* [U-Boot-Users] Update on the fdt command
@ 2007-03-21 14:08 Jerry Van Baren
2007-03-21 14:49 ` Wolfgang Denk
0 siblings, 1 reply; 4+ messages in thread
From: Jerry Van Baren @ 2007-03-21 14:08 UTC (permalink / raw)
To: u-boot
Hi Wolfgang, et al.,
I'm getting close to having a functional "fdt" command. For a sneak
peek you can look here:
<http://www.cideas.us/cgi-bin/gitweb.cgi?p=u-boot/u-boot-mpc83xx;a=summary>
The new fdt command can be found here:
<http://www.cideas.us/cgi-bin/gitweb.cgi?p=u-boot/u-boot-mpc83xx;a=blob;f=common/cmd_fdt.c;h=8cce35b137f4214a2c6cbab026460b54d72bb6e0;hb=HEAD>
(see line 777 ff. to see the subcommand list). Please critique my
subcommands - it will hurt my feelings, but I'll get over it.
Remaining improvements before it is Ready For Consideration[tm]:
* Implement mknod and the rm node/property subcommands
* Fix up the board-specific OF extensions (primarily PCI-related) to
use libfdt instead of the current ft_build.c routines
* Add/modify a configuration #define to select between the ft_build.c
methodology & support and the new fdt/libfdt methodology & support.
* Remove the auto-OF generation calls from the "bootm" command
- This could be controversial: my proposal is to replace the automagic
generation of the "chosen", "u-boot-env", and "bd_t" nodes
currently done by the "bootm" command with explicit calls to the
new "fdt" command.
My proposal to Wolfgang is that denx.de create a repository for my
changes so I can submit them and ultimately he can pull the changes into
the master repo. The changes to add David Gibson's libfdt to the source
tree will be fairly large (I'm assuming too large to review as an
emailed patch). The changes involving the new "fdt" command are more
manageable, but still will be non-trivial since the new command is 800++
lines.
Best regards,
gvb
^ permalink raw reply [flat|nested] 4+ messages in thread* [U-Boot-Users] Update on the fdt command
2007-03-21 14:08 [U-Boot-Users] Update on the fdt command Jerry Van Baren
@ 2007-03-21 14:49 ` Wolfgang Denk
2007-03-21 15:15 ` Jerry Van Baren
0 siblings, 1 reply; 4+ messages in thread
From: Wolfgang Denk @ 2007-03-21 14:49 UTC (permalink / raw)
To: u-boot
Hi,
in message <46013C58.2040300@smiths-aerospace.com> you wrote:
>
> I'm getting close to having a functional "fdt" command. For a sneak
> peek you can look here:
> <http://www.cideas.us/cgi-bin/gitweb.cgi?p=u-boot/u-boot-mpc83xx;a=summary>
>
> The new fdt command can be found here:
> <http://www.cideas.us/cgi-bin/gitweb.cgi?p=u-boot/u-boot-mpc83xx;a=blob;f=common/cmd_fdt.c;h=8cce35b137f4214a2c6cbab026460b54d72bb6e0;hb=HEAD>
> (see line 777 ff. to see the subcommand list). Please critique my
> subcommands - it will hurt my feelings, but I'll get over it.
Looks fine to me, but I cannot understand this part:
786 "fdt rm <path> <prop> - Delete <property>\n"
787 "fdt rm <path> <node> - Delete the node *NOTE THE SPACE*\n"
NOTE THE SPACE? Which one? And where is the difference?
> My proposal to Wolfgang is that denx.de create a repository for my
> changes so I can submit them and ultimately he can pull the changes into
> the master repo. The changes to add David Gibson's libfdt to the source
Is the name u-boot-fdt ok? Can you please send me your public SSH key?
Best regards,
Wolfgang Denk
--
DENX Software Engineering GmbH, HRB 165235 Munich, CEO: Wolfgang Denk
Office: Kirchenstr. 5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
"Just think, with VLSI we can have 100 ENIACS on a chip!"
- Alan Perlis
^ permalink raw reply [flat|nested] 4+ messages in thread
* [U-Boot-Users] Update on the fdt command
2007-03-21 14:49 ` Wolfgang Denk
@ 2007-03-21 15:15 ` Jerry Van Baren
2007-03-21 16:41 ` Wolfgang Denk
0 siblings, 1 reply; 4+ messages in thread
From: Jerry Van Baren @ 2007-03-21 15:15 UTC (permalink / raw)
To: u-boot
Wolfgang Denk wrote:
> Hi,
>
> in message <46013C58.2040300@smiths-aerospace.com> you wrote:
>> I'm getting close to having a functional "fdt" command. For a sneak
>> peek you can look here:
>> <http://www.cideas.us/cgi-bin/gitweb.cgi?p=u-boot/u-boot-mpc83xx;a=summary>
>>
>> The new fdt command can be found here:
>> <http://www.cideas.us/cgi-bin/gitweb.cgi?p=u-boot/u-boot-mpc83xx;a=blob;f=common/cmd_fdt.c;h=8cce35b137f4214a2c6cbab026460b54d72bb6e0;hb=HEAD>
>> (see line 777 ff. to see the subcommand list). Please critique my
>> subcommands - it will hurt my feelings, but I'll get over it.
>
> Looks fine to me, but I cannot understand this part:
>
> 786 "fdt rm <path> <prop> - Delete <property>\n"
> 787 "fdt rm <path> <node> - Delete the node *NOTE THE SPACE*\n"
>
> NOTE THE SPACE? Which one? And where is the difference?
The space between <path> and <node>... everywhere else, <path> includes
the last node and a space delimits the <property>. I propose this to
make parsing easier: I would not have to find the last "/" to find the
node that was intended to be deleted. The alternative is:
"fdt rm <path> <prop> - Delete <property>\n"
"fdt rm <path> - Delete the last node in <path>\n"
I have not implemented the rm command yet, so I'm flexible. It also
means I have not tried it in real life... I wouldn't be surprised if I
find the extra space between <path> and <node> is so annoying that it is
worth finding the last '/' (which really isn't that difficult to parse).
The advantage of specifying "<path> <node>" as I see it is that it is
harder to screw up and delete major parts of the tree inadvertently
rm /cpus # poof, all gone!
when you really meant to do
rm /cpus PowerPC,8560 at 0
Either way, I would not allow the deletion of the root node. For the
<path>-only version
rm /
is valid syntax but would be a problem - would have to test for this as
a special case. For the <path> <node>, this becomes invalid syntax and
would not be permitted. (Note that, either way, you can delete all the
subnodes of the root '/' individually. Deleting the subnodes makes
sense, deleting the root node itself does not.)
>> My proposal to Wolfgang is that denx.de create a repository for my
>> changes so I can submit them and ultimately he can pull the changes into
>> the master repo. The changes to add David Gibson's libfdt to the source
>
> Is the name u-boot-fdt ok? Can you please send me your public SSH key?
u-boot-fdt is great. I'll send you my key.
> Best regards,
>
> Wolfgang Denk
Thanks,
gvb
^ permalink raw reply [flat|nested] 4+ messages in thread
* [U-Boot-Users] Update on the fdt command
2007-03-21 15:15 ` Jerry Van Baren
@ 2007-03-21 16:41 ` Wolfgang Denk
0 siblings, 0 replies; 4+ messages in thread
From: Wolfgang Denk @ 2007-03-21 16:41 UTC (permalink / raw)
To: u-boot
In message <46014C02.1050103@smiths-aerospace.com> you wrote:
>
> >> My proposal to Wolfgang is that denx.de create a repository for my
> >> changes so I can submit them and ultimately he can pull the changes into
> >> the master repo. The changes to add David Gibson's libfdt to the source
> >
> > Is the name u-boot-fdt ok? Can you please send me your public SSH key?
>
> u-boot-fdt is great. I'll send you my key.
Done - see
http://www.denx.de/cgi-bin/gitweb.cgi?p=u-boot/u-boot-fdt.git;a=summary
I also got your key, and all should be set up.
Best regards,
Wolfgang Denk
--
DENX Software Engineering GmbH, HRB 165235 Munich, CEO: Wolfgang Denk
Office: Kirchenstr. 5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
Our business is run on trust. We trust you will pay in advance.
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2007-03-21 16:41 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2007-03-21 14:08 [U-Boot-Users] Update on the fdt command Jerry Van Baren
2007-03-21 14:49 ` Wolfgang Denk
2007-03-21 15:15 ` Jerry Van Baren
2007-03-21 16:41 ` Wolfgang Denk
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.