public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug c++/59558] New: [MSP430-gcc] CPU setup, pre start program
@ 2013-12-19 10:57 jkroby at p3r dot info
  2014-11-16 20:00 ` [Bug c++/59558] " pab at pabigot dot com
  0 siblings, 1 reply; 2+ messages in thread
From: jkroby at p3r dot info @ 2013-12-19 10:57 UTC (permalink / raw)
  To: gcc-bugs

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59558

            Bug ID: 59558
           Summary: [MSP430-gcc] CPU setup, pre start program
           Product: gcc
           Version: unknown
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: jkroby at p3r dot info

Gcc: msp430-gcc (GNU GCC patched mspgcc-20110716) 4.5.3
Run on msp430G2553
Library used: Energia 0101E0009
compiled on host: Linux version 3.2.0-56-generic (buildd@roseapple) (gcc
version 4.6.3 (Ubuntu/Linaro 4.6.3-1ubuntu5) ) #86-Ubuntu SMP Wed Oct 23
09:20:45 UTC 2013

I have wrote a big program with many library and i test it from some month, all
OK but, when i add new code i found a strange think, the CPU do nothink so i
use debugger and i find the problem:

using msp430g2553 with 510 bytes of ram memory used (512 max), (calculation
way: i enlarge the variable and i use the linker error to calculate the size, i
reduce the size for stay in the limit)

With debugger i discover:
the first instruction CPU load from 0x0120 WDTCTL in 0x03fc address ram and
make an or so the location becam 0x5408, it use this value for reset wdt, i
think before energia initialize it, SP is initialized to 0x0400 when the code
use CALL and RET (for init) overwrite that value at 0x03fc so when it use the
value for clear Watchdog happen a PUC because the write pin(password) is not
correct (WDTPW that has to be 5A high byte is 0xda, call return address of the
stack).
i reduce the code of 4 bytes for be sure and it works.
All is ok only the space conflict with stack and temporary variable is a
problem.

I hope to be clear, i'm ready to give more detail.

I think is a problem of the library for initialization but i don't know which
component is.

Roberto Pomo <jkroby@p3r.info>


^ permalink raw reply	[flat|nested] 2+ messages in thread

* [Bug c++/59558] [MSP430-gcc] CPU setup, pre start program
  2013-12-19 10:57 [Bug c++/59558] New: [MSP430-gcc] CPU setup, pre start program jkroby at p3r dot info
@ 2014-11-16 20:00 ` pab at pabigot dot com
  0 siblings, 0 replies; 2+ messages in thread
From: pab at pabigot dot com @ 2014-11-16 20:00 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59558

Peter A. Bigot <pab at pabigot dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |pab at pabigot dot com

--- Comment #1 from Peter A. Bigot <pab at pabigot dot com> ---
This report is against the out-of-tree target implemented in mspgcc, not the
subsequent integrated port to MSP430 referred to as msp430-elf.  C++ was never
officially supported in mspgcc, and mspgcc itself is no longer supported.


^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2014-11-16 20:00 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2013-12-19 10:57 [Bug c++/59558] New: [MSP430-gcc] CPU setup, pre start program jkroby at p3r dot info
2014-11-16 20:00 ` [Bug c++/59558] " pab at pabigot dot com

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).