Question concerning Assembler error

Question concerning Assembler error

[Jeff Harrell]

[-- Attachment #1.1: Type: text/plain, Size: 1321 bytes --]

I am getting a strange error from the assembler that I am not quite sure
what to make of it.  Here
is the  Error and code snippet:


     >make
     mipsel-linux-gcc -D_ASSEMBLER_ -mcpu=r4600 -mips2 -Wall
     -Wstrict-prototypes -O2 -fomit-frame-pointer -G -0
     -mno-abicalls -fno-pic -pipe -mlong-calls -Wimplicit -c
     avreset.S
     avreset.S: Assembler messages:
     avreset.S:262: Error: Rest of line ignored. First ignored
     character is `0'.
     avreset.S:1006: Error: Rest of line ignored. First ignored
     character is `0'.
     gmake: *** [avreset.o] Error 1
     gmake: Target `all' not remade because of errors.


     And here is the code that seems to be causing the problem:

        /* Interrupt : For now we simply disable interrupts and
     return */

             MFC0(   k0, C0_STATUS)
             srl     k0, 1
             sll     k0, 1
             MTC0(   k0, C0_STATUS)
             nop
             .set    mips3
     ==>  eret   <==
             .set    mips2
             nop


Any information that anyone might have would be greatly appreciated.

Regards,
Jeff Harrell

--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Jeff Harrell                    Work:  (801) 619-6104
Broadband Access group/TI
jharrell@ti.com
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



[-- Attachment #1.2: Type: text/html, Size: 1992 bytes --]

[-- Attachment #2: Card for Jeff Harrell --]
[-- Type: text/x-vcard, Size: 376 bytes --]

begin:vcard 
n:Harrell;Jeff
tel;cell:(801) 597-6268
tel;fax:(801) 619-6150
tel;work:(801) 619-6104
x-mozilla-html:TRUE
url:http://www.ti.com
org:Broadband Access Group
version:2.1
email;internet:jharrell@ti.com
title:Texas Instruments
adr;quoted-printable:;;170 West Election Rd. Suite 100	=0D=0AMS 4106		;Draper;Utah;84020-6410;USA
x-mozilla-cpt:;0
fn:Jeff Harrell
end:vcard

Re: Question concerning Assembler error

[H.Heinold]

On Thu, Mar 08, 2001 at 09:20:15AM -0700, Jeff Harrell wrote:
> I am getting a strange error from the assembler that I am not quite sure
> what to make of it.  Here
> is the  Error and code snippet:
> 
> 
>      >make
>      mipsel-linux-gcc -D_ASSEMBLER_ -mcpu=r4600 -mips2 -Wall
>      -Wstrict-prototypes -O2 -fomit-frame-pointer -G -0
>      -mno-abicalls -fno-pic -pipe -mlong-calls -Wimplicit -c

Hm sorry cant help with the assembler problem, but which machine
has a 4600 Prozessor and run mipsel on it?

-- 


Henning Heinold

Re: Question concerning Assembler error

[Ralf Baechle]

On Thu, Mar 08, 2001 at 05:30:03PM +0100, H.Heinold wrote:

> Hm sorry cant help with the assembler problem, but which machine
> has a 4600 Prozessor and run mipsel on it?

RM200C.

  Ralf

Re: Question concerning Assembler error

[Kevin D. Kissell]

> > Hm sorry cant help with the assembler problem, but which machine
> > has a 4600 Prozessor and run mipsel on it?
> 
> RM200C.

Actually, It's far more likely that Jeff is working with a
MIPS 4KC or 5KC CPU core.  -mcpu=r4600 ends up
being the closest fit to the MIPS32 ISA and pipeline
among the options available for the Linux-capable
gcc compilers.

            Kevin K.

Re: Question concerning Assembler error

[Kevin D. Kissell]

> > Hm sorry cant help with the assembler problem, but which machine
> > has a 4600 Prozessor and run mipsel on it?
> 
> RM200C.

Actually, It's far more likely that Jeff is working with a
MIPS 4KC or 5KC CPU core.  -mcpu=r4600 ends up
being the closest fit to the MIPS32 ISA and pipeline
among the options available for the Linux-capable
gcc compilers.

            Kevin K.

Re: Question concerning Assembler error

[Ralf Baechle]

On Thu, Mar 08, 2001 at 05:54:29PM +0100, Kevin D. Kissell wrote:

> > > Hm sorry cant help with the assembler problem, but which machine
> > > has a 4600 Prozessor and run mipsel on it?
> > 
> > RM200C.
> 
> Actually, It's far more likely that Jeff is working with a
> MIPS 4KC or 5KC CPU core.  -mcpu=r4600 ends up
> being the closest fit to the MIPS32 ISA and pipeline
> among the options available for the Linux-capable
> gcc compilers.

Maybe your answer is right, but then the question is wrong :-)

  Ralf

Re: Question concerning Assembler error

[Ralf Baechle]

On Thu, Mar 08, 2001 at 05:54:29PM +0100, Kevin D. Kissell wrote:

> > > Hm sorry cant help with the assembler problem, but which machine
> > > has a 4600 Prozessor and run mipsel on it?
> > 
> > RM200C.
> 
> Actually, It's far more likely that Jeff is working with a
> MIPS 4KC or 5KC CPU core.  -mcpu=r4600 ends up
> being the closest fit to the MIPS32 ISA and pipeline
> among the options available for the Linux-capable
> gcc compilers.

Maybe your answer is right, but then the question is wrong :-)

  Ralf

Re: Question concerning Assembler error

[Ralf Baechle]

On Thu, Mar 08, 2001 at 09:20:15AM -0700, Jeff Harrell wrote:

>         /* Interrupt : For now we simply disable interrupts and
>      return */
> 
>              MFC0(   k0, C0_STATUS)
>              srl     k0, 1
>              sll     k0, 1
>              MTC0(   k0, C0_STATUS)
>              nop
>              .set    mips3
>      ==>  eret   <==
>              .set    mips2
>              nop
> 
> 
> Any information that anyone might have would be greatly appreciated.

I suggest to run this code through the preprocessor only using the
-E -C options.  The output will be somewhat cryptic but explain much better
what's wrong.

  Ralf

mips gcc 2.95.2 and 2.91.66 bug

[Fabrice Bellard]

Hi!

Maybe this bug can interest you: when using byte swaping in le16_to_cpu
for example, mips gcc 2.95.2 and 2.91.66 sometime do not generate correct
code : the u16 to u32 convertion is missing. I found this bug while
compiling drivers/mtd/ftl.c in build_maps(). Here is a sample source to
reproduce the bug:

typedef unsigned short __u16;

extern __inline__ __const__ __u16 le16_to_cpu(__u16 x)
{
    return ((__u16)( \
		(((__u16)(x) & (__u16)0x00ffU) << 8) | \
		(((__u16)(x) & (__u16)0xff00U) >> 8) ));
}

int test(int xtrans, int xvalid, __u16 *ptr)
{
    if ((xvalid+xtrans != le16_to_cpu(*ptr))) {
	return -1;
    }
    return 0;
}

The generated asm is :

test:
	.frame	$sp,0,$31		# vars= 0, regs= 0/0, args= 0,
extra= 0
	.mask	0x00000000,0
	.fmask	0x00000000,0
	.set	noreorder
	.cpload	$25
	.set	reorder
	lhu	$3,0($7)
	addu	$5,$6,$5
	sll	$4,$3,8
	srl	$3,$3,8
	or	$4,$4,$3
	.set	noreorder
	.set	nomacro
	bne	$5,$4,$L7
	li	$2,-1			# 0xffffffff
	.set	macro
	.set	reorder

	move	$2,$0
$L7:
	j	$31
	.end	test

The andi op is missing.

egcs-1.0.3 seems to be OK.

I have no experience in submitting gcc bugs, so if someone could forward
this mail to the relevant gcc mailing list...

Fabrice.

Re: mips gcc 2.95.2 and 2.91.66 bug

[Hiroyuki Machida]

From: Fabrice Bellard <bellard@email.enst.fr>
Subject: mips gcc 2.95.2 and 2.91.66 bug
Date: Thu, 8 Mar 2001 17:43:31 +0100 (MET)

> Hi!
> 
> Maybe this bug can interest you: when using byte swaping in le16_to_cpu
> for example, mips gcc 2.95.2 and 2.91.66 sometime do not generate correct
> code : the u16 to u32 convertion is missing. I found this bug while
> compiling drivers/mtd/ftl.c in build_maps(). Here is a sample source to
> reproduce the bug:

This bug is fixed by following chage in GCC 3.0. 

gcc/gcc/ChangeLog.3:

2000-04-24  Hiroyuki Machida <machida@sm.sony.co.jp>

        * combine.c (try_combine): Update reg_nonzero_bits of
        newi2pat before newpat.


Please refer mail archives at gcc.gnu.org for details.
You can easily apply this fix to gcc-2.95.x.

---
Hiroyuki Machida
Creative Station		SCE Inc.