Hi Eric,
Only now I was able to get back at working with RTEMS/EPICS. I tried your recipe. I used the exact versions you mentioned and your updated script. Everything compiled, except for EPICS base (step 8). I skipped step 7 because I'll only use the pc386 BSP. I am compiling on a Ubuntu 12.04 64 bits machine.
The compilation process fails with:
~/rtems/rtems-4.10/bin/i386-rtems4.10-g++ -B~/rtems/rtems-4.10/i386-rtems4.10/pc386/lib/ -specs bsp_specs -qrtems -c -mtune=i386 -DUNIX -O2 -g -g -Wall -MMD -I. -I../O.Common -I. -I.. -I../../../../src/cas/generic -I../../../../src/cas/io/bsdSocket -I../../../../src/cas/generic/st -I../../../../src/cas/../ca -I../../../../include/os/RTEMS -I../../../../include ../../../../src/cas/generic/casDGClient.cc
/tmp/ccGTLarr.s: Assembler messages:
/tmp/ccGTLarr.s:665: Error: junk at end of line, first unrecognized character is `)'
/tmp/ccGTLarr.s:665: Error: junk at end of line, first unrecognized character is `:'
/tmp/ccGTLarr.s:665: Error: junk at end of line, first unrecognized character is `2'
(600 similar lines ...)
/tmp/ccGTLarr.s:665: Error: junk at end of line, first unrecognized character is `1'
/tmp/ccGTLarr.s:665: Error: junk at end of line, first unrecognized character is `:'
/tmp/ccGTLarr.s:665: Error: junk at end of line, first unrecognized character is `0'
/tmp/ccGTLarr.s:665: Error: junk at end of line, first unrecognized character is `"'
make[4]: ** [casDGClient.o] Erro 1
make[4]: Saindo do diretório `~/epics/base-3.14.12.2/src/cas/build/O.RTEMS-pc386'
make[3]: ** [install.RTEMS-pc386] Erro 2
make[3]: Saindo do diretório `~/epics/base-3.14.12.2/src/cas/build'
make[2]: ** [build.install] Erro 2
make[2]: Saindo do diretório `~/epics/base-3.14.12.2/src/cas'
make[1]: ** [cas.install] Erro 2
make[1]: Saindo do diretório ~/epics/base-3.14.12.2/src'
make: ** [src.install] Erro 2
Where "Saindo do diretório" means "Leaving directory"
I don't believe the problem lies within the toolchain, because the example ticker.exe runs fine.
I have no clue where the problem might be...
Bruno Seiva Martins
On Tue, Sep 11, 2012 at 12:31 PM, Joel Sherrill
<[email protected]> wrote:
Why doesn't libbspExt compile? I don't see any output.
I have never understood why you all didn't work to
get the libbspExt capabilities into the PIC interrupt
model code. In particular, the new shared implementation
by Sebastian Huber.
FWIW I am not even sure at this point what capabilities
it offers that you don't have in the shared implementation
of the PIC interrupt model.
--joel
On 09/11/2012 10:17 AM, Eric Norum wrote:
AFAIK the RTEMS-4.11 stuff is still a development branch.
I just checked and found that I could use EPICS/RTEMS-4.10 on
my OS X "Mountain Lion" host as follows:
1) Download and
install gmp-5.0.5, mpc-1.0., mpfr-3.1.1, automake-1.11.1and autoconf-2.68
with PREFX=/usr/llocal/rtems/rtems-4.10.
4) Ran the bootstrap script in the top RTEMS source
directory.
5) Build RTEMS:
/usr/local/rtems/rtems-4.10/src/rtems-4.10.2/configure
--prefix=/usr/local/rtems/rtems-4.10 --enable-cxx
--disable-rdbg --enable-tests=samples --enable-networking
--enable-posix --target=powerpc-rtems4.10
'--enable-rtemsbsp=mvme3100 mvme2100 beatnik'
6) Build the RTEMS add-on packages (to get readline)
7) Build 'extended BSP' support.
8) Build EPICS with the attached
configure/os/CONFIG_SITE.Common.RTEMS
I tried the above with the RTEMS-4.11 stuff but ran into
problems at step (7) -- I couldn't get the the 'libbspExt'
stuff to build. I don't think that there's anything wrong
with the source, just the Makefile arrangements seem to have
changed between 4.10 and 4.11. It would be nice if the bspExt
stuff could migrate to RTEMS itself….
--
Joel Sherrill, Ph.D. Director of Research& Development
[email protected] On-Line Applications Research
Ask me about RTEMS: a free RTOS Huntsville AL 35805
Support Available (256) 722-9985