removing -S option from dtc?

removing -S option from dtc?

[Kumar Gala]

Jon,

Any thoughts on removing the -S option from dtc.  It always seemed  
confusing to me and not sure if any really uses it.

- k

Re: removing -S option from dtc?

[Jon Loeliger]

> Jon,
> 
> Any thoughts on removing the -S option from dtc.  It always seemed  
> confusing to me and not sure if any really uses it.
> 
> - k

Well, not yet.

jdl

Re: removing -S option from dtc?

[Jon Loeliger]

> > Jon,
> > 
> > Any thoughts on removing the -S option from dtc.  It always seemed  
> > confusing to me and not sure if any really uses it.
> > 
> > - k
> 
> Well, not yet.

Well, that was vague in its conciseness.

I meant, it's not time to remove it yet.
There are still plenty of lingering users that
have not figured out U-Boot has improved yet.

jdl

Re: removing -S option from dtc?

[Kumar Gala]

On Jul 30, 2008, at 8:12 AM, Jon Loeliger wrote:

>>> Jon,
>>>
>>> Any thoughts on removing the -S option from dtc.  It always seemed
>>> confusing to me and not sure if any really uses it.
>>>
>>> - k
>>
>> Well, not yet.
>
> Well, that was vague in its conciseness.
>
> I meant, it's not time to remove it yet.
> There are still plenty of lingering users that
> have not figured out U-Boot has improved yet.

I was thinking we'd keep -p (for padding) but remove -S (pad to a  
fixed size).

- k

Re: removing -S option from dtc?

[Jon Loeliger]

> 
> I was thinking we'd keep -p (for padding) but remove -S (pad to a  
> fixed size).

Oh, we can deprecate it and start beating people over
the head with removal and upgrade avice tell we are blue.

Start by telling Stuart H to remove it.  I already tried.

jdl

[PATCH] dtc: give advance warning that "-S" is going away.

[Paul Gortmaker]

The "-S" option allowed the specification of a minimum size for
the blob, however the main reason for caring about the size is
so there is enough padding to add a chosen node by u-boot or
whoever.  In which case, folks don't really care about the absolute
size, but rather the size of the padding added for this -- which
is what the "-p" option does.  Having the "-S" just confuses people.

Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
---
 dtc.c |    3 +++
 1 files changed, 3 insertions(+), 0 deletions(-)

diff --git a/dtc.c b/dtc.c
index d8fd43b..84bee2d 100644
--- a/dtc.c
+++ b/dtc.c
@@ -182,6 +182,9 @@ int main(int argc, char *argv[])
 	if (minsize && padsize)
 		die("Can't set both -p and -S\n");
 
+	if (minsize)
+		fprintf(stderr, "DTC: Use of \"-S\" is deprecated; it will be removed soon, use \"-p\" instead\n");
+
 	fprintf(stderr, "DTC: %s->%s  on file \"%s\"\n",
 		inform, outform, arg);
 
-- 
1.5.6.2

Re: removing -S option from dtc?

[Jon Loeliger]

> > 
> > I was thinking we'd keep -p (for padding) but remove -S (pad to a  
> > fixed size).
> 
> Oh, we can deprecate it and start beating people over
> the head with removal and upgrade avice tell we are blue.

s/avice/advice
s/tell/until/

Rough day.

jdl

Re: [PATCH] dtc: give advance warning that "-S" is going away.

[Jon Loeliger]

> The "-S" option allowed the specification of a minimum size for
> the blob, however the main reason for caring about the size is
> so there is enough padding to add a chosen node by u-boot or
> whoever.  In which case, folks don't really care about the absolute
> size, but rather the size of the padding added for this -- which
> is what the "-p" option does.  Having the "-S" just confuses people.
> 
> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>

You rock.

> +	if (minsize)
> +		fprintf(stderr, "DTC: Use of \"-S\" is deprecated; it will be r
> emoved soon, use \"-p\" instead\n");
> +

Or use a U-boot that handles re-sizing automatically.

jdl

Re: [PATCH] dtc: give advance warning that "-S" is going away.

[Jon Loeliger]

> The "-S" option allowed the specification of a minimum size for
> the blob, however the main reason for caring about the size is
> so there is enough padding to add a chosen node by u-boot or
> whoever.  In which case, folks don't really care about the absolute
> size, but rather the size of the padding added for this -- which
> is what the "-p" option does.  Having the "-S" just confuses people.
> 
> Signed-off-by: Paul Gortmaker <paul.gortmaker-CWA4WttNNZF54TAoqtyWWQ@public.gmane.org>

You rock.

> +	if (minsize)
> +		fprintf(stderr, "DTC: Use of \"-S\" is deprecated; it will be r
> emoved soon, use \"-p\" instead\n");
> +

Or use a U-boot that handles re-sizing automatically.

jdl

Re: [PATCH] dtc: give advance warning that "-S" is going away.

[Jon Loeliger]

> The "-S" option allowed the specification of a minimum size for
> the blob, however the main reason for caring about the size is
> so there is enough padding to add a chosen node by u-boot or
> whoever.  In which case, folks don't really care about the absolute
> size, but rather the size of the padding added for this -- which
> is what the "-p" option does.  Having the "-S" just confuses people.
> 
> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>

Applied.

jdl

Re: [PATCH] dtc: give advance warning that "-S" is going away.

[Jon Loeliger]

> The "-S" option allowed the specification of a minimum size for
> the blob, however the main reason for caring about the size is
> so there is enough padding to add a chosen node by u-boot or
> whoever.  In which case, folks don't really care about the absolute
> size, but rather the size of the padding added for this -- which
> is what the "-p" option does.  Having the "-S" just confuses people.
> 
> Signed-off-by: Paul Gortmaker <paul.gortmaker-CWA4WttNNZF54TAoqtyWWQ@public.gmane.org>

Applied.

jdl