2.4.31 & latest binutils: asm-problems still there

2.4.31 & latest binutils: asm-problems still there

[Herbert Rosmanith]

good morning,

I've just tried to compile 2.4.31 and it still doesn't compile
cleanly with the latest binutils release.

gcc -D__KERNEL__ -I/data/root/linux-2.4.31/include -Wall -Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common -fomit-frame-pointer -pipe -mpreferred-stack-boundary=2 -march=i686 -malign-functions=4    -nostdinc -iwithprefix include -DKBUILD_BASENAME=process  -c -o process.o process.c
{standard input}: Assembler messages:
{standard input}:750: Error: suffix or operands invalid for `mov'
{standard input}:751: Error: suffix or operands invalid for `mov'
{standard input}:845: Error: suffix or operands invalid for `mov'
{standard input}:846: Error: suffix or operands invalid for `mov'
{standard input}:897: Error: suffix or operands invalid for `mov'
{standard input}:898: Error: suffix or operands invalid for `mov'
{standard input}:900: Error: suffix or operands invalid for `mov'
{standard input}:912: Error: suffix or operands invalid for `mov'

alessandro suardi told me that this problem is solved using the
patch from:
  http://www.kernel.org/pub/linux/devel/binutils/linux-2.4-seg-4.patch

which are dated from march (2005-03-27) and therefore, about 3 months
old.

it's about time this gets into the official kernel. who is in charge
of it? (it's obviously not sufficient to report to lkml).

best regards,
herbert rosmanith

Re: 2.4.31 & latest binutils: asm-problems still there

[Marcelo Tosatti]

On Sat, Jun 04, 2005 at 05:29:31AM +0200, Herbert Rosmanith wrote:
> 
> good morning,
> 
> I've just tried to compile 2.4.31 and it still doesn't compile
> cleanly with the latest binutils release.
> 
> gcc -D__KERNEL__ -I/data/root/linux-2.4.31/include -Wall -Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common -fomit-frame-pointer -pipe -mpreferred-stack-boundary=2 -march=i686 -malign-functions=4    -nostdinc -iwithprefix include -DKBUILD_BASENAME=process  -c -o process.o process.c
> {standard input}: Assembler messages:
> {standard input}:750: Error: suffix or operands invalid for `mov'
> {standard input}:751: Error: suffix or operands invalid for `mov'
> {standard input}:845: Error: suffix or operands invalid for `mov'
> {standard input}:846: Error: suffix or operands invalid for `mov'
> {standard input}:897: Error: suffix or operands invalid for `mov'
> {standard input}:898: Error: suffix or operands invalid for `mov'
> {standard input}:900: Error: suffix or operands invalid for `mov'
> {standard input}:912: Error: suffix or operands invalid for `mov'
> 
> alessandro suardi told me that this problem is solved using the
> patch from:
>   http://www.kernel.org/pub/linux/devel/binutils/linux-2.4-seg-4.patch
> 
> which are dated from march (2005-03-27) and therefore, about 3 months
> old.
> 
> it's about time this gets into the official kernel. who is in charge
> of it? (it's obviously not sufficient to report to lkml).

Looks OK except that one "movl" conversion was forgotten in 
the x86-64 diff:

@@ -609,7 +609,7 @@ struct task_struct *__switch_to(struct t
 	}
 	{
 		unsigned gsindex;
-		asm volatile("movl %%gs,%0" : "=g" (gsindex)); 
+		asm volatile("movl %%gs,%0" : "=r" (gsindex)); 
 		if (unlikely((gsindex | next->gsindex) || prev->gs)) {

Who wrote the patch? 

Re: 2.4.31 & latest binutils: asm-problems still there

[Willy Tarreau]

On Sun, Jun 05, 2005 at 03:16:32PM -0300, Marcelo Tosatti wrote:
> On Sat, Jun 04, 2005 at 05:29:31AM +0200, Herbert Rosmanith wrote:
(...)
> > alessandro suardi told me that this problem is solved using the
> > patch from:
> >   http://www.kernel.org/pub/linux/devel/binutils/linux-2.4-seg-4.patch
> > 
> > which are dated from march (2005-03-27) and therefore, about 3 months
> > old.
> > 
> > it's about time this gets into the official kernel. who is in charge
> > of it? (it's obviously not sufficient to report to lkml).
> 
> Looks OK except that one "movl" conversion was forgotten in 
> the x86-64 diff:
> 
> @@ -609,7 +609,7 @@ struct task_struct *__switch_to(struct t
>  	}
>  	{
>  		unsigned gsindex;
> -		asm volatile("movl %%gs,%0" : "=g" (gsindex)); 
> +		asm volatile("movl %%gs,%0" : "=r" (gsindex)); 
>  		if (unlikely((gsindex | next->gsindex) || prev->gs)) {
> 
> Who wrote the patch? 

I believe it's H.J. Lu (CC'd).

Willy

Re: 2.4.31 & latest binutils: asm-problems still there

[H. J. Lu]

On Mon, Jun 06, 2005 at 01:34:59AM +0200, Willy Tarreau wrote:
> On Sun, Jun 05, 2005 at 03:16:32PM -0300, Marcelo Tosatti wrote:
> > On Sat, Jun 04, 2005 at 05:29:31AM +0200, Herbert Rosmanith wrote:
> (...)
> > > alessandro suardi told me that this problem is solved using the
> > > patch from:
> > >   http://www.kernel.org/pub/linux/devel/binutils/linux-2.4-seg-4.patch
> > > 
> > > which are dated from march (2005-03-27) and therefore, about 3 months
> > > old.
> > > 
> > > it's about time this gets into the official kernel. who is in charge
> > > of it? (it's obviously not sufficient to report to lkml).
> > 
> > Looks OK except that one "movl" conversion was forgotten in 
> > the x86-64 diff:
> > 
> > @@ -609,7 +609,7 @@ struct task_struct *__switch_to(struct t
> >  	}
> >  	{
> >  		unsigned gsindex;
> > -		asm volatile("movl %%gs,%0" : "=g" (gsindex)); 
> > +		asm volatile("movl %%gs,%0" : "=r" (gsindex)); 
> >  		if (unlikely((gsindex | next->gsindex) || prev->gs)) {
> > 
> > Who wrote the patch? 
> 
> I believe it's H.J. Lu (CC'd).
> 

Is there a problem?


H.J.