Re: arm intr questions

> - who uses irqblock[]?

as it seems there is nothing using irqblock, i'd like to remove it.
can anyone review the attached patch?  thanks.

YAMAMOTO Takashi

Index: arch/shark/isa/isa_irqhandler.c
===================================================================
RCS file: /cvsroot/src/sys/arch/shark/isa/isa_irqhandler.c,v
retrieving revision 1.7
diff -u -p -r1.7 isa_irqhandler.c
--- arch/shark/isa/isa_irqhandler.c	11 Dec 2005 12:19:02 -0000	1.7
+++ arch/shark/isa/isa_irqhandler.c	4 May 2006 12:45:52 -0000
 <at>  <at>  -97,7 +97,6  <at>  <at>  u_int actual_mask;
 u_int disabled_mask;
 u_int spl_mask;
 u_int irqmasks[IPL_LEVELS];
-u_int irqblock[NIRQS];

 extern u_int soft_interrupts;	/* Only so we can initialise it */

 <at>  <at>  -125,7 +124,6  <at>  <at>  irq_init()
 	/* Clear all the IRQ handlers and the irq block masks */
 	for (loop = 0; loop < NIRQS; ++loop) {
 		irqhandlers[loop] = NULL;
-		irqblock[loop] = 0;
 	}

Re: arm intr questions

On Thu, 4 May 2006, YAMAMOTO Takashi wrote:

> > - who uses irqblock[]?
> 
> as it seems there is nothing using irqblock, i'd like to remove it.
> can anyone review the attached patch?  thanks.

It looks reasonable to me.

--

-- 
Ben Harris                                                   <bjh21 <at> NetBSD.org>
Portmaster, NetBSD/acorn26           <URL:http://www.NetBSD.org/Ports/acorn26/>

Nie wieder pleite ...

Hi Matti,

hier der Link zur Lottoseite von der ich dir erzählt habe. Wolln mal schauen, ob einer von uns beiden Glück hat?

http://www.cashmechanic.com

Der Deal steht? wenn einer von uns beiden gewinnt, lädt er den anderen zu einem Cabrio ein. Bei den Gewinnen
sind das ja Peanuts. Ich reserviere mir schon mal meinen neuen Mazda 

Ciao,

Andi

Freu dich auf die Zigarette danach ;)

hallo, heute ist es endlich soweit, ich gebe dir meine handynummer:
01604591711, du bist einer der ersten menschen denen ich sie anvertraue, so können wir immer per sms in
kontakt bleiben, denn ich habe es immer dabei! wirst du mir ganz schnell schreiben, ja?
liebe grüsse, deine anja
> 

gzboot booting problem

Re: gzboot booting problem

Danny Lau wrote:
> Hi,
> 
> I'm trying to run netbsd on the sumsung s3c2410.
> I have been noticed from the gzboot source code(gzboot.c:main()),
> that gzboot will booting kernel image which just concatenated on to the end
> of it when md_root_size = 0.
> So I have concatenated the gzboot image, named gzboot.bin, and the kernel
> image, named netbsd.bin, by the command as follow:
> 
> $ cat gzboot.bin netbsd.bin > netbsd
> 
> But I just can only get messages from gzboot. I track the instructions
> step-by-step via Jtag, and get expected value before jumping to the kernel,
> that is the address where to jump is pointed to the kernel entry.
> Strangely,
> when jumping, it doesn't jump to the address - ENTRY_NP(kernel_text).
> Can anybody help me!

Where does it jump to?

--

-- 
Gillette - the best a man can forget

Re: gzboot booting problem

>> step-by-step via Jtag, and get expected value before jumping to the kernel,
>> that is the address where to jump is pointed to the kernel entry.
>
> Where does it jump to?

And is it the exact instruction expected to run after all?  You have a JTAG tool
at hand.  Have you ever tried to run "plain" netbsd kernel image manually
downloaded at "gunzipped" RAM address?  If it runs ok, gzboot fails to work.

Toru Nishimura/ALKYL Technology

Re: gzboot booting problem

Re: gzboot booting problem

On Mon, May 22, 2006 at 09:22:03PM +0800, Danny Lau wrote:
>    I don't know, it jump to a invalid adress. I am now checking the board
>    to see whether this phenomena is caused by the hardware fault.

Just to be sure...  You're loading 'netbsd' somewhere other
than where the decompressed kernel should go, right?

Have you checked to see that the data at the kernel entry
point is what you expect it to be, and that the jump instruction
is trying to jump to that address?

-allen

--

-- 
                  Use NetBSD!  http://www.NetBSD.org/

Re: gzboot booting problem

On Sat, May 27, 2006 at 12:19:16AM +0800, Danny Lau wrote:
>    The  develop  board  ---NOT  SMDK2410  board,  just engage the s3c2410
>    chip --- I  use  is booted from NAND Flash when power on, so I wrote a
>    piece  of  code  to  copy image from NAND Flash to SDRAM since I can't
>    find similar function in the gzboot.
> 
>    I'm sure the R15 is 0x30200000 (via Jtag) just before jumping, and the
>    plain    kernel   can   be successfully   executed   when   downloaded
>    directly into  that  address by some utilities. So I guess that gzboot
>    decompress  kernel  in  a  wrong  way,  or  the  code I wrote has some
>    problems.

Can you read the RAM as 0x30200000 via JTag?  If so, does it
correspond to the first bytes of the kernel?

Is there other system initialization happening when you boot
with "some utilities" ?

-allen

--

-- 
                  Use NetBSD!  http://www.NetBSD.org/