[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [openrisc] problems with or1ksim



lavender_uc@yahoo.com wrote:
> 
> Thank Damjan and Simon for the prompt reply. I'm very much impressed
> that you always take care of our questions in such a serious manner.
> 
> Based on Simon's reply, I assume the simulation built with
> or32-uclinux or or32-rtems should make no difference. Is this correct?

Yes.

> So what's going on with RTL regression test? The makefile uses
> or32-rtems. Could I change it to or32-uclinux  and it should work
> well? Thanks in advance.

Well, this should be changed to or32-uclinux, but I'm not sure, if it
will work right away. Maybe there will be the need for changing some
preprocessor directives in the code (.section directives particularly).
I suggest that you build or32-rtems tool chain also and compile these
programs with it. We will change the this to or32-uclinux soon.

> 
> nan
> 
> ----- Original Message -----
> From: "Damjan Lampret" <lampret@o... >
> To: <openrisc@o... >
> Date: Mon, 18 Nov 2002 14:00:13 +0100
> Subject: Re: [openrisc] problems with or1ksim
> 
> >
> >
> > BTW you can go to http://www.opencores.org/projects/or1k/ATS/ and
> > see that
> > or1ksim testbench works, although I used or32-uclibc in the script
> > (or32-uclinux didn't work, will change now since it works now). You
> > can also
> > see testbench log there.
> >
> > regards,
> > Damjan
> >
> > ----- Original Message -----
> > From: "Simon Srot" <simons@o... >
> > To: <openrisc@o... >
> > Sent: Sunday, November 17, 2002 4:09 PM
> > Subject: Re: [openrisc] problems with or1ksim
> >
> >
> > > Nan Zhou wrote:
> > > >
> > > > Hi,
> > > >
> > > > I'm back with the old question posted last month. My
> > question is how
> > > > should I build the toolchain configuring for or32-rtems?
> > Should I follow
> > > > the instuction page:
> > > >
> > http://www.opencores.org/projects/or1k/GNU%20Toolchain%20Port
> > > > and change all "or32-uclinux" to "or32-rtems"?
> > >
> > > Now the testbench is written in such way, that it can be
> > compiled with
> > > or32-uclinux targeted compiler. You can also configure the
> > testbench for
> > > or32-rtems target, but this was not tested. Testbench is meant
> > for
> > > testing the or1ksim and for that reason I don't see the need
> > for
> > > compiling it with or32-rtems. We just need the code to test
> > the or1ksim
> > > and actually and or32-rtems would produce the same code.
> > >
> > > BTW, I noticed that dhry, mycompress, basic and cbasic
> > currently are not
> > > working. I will check what is wrong someday in the near
> > future, but I
> > > know that if you checkout the revision from 16-8-2002 the all
> > tests will
> > > pass.
> > >
> > > >
> > > > But there's error when doing this. What's the directory
> > or1k/rtems and
> > > > or1k/rtems-20020807 for? Do I need it when building
> > toolchain? The
> > > > documentation is not clear to me. Thanks a lot for your
> > kindly help.
> > >
> > > These two folders contains the rtems OS code, as far as I know
> > > or1k/rtems is old version, and or1k/rtems-20020807 is the one
> > to be
> > > used. But you don't need them to run testbench.
> > >
> > > >
> > > > Sicerely,
> > > > nan
> > > >
> > > > ----- Original Message -----
> > > > From: Marko Mlinar <markom@o... >
> > > > To: openrisc@o...
> > > > Date: Thu, 24 Oct 2002 08:01:12 +0200
> > > > Subject: Re: [openrisc] problems with or1ksim
> > > >
> > > > >
> > > > >
> > > > > I suspect you forgot to set your $PATH, since
> > testbench/configure
> > > > > did not
> > > > > detect your C compiler. The other simple problem
> > would be that you
> > > > > have
> > > > > toolchain for or32-uclinux and you are configuring
> > for or32-rtems.
> > > > >
> > > > > basic.S is source file. simulator should be executed
> > with binary
> > > > > files.
> > > > >
> > > > > Marko
> > > > >
> > > > > On Thursday 24 October 2002 01:39, Nan Zhou wrote:
> > > > > > Thanks for your prompt response, Marko and
> > Simon. I
> > > > > > went throught the whole building process of
> > toolchain
> > > > > > and or1ksim. Everything went smoothly until I
> > entered
> > > > > > or1ksim/testbench/:
> > > > > > ./configure --target=or32-rtems
> > > > > > make all
> > > > > >
> > > > > > Here's what happened:
> > > > > >
> > > > > > make[1]: Entering directory
> > > > > >
> > `/usr/home/nzhou/or1k/or1ksim/testbench/support'
> > > > > > cc -DPACKAGE=\"or1ksimtest\" -DVERSION=\"1.3\"
> > > > > > -DSTDC_HEADERS=1 -Dconst= -Dinline=  -I. -I.
> > > > > > -I../support    -Wall -g -nostdlib -mhard-div
> > -c
> > > > > > support.c
> > > > > > cc1: Invalid option `hard-div'
> > > > > > make[1]: *** [support.o] Error 1
> > > > > > make[1]: Leaving directory
> > > > > >
> > `/usr/home/nzhou/or1k/or1ksim/testbench/support'
> > > > > > make: *** [all-recursive] Error 1
> > > > > >
> > > > > >
> > > > > > Then I run ../sim -f ../sim.cfg basic.S:
> > > > > >
> > > > > > Reading script file from '../sim.cfg'...
> > > > > > Verbose on, simdebug off, interactive prompt
> > off
> > > > > > Machine initialization...
> > > > > > Clock cycle: 100ns
> > > > > > Data cache present.
> > > > > > Insn cache tag present.
> > > > > > BPB simulation off.
> > > > > > BTIC simulation off.
> > > > > >
> > > > > > Building automata... done, num uncovered:
> > 0/212.
> > > > > > Parsing operands data... done.
> > > > > > loadcode: filename basic.S  startaddr=0
> > > > > > virtphy_transl=0
> > > > > > Not COFF file format
> > > > > > Not ELF file format.
> > > > > > identifyfile2: Success
> > > > > > WARNING: dependstats stats must be enabled to
> > do
> > > > > > history analisis.
> > > > > > Resetting 1 UART(s).
> > > > > > WARNING: UART0 has problems with RX file
> > stream.
> > > > > > WARNING: Keyboard has problems with RX file
> > stream.
> > > > > > NOTE   : ata_device, using device type FILE.
> > > > > > file /tmp/sim_atadev0 already exists. Using
> > existing
> > > > > > file.
> > > > > > ata_device_debug: requested filesize was: 1
> > (MBytes).
> > > > > > ata_device_debug: actual filesize is: 1
> > (MBytes).
> > > > > > NOTE   : ata_device, using type NO_CONNECT.
> > > > > > ata_device_debug: ata_devices_hw_reset.
> > > > > > setting status register BSY 0x80
> > > > > > Resetting Tick Timer.
> > > > > > Resetting Power Management.
> > > > > > Resetting PIC.
> > > > > > Resetting memory controller.
> > > > > > Starting at 0x00000000
> > > > > > Exception 0x100 (Reset) at 0x0, EA: 0x0, ppc:
> > 0x0,
> > > > > > npc: 0x4
> > > > > >
> > > > > >
> > > > > > It stopped here. I used Ctl+C and it entered
> > prompt
> > > > > > (sim).
> > > > > >
> > > > > > I wonder what did I do wrong and how should I
> > have the
> > > > > > simulator running. By the way, there's no dhry
> > or exit
> > > > > > binary file under testbench.
> > > > > >
> > > > > > Thanks a lot. I appreciate your time and input.
> > > > > >
> > > > > > - nan
> > > > > >
> > > > > > --- Marko Mlinar <markom@o... > wrote:
> > > > > > > Hi!
> > > > > > >
> > > > > > > I was unable to reconstruct your problem,
> > I have
> > > > > > > some
> > > > > > > advices below, otherwise please send more
> > detailed
> > > > > > > bug report.
> > > > > > >
> > > > > > > > WARNING: UARTO has problems with RX
> > file stream.
> > > > > > > >
> > > > > > > > WARNING: Keyboard has problems with
> > RX file
> > > > > > >
> > > > > > > system.)
> > > > > > > It is possible that lack of uart0.rx or
> > keyboard0.rx
> > > > > > > files cause segmentation
> > > > > > > fault. Please create the files and try if
> > it helps.
> > > > > > >
> > > > > > > > The same problem happened when I ran
> > > > > > >
> > > > > > > or32-uclinux-sim from
> > > > > > >
> > > > > > > > /opt/or32-uclinux/bin/ instead with
> > inputing
> > > > > > >
> > > > > > > sim.cfg. But if sim.cfg is not
> > > > > > >
> > > > > > > > given here, I will get exception and
> > when
> > > > > > >
> > > > > > > canceling it, I can get into
> > > > > > >
> > > > > > > > simulator as (sim).
> > > > > > >
> > > > > > > Also try with sim "exit" and "dhry"
> > binary.
> > > > > > >
> > > > > > > > I got confused why the configuration
> > given in
> > > > > > >
> > > > > > > toolchains instruction page,
> > > > > > >
> > > > > > > > Architectural Simulator page, and the
> > README file
> > > > > > >
> > > > > > > are different. They are:
> > > > > > > > ./configure --target=or32-uclinux
> > > > > > >
> > > > > > > --prefix=/opt/or32-uclinux
> > > > > > >
> > > > > > > > ./configure --target=or32-rtems
> > > > > > >
> > > > > > > --prefix=/opt/or32-rtems
> > > > > > > Depends on toolchain you have.
> > > > > > >
> > > > > > > > ./configure --target=or32-uclinux
> > > > > > > >
> > > > > > > > respectively.
> > > > > > > >
> > > > > > > > I wonder if there're different
> > simulators for RTL
> > > > > > >
> > > > > > > and architecture level.
> > > > > > >
> > > > > > > > Also, the README mentions there're
> > drhy.or32 test
> > > > > > >
> > > > > > > file under testbench
> > > > > > >
> > > > > > > > directory but I can't find it.
> > > > > > >
> > > > > > > Yes, it was renamed to "dhry", but README
> > was not
> > > > > > > yet updated.
> > > > > > >
> > > > > > > hope this helps,
> > > > > > > Marko
> > > > > >
> > > > > >
> > __________________________________________________
> > > > > > Do you Yahoo!?
> > > > > > Y! Web Hosting - Let the expert host your web
> > site
> > > > > > http://webhosting.yahoo.com/
> > > > >
> > > >
> > >
> > >
> >
> --
> To unsubscribe from openrisc mailing list please visit http://www.opencores.org/mailinglists.shtml
--
To unsubscribe from openrisc mailing list please visit http://www.opencores.org/mailinglists.shtml