./wip/or1ksim, OpenRISC emulator

[ CVSweb ] [ Homepage ] [ RSS ] [ Required by ] [ Add to tracker ]

Branch: CURRENT, Version: 0.5.1rc1, Package name: or1ksim-0.5.1rc1, Maintainer: pkgsrc-users

Or1ksim is a generic OpenRISC 1000 architecture simulator capable of
emulating OpenRISC based computer systems.
The goal is to simulate 32-bit and 64-bit OpenRISC CPUs and enough surrounding
hardware to fake real machines, capable of running real (unmodified) operating
systems, such as NetBSD, RTEMS, eCos and Linux.

Or1ksim provides several unique features:

* Free, open source code.
* High level, fast architecture simulation that allows early code
analysis and system performance evaluation.
* All major models of OpenCores peripheral and system controller
cores supportedi.
* Easy addition of new peripheral models.
* Remote debugging through a network socket with the GNU Debugger (GDB).
* Support for different environments (memory configurations and sizes,
OR1K processor model, configuration of peripheral devices).

Required to run:

Required to build:

Master sites:

RMD160: 24aefc2431b94f2d1ae4a348e2eb21ab8b4bb824
Filesize: 2203.57 KB

Version history: (Expand)

CVS history: (Expand)

   2014-06-01 14:49:35 by Thomas Klausner | Files touched by this commit (208)
Log message:
It is a user-defined variable and should NOT be set in Makefiles.
   2013-10-14 15:00:04 by Ryo ONODERA | Files touched by this commit (19) | Package removed
Log message:
Update to 0.5.1rc1

* Compiled but not working properly for me.
  vmlinux kernel that is built from openrisc.net Linux git repository
  does not boot. 
  Infinite(?) ERR: 8-bit program load out of memory area: [hexnumer].

* Linux/openrisc does not for me, so network support code is not verified.

* For the record, 0.2.0 also does not works for me.

New in release 0.5.1rc1

New features (shouldn't be there during a release cycle, but prompted by debug
needs elsewhere in the tool chain).

New option --trace provides a one line dump of instruction executed and any
register or memory location changed after each instruction.

A new configuration option "use_nmi" is added to the programmable interrupt
controller (PIC). This causes interrupt lines 0 and 1 to be non-maskable, but
only in the sense that the corresponding bits in PICMR are hard-wired to 1.

New config setting for memory initialization "exitnops" fills memory with
"l.nop 1", which will cause the simulator to exit. Good for tracking \ 

New l.nop opcodes to turn instruction tracing on and off. Good for debugging

New l.nop opcode to return a random number. Good for applications (like the
Linux kernel) which need access to the host entropy to ensure randomness.

New l.nop opcode to identify that this is an Or1ksim session.  Allows
applications to determine if they are running on Or1ksim or hardware.

No bugs are fixed.

The following bugs are outstanding
* Bug 1973: Or1ksim needs PIC tests.
* But 1931: Or1ksim needs test for l.nop opcodes.
* Bug 1823: Configuration file error line numbers are wrong.
* Bug 1822: ATA configuration is broken. Documented in user guide.
* Bug 1758: Memory controller issues. Workaround in the user guide.
   2012-10-03 23:26:15 by Aleksej Saushev | Files touched by this commit (1)
Log message:
Support staged installation.
   2009-10-11 12:44:55 by Thomas Klausner | Files touched by this commit (222)
Log message:
Remove obsolete @dirrm lines.
   2009-05-20 02:58:40 by Thomas Klausner | Files touched by this commit (75) | Package updated
Log message:
Recursive ABI depends update and PKGREVISION bump for readline-6.0 shlib
major change.

Reported by Robert Elz in PR 41345.
   2006-10-30 23:45:30 by Sergey Svishchev | Files touched by this commit (17)
Log message:
Delint more low-hanging fruit:
- whitespace, COMMENT, PLIST police
- ${PAX} -pe -> pax -pp
- other minor fixes.
   2006-05-19 18:51:13 by Yoshito Komatsu | Files touched by this commit (107)
Log message:
Replace tech-pkg@NetBSD.org with pkgsrc-users@NetBSD.org.
   2006-02-26 14:32:24 by Klaus Heinz | Files touched by this commit (1)
Log message:
Building and running (at least starting) on DragonFlyBSD confirmed.