[Buildroot] microcom broken with latest uclibc snapshot.

[Buildroot] microcom broken with latest uclibc snapshot.

[Ulf Samuelsson]

When using the latest uClibc snapshot to build the toolchain,
It looks like the high speeds are not acceptable (> 1151200) 
and the generated <staging_dir>/arm-linux/sys-include/bits/termios.h
has higher speeds commented away.

#if 0 /* limited on uClibc, keep in sync w/ cfsetspeed.c */
#define B230400 0010003
...
#endif

This is not a good idea, if the MCU can handle higher speed!

It also breaks microcom-1.02.

Best Regards,
Ulf Samuelsson
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://busybox.net/lists/buildroot/attachments/20070123/d9c4616a/attachment.htm 

[Buildroot] microcom broken with latest uclibc snapshot.

[Bernhard Fischer]

On Tue, Jan 23, 2007 at 12:32:37AM +0100, Ulf Samuelsson wrote:
>When using the latest uClibc snapshot to build the toolchain,
>It looks like the high speeds are not acceptable (> 1151200) 
>and the generated <staging_dir>/arm-linux/sys-include/bits/termios.h
>has higher speeds commented away.
>
>#if 0 /* limited on uClibc, keep in sync w/ cfsetspeed.c */
>#define B230400 0010003
>...
>#endif
>
>This is not a good idea, if the MCU can handle higher speed!
>
>It also breaks microcom-1.02.

I'm pretty sure that i fixed microcom, please update and confirm since i
only compile tested it..

[Buildroot] microcom broken with latest uclibc snapshot.

[Ulf Samuelsson]

Bernhard Fischer wrote:
> On Tue, Jan 23, 2007 at 12:32:37AM +0100, Ulf Samuelsson wrote:
>> When using the latest uClibc snapshot to build the toolchain,
>> It looks like the high speeds are not acceptable (> 1151200)
>> and the generated <staging_dir>/arm-linux/sys-include/bits/termios.h
>> has higher speeds commented away.
>>
>> #if 0 /* limited on uClibc, keep in sync w/ cfsetspeed.c */
>> #define B230400 0010003
>> ...
>> #endif
>>
>> This is not a good idea, if the MCU can handle higher speed!
>>
>> It also breaks microcom-1.02.
>
> I'm pretty sure that i fixed microcom, please update and confirm
> since i only compile tested it..

My Buildroot copy is from 2007-01-16.

But why limit the UART speed?
An AT91RM9200 can do 460kBAUD.
Have several customer which would like to avoid
such artificial limitation.
Isn't it better to make the max speed a configuration item?

Best Regards,
Ulf Samuelsson
ulf at atmel.com
GSM:  +46 (706) 22 44 57
Tel:     +46  (8) 441 54 22
Fax:     +46 (8) 441 54 29
Mail: Box 2033  174 02 Sundbyberg
Visit: Kavalleriv?gen 24
          174 58 Sundbyberg'
Sweden

[Buildroot] microcom broken with latest uclibc snapshot.

[Bernhard Fischer]

On Tue, Jan 23, 2007 at 10:06:16AM +0100, Ulf Samuelsson wrote:
>Bernhard Fischer wrote:
>>On Tue, Jan 23, 2007 at 12:32:37AM +0100, Ulf Samuelsson wrote:
>>>When using the latest uClibc snapshot to build the toolchain,
>>>It looks like the high speeds are not acceptable (> 1151200)
>>>and the generated <staging_dir>/arm-linux/sys-include/bits/termios.h
>>>has higher speeds commented away.
>>>
>>>#if 0 /* limited on uClibc, keep in sync w/ cfsetspeed.c */
>>>#define B230400 0010003
>>>...
>>>#endif
>>>
>>>This is not a good idea, if the MCU can handle higher speed!
>>>
>>>It also breaks microcom-1.02.
>>
>>I'm pretty sure that i fixed microcom, please update and confirm
>>since i only compile tested it..
>
>My Buildroot copy is from 2007-01-16.

today is something like 2007-01-23 (for me) ;)

$ svn log -r "{2007-01-16}:HEAD" | grep -e ^r[[:digit:]] | wc -l
78

We had 78 patches going in since then.

>
>But why limit the UART speed?
>An AT91RM9200 can do 460kBAUD.
>Have several customer which would like to avoid
>such artificial limitation.
>Isn't it better to make the max speed a configuration item?

I don't see a reason why it shouldn't be configurable, agree.
Perhaps send a proposed patch along the rational to the uClibc list..