Commit Graph

26 Commits

Author SHA1 Message Date
f359364888 sal/gem5: getTimerTicks(), getTimerTicksPerSecond() implemented
Change-Id: I01fdb5e4bdd61fc761e93ef77904c830131c9ed6
2014-01-14 12:13:55 +01:00
45e0b41022 gem5: restore works now
The function restore(PATH) can now be used to restore a checkpoint.

Change-Id: I25faf9f6335261d2b3ade4185eae93983ece9f97
2013-11-13 17:15:19 +01:00
aecb353087 cmake: gem5-related build system updates
The build system now allows incremental gem5 builds. Unfortunately,
the current solution always requires re-linking the executable.
Without the enforcement of re-linking, the fail code will be rebuilt
but not linked into gem5.

The number of cores for building gem5 is derived from /proc/cpuinfo.
As before, only the gem5.debug configuration is supported.

Change-Id: Ib13b15d1ecd62196eb251e0fd00953f4eb052feb
2013-07-04 15:01:22 +02:00
043e264677 gem5: The serializeAll(path) method don't have to be invoked on the root object.
The checkpoint which is produced by this save method is a little bit
different to the checkpoint which is produced by the --take-checkpoint
command. It differs in the save-parameters so_state, funcExeInst, intRegs,
_upc, _nupc, _when. Tests have shown that it probably does not affect the
course of the program execution.

Change-Id: Id776a10f2d40f71643e9edbb45d7368609309df4
2013-05-23 18:10:14 +02:00
503a40be7b gem5: save/restore moved to Gem5Controller
Change-Id: I5bda346f29d74fdfc67f73f0af83ed1e006d8616
2013-05-23 17:17:36 +02:00
a6c24489f6 Revert "gem5: save/restore moved to Gem5Controller"
This reverts commit d525005990.
2013-05-22 17:38:56 +02:00
dd81913f2c Revert "gem5: The serializeAll(path) method don't have to be invoked on the root object."
This reverts commit 60735f254e.
2013-05-22 17:38:39 +02:00
60735f254e gem5: The serializeAll(path) method don't have to be invoked on the root object.
The checkpoint which is produced by this save method is a little bit
different to the checkpoint which is produced by the --take-checkpoint
command. It differs in the save-parameters so_state, funcExeInst, intRegs,
_upc, _nupc, _when. Tests have shown that it probably does not affect the
course of the program execution.

Change-Id: I19b3fc809288224532e0ed6b7910a45115cb1c5d
2013-05-22 15:56:04 +02:00
d525005990 gem5: save/restore moved to Gem5Controller
Change-Id: I0ce72f04a8de88f4f8b6c1f0bba7a9097c82e107
2013-05-22 13:54:55 +02:00
08febe5819 gem5 build system improved
Encapsulated gem5-specific code into wrapper functions to separate the
build process (Fail: CMake, gem5: scons). Added some gem5-related FIXMEs.

Another CMake related FIXME added. +some cosmetics.

Change-Id: Id84b480127b1f13aed6a0ee97f3583f410d531c5
2013-04-08 17:57:06 +02:00
794466d7d7 ~Gem5Controller() should delete it's MemoryManager object 2013-03-19 13:39:18 +01:00
eb17e9ef82 core/sal: move command-line parameter passing to SC::startup() 2013-03-14 22:29:43 +01:00
5c4b132550 ~SimulatorController: do not free ConcreteCPU object ptr in the base class
In fact, delete should be called in the destructor of each derived class (BochsController and Gem5Controller at the moment).

Additionally, this is the reason why ~SimulatorController is declared as virtual.

git-svn-id: https://www4.informatik.uni-erlangen.de/i4svn/danceos/trunk/devel/fail@2064 8c4709b5-6ec9-48aa-a5cd-a96041d1645a
2013-02-06 15:39:50 +00:00
552a5fb4ac coding-style++, comments++, FIXMEs++
git-svn-id: https://www4.informatik.uni-erlangen.de/i4svn/danceos/trunk/devel/fail@2063 8c4709b5-6ec9-48aa-a5cd-a96041d1645a
2013-02-06 15:39:46 +00:00
b052c0494b Architecture changes (only gem5 implementation right now):
- The register manager is gone. It's functionality is now encapsulated in the
  CPU classes.
- For the client, there is the ConcreteCPU class that encapsulates the access
  to the CPU state (including registers) and architecture details. The
  correspondig objects for the CPUs inside the simulator can be accessed
  through the SimulatorController.getCPU() function.
- Listener got a new ConcreteCPU* member to filter for which CPU the events
  should fire. The default NULL is used as wildcard for all aviable CPUs. The
  events respectively got a ConcreteCPU* member to indicate which CPU really
  fired the event.
- For the server, there is CPUArchitecture to access the architecture details.

git-svn-id: https://www4.informatik.uni-erlangen.de/i4svn/danceos/trunk/devel/fail@1966 8c4709b5-6ec9-48aa-a5cd-a96041d1645a
2012-12-02 17:50:46 +00:00
e0e95faa5b Restructured the gem5 backend:
- FailGem5Device is gone.
- There are now changes directly made to the gem5 source.
- Gem5Connector is a helper class that is compiled inside the gem5 context to workaround problems with gem5 header in fail.

Things that are working:
- BPSingleListener
- MemAccessListener
- Save and restore simulator state

git-svn-id: https://www4.informatik.uni-erlangen.de/i4svn/danceos/trunk/devel/fail@1820 8c4709b5-6ec9-48aa-a5cd-a96041d1645a
2012-10-24 19:19:14 +00:00
171d178309 SimCon interface update: save returns a boolean, now (+ redundant virtual keywords removed).
git-svn-id: https://www4.informatik.uni-erlangen.de/i4svn/danceos/trunk/devel/fail@1724 8c4709b5-6ec9-48aa-a5cd-a96041d1645a
2012-10-05 12:12:56 +00:00
4b8a417eda Added missing include of Listener.hpp in Gem5Controller.cc
git-svn-id: https://www4.informatik.uni-erlangen.de/i4svn/danceos/trunk/devel/fail@1709 8c4709b5-6ec9-48aa-a5cd-a96041d1645a
2012-10-02 12:58:45 +00:00
7d49b6f063 Simulator specific listener are now implemented using aspects instead of an additional inheritance level
git-svn-id: https://www4.informatik.uni-erlangen.de/i4svn/danceos/trunk/devel/fail@1706 8c4709b5-6ec9-48aa-a5cd-a96041d1645a
2012-10-02 11:42:18 +00:00
057c216dc0 Removed BufferCache stuff (fast breakpoints will do the job, see r1685).
git-svn-id: https://www4.informatik.uni-erlangen.de/i4svn/danceos/trunk/devel/fail@1686 8c4709b5-6ec9-48aa-a5cd-a96041d1645a
2012-09-25 11:36:19 +00:00
hsc
2fed94ae49 global interrupt_to_fire seems not to be Bochs-specific at the moment
FIXME: This shouldn't be a global variable (probably a SimulatorController
member?).  Reconsider interrupt generation being part of the generic
interface.

git-svn-id: https://www4.informatik.uni-erlangen.de/i4svn/danceos/trunk/devel/fail@1612 8c4709b5-6ec9-48aa-a5cd-a96041d1645a
2012-09-11 17:10:11 +00:00
918eb94523 Removed unnecessary commented out code from previous commit.
git-svn-id: https://www4.informatik.uni-erlangen.de/i4svn/danceos/trunk/devel/fail@1605 8c4709b5-6ec9-48aa-a5cd-a96041d1645a
2012-09-07 22:39:54 +00:00
1403fa3828 Step 1 from experiment hsc-simple is now working with gem5
git-svn-id: https://www4.informatik.uni-erlangen.de/i4svn/danceos/trunk/devel/fail@1604 8c4709b5-6ec9-48aa-a5cd-a96041d1645a
2012-09-07 22:35:52 +00:00
402ca4ec45 Added mechanism to specialize listener + basic BPSingleListener for gem5
git-svn-id: https://www4.informatik.uni-erlangen.de/i4svn/danceos/trunk/devel/fail@1603 8c4709b5-6ec9-48aa-a5cd-a96041d1645a
2012-09-07 00:57:05 +00:00
c30011fc01 Gem5 is now linked against fail (not fully working)
git-svn-id: https://www4.informatik.uni-erlangen.de/i4svn/danceos/trunk/devel/fail@1466 8c4709b5-6ec9-48aa-a5cd-a96041d1645a
2012-08-01 16:40:16 +00:00
c06565aa4e Basic SAL files and makefile modifications for adding gem5.
git-svn-id: https://www4.informatik.uni-erlangen.de/i4svn/danceos/trunk/devel/fail@1457 8c4709b5-6ec9-48aa-a5cd-a96041d1645a
2012-07-17 15:35:29 +00:00