Hello,
To debug this we wrote a routine to enable the RISC timer on the CPM.
We verify it that this timer is actually getting counted down. This
tells us that CPM is running.
However in the file cpm/mpc8260/serial_scc.c as soon as following line of
code is executed
tbdf->cdb_sc |= BD_SC_READY;
the RISC timers stops counting.
So somehow it appears that the transmit buffer descriptor that we are
setting up for CPM to transmit data on SCC1 appears to be "invalid"?
The CPM-RAM address that can be used is defined in
include/asm-ppc/cpm_8260.h (definition of CPM_DATAONLY_BASE) so we moved its
location around to 0x3000, also changing
CPM_DATA_ONLY_SIZE also, but it had no effect.
I am sending question to folks at Motorola but if someone has encountered
this before, please do let me know what could be the possible reason, that
CPM gets stuck momemnt we mark tx descriptor is valid. Since the
CPM at this point processing the RISC timer, I think CPM is getting the
clocked correctly...
-Subodh Nijsure
-----Original Message-----
From: Wolfgang Denk
To: Subodh Nijsure
Cc: U-Boot Mailing List
Sent: 5/16/04 1:09 PM
Subject: Re: [U-Boot-Users] CPM on 8250 board seems stuck.
In message
<F0C0133D6BE53342B2A78DE5D200E0A0074577(a)ssmail2.corp.skystream.com> you
wrote:
>
> When I step through the code using BDI, I see is the code is stuck in
a
> while
> loop in serial_putc waiting for the Buffer descriptor ready bit to get
> cleared
> in putc function.
>
> while ( tbdf->cbd_sc & BD_SC_READY )
> ;
...
> It appears to me that CPM is "dead". Now how do I prove things either
way,
> that it is not dead or its is working. Any CLUES I would really
appreciate
> them.
Most probably the CPM is _not_ dead.
This is a typical case where you should lean back and think about the
problem. So you programmed an UART, and wait for it to complete send-
ing a character. When will this complete? After the last bit and the
stop bit(s) have been clocked out. How long will this take? Well,
this obviously depends on the clock. Why might it hang forever?
Ummm... Maybe there is no clock???
Maybe there is no clock???
Check your BRG setup (probably OK, since not board dependent), and
double-check your clock routing!
> Anybody here that has done 8250 board, and booted u-boot on it, care
to
> share your
> config.h file?
Why don't you just use one of the files in the include/configs
directory? Checking for example the MAINTAINERS file will reveal at
least two MPC8250 based boards.
Best regards,
Wolfgang Denk
--
Software Engineering: Embedded and Realtime Systems, Embedded Linux
Phone: (+49)-8142-4596-87 Fax: (+49)-8142-4596-88 Email: wd(a)denx.de
"The whole world is about three drinks behind." - Humphrey Bogart
