Skip to contentSkip to conversation

show_boot_progess @ ppc not working

U-Boot Applied Andre Schwarz 5 participants 15 messages Dec 9, 2008 – Jul 9, 2009

lore
Share & tools
lore
#0 lore raw

All,

can someone tell me why the board specific function "void show_boot_progress(int arg)"
is no longer called (at least on MPC5200).

Of course this line is present in the board config :

#define CONFIG_SHOW_BOOT_PROGRESS 1

What have I missed ?

regards,
Andr?

MATRIX VISION GmbH, Talstra?e 16, DE-71570 Oppenweiler - Registergericht: Amtsgericht Stuttgart, HRB 271090 Gesch?ftsf?hrer: Gerhard Thullner, Werner Armingeon, Uwe Furtner

#1 lore raw

Hello Andre,

Quoted 6 lines · Andre Schwarz

All,

can someone tell me why the board specific function "void show_boot_progress(int arg)"
is no longer called (at least on MPC5200).

Quoted 7 lines

Of course this line is present in the board config :

#define CONFIG_SHOW_BOOT_PROGRESS 1

What have I missed ?

The show_boot_progress () function was changed in a weak function, see commit:

make show_boot_progress () weak.

author Heiko Schocher <[email protected]>

     Fri, 13 Jul 2007 07:54:17 +0000 (09:54 +0200)
committer    Heiko Schocher <[email protected]>
     Fri, 13 Jul 2007 07:54:17 +0000 (09:54 +0200)

commit fad63407154f46246ce80d53a9c669a44362ac67

For what board exactly is it not working?

Maybe a problem with your compiler?

bye
Heiko

#2 lore raw
Quoted 23 lines · Heiko Schocher

Hello Andre,

Quoted 6 lines · Andre Schwarz

All,

can someone tell me why the board specific function "void show_boot_progress(int arg)"
is no longer called (at least on MPC5200).

Quoted 8 lines

Of course this line is present in the board config :

#define CONFIG_SHOW_BOOT_PROGRESS 1

What have I missed ?

The show_boot_progress () function was changed in a weak function, see commit:

make show_boot_progress () weak.

Quoted 8 lines

author Heiko Schocher <[email protected]>

     Fri, 13 Jul 2007 07:54:17 +0000 (09:54 +0200)
committer    Heiko Schocher <[email protected]>
     Fri, 13 Jul 2007 07:54:17 +0000 (09:54 +0200)

commit fad63407154f46246ce80d53a9c669a44362ac67

For what board exactly is it not working?

mvBC-P (5200 based) -> board/matrix_vision/mvbc_p

Quoted 2 lines

Maybe a problem with your compiler?

Don't know - it's ELDK-4.2 :

ppc_6xx-gcc -v gives :

Reading specs from
/home/eldk-4.2/usr/bin/../lib/gcc/powerpc-linux/4.2.2/specs
Target: powerpc-linux
Configured with:
/opt/eldk/build/ppc-2008-04-01/work/usr/src/denx/BUILD/crosstool-0.43/build/gcc-4.2.2-glibc-20070515T2025-eldk/powerpc-linux/gcc-4.2.2/configure --target=powerpc-linux --host=i686-host_pc-linux-gnu --prefix=/var/tmp/eldk.UZpAG7/usr/crosstool/gcc-4.2.2-glibc-20070515T2025-eldk/powerpc-linux
--disable-hosted-libstdcxx --with-headers=/var/tmp/eldk.UZpAG7/usr/crosstool/gcc-4.2.2-glibc-20070515T2025-eldk/powerpc-linux/powerpc-linux/include --with-local-prefix=/var/tmp/eldk.UZpAG7/usr/crosstool/gcc-4.2.2-glibc-20070515T2025-eldk/powerpc-linux/powerpc-linux --disable-nls --enable-threads=posix --enable-symvers=gnu
--enable-__cxa_atexit --enable-languages=c,c++,java --enable-shared
--enable-c99 --enable-long-long --without-x
Thread model: posix
gcc version 4.2.2

Quoted 3 lines

bye
Heiko

MATRIX VISION GmbH, Talstra?e 16, DE-71570 Oppenweiler - Registergericht: Amtsgericht Stuttgart, HRB 271090 Gesch?ftsf?hrer: Gerhard Thullner, Werner Armingeon, Uwe Furtner

#3 lore raw
In reply to Andre Schwarz

Hello Andre,

Quoted 6 lines · Andre Schwarz

All,

can someone tell me why the board specific function "void show_boot_progress(int arg)"
is no longer called (at least on MPC5200).

Quoted 7 lines

Of course this line is present in the board config :

#define CONFIG_SHOW_BOOT_PROGRESS 1

What have I missed ?

Nothing! Good catch. I tried it on a mpc8xx and on a mpx82xx based board and it also didnt worked :-(

Can you try the following patch? (This patch solved it on my boards ...)

thanks,
Heiko

[PATCH] all platforms: make show_boot_progress () again working.

---
 include/common.h |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
include/common.h +0 -0
index 5968036..6583f00 100644
@@ -693,7 +693,7 @@ int    pcmcia_init (void);
 /*
  * Board-specific Platform code can reimplement show_boot_progress ()

if needed
*/
-void __attribute__((weak)) show_boot_progress (int val); +void show_boot_progress (int val);

#ifdef CONFIG_INIT_CRITICAL
#error CONFIG_INIT_CRITICAL is deprecated!

Signature · 1 line
1.5.6.1
#4 lore raw
Quoted 45 lines · Heiko Schocher

Hello Andre,

Quoted 6 lines · Andre Schwarz

All,

can someone tell me why the board specific function "void show_boot_progress(int arg)"
is no longer called (at least on MPC5200).

Quoted 8 lines

Of course this line is present in the board config :

#define CONFIG_SHOW_BOOT_PROGRESS 1

What have I missed ?

Nothing! Good catch. I tried it on a mpc8xx and on a mpx82xx based board and it also didnt worked :-(

Can you try the following patch? (This patch solved it on my boards ...)

thanks,
Heiko

[PATCH] all platforms: make show_boot_progress () again working.

---
 include/common.h |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
include/common.h +0 -0
index 5968036..6583f00 100644
@@ -693,7 +693,7 @@ int    pcmcia_init (void);
 /*
  * Board-specific Platform code can reimplement show_boot_progress ()

if needed
*/

include/common.h

Quoted 5 lines
include/common.h +1 -1
-void __attribute__((weak)) show_boot_progress (int val);
+void show_boot_progress (int val);
 
 #ifdef CONFIG_INIT_CRITICAL
 #error CONFIG_INIT_CRITICAL is deprecated!

Heiko,

of course this patch makes it work again.
But the question is : Why is this specific weak function not replaced by the board specific one ?

This gives me some pain when looking at the other weak functions....

In order to be able to compile u-boot I also (still) have to change <config.h> into "../include/config.h" inside common/env_embedded.c. Otherwise the config file is not taken into account and complains about CFG_ENV_SIZE undeclared.

I've discussed this with Wolfgang already a few weeks ago - he blamed my system configuration.
This issues occur on my gentoo and ubuntu system.
There's nothing special about this systems (out of the box + ELDK-4.2).

How can we solve this without making a guessing game out of it ?

regards,
Andr?

MATRIX VISION GmbH, Talstra?e 16, DE-71570 Oppenweiler - Registergericht: Amtsgericht Stuttgart, HRB 271090 Gesch?ftsf?hrer: Gerhard Thullner, Werner Armingeon, Uwe Furtner

#5 lore raw

Hi All,

I had some 'fun' with weak functions a little while ago, and I'm wondering if maybe we should fix up a few inconsistencies...

Quoted 10 lines · Andre Schwarz
Quoted 9 lines · Heiko Schocher

Hello Andre,

Quoted 5 lines · Andre Schwarz

All,

can someone tell me why the board specific function "void show_boot_progress(int arg)"
is no longer called (at least on MPC5200).

<snip>

Quoted 4 lines
Quoted 4 lines

Can you try the following patch? (This patch solved it on my boards ...)

<snip>

Quoted 2 lines

Review

Quoted 2 lines
(quoted diff) +1 -1
-void __attribute__((weak)) show_boot_progress (int val);
+void show_boot_progress (int val);

This makes show_boot_progress () consistent with nearly every other weak function (see below)

Interesting that this is the only function defined weak in this way. All other weak functions are defined normally (no attributes) and then declared weak with an alias. Good examples can be found in /include/status_led.h lines 391-396 and /lib_arm/board.c lines 125-138

<snip>

Quoted 3 lines

Heiko,

of course this patch makes it work again.

This patch leaves only a handful of inconsistent weak functions:

/include/asm-avr32/arch-at32ap700x/clk.h line 85:

extern void gclk_init(void) __attribute__((weak));

But this is not designed to be overridable - It is designed to not require a declaration - See /cpu/at32ap/cpu.c lines 68-69:

	if(gclk_init)
		gclk_init();

/common/cmd_boot.c lines 32-36: (/common/cmd_elf.c lines 31-54 and /cpu/mips/cpu.c lines 41-43 are same semantics)

__attribute__((weak))
unsigned long do_go_exec (ulong (*entry)(int, char *[]), int argc, char *argv[])

{
	return entry (argc, argv);
}

<snip>

Quoted 2 lines

But the question is : Why is this specific weak function not replaced by the board specific one ?

Good question - maybe because this function is inlined? (just a wild guess)
- could be any of a number of reasons as far as I can tell.

Quoted 2 lines

This gives me some pain when looking at the other weak functions....

Is there any reason _not_ to fix up these last four weak functions so that all weak functions are defined and declared the same way? I can do up a patch if the consensus is to make them consistent.

Regards,

Graeme

#6 lore raw
Quoted 85 lines · Graeme Russ

Hi All,

I had some 'fun' with weak functions a little while ago, and I'm wondering if maybe we should fix up a few inconsistencies...

Quoted 13 lines · Andre Schwarz
Quoted 11 lines · Heiko Schocher

Hello Andre,

Quoted 6 lines · Andre Schwarz

All,

can someone tell me why the board specific function "void show_boot_progress(int arg)"
is no longer called (at least on MPC5200).

<snip>

Quoted 4 lines
Quoted 4 lines

Can you try the following patch? (This patch solved it on my boards ...)

<snip>

Quoted 3 lines

Review

Quoted 3 lines
(quoted diff) +1 -1
-void __attribute__((weak)) show_boot_progress (int val);
+void show_boot_progress (int val);

This makes show_boot_progress () consistent with nearly every other weak function (see below)

Interesting that this is the only function defined weak in this way. All other weak functions are defined normally (no attributes) and then declared weak with an alias. Good examples can be found in /include/status_led.h lines 391-396 and /lib_arm/board.c lines 125-138

<snip>

Quoted 4 lines

Heiko,

of course this patch makes it work again.

This patch leaves only a handful of inconsistent weak functions:

/include/asm-avr32/arch-at32ap700x/clk.h line 85:

extern void gclk_init(void) __attribute__((weak));

But this is not designed to be overridable - It is designed to not require a declaration - See /cpu/at32ap/cpu.c lines 68-69:

	if(gclk_init)
		gclk_init();

/common/cmd_boot.c lines 32-36: (/common/cmd_elf.c lines 31-54 and /cpu/mips/cpu.c lines 41-43 are same semantics)

__attribute__((weak))
unsigned long do_go_exec (ulong (*entry)(int, char *[]), int argc, char *argv[])

{
	return entry (argc, argv);
}

<snip>

Quoted 3 lines

But the question is : Why is this specific weak function not replaced by the board specific one ?

Good question - maybe because this function is inlined? (just a wild guess)
- could be any of a number of reasons as far as I can tell.

Quoted 8 lines
Quoted 2 lines

This gives me some pain when looking at the other weak functions....

Is there any reason _not_ to fix up these last four weak functions so that all weak functions are defined and declared the same way? I can do up a patch if the consensus is to make them consistent.

Please send a proper patch with explanation.
Since I can not explain this effect 100% I'm not able to fix it.

Quoted 4 lines

Regards,

Graeme

MATRIX VISION GmbH, Talstra?e 16, DE-71570 Oppenweiler - Registergericht: Amtsgericht Stuttgart, HRB 271090 Gesch?ftsf?hrer: Gerhard Thullner, Werner Armingeon, Uwe Furtner

#7 lore raw
In reply to Heiko Schocher

Heiko,

this issue looks still unfixed and is somewhat annoying.

Im unable to use show_boot_progress with this weak definition in include/common.h.

Simply re-defining the prototype in my local repo is not reasonable.

How did you solve it ?

Regards,
Andr??

Quoted 42 lines · Heiko Schocher

Hello Andre,

Quoted 6 lines · Andre Schwarz

All,

can someone tell me why the board specific function "void show_boot_progress(int arg)"
is no longer called (at least on MPC5200).

Quoted 7 lines

Of course this line is present in the board config :

#define CONFIG_SHOW_BOOT_PROGRESS 1

What have I missed ?

Nothing! Good catch. I tried it on a mpc8xx and on a mpx82xx based board and it also didnt worked :-(

Can you try the following patch? (This patch solved it on my boards ...)

thanks,
Heiko

[PATCH] all platforms: make show_boot_progress () again working.

---
 include/common.h |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
include/common.h +0 -0
index 5968036..6583f00 100644
@@ -693,7 +693,7 @@ int    pcmcia_init (void);
 /*
  * Board-specific Platform code can reimplement show_boot_progress ()

if needed
*/

include/common.h

Quoted 5 lines
include/common.h +1 -1
-void __attribute__((weak)) show_boot_progress (int val);
+void show_boot_progress (int val);
 
 #ifdef CONFIG_INIT_CRITICAL
 #error CONFIG_INIT_CRITICAL is deprecated!

MATRIX VISION GmbH, Talstra?e 16, DE-71570 Oppenweiler
Registergericht: Amtsgericht Stuttgart, HRB 271090
Gesch?ftsf?hrer: Gerhard Thullner, Werner Armingeon, Uwe Furtner, Hans-Joachim Reich

#8 lore raw

Hello Andr?,

sorry for the late reply, I was on vacation ...

Quoted 1 line · Andr? Schwarz

this issue looks still unfixed and is somewhat annoying.

Seems that this fix didn;t find the way in mainline.

Quoted 4 lines

Im unable to use show_boot_progress with this weak definition in include/common.h.

Simply re-defining the prototype in my local repo is not reasonable.

Hmm... as I can remember, the patch I posted solved it on my boards on which I tried it ...

Quoted 1 line

How did you solve it ?

With the Patch I posted ... I try it with actual code tomorrow again ...

bye
Heiko

Signature · 2 lines
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
#9 lore raw
Quoted 8 lines · Heiko Schocher

Hello Andr??,

sorry for the late reply, I was on vacation ...

Quoted 1 line · Andr?? Schwarz

this issue looks still unfixed and is somewhat annoying.

Seems that this fix didn;t find the way in mainline.

What fix ? removing the _weak_ attribute ?

Quoted 8 lines
Quoted 5 lines

Im unable to use show_boot_progress with this weak definition in include/common.h.

Simply re-defining the prototype in my local repo is not reasonable.

Hmm... as I can remember, the patch I posted solved it on my boards on which I tried it ...

removing the weak definition works for me ... but of course I don't want to need an extra patch. It should simply work out-of-the-box.

Quoted 5 lines
Quoted 2 lines

How did you solve it ?

With the Patch I posted ... I try it with actual code tomorrow again ...

yes, please.

Regards,
Andr??

Quoted 3 lines

bye
Heiko

MATRIX VISION GmbH, Talstra?e 16, DE-71570 Oppenweiler
Registergericht: Amtsgericht Stuttgart, HRB 271090
Gesch?ftsf?hrer: Gerhard Thullner, Werner Armingeon, Uwe Furtner, Hans-Joachim Reich

#10 lore raw

Hello Andr?,

Quoted 10 lines · Andr? Schwarz
Quoted 7 lines · Heiko Schocher

Hello Andr?,

sorry for the late reply, I was on vacation ...

Quoted 1 line · Andr? Schwarz

this issue looks still unfixed and is somewhat annoying.

Seems that this fix didn;t find the way in mainline.

What fix ? removing the _weak_ attribute ?

Yep.

Quoted 9 lines
Quoted 6 lines
Quoted 4 lines

Im unable to use show_boot_progress with this weak definition in include/common.h.

Simply re-defining the prototype in my local repo is not reasonable.

Hmm... as I can remember, the patch I posted solved it on my boards on which I tried it ...

removing the weak definition works for me ... but of course I don't want to need an extra patch. It should simply work out-of-the-box.

So, the only problem is/was, that this patch didn;t found his way in mainline ... Wolfgang, could you please pick up this patch, when you are back from vacation?

Thanks

bye
Heiko

Signature · 2 lines
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
#11 lore raw

Dear Heiko,

Quoted 7 lines · In message <[email protected]> you wrote:
Quoted 3 lines

removing the weak definition works for me ... but of course I don't want to need an extra patch. It should simply work out-of-the-box.

So, the only problem is/was, that this patch didn;t found his way in mainline ... Wolfgang, could you please pick up this patch, when you are back from vacation?

Ummm... seems I lost traces of this patch. Do you have a reference for me (date and subject, or URL to ML archive or similar)?

Thanks.

Best regards,

Wolfgang Denk

Signature · 5 lines
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
Well, the way I see it, logic is only a way of being ignorant by num-
bers.                                 - Terry Pratchett, _Small Gods_
#12 lore raw

Hello Wolfgang,

Quoted 9 lines · Wolfgang Denk
Quoted 5 lines · In message <[email protected]> you wrote:
Quoted 2 lines

removing the weak definition works for me ... but of course I don't want to need an extra patch. It should simply work out-of-the-box.

So, the only problem is/was, that this patch didn;t found his way in mainline ... Wolfgang, could you please pick up this patch, when you are back from vacation?

Ummm... seems I lost traces of this patch. Do you have a reference for me (date and subject, or URL to ML archive or similar)?

http://lists.denx.de/pipermail/u-boot/2008-December/044556.html

bye
Heiko

Signature · 2 lines
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
#13 lore raw
In reply to Heiko Schocher

Dear Heiko Schocher,

Quoted 4 lines · In message <[email protected]> you wrote:

[PATCH] all platforms: make show_boot_progress () again working.

Applied, but...

Quoted 11 lines
 include/common.h |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
include/common.h +0 -0
index 5968036..6583f00 100644
@@ -693,7 +693,7 @@ int    pcmcia_init (void);
 /*
  * Board-specific Platform code can reimplement show_boot_progress ()

if needed

^^^^^^^^^^^^^^^^^

... your patch was line-wrapped and I had to manually apply it.

Best regards,

Wolfgang Denk

Signature · 6 lines
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
"Where humor is concerned there are no standards -- no  one  can  say
what is good or bad, although you can be sure that everyone will.
- John Kenneth Galbraith
#14 lore raw

Hello Wolfgang,

Quoted 20 lines · Wolfgang Denk
Quoted 3 lines · In message <[email protected]> you wrote:

[PATCH] all platforms: make show_boot_progress () again working.

Applied, but...

Quoted 11 lines
 include/common.h |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)
include/common.h +0 -0
index 5968036..6583f00 100644
@@ -693,7 +693,7 @@ int    pcmcia_init (void);
 /*
  * Board-specific Platform code can reimplement show_boot_progress ()

if needed

^^^^^^^^^^^^^^^^^

Quoted 1 line

... your patch was line-wrapped and I had to manually apply it.

Ups, sorry. Seems at this time I made this patch, I didn;t had setup correct my mailer. Thanks for fixing.

bye
Heiko

Signature · 2 lines
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany