Commit graph

4109 commits

Author SHA1 Message Date
Gabe Black
10871b7342 Add build hooks for x86.
--HG--
extra : convert_revision : 438eb74f14e6ea60bab5012110f3946c9213786e
2007-03-03 16:01:48 +00:00
Gabe Black
94133657a1 Merge zizzer.eecs.umich.edu:/bk/newmem
into  ahchoo.blinky.homelinux.org:/home/gblack/m5/newmem

--HG--
extra : convert_revision : dcb1fc0c6252fb96a956640c6d7995679da725e5
2007-03-03 03:34:52 +00:00
Ali Saidi
4e8d2d1593 make ldtw(a) -- Twin 32 bit load work correctly -- by doing it the same way as the twin 64 bit loads
src/arch/isa_parser.py:
src/arch/sparc/isa/decoder.isa:
src/arch/sparc/isa/operands.isa:
src/base/bigint.hh:
src/cpu/simple/atomic.cc:
src/cpu/simple/timing.cc:
src/mem/packet_access.hh:
    make ldtw(a) Twin 32 bit load work correctly

--HG--
extra : convert_revision : 2646b269d58cc1774e896065875a56cf5e313b42
2007-03-02 22:34:51 -05:00
Gabe Black
d8ada247f4 Forgot to commit this new file last earlier.
--HG--
extra : convert_revision : f2d80ae551b7e29426141d5c9fe355b43a0b9c7d
2007-03-02 14:43:27 +00:00
Gabe Black
ececf101c7 Make the m5 psuedo instructions use the BasicOperate format
--HG--
extra : convert_revision : f02da702ab9b99da124fac7e10a07386b04f3a0f
2007-02-28 16:49:17 +00:00
Gabe Black
eb57b4f214 Merge zizzer.eecs.umich.edu:/bk/newmem
into  ahchoo.blinky.homelinux.org:/home/gblack/m5/newmem-sparc32

--HG--
extra : convert_revision : 88d1401f6e6b7c82344abef2c81b3c22bf6a0499
2007-02-28 16:39:42 +00:00
Gabe Black
29e5df890d Make trap instructions always generate TrapInstruction Fault objects which call into the Process object to handle system calls. Refactored the Process objects, and move the handler code into it's own file, and add some syscalls which are used in a natively compiled hello world. Software traps with trap number 3 (not syscall number 3) are supposed to cause the register windows to be flushed but are ignored right now. Finally, made uname for SPARC report a 2.6.12 kernel which is what m22-018.pool happens to be running.
--HG--
extra : convert_revision : ea873f01c62234c0542f310cc143c6a7c76ade94
2007-02-28 16:36:38 +00:00
Gabe Black
99948060b2 The "hostname" variable isn't used in the process classes. It should be removed from the other ones as well.
--HG--
extra : convert_revision : 0c07534de42d6c32ac26d9e43709111e3ab30d57
2007-02-28 16:29:25 +00:00
Gabe Black
dc8cc8de3d Fix issue with twolf where the presence or absence of two files, smred.sav or smred.sv2, would affect the outcome of the program. These names are based on the input file names which are in turn based on the input set selected. There may be more files like this generated for larger input sets, for example "mdred.sv3"
--HG--
extra : convert_revision : f4f1d3fd0fb28468b0ee507aaadf3c14aa9cf924
2007-02-27 10:37:48 +00:00
Gabe Black
a86ad3a512 Merge zizzer.eecs.umich.edu:/bk/newmem
into  ahchoo.blinky.homelinux.org:/home/gblack/m5/newmem

--HG--
extra : convert_revision : 7e8c3572ede7d93910fc3e2a2e76d9a38b1f4243
2007-02-26 10:34:45 +00:00
Ali Saidi
f892608ff7 Merge zizzer:/bk/newmem
into  zeep.pool:/z/saidi/work/m5.newmem

--HG--
extra : convert_revision : a4f80ce975a23ba9858e6bf2dbbfed8897dd1810
2007-02-24 22:10:06 -05:00
Ali Saidi
cf0e202cba make m5 readfile work on solaris... we can have a solaris regression soon!
src/arch/sparc/isa/decoder.isa:
    add readfile and break to sparc decoder
src/arch/sparc/isa/operands.isa:
    fix O0-O5 operands registers
util/m5/Makefile.sparc:
    Make sparc makefile compile a 64bit binary
util/m5/m5.c:
    readfile was in here twice, once will be sufficient I think
util/m5/m5op_sparc.S:
    implement readfile and debugbreak

--HG--
extra : convert_revision : 139b3f480ee6342b37b5642e072c8486d91a3944
2007-02-24 22:05:01 -05:00
Gabe Black
c0c3a3f491 Merge zizzer.eecs.umich.edu:/bk/newmem
into  ahchoo.blinky.homelinux.org:/home/gblack/m5/newmem

--HG--
extra : convert_revision : e0eb0240848698496bd55093a313eb2e0f512ebc
2007-02-23 12:54:07 +00:00
Gabe Black
6ae4cae971 Ali and I both made the same change and we only need it once. I liked mine a little better.
--HG--
extra : convert_revision : 3a1b7856e6143ca089fd6e36492608377dfede19
2007-02-23 01:05:34 +00:00
Gabe Black
187cc99e4e Merge zizzer.eecs.umich.edu:/bk/newmem
into  ahchoo.blinky.homelinux.org:/home/gblack/m5/newmem-sparc32

--HG--
extra : convert_revision : a7697ea8457a03318e3fcf34775bf3ecc4786e8a
2007-02-23 01:05:33 +00:00
Ali Saidi
a5b73a6e33 Merge zizzer:/bk/newmem
into  pb15.local:/Users/ali/work/m5.newmem

--HG--
extra : convert_revision : 887b278dac6db5ea17ade641de84d0ab8b05db96
2007-02-22 20:05:32 -05:00
Gabe Black
341a1eed6c Merge zizzer.eecs.umich.edu:/bk/newmem
into  ahchoo.blinky.homelinux.org:/home/gblack/m5/newmem-sparc32

--HG--
extra : convert_revision : 70dcd9d1d669c1c619411389487b7910861550e3
2007-02-22 13:18:23 +00:00
Gabe Black
34b4722aee Make the m5 pseudo instructions only work in FS. Also, make sure any undefined opcodes in impdep2 (which in SE is all of them) trap with an illegal_instruction exception.
--HG--
extra : convert_revision : dd7848d0685e4cc6f5fd5e3b846a3f70b62ee30a
2007-02-22 13:17:51 +00:00
Nathan Binkert
c003dda793 Make it easier to turn off the remote debugger
--HG--
extra : convert_revision : d88784736df5f9b498770fb7e98f52715669c0e1
2007-02-21 22:25:48 -08:00
Ali Saidi
b750d6a597 Merge zizzer:/bk/newmem
into  pb15.local:/Users/ali/work/m5.newmem

--HG--
extra : convert_revision : e0057583132ce545eb1867b446484e8984b97282
2007-02-22 01:15:16 -05:00
Nathan Binkert
fa4c3d74fe Get rid of the ConsoleListener SimObject and just fold the
relevant code directly into the SimConsole object.  Now,
you can easily turn off the listen port by just specifying
0 as the port.

--HG--
extra : convert_revision : c8937fa45b429d8a0728e6c720a599e38972aaf0
2007-02-21 22:14:11 -08:00
Ali Saidi
63fef6b011 fix se compiling oops
--HG--
extra : convert_revision : ce7ac94da0ed6bad457a8a9e4c949b0c3b09c2ae
2007-02-22 01:11:04 -05:00
Nathan Binkert
783e642ed8 Make sure that all variables in the NSGigE device model are
initialized.

--HG--
extra : convert_revision : b4b156ed8e3c0c4c4f8043ff86dc232ebad38668
2007-02-21 20:45:05 -08:00
Nathan Binkert
8e77d771f9 Make comments refer to ticks not cycles
--HG--
extra : convert_revision : 4970a76890a3256073423a827dd0c55cfcb19a08
2007-02-21 20:35:30 -08:00
Ali Saidi
f01f8f1be6 Merge zizzer:/bk/newmem
into  zeep.pool:/z/saidi/work/m5.newmem

--HG--
extra : convert_revision : 4105ebbeca59206bece27f229ee810d594fb4310
2007-02-21 21:06:29 -05:00
Ali Saidi
7a2ecf9e26 add pseduo instruction support for sparc
util/m5/Makefile.alpha:
    Clean up to make it a bit easier to muck with
util/m5/Makefile.alpha:
    Make the makefile more reasonable
util/m5/Makefile.alpha:
    Remove authors from copyright.
util/m5/Makefile.alpha:
    Updated Authors from bk prs info
util/m5/Makefile.alpha:
    bk cp Makefile Makefile.alpha
src/arch/sparc/tlb.cc:
    Clean up the cache code a little bit and make sure the uncacbale bit is set when appropriate
src/arch/alpha/isa/decoder.isa:
src/sim/pseudo_inst.cc:
src/sim/pseudo_inst.hh:
    Rename AlphaPseudo -> PseudoInst since it's all generic
src/arch/sparc/isa/bitfields.isa:
src/arch/sparc/isa/decoder.isa:
src/arch/sparc/isa/includes.isa:
src/arch/sparc/isa/operands.isa:
    Add support for pseudo instructions in sparc
util/m5/Makefile.alpha:
util/m5/Makefile.sparc:
    split off alpha make file and sparc make file for m5 app
util/m5/m5.c:
    ivle and ivlb aren't used anymore
util/m5/m5op.h:
    stdint seems like a more generic better fit here
util/m5/m5op_alpha.S:
    move the op ids into their own header file since we can share them between sparc and alpha

--HG--
rename : util/m5/Makefile => util/m5/Makefile.sparc
rename : util/m5/m5op.S => util/m5/m5op_alpha.S
extra : convert_revision : 490ba2e8b8bc6e28bfc009cedec6b686b28e7834
2007-02-21 21:06:17 -05:00
Nathan Binkert
06ae2d0445 Fix compile issues on gcc 4.1.x related to namespaces.
This basically involves moving the builder code outside of any
namespace.  While we're at it, move a few braces outside of
a couple #if/#else/#endif blocks so it's easier to match up
the braces.

--HG--
extra : convert_revision : a7834532aadc63b0e0ff988dd5745049e02e6312
2007-02-21 16:42:16 -08:00
Nathan Binkert
2a67f2b08c Fix tracing so it starts right away if --trace-start is not
specified.

--HG--
extra : convert_revision : 49c1ea0b8c313949124aed84b1055db0b3c55bd8
2007-02-21 14:08:13 -08:00
Gabe Black
f8ffc84d0b Merge zizzer.eecs.umich.edu:/bk/newmem
into  ahchoo.blinky.homelinux.org:/home/gblack/m5/newmem

--HG--
extra : convert_revision : bab45577db1967de1dd88ec9b228f106a4ab7479
2007-02-21 18:30:52 +00:00
Nathan Binkert
a329631edb Automatically generate m5/internal/__init__.py and swig/init.cc
based on the swig modules that we have

--HG--
extra : convert_revision : 2fd12db39d46608a62b9df36c2b36189f1d2bc30
2007-02-21 10:30:51 -08:00
Nathan Binkert
3fb3616be4 Fix majory brokenness in my previous MySQL commit, basically
this is just a shuffling around of code and fixes to make
stuff commit properly

--HG--
extra : convert_revision : a057f7fe4962cfc6200781ff66d2c26bf9c6eb8c
2007-02-21 10:15:17 -08:00
Nathan Binkert
5000c4d878 #include needed for compile
--HG--
extra : convert_revision : fda9ab0d04f77f27810018a8639d6ea8abb59326
2007-02-21 10:13:10 -08:00
Gabe Black
3afbf306f4 Merge zizzer.eecs.umich.edu:/bk/newmem
into  ahchoo.blinky.homelinux.org:/home/gblack/m5/newmem

tests/long/30.eon/ref/alpha/linux/simple-timing/m5stats.txt:
    "Hand merge" that just used the local file.

--HG--
extra : convert_revision : 24c7fa192094958be5a9d17c3461f3328079fd3b
2007-02-21 00:38:11 +00:00
Gabe Black
7ae834446d Update 50.vortex simple-timing for 8k blk_size
--HG--
extra : convert_revision : 73f8c4f8f6da901021ea38e5ac053d905454a3ff
2007-02-21 00:25:50 +00:00
Gabe Black
fb7169848c Update 50.vortex simple-atomic for 8k blk_size
--HG--
extra : convert_revision : 82a131bf16b856dadf62a678ce74350079433692
2007-02-21 00:24:45 +00:00
Gabe Black
4674fce113 Update 40.perlbmk simple-timing for 8k blk_size
--HG--
extra : convert_revision : 461751061dc5db076f11e9c3b37da25cc47c583e
2007-02-21 00:23:52 +00:00
Gabe Black
e574b42d9c Update 40.perlbmk simple-atomic for 8k blk_size
--HG--
extra : convert_revision : 135274a64ead4962faa4f34b2df4e9de453cbe7f
2007-02-21 00:23:02 +00:00
Gabe Black
3d4fe3f459 Update 30.eon simple-timing for 8k blk_size. It's strange this is necessary because simple-atomic doesn't seem affected.
--HG--
extra : convert_revision : 6a8c77f0ca76eb06ac7eb5216af6adba3759c4c7
2007-02-21 00:21:43 +00:00
Ali Saidi
9062525c23 Merge zizzer:/bk/newmem
into  zeep.pool:/z/saidi/work/m5.newmem

--HG--
extra : convert_revision : 10d4dc08411c7a433a7194e94f69ca1d639a1ce7
2007-02-18 19:57:58 -05:00
Ali Saidi
bd367d4825 implement vtophys and 32bit gdb support
src/arch/alpha/vtophys.cc:
src/arch/alpha/vtophys.hh:
src/arch/sparc/arguments.hh:
    move Copy* to vport since it's generic for all the ISAs
src/arch/sparc/isa_traits.hh:
    the Solaris kernel sets up a virtual-> real mapping for all memory starting at SegKPMBase
src/arch/sparc/pagetable.hh:
    add a class for getting bits out of the TteTag
src/arch/sparc/remote_gdb.cc:
    add 32bit support kinda.... If its 32 bit
src/arch/sparc/remote_gdb.hh:
    Add 32bit register offsets too.
src/arch/sparc/tlb.cc:
    cleanup generation of tsb pointers
src/arch/sparc/tlb.hh:
    add function to return tsb pointers for an address
    make lookup public so vtophys can use it
src/arch/sparc/vtophys.cc:
src/arch/sparc/vtophys.hh:
    write vtophys for sparc
src/base/bitfield.hh:
    return a mask of bits first->last
src/mem/vport.cc:
src/mem/vport.hh:
    move Copy* here since it's ISA generic

--HG--
extra : convert_revision : c42c331e396c0d51a2789029d8e232fe66995d0f
2007-02-18 19:57:46 -05:00
Nathan Binkert
4e7f8c0885 Get rid of the stand alone ParamContext since all of the
relevant stuff has now been moved to python.

--HG--
extra : convert_revision : 608e5ffd0e2b33949a2b183117216f136cfa4484
2007-02-18 09:31:25 -08:00
Nathan Binkert
ee93b48314 Get rid of the Serialize and IntervalStats Param contexts
since they're no longer used

--HG--
extra : convert_revision : e39590aa03cc4c961d2eb5dab57862811f431e4d
2007-02-18 09:08:32 -08:00
Nathan Binkert
80b4347d87 The trace_mem function is really part of the encumbered
FullCPU

--HG--
extra : convert_revision : d7fc876333fc3474bc3827f710aa472e2ad847f4
2007-02-18 08:50:21 -08:00
Nathan Binkert
e94103397c Get rid of the Statistics and Statreset ParamContexts, and
expose all of the relevant functionality to python.  Clean
up the mysql code while we're at it.

--HG--
extra : convert_revision : 5b711202a5a452b8875ebefb136a156b65c24279
2007-02-17 22:52:32 -08:00
Nathan Binkert
01f32efa4b Check that there is a param context list before trying
to loop through it.  This is more important as we get rid
of param contexts

--HG--
extra : convert_revision : 5a24048b5c3d609285da83dfcb106910afad6919
2007-02-17 22:36:39 -08:00
Nathan Binkert
a41f17b40e Remove the event_ignore stuff since it was never really used
--HG--
extra : convert_revision : ef5f3492e8232d08af7e1eae64ba96c79ca14b6f
2007-02-17 22:11:21 -08:00
Nathan Binkert
8c1c68a31e Give the progress event its own priority
--HG--
extra : convert_revision : 6357ade64deb42fae68b2766545b1c4cdc673fc9
2007-02-17 22:07:50 -08:00
Nathan Binkert
08f024d3ff Default to tracing being disabled in C++, it will be turned
on in python.  Fix the trace start code so it actually starts
when it is suppsed to.  Make the Exec tracing stuff obey the
trace enabled flag.

--HG--
extra : convert_revision : 634ba0b4f52345d4bf40d43e239cef7ef43e7691
2007-02-17 20:32:39 -08:00
Nathan Binkert
18e245ad0b Pass an exception from a python event through the event queue
back into python so we don't just silently ignore those errors

--HG--
extra : convert_revision : e2f5566a4681f1b8ea80af50071119118afa7d8a
2007-02-17 20:27:11 -08:00
Gabe Black
3cb2686828 Update for 8k block size.
--HG--
extra : convert_revision : 6249d3697c74a0f9f9b2f03857b75785b02bfc8b
2007-02-17 20:35:00 -05:00