Re: [U-Boot] Multiple device support -- sorry state :(

2009-04-08 Thread Jerry Van Baren
k...@koi8.net wrote:
 OK, this is _NOT_ just multiple I2C adapters... The entire thing is
 fundamentally broken.
 
 One supposed to have _THE_ device and only this device is somehow supported.
 
 Now it is USB. Each and every USB driver exports the same set of functions,
 submit_XXX_msg(...) That means there can be one and only one USB device in
 the system.

[snip the badly scorched spot ;-)]

 USB keyboard is another grand kludge deserving its own chapter... As of now
 one can only switch to it from command line because USB is not even
 initialized until do_usb() from cmd_usb.c is called... What if we do NOT
 have a serial console at all?

Dumb question because I have not looked seriously at the v2 fork of 
u-boot: how does the v2 fork handle this?  Better?  Since the v2 fork 
uses (or is close to) the linux driver model, I would expect it to be 
better.

Best regards,
gvb
___
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot


Re: [U-Boot] Multiple device support -- sorry state :(

2009-04-08 Thread ksi
On Wed, 8 Apr 2009, Jerry Van Baren wrote:

 k...@koi8.net wrote:
  OK, this is _NOT_ just multiple I2C adapters... The entire thing is
  fundamentally broken.
  
  One supposed to have _THE_ device and only this device is somehow
 supported.
  
  Now it is USB. Each and every USB driver exports the same set of
 functions,
  submit_XXX_msg(...) That means there can be one and only one USB
 device in
  the system.
 
 [snip the badly scorched spot ;-)]
 
  USB keyboard is another grand kludge deserving its own chapter... As
 of now
  one can only switch to it from command line because USB is not even
  initialized until do_usb() from cmd_usb.c is called... What if we do
 NOT
  have a serial console at all?
 
 Dumb question because I have not looked seriously at the v2 fork of
 u-boot:
 how does the v2 fork handle this?  Better?  Since the v2 fork uses (or
 is
 close to) the linux driver model, I would expect it to be better.

Nothing's changed here...

And that is _NOT_ just USB, it is everywhere -- I2C, SPI, etc. It is stiff,
inflexible, everything written in stone and that stone is permanently set in
concrete.

---
**
*  k...@homeKOI8 Net The impossible we do immediately.  *
*  Las Vegas   NV, USA  Miracles require 24-hour notice.   *
**
___
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot


Re: [U-Boot] Multiple device support -- sorry state :(

2009-04-08 Thread Robert Schwebel
Hi Jerry,

On Wed, Apr 08, 2009 at 03:19:19PM -0400, Jerry Van Baren wrote:
 k...@koi8.net wrote:
  OK, this is _NOT_ just multiple I2C adapters... The entire thing is
  fundamentally broken.
 
  One supposed to have _THE_ device and only this device is somehow supported.
 
  Now it is USB. Each and every USB driver exports the same set of functions,
  submit_XXX_msg(...) That means there can be one and only one USB device in
  the system.

 [snip the badly scorched spot ;-)]

  USB keyboard is another grand kludge deserving its own chapter... As of now
  one can only switch to it from command line because USB is not even
  initialized until do_usb() from cmd_usb.c is called... What if we do NOT
  have a serial console at all?

 Dumb question because I have not looked seriously at the v2 fork of
 u-boot: how does the v2 fork handle this? Better? Since the v2 fork
 uses (or is close to) the linux driver model, I would expect it to be
 better.

Right, we use something close to the driver model, so no problem with
that in v2 land. Other than linux, we have per-device parameters.

Example, with a freshly checked out tree:

  r...@thebe:u-boot-v2$ make sandbox_defconfig
  r...@thebe:u-boot-v2$ make
  r...@thebe:u-boot-v2$ ./scripts/ubootenv -s -p 0x1 examples/environment/ 
env.bin

That gets you an u-boot-v2 (user mode u-boot) plus the example
environment.

Let's start it:

  r...@thebe:u-boot-v2-denx$ ./uboot -e env.bin
  add file env.bin()


  U-Boot 2.0.0-rc8-00084-g97568fd (Apr  8 2009 - 21:54:30)

  Board: sandbox
  Malloc space: 0xb761f008 - 0xb7e1f008 (size  8 MB)
  Open /dev/env0 Function not implemented
  running /env/bin/init...
  not found
  uboot:/

Let's check which devices we have:

  uboot:/ devinfo
  devices:
  |env0
  |console0
  |mem
  |eth0
  |filesystem: /
  |filesystem: /dev

  drivers:
 console
   cfi_flash
 mem
 ram
 rom
  cramfs
   ramfs
   devfs
hostfile

Let's check what parameters are there:

  uboot:/ devinfo /dev/console0
  base  : 0x
  size  : 0x
  driver: console

  no info available for console0
  Parameters:
active = ioe

So you can have as many devices as you want, plus each device has parameters
attached to it, so it's completely object oriented, without any global variable
crap.

rsc
-- 
Pengutronix e.K.   | |
Industrial Linux Solutions | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0|
Amtsgericht Hildesheim, HRA 2686   | Fax:   +49-5121-206917- |
___
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot


Re: [U-Boot] Multiple device support -- sorry state :(

2009-04-08 Thread Robert Schwebel
On Wed, Apr 08, 2009 at 12:25:16PM -0700, k...@koi8.net wrote:
 On Wed, 8 Apr 2009, Jerry Van Baren wrote:
 
  k...@koi8.net wrote:
   OK, this is _NOT_ just multiple I2C adapters... The entire thing is
   fundamentally broken.
   
   One supposed to have _THE_ device and only this device is somehow
  supported.
   
   Now it is USB. Each and every USB driver exports the same set of
  functions,
   submit_XXX_msg(...) That means there can be one and only one USB
  device in
   the system.
  
  [snip the badly scorched spot ;-)]
  
   USB keyboard is another grand kludge deserving its own chapter... As
  of now
   one can only switch to it from command line because USB is not even
   initialized until do_usb() from cmd_usb.c is called... What if we do
  NOT
   have a serial console at all?
  
  Dumb question because I have not looked seriously at the v2 fork of
  u-boot:
  how does the v2 fork handle this?  Better?  Since the v2 fork uses (or
  is
  close to) the linux driver model, I would expect it to be better.
 
 Nothing's changed here...
 
 And that is _NOT_ just USB, it is everywhere -- I2C, SPI, etc. It is stiff,
 inflexible, everything written in stone and that stone is permanently set in
 concrete.

I suppose you didn't look in the right place. We don't even have support
for i2c and spi in v2 :-)

rsc
-- 
Pengutronix e.K.   | |
Industrial Linux Solutions | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0|
Amtsgericht Hildesheim, HRA 2686   | Fax:   +49-5121-206917- |
___
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot


Re: [U-Boot] Multiple device support -- sorry state :(

2009-04-08 Thread ksi
On Wed, 8 Apr 2009, Robert Schwebel wrote:

 On Wed, Apr 08, 2009 at 12:25:16PM -0700, k...@koi8.net wrote:
  On Wed, 8 Apr 2009, Jerry Van Baren wrote:
  
   k...@koi8.net wrote:
OK, this is _NOT_ just multiple I2C adapters... The entire thing
 is
fundamentally broken.

One supposed to have _THE_ device and only this device is somehow
   supported.

Now it is USB. Each and every USB driver exports the same set of
   functions,
submit_XXX_msg(...) That means there can be one and only one USB
   device in
the system.
   
   [snip the badly scorched spot ;-)]
   
USB keyboard is another grand kludge deserving its own chapter...
 As
   of now
one can only switch to it from command line because USB is not
 even
initialized until do_usb() from cmd_usb.c is called... What if we
 do
   NOT
have a serial console at all?
   
   Dumb question because I have not looked seriously at the v2 fork of
   u-boot:
   how does the v2 fork handle this?  Better?  Since the v2 fork uses
 (or
   is
   close to) the linux driver model, I would expect it to be better.
  
  Nothing's changed here...
  
  And that is _NOT_ just USB, it is everywhere -- I2C, SPI, etc. It is
 stiff,
  inflexible, everything written in stone and that stone is permanently
 set in
  concrete.
 
 I suppose you didn't look in the right place. We don't even have support
 for i2c and spi in v2 :-)

Ah, that's that forked one! Sorry, my bad... I thought about the new version
of a legacy one that just shuffled source files to put device drivers under
devices/*...

Can you tell me where it lives to clone the repository? I'd rather give up
on v1 entirely because it is not going anywhere...

Do you have a separate mailing list for v2?

---
**
*  k...@homeKOI8 Net The impossible we do immediately.  *
*  Las Vegas   NV, USA  Miracles require 24-hour notice.   *
**
___
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot


Re: [U-Boot] Multiple device support -- sorry state :(

2009-04-08 Thread Robert Schwebel
On Wed, Apr 08, 2009 at 01:18:45PM -0700, k...@koi8.net wrote:
  I suppose you didn't look in the right place. We don't even have support
  for i2c and spi in v2 :-)

 Ah, that's that forked one! Sorry, my bad... I thought about the new version
 of a legacy one that just shuffled source files to put device drivers under
 devices/*...

 Can you tell me where it lives to clone the repository? I'd rather give up
 on v1 entirely because it is not going anywhere...

git clone git://www.denx.de/git/u-boot-v2

 Do you have a separate mailing list for v2?

No, just use this one. As long as the traffic keeps low, I suppose there
is no need for a separate list. However, it's probably a good idea to
have Sascha on Cc: :-)

rsc
-- 
Pengutronix e.K.   | |
Industrial Linux Solutions | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0|
Amtsgericht Hildesheim, HRA 2686   | Fax:   +49-5121-206917- |
___
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot


Re: [U-Boot] Multiple device support -- sorry state :(

2009-04-08 Thread ksi
On Wed, 8 Apr 2009, Robert Schwebel wrote:

 On Wed, Apr 08, 2009 at 01:18:45PM -0700, k...@koi8.net wrote:
   I suppose you didn't look in the right place. We don't even have
 support
   for i2c and spi in v2 :-)
 
  Ah, that's that forked one! Sorry, my bad... I thought about the new
 version
  of a legacy one that just shuffled source files to put device drivers
 under
  devices/*...
 
  Can you tell me where it lives to clone the repository? I'd rather
 give up
  on v1 entirely because it is not going anywhere...
 
 git clone git://www.denx.de/git/u-boot-v2
 
  Do you have a separate mailing list for v2?
 
 No, just use this one. As long as the traffic keeps low, I suppose there
 is no need for a separate list. However, it's probably a good idea to
 have Sascha on Cc: :-)

OK, thanks. Cloning now :)

---
**
*  k...@homeKOI8 Net The impossible we do immediately.  *
*  Las Vegas   NV, USA  Miracles require 24-hour notice.   *
**
___
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot


Re: [U-Boot] Multiple device support -- sorry state :(

2009-04-08 Thread ksi
On Wed, 8 Apr 2009, k...@koi8.net wrote:

 On Wed, 8 Apr 2009, Robert Schwebel wrote:
 
  On Wed, Apr 08, 2009 at 01:18:45PM -0700, k...@koi8.net wrote:
I suppose you didn't look in the right place. We don't even have
  support
for i2c and spi in v2 :-)
  
   Ah, that's that forked one! Sorry, my bad... I thought about the new
  version
   of a legacy one that just shuffled source files to put device
 drivers
  under
   devices/*...
  
   Can you tell me where it lives to clone the repository? I'd rather
  give up
   on v1 entirely because it is not going anywhere...
  
  git clone git://www.denx.de/git/u-boot-v2
  
   Do you have a separate mailing list for v2?
  
  No, just use this one. As long as the traffic keeps low, I suppose
 there
  is no need for a separate list. However, it's probably a good idea to
  have Sascha on Cc: :-)
 
 OK, thanks. Cloning now :)

OK, got a look at it. Looks promising but it is in very early stage yet... I
wouldn't say in pre-conception stage, but definitely on a very beginning of
the first trimester :)

I have the first prototype of MPC8548E-based big motherboard sitting on my
desk right now and it is in full bringup stage. All the hardware is checked,
bunch of smoked parts replaced and fixed, all clocks are ticking, CPU
fetching etc.

I will definitely join V2 development but for now I'll probably make a huge
set of horrible hacks over V1 to test everything and get my revision 2
started. Then I will be able to work on U-Boot V2 while my revision 2 is
made (schematic rework, re-layout, PCB, missing parts sourcing, assembly
etc.)

---
**
*  k...@homeKOI8 Net The impossible we do immediately.  *
*  Las Vegas   NV, USA  Miracles require 24-hour notice.   *
**
___
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot


Re: [U-Boot] Multiple device support -- sorry state :(

2009-04-08 Thread Robert Schwebel
On Wed, Apr 08, 2009 at 02:25:57PM -0700, k...@koi8.net wrote:
  OK, thanks. Cloning now :)
 
 OK, got a look at it. Looks promising but it is in very early stage yet... I
 wouldn't say in pre-conception stage, but definitely on a very beginning of
 the first trimester :)

Well, you are free to send patches. We use it mainly for ARM systems
which happen to be flash based (NAND, NOR), without PCI. So what was
implemented was what we needed. No reason for you not to send us your
favourite and fancy new features :-)

 I have the first prototype of MPC8548E-based big motherboard sitting
 on my desk right now and it is in full bringup stage. All the hardware
 is checked, bunch of smoked parts replaced and fixed, all clocks are
 ticking, CPU fetching etc.
 
 I will definitely join V2 development but for now I'll probably make a
 huge set of horrible hacks over V1 to test everything and get my
 revision 2 started. Then I will be able to work on U-Boot V2 while my
 revision 2 is made (schematic rework, re-layout, PCB, missing parts
 sourcing, assembly etc.)

Go ahead, we definitely are looking forward to more community
contribution.

rsc
-- 
Pengutronix e.K.   | |
Industrial Linux Solutions | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0|
Amtsgericht Hildesheim, HRA 2686   | Fax:   +49-5121-206917- |
___
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot


Re: [U-Boot] Multiple device support -- sorry state :(

2009-04-08 Thread ksi
On Wed, 8 Apr 2009, Robert Schwebel wrote:

 On Wed, Apr 08, 2009 at 02:25:57PM -0700, k...@koi8.net wrote:
   OK, thanks. Cloning now :)
  
  OK, got a look at it. Looks promising but it is in very early stage
 yet... I
  wouldn't say in pre-conception stage, but definitely on a very
 beginning of
  the first trimester :)
 
 Well, you are free to send patches. We use it mainly for ARM systems
 which happen to be flash based (NAND, NOR), without PCI. So what was
 implemented was what we needed. No reason for you not to send us your
 favourite and fancy new features :-)

Sure I will :) Ah, ARM, my old love, still remeber that talk -- ldmia,
stmfd... Romantic times, eh...

  I have the first prototype of MPC8548E-based big motherboard sitting
  on my desk right now and it is in full bringup stage. All the hardware
  is checked, bunch of smoked parts replaced and fixed, all clocks are
  ticking, CPU fetching etc.
  
  I will definitely join V2 development but for now I'll probably make a
  huge set of horrible hacks over V1 to test everything and get my
  revision 2 started. Then I will be able to work on U-Boot V2 while my
  revision 2 is made (schematic rework, re-layout, PCB, missing parts
  sourcing, assembly etc.)
 
 Go ahead, we definitely are looking forward to more community
 contribution.

OK, I probably will. V1, IMHO, is beyond repair; I even contemplated forking
off my own...

---
**
*  k...@homeKOI8 Net The impossible we do immediately.  *
*  Las Vegas   NV, USA  Miracles require 24-hour notice.   *
**
___
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot