* [U-Boot-Users] Problems with the port to PPC 405EP ev. board
@ 2003-03-17 15:11 llandre
2003-03-17 15:23 ` Vladimir Gurevich
[not found] ` <5.1.1.6.0.20030317171757.00a81480@192.168.2.1>
0 siblings, 2 replies; 4+ messages in thread
From: llandre @ 2003-03-17 15:11 UTC (permalink / raw)
To: u-boot
I'm porting U-Boot to this board.
I compared the boot code (start.S) with the OpenBios' startup (entry.S).
Openbios, in case PLL is not initialized, writes default values into the
PLLMR0 and PLLMR1 registers. U-Boot, in my understanding, assumes PLL is
already initialized and read its registers in order to determine the
frequencies and other information (get_sys_info in file speed.c). Is it
correct? If yes, why does not it initialize the PLL?
I tried to add some code to start.S in order to do the same thing openbios
does, but the assembler does not like it:
---------------------------------------------------------------------------
ppc-linux-gcc -Wa,-gstabs -D__ASSEMBLY__ -g -Os -mrelocatable -ffixed-r14
-meabi -mrelocatable -ffixed-r14 -meabi -D__KERNEL__ -DTEXT_BASE=0xFFF80000
-I/proj/ppc405ep/sw/u-boot/include -fno-builtin -pipe -DCONFIG_PPC
-D__powerpc__ -DCONFIG_4xx -ffixed-r2 -ffixed-r29 -mstring -Wa,-m405
-mcpu=405 -msoft-float -DCONFIG_PPC -D__powerpc__ -DCONFIG_4xx -ffixed-r2
-ffixed-r29 -mstring -Wa,-m405 -mcpu=405 -msoft-float -c -o start.o
/proj/ppc405ep/sw/u-boot/cpu/ppc4xx/start.S
/proj/ppc405ep/sw/u-boot/cpu/ppc4xx/start.S: Assembler messages:
/proj/ppc405ep/sw/u-boot/cpu/ppc4xx/start.S:849: Error: attempt to
.org/.space backwards? (-8)
/proj/ppc405ep/sw/u-boot/cpu/ppc4xx/start.S:1582: Internal error!
Assertion failure in write_contents at ../../gas/write.c line 1187.
Please report this bug.
make[1]: *** [start.o] Error 1
make[1]: Leaving directory `/proj/ppc405ep/sw/u-boot/cpu/ppc4xx'
make: *** [subdirs] Error 1
---------------------------------------------------------------------------
I never saw an error like that. Anybody can tell me what is wrong?
Thanx in advance.
llandre
^ permalink raw reply [flat|nested] 4+ messages in thread
* [U-Boot-Users] Problems with the port to PPC 405EP ev. board
2003-03-17 15:11 [U-Boot-Users] Problems with the port to PPC 405EP ev. board llandre
@ 2003-03-17 15:23 ` Vladimir Gurevich
[not found] ` <5.1.1.6.0.20030317171757.00a81480@192.168.2.1>
1 sibling, 0 replies; 4+ messages in thread
From: Vladimir Gurevich @ 2003-03-17 15:23 UTC (permalink / raw)
To: u-boot
Hello,
I'd guess that attaching your new start.S or, at least,
the diff to the original one will definitely help...
Best regards,
Vladimir
llandre wrote:
> ppc-linux-gcc -Wa,-gstabs -D__ASSEMBLY__ -g -Os -mrelocatable
> -ffixed-r14 -meabi -mrelocatable -ffixed-r14 -meabi -D__KERNEL__
> -DTEXT_BASE=0xFFF80000 -I/proj/ppc405ep/sw/u-boot/include -fno-builtin
> -pipe -DCONFIG_PPC -D__powerpc__ -DCONFIG_4xx -ffixed-r2 -ffixed-r29
> -mstring -Wa,-m405 -mcpu=405 -msoft-float -DCONFIG_PPC -D__powerpc__
> -DCONFIG_4xx -ffixed-r2 -ffixed-r29 -mstring -Wa,-m405 -mcpu=405
> -msoft-float -c -o start.o /proj/ppc405ep/sw/u-boot/cpu/ppc4xx/start.S
> /proj/ppc405ep/sw/u-boot/cpu/ppc4xx/start.S: Assembler messages:
> /proj/ppc405ep/sw/u-boot/cpu/ppc4xx/start.S:849: Error: attempt to
> .org/.space backwards? (-8)
> /proj/ppc405ep/sw/u-boot/cpu/ppc4xx/start.S:1582: Internal error!
> Assertion failure in write_contents at ../../gas/write.c line 1187.
> Please report this bug.
> make[1]: *** [start.o] Error 1
> make[1]: Leaving directory `/proj/ppc405ep/sw/u-boot/cpu/ppc4xx'
> make: *** [subdirs] Error 1
> ---------------------------------------------------------------------------
>
> I never saw an error like that. Anybody can tell me what is wrong?
^ permalink raw reply [flat|nested] 4+ messages in thread
* [U-Boot-Users] Problems with the port to PPC 405EP ev. board
[not found] ` <5.1.1.6.0.20030317171757.00a81480@192.168.2.1>
@ 2003-03-17 18:25 ` Vladimir Gurevich
2003-03-18 15:49 ` llandre
0 siblings, 1 reply; 4+ messages in thread
From: Vladimir Gurevich @ 2003-03-17 18:25 UTC (permalink / raw)
To: u-boot
Hello,
My suggestion will be to move your pll_write and reset_cpu
functions to the bottom of the file, at least past the exception
vectors. Remember, that the execution starts at 0x100
and there are only 256 bytes before the next exception
vector. Your problem is that your code is too big to
be placed in the exception vector space and thus overlapped with
the next vector.
Hope, this helps.
Vladimir
llandre wrote:
>
>> Hello,
>>
>> I'd guess that attaching your new start.S or, at least,
>> the diff to the original one will definitely help...
>
>
> Sorry, I thought we were not allowed to send attachments to the mailing
> list.
> I attached the new start.S. The new code is enclosed by the "#if
> (defined(CONFIG_BUBINGA405EP) && defined(CONFIG_NO_SERIAL_EEPROM))"
> directives.
>
>
>
> llandre
^ permalink raw reply [flat|nested] 4+ messages in thread
* [U-Boot-Users] Problems with the port to PPC 405EP ev. board
2003-03-17 18:25 ` Vladimir Gurevich
@ 2003-03-18 15:49 ` llandre
0 siblings, 0 replies; 4+ messages in thread
From: llandre @ 2003-03-18 15:49 UTC (permalink / raw)
To: u-boot
>My suggestion will be to move your pll_write and reset_cpu functions to
>the bottom of the file, at least past the exception
>vectors. Remember, that the execution starts at 0x100 and there are only
>256 bytes before the next exception vector. Your problem is that your code
>is too big to
>be placed in the exception vector space and thus overlapped with
>the next vector.
You were right. I moved the new code at the end of the file and now the
build is ok.
Many thanks again and best regards.
llandre
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2003-03-18 15:49 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2003-03-17 15:11 [U-Boot-Users] Problems with the port to PPC 405EP ev. board llandre
2003-03-17 15:23 ` Vladimir Gurevich
[not found] ` <5.1.1.6.0.20030317171757.00a81480@192.168.2.1>
2003-03-17 18:25 ` Vladimir Gurevich
2003-03-18 15:49 ` llandre
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.