Commit graph

545 commits

Author SHA1 Message Date
Nathan Binkert
4bbe8c1058 Merge zizzer.eecs.umich.edu:/bk/m5
into ziff.eecs.umich.edu:/z/binkertn/research/m5/kernel

--HG--
extra : convert_revision : 9ee9b8550b5761fa400df83caf268bd1c8ddc7ed
2004-02-05 02:53:03 -05:00
Andrew Schultz
daa579fc0f Fix PCI code so it builds properly now
dev/pciconfigall.cc:
dev/pciconfigall.hh:
dev/pcidev.cc:
dev/pcidev.hh:
dev/tsunami.cc:
dev/tsunami.hh:
    A bunch of changes to clean up new PCI code and to fix build

--HG--
extra : convert_revision : 71063bcc565c50fc293b323ddce2c8e701f544ff
2004-02-05 02:25:45 -05:00
Ali Saidi
69e1e10f5d Added code so that BAR writes will result in an updated memory mapping
dev/pcidev.cc:
dev/pcidev.hh:
    BAR changes should now change the mmu mapping

--HG--
extra : convert_revision : 2d5c60ef076ab0588a25def1ecd9dbb90c9144d7
2004-02-04 19:56:24 -05:00
Andrew Schultz
aaaa782165 Removed the tsunami_pci* files
--HG--
extra : convert_revision : e4a65d2664f2cd952d206d2079e3d093a35c20aa
2004-02-04 15:04:36 -05:00
Andrew Schultz
f2b8df2420 Changes to support PciDev as a new base class for devices which use PCI
--HG--
extra : convert_revision : fc2bc4399665344220b1a7e345ae6d2df03c9960
2004-02-04 15:03:50 -05:00
Steve Reinhardt
98948b2e57 More reformatting of reference parameter declarations.
The last change only caught the ones with types that
started with capitals.  This pass catches the rest
(mostly STL and uint*_t types).

base/cprintf_formats.hh:
cpu/simple_cpu/simple_cpu.cc:
sim/serialize.cc:
sim/serialize.hh:
    Change "foo_t& foo" to "foo_t &foo".

--HG--
extra : convert_revision : fc7f7425db2aef33e490f952b5ce74c8c36d0d41
2004-02-04 11:16:30 -08:00
Andrew Schultz
579c5b59e1 Removed pci_dev from AdaptecController
--HG--
extra : convert_revision : 2e2b388ff48bd2718cc2208983e6ba5626317ee8
2004-02-03 18:56:39 -05:00
Andrew Schultz
9b18af0887 Forgot to fixup the boot_linux.ini file to use AdaptecController
--HG--
extra : convert_revision : a4d7aa7b795f3495dd1272a1014a492f916b4f38
2004-02-03 17:09:49 -05:00
Andrew Schultz
0a585921f0 Mostly changes to point Tsunami related stuff to use AdaptecController
dev/tsunami.cc:
dev/tsunami.hh:
    Change to use AdaptecController with Tsunami

--HG--
extra : convert_revision : 6ff4f9f7ae1959589340b2ab23755b4dbfec4dbc
2004-02-03 17:02:48 -05:00
Andrew Schultz
d08e1bc569 Fix to support redefinition of functional_memory base class
--HG--
extra : convert_revision : c06b2cfd2787022ee5dc664886873a9afa459434
2004-02-03 16:59:40 -05:00
Andrew Schultz
8a28933c24 Automerged
--HG--
extra : convert_revision : 7925eb08f4ebc3e846a1a304eeeb832b48862fca
2004-02-03 16:35:26 -05:00
Ali Saidi
53d81cfe32 Merge zizzer.eecs.umich.edu:/.automount/ziff/z/binkertn/research/m5/kernel
into zizzer.eecs.umich.edu:/y/saidi/work/m5

--HG--
extra : convert_revision : 2a71828fc732961ab2fcee1afee0f6daa092e4c9
2004-02-03 16:28:48 -05:00
Ali Saidi
0476a4ee12 Aded tsunami_pcidev and associated updates
--HG--
extra : convert_revision : f29f421c966e36aa372d0faef6257c1aea205a03
2004-02-03 16:23:51 -05:00
Nathan Binkert
85bc028185 merge
--HG--
extra : convert_revision : 47425264e672f727cbb13aa7b9bb2a67790b25e8
2004-02-03 11:24:03 -05:00
Nathan Binkert
368e6e3e57 Merge zizzer.eecs.umich.edu:/bk/m5
into ziff.eecs.umich.edu:/z/binkertn/research/m5/latest

--HG--
extra : convert_revision : 9b8a94cbaacbbb4a53d15c0893ee4d3183f29ce0
2004-02-03 10:51:10 -05:00
Nathan Binkert
b1221a3395 Remote debugging cleanup and internal debugging support
base/kgdb.h:
    Remove flags that aren't used
base/remote_gdb.cc:
    Better debugging:
    - Give each class a name() function so that the trace infrastructure
    knows the correct object name.
    - Make the remote debugger capable of detach.
    - Split out the RGDB trace flag into a bunch of specific flags.
    Remove dead code
    Add a new trap type
base/remote_gdb.hh:
    Add a name() to the various objects for the trace system
base/trace.hh:
    don't need a using directive
    add DPRINTFNR: debug printf, no flag, raw output
kern/tru64/tru64_system.cc:
    use the INT trap type instead of IF

--HG--
extra : convert_revision : 25e610216c6f43d5d328651bba915f71bade059e
2004-02-03 10:50:04 -05:00
Steve Reinhardt
3e5070a3f1 Fix bug: forgot branchTarget() method on indirect branches.
arch/alpha/isa_desc:
    Add missing branchTarget() method for indirect branches.
cpu/static_inst.hh:
    Add comment clarifying when branchTarget() can be used
    on indirect branches.

--HG--
extra : convert_revision : 0dcfb36a9792a338cefceb3d1501825abace7ac5
2004-02-03 07:19:05 -08:00
Steve Reinhardt
2f369ee5d5 Fix handling of uncacheable loads. Make them wait until they're at the tail of
the ROB (so we know they're correct-path) before we mark them ready to issue.

--HG--
extra : convert_revision : be1d431bdc939100d79a5fea3e415b3acf9e8d75
2004-02-02 19:09:47 -08:00
Steve Reinhardt
19cc144960 Get rid of LVQ-port function unit... only used for redundant
multithreading, which has been gone for a long time.

--HG--
extra : convert_revision : 717d13455483c3adc7b7cbbcc5f7493d51867974
2004-02-02 18:58:23 -08:00
Steve Reinhardt
9bfd18a155 Clean up some confusing ROBStation fields left over from SS (ea_comp, in_LSQ).
Eliminate indirect decoder.cc dependence on rob_station.hh via full_cpu.hh.

--HG--
extra : convert_revision : 9355cddd6cd8bbe100aae1b2d45d2752e76ba3f6
2004-02-02 18:53:27 -08:00
Steve Reinhardt
b79f6b3273 insn->inst
--HG--
extra : convert_revision : fcc556fb7e65855ec3c04ef272177c8e7a38fff9
2004-02-02 18:14:11 -08:00
Steve Reinhardt
548b4a1cb0 Change "Foo& foo" declarations to "Foo &foo".
This primarily to be internally consistent (sometimes we used one,
sometimes the other, even within the same line of code!).
I picked the latter to be symmetric with "Foo *foo".

base/cprintf_formats.hh:
base/range.hh:
base/refcnt.hh:
base/res_list.hh:
base/statistics.hh:
base/str.hh:
cpu/exec_context.hh:
cpu/simple_cpu/simple_cpu.cc:
cpu/simple_cpu/simple_cpu.hh:
sim/serialize.cc:
sim/serialize.hh:
sim/syscall_emul.hh:
    Change "Foo& foo" declarations to "Foo &foo".

--HG--
extra : convert_revision : ca1b0e85a578b539214bda3b8d61ac23792f2e87
2004-02-02 15:55:35 -08:00
Steve Reinhardt
956aff1291 More conversions of ref-counted parameters to references:
Statistics::NodePtr, StaticInstPtr, ethernet PacketPtr.

base/statistics.hh:
    Change NodePtr parameters to Nodeptr &.
cpu/exetrace.hh:
    Change StaticInstPtr parameters to StaticInstPtr &.
dev/etherbus.cc:
dev/etherbus.hh:
dev/etherdump.cc:
dev/etherdump.hh:
dev/etherint.hh:
dev/etherlink.cc:
dev/etherlink.hh:
dev/ethertap.cc:
dev/ethertap.hh:
    change PacketPtr parameters to PacketPtr &.

--HG--
extra : convert_revision : a778efdca33b0ec5beb76cf47db0e9e4728897ee
2004-02-02 15:45:22 -08:00
Steve Reinhardt
7b07b0877f Change MemReqPtr parameters to references.
This avoids incrementing and decrementing the MemReq
reference counters on every call and return.

arch/alpha/alpha_memory.cc:
arch/alpha/alpha_memory.hh:
cpu/exec_context.hh:
cpu/memtest/memtest.cc:
cpu/memtest/memtest.hh:
dev/alpha_console.cc:
dev/alpha_console.hh:
    Change MemReqPtr parameters to references.

--HG--
extra : convert_revision : 3ba18bdd9f996563988402576bfdd3430e1ab1e5
2004-02-02 10:47:21 -08:00
Steve Reinhardt
7c8413db10 Make stats:dump_period work even if dump_cycle isn't set.
--HG--
extra : convert_revision : 9e85c0fa0f6c5393dd8e91034a49975628cf825d
2004-02-02 07:19:45 -08:00
Andrew Schultz
64b6bad6ab Change configs to use SimpleController instead of ScsiController
--HG--
extra : convert_revision : f8288b13cc0d4ad8ef5cab458012cb5a5cb242c4
2004-01-31 05:38:16 -05:00
Andrew Schultz
240a434e88 Split the SCSI controller into a base class (ScsiController) and
two seperate implementations (SimpleController for Tru64 and
AdaptecController for Linux)

--HG--
extra : convert_revision : 5458a42d34e6001b3ea11b87511b1f9ee853b4f7
2004-01-31 05:36:14 -05:00
Ali Saidi
5164de4a03 added tsunami_pcidev
--HG--
extra : convert_revision : 14ad0f7afd930c29ff45f6edd51db8d8f8a07be0
2004-01-31 01:25:13 -05:00
Nathan Binkert
bbf952376e Update for changes in the console
This is basically the same thing that was done for tlaser_uart

--HG--
extra : convert_revision : e0de84b83d1ee4fe0ef769b4de232dbbd7d9c228
2004-01-30 18:12:48 -05:00
Nathan Binkert
35fb439c33 Merge zizzer.eecs.umich.edu:/bk/m5
into ziff.eecs.umich.edu:/z/binkertn/research/m5/kernel

--HG--
extra : convert_revision : 9bb172c13912f69c53897ae215e077c9b8194dd9
2004-01-30 18:10:15 -05:00
Andrew Schultz
60279bb67e Fixed the spec2000.ini file and made some cosmetic changes to the
setup-spec script

util/setup-spec:
    Made some changes so the information contained in the ini file explains
    what the init_param value corresponds to

--HG--
extra : convert_revision : 4082209f920244c262150c224a92ea21a55febd5
2004-01-30 17:38:25 -05:00
Andrew Schultz
74ae052b69 Added the spec2000 ini file
--HG--
extra : convert_revision : 981a5a536a25b07dd54ed0166b291876b0a498c0
2004-01-30 17:20:03 -05:00
Nathan Binkert
4ae64216c3 Merge
--HG--
extra : convert_revision : 8690e31b64235874d74ea4a1123a408610fb115b
2004-01-30 16:01:26 -05:00
Nathan Binkert
a0c3e61d32 Merge Sampler
--HG--
extra : convert_revision : d10232f4602790f3b325e2e10e3098caaeaa53b3
2004-01-30 15:59:06 -05:00
Ali Saidi
0035440536 Linux boots with no devices
dev/tsunami.hh:
    Started commenting code
dev/tsunami_cchip.cc:
    removed unneccessary config files
dev/tsunami_io.cc:
    Added code to see the value written
dev/tsunami_uart.cc:
    conviently one of the addresses the SuperI/O southbridge can be is the same space
    as the UART. This stops the simulator from panicing although it should probably be
    changed a bit.

--HG--
extra : convert_revision : a3334a2c418ee8228089d0e1791fa78bbb276fe5
2004-01-30 15:24:50 -05:00
Steve Raasch
2f4f7aacf7 major rework/rewrite of the breakdown script
--HG--
extra : convert_revision : 2e1d186e0f20b6089300b82a3f6dcb1d0b44c09d
2004-01-30 13:51:09 -05:00
Steve Reinhardt
7d64183b44 Use $(MAKE) instead of 'make' when building libelf (suppresses an error message about -j).
Check for, delete, and die on zero-length .d files.

--HG--
extra : convert_revision : 71b7f9e639d96fb7ed30952b142d69c34c725774
2004-01-30 09:16:34 -08:00
Steve Reinhardt
e7e6717a1a Fix libelf generation so it works on a clean make -j.
--HG--
extra : convert_revision : 377633aa2c04695c598ee82a9529d2e40f0ffef7
2004-01-30 08:18:20 -08:00
Nathan Binkert
9374efa725 wrap debugging stuff with #ifdef DEBUG
--HG--
extra : convert_revision : 8205633c1c1b49b04f389aa40f95fbbc03a43fb6
2004-01-29 22:30:14 -05:00
Nathan Binkert
cb35f819c5 delete the data in the arglist when the list is destroyed,
not while printing out the data.  This allows the data
to be dumped more than once.

base/cprintf.hh:
    need a destructor

--HG--
extra : convert_revision : 235e9fe24488ac4c0ae1b562ef9fa6e0bd1e899c
2004-01-29 17:44:08 -05:00
Nathan Binkert
fe4d9f124f whack debugging statement
--HG--
extra : convert_revision : f623d35a6e5a234e2b203974012c653b91826dbf
2004-01-29 16:33:03 -05:00
Nathan Binkert
ee4263f72e Fix character input by handling the character and the
special console values separately.

dev/alpha_console.cc:
    use new console specific input function

--HG--
extra : convert_revision : 08997d6115d2aac3a26cac2774b3c3fc0cd76ab0
2004-01-29 16:32:03 -05:00
Nathan Binkert
639037d127 remove #include of nonexistent file
sim/universe.cc:
    oops.  this doesn't exist

--HG--
extra : convert_revision : 2cfb3680e4ebe3f27f22a79f853d4d6df445e65a
2004-01-29 10:32:01 -05:00
Nathan Binkert
2db1b6ea1b provide an output stream for simulator output. (This takes place of the
statStream catchall that we had before)
Also provide an optional output directory that multiple simulator output
files can be written to.
Make most output files use the output directory

base/misc.cc:
    send warnings to the outputStream as well
base/trace.cc:
    dprintf_stream defaults to cerr
dev/console.cc:
    use the output directory for the console output if it exists
dev/etherdump.cc:
    dump to the output directory if it exists
sim/builder.cc:
sim/builder.hh:
    move constructor and destructor to .cc file
    use a function to get the stream that the builder dumps its
    output to, and create a separate file in the output directory
    if able
sim/main.cc:
    statStream -> outputStream
sim/serialize.cc:
    dump checkpoints to the output directory if specified
sim/universe.cc:
    provide an output stream for simulator output.  (This takes place of the
    statStream catchall that we had before)
    Also provide an optional output directory that multiple simulator output
    files can be written to.

--HG--
extra : convert_revision : 03abce20edbbf7ec19c9ddd8d69ec8485c383532
2004-01-29 00:38:18 -05:00
Nathan Binkert
f994cf4b5e use $ENV{PWD} so that the amd paths work ok.
Also, compare real paths to ensure that the links point
to the right place

--HG--
extra : convert_revision : f4127a67f02b714808c6d2181fabf245ed2309c0
2004-01-29 00:37:07 -05:00
Nathan Binkert
dc70ce3a60 cleanup on aisle 5
base/statistics.cc:
    dead code

--HG--
extra : convert_revision : 8f3fd638acdf7a704475ea90b607a3225a3c174d
2004-01-29 00:36:22 -05:00
Andrew Schultz
2fb876e55d Remove references to dep_tree to fix build
--HG--
extra : convert_revision : dff311c1a18bd01750c1b5902410c23e0f43b798
2004-01-28 23:00:30 -05:00
Ali Saidi
00c49783ef moved devices around in config space and support figuring out cpuid
from cchip register

dev/tsunami_cchip.cc:
    added support for figuring out which cpu you are

--HG--
extra : convert_revision : 7862678a259931bb0a5b2ca8ad298a704bd272ec
2004-01-28 20:23:40 -05:00
Ron Dreslinski
d41c904402 Add support for PIC interrupts in IO, and DIRx interrupts in CChip
dev/tsunami_cchip.cc:
dev/tsunami_cchip.hh:
    Add Interrupt capabilities for DIRx, added postDRIR and clearDRIR
    functions
dev/tsunami_io.cc:
    Add PIC interrupts, and post/clearPIC functions
dev/tsunami_io.hh:
    Add support for PIC interrupts, added post/clearPIC functions

--HG--
extra : convert_revision : b705568670b157c1a4496c365226526fa96e21e0
2004-01-28 19:18:29 -05:00
Ali Saidi
b6fba57065 our first interrupt
cpu/exetrace.cc:
    added looking for symbols at PC+4 and PC+8 thanks to gcc skiping
    setting the gp where it can and jumping <func>+8
dev/console.cc:
    commented out weird interrupt per nate's suggestion
dev/tsunami_cchip.cc:
    moved rtc flag to correct bit
dev/tsunami_io.cc:
    time interrupt will be 1024Hz and at some point be configurable by
    linux
dev/tsunami_io.hh:
    Timer interrupt will be 1024hz for now and in the future be
    configurable by linux

--HG--
extra : convert_revision : 2fcc924c8848eb3c6166d9d517617ed193a2b89a
2004-01-28 18:12:52 -05:00