > 2.95.*? Can anyone confirm if gcc 2.96 accepts this?
> 
> What *requires* 2.96 still?  Is it a large number of people or obscure
> architecture?

I don't know, I was just trying to figure out the impact of requiring gcc 3
to compile the kernel. I never used gcc 2.96 btw.

- Arnaldo

Re: [PATCH] Isapnp warning

[Chris Wedgwood]

On Sat, Jun 21, 2003 at 09:11:01PM -0300, Arnaldo Carvalho de Melo wrote:

> Humm, I'd love to do that, i.e. to make gcc 3 required, lots of good
> stuff like this one, anonymous structs, etc, etc, lots of stuff
> could be done in an easier way, but are we ready to abandon gcc
> 2.95.*? Can anyone confirm if gcc 2.96 accepts this?

What *requires* 2.96 still?  Is it a large number of people or obscure
architecture?


  --cw

Re: [PATCH] Isapnp warning

[Arnaldo Carvalho de Melo]

Em Sat, Jun 21, 2003 at 04:53:32PM -0700, Linus Torvalds escreveu:
> 
> On Sat, 21 Jun 2003, Andrew Morton wrote:
> > 
> > Meanwhile, let's do this:
> 
> I'd prefer the C99 thing, ie
> 
> 	for (int i = xxx ...)
> 
> syntax. I know gcc-3.x supports it, maybe 2.96 does too? If so, we could
> just add "-std=c99" or whatever, and start using that.

Humm, I'd love to do that, i.e. to make gcc 3 required, lots of good stuff
like this one, anonymous structs, etc, etc, lots of stuff could be done in
an easier way, but are we ready to abandon gcc 2.95.*? Can anyone confirm
if gcc 2.96 accepts this?

- Arnaldo

Re: [PATCH] Isapnp warning

[Linus Torvalds]

On Sat, 21 Jun 2003, Andrew Morton wrote:
> 
> Meanwhile, let's do this:

I'd prefer the C99 thing, ie

	for (int i = xxx ...)

syntax. I know gcc-3.x supports it, maybe 2.96 does too? If so, we could
just add "-std=c99" or whatever, and start using that.

		Linus

Re: [PATCH] Isapnp warning

[Andrew Morton]

Geert Uytterhoeven <geert@linux-m68k.org> wrote:
>
> > >  int pnp_add_irq_resource(struct pnp_dev *dev, int depnum, struct pnp_irq *data)
>  > >  {
>  > > +#ifdef CONFIG_PCI
>  > >  	int i;
>  > > +#endif
>  > 
>  > This is far uglier than te warning
> 
>  It depends on your goals. These warnings distract us from the real harmful
>  warnings. Will we ever have a kernel that compiles with -Werror?

It would be nice.  But as soon as we do that, some gcc guy will have a
brainfart and we'll get a whole new batch of warnings which we cannot turn
off.  Again.  I've been involved in projects where it was unacceptable to
upgrade the gcc version for this sole reason.


Meanwhile, let's do this:

diff -puN drivers/pnp/resource.c~misc6 drivers/pnp/resource.c
--- 25/drivers/pnp/resource.c~misc6	2003-06-21 12:47:23.000000000 -0700
+++ 25-akpm/drivers/pnp/resource.c	2003-06-21 12:47:44.000000000 -0700
@@ -97,7 +97,6 @@ int pnp_get_max_depnum(struct pnp_dev *d
 
 int pnp_add_irq_resource(struct pnp_dev *dev, int depnum, struct pnp_irq *data)
 {
-	int i;
 	struct pnp_resources *res;
 	struct pnp_irq *ptr;
 	res = pnp_find_resources(dev,depnum);
@@ -113,9 +112,13 @@ int pnp_add_irq_resource(struct pnp_dev 
 	else
 		res->irq = data;
 #ifdef CONFIG_PCI
-	for (i=0; i<16; i++)
-		if (data->map & (1<<i))
-			pcibios_penalize_isa_irq(i);
+	{
+		int i;
+
+		for (i=0; i<16; i++)
+			if (data->map & (1<<i))
+				pcibios_penalize_isa_irq(i);
+	}
 #endif
 	return 0;
 }

_

Re: [PATCH] Isapnp warning

[Sean Neakums]

Geert Uytterhoeven <geert@linux-m68k.org> writes:

> On 21 Jun 2003, Alan Cox wrote:
>> On Sul, 2003-06-15 at 19:36, Geert Uytterhoeven wrote:
>> > Isapnp: Kill warning if CONFIG_PCI is not set
>> > 
>> > --- linux-2.5.x/drivers/pnp/resource.c	Tue May 27 19:03:04 2003
>> > +++ linux-m68k-2.5.x/drivers/pnp/resource.c	Sun Jun  8 13:31:20 2003
>> > @@ -97,7 +97,9 @@
>> >  
>> >  int pnp_add_irq_resource(struct pnp_dev *dev, int depnum, struct pnp_irq *data)
>> >  {
>> > +#ifdef CONFIG_PCI
>> >  	int i;
>> > +#endif
>> 
>> This is far uglier than te warning
>
> It depends on your goals. These warnings distract us from the real harmful
> warnings. Will we ever have a kernel that compiles with -Werror?

Unless GCC never emits an invalid warning, no.

Re: [PATCH] Isapnp warning

[Geert Uytterhoeven]

On 21 Jun 2003, Alan Cox wrote:
> On Sul, 2003-06-15 at 19:36, Geert Uytterhoeven wrote:
> > Isapnp: Kill warning if CONFIG_PCI is not set
> > 
> > --- linux-2.5.x/drivers/pnp/resource.c	Tue May 27 19:03:04 2003
> > +++ linux-m68k-2.5.x/drivers/pnp/resource.c	Sun Jun  8 13:31:20 2003
> > @@ -97,7 +97,9 @@
> >  
> >  int pnp_add_irq_resource(struct pnp_dev *dev, int depnum, struct pnp_irq *data)
> >  {
> > +#ifdef CONFIG_PCI
> >  	int i;
> > +#endif
> 
> This is far uglier than te warning

It depends on your goals. These warnings distract us from the real harmful
warnings. Will we ever have a kernel that compiles with -Werror?

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Re: [PATCH] Isapnp warning

[Alan Cox]

On Sul, 2003-06-15 at 19:36, Geert Uytterhoeven wrote:
> Isapnp: Kill warning if CONFIG_PCI is not set
> 
> --- linux-2.5.x/drivers/pnp/resource.c	Tue May 27 19:03:04 2003
> +++ linux-m68k-2.5.x/drivers/pnp/resource.c	Sun Jun  8 13:31:20 2003
> @@ -97,7 +97,9 @@
>  
>  int pnp_add_irq_resource(struct pnp_dev *dev, int depnum, struct pnp_irq *data)
>  {
> +#ifdef CONFIG_PCI
>  	int i;
> +#endif

This is far uglier than te warning

[PATCH] Isapnp warning

[Geert Uytterhoeven]

Isapnp: Kill warning if CONFIG_PCI is not set

--- linux-2.4.x/drivers/pnp/isapnp.c	Sat Aug 17 14:10:39 2002
+++ linux-m68k-2.4.x/drivers/pnp/isapnp.c	Fri Jun  6 12:27:42 2003
@@ -510,7 +510,9 @@
                                                int dependent, int size)
 {
 	unsigned char tmp[3];
+#ifdef CONFIG_PCI
 	int i;
+#endif
 	struct isapnp_irq *irq, *ptr;
 
 	isapnp_peek(tmp, size);

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

[PATCH] Isapnp warning

[Geert Uytterhoeven]

Isapnp: Kill warning if CONFIG_PCI is not set

--- linux-2.5.x/drivers/pnp/resource.c	Tue May 27 19:03:04 2003
+++ linux-m68k-2.5.x/drivers/pnp/resource.c	Sun Jun  8 13:31:20 2003
@@ -97,7 +97,9 @@
 
 int pnp_add_irq_resource(struct pnp_dev *dev, int depnum, struct pnp_irq *data)
 {
+#ifdef CONFIG_PCI
 	int i;
+#endif
 	struct pnp_resources *res;
 	struct pnp_irq *ptr;
 	res = pnp_find_resources(dev,depnum);

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds