Commit graph

7382 commits

Author SHA1 Message Date
Nathan Binkert
f08cb65703 clean up benchmarks differentiate the two surge benchmarks
--HG--
extra : convert_revision : 8fd9717b34a5cdfa9e36a421e3df7cd957becb9c
2004-11-15 16:11:30 -05:00
Nathan Binkert
63dbaccdc4 more whitespace fixes
configs/boot/spec-surge-client.rcS:
configs/boot/spec-surge-server.rcS:
    whitespace fixes

--HG--
extra : convert_revision : 90eb2a325a3cf4436081c992d6f3e0f903980ff7
2004-11-15 16:07:28 -05:00
Nathan Binkert
abbbea57a7 more formatting
configs/boot/nfs-server.rcS:
    fix whitespace

--HG--
extra : convert_revision : d421d5fa1e041b041d7c47d36abe55a2dd3e7487
2004-11-15 15:45:55 -05:00
Nathan Binkert
2cc34a34df formatting
configs/boot/nfs-client.rcS:
    fix whitespace

--HG--
extra : convert_revision : dc384870de9431fe30b13a2c8c69d97b2d06dca9
2004-11-15 15:44:25 -05:00
Nathan Binkert
49d470a3b5 cleanup the nfs server script
configs/boot/nfs-server.rcS:
    don't need extra IP aliases

    re-order the nfs daemons so they happen in the right order
    and don't have the shell put them in the background since they
    will auto detach anyway.

--HG--
extra : convert_revision : a6b13ac91d1649878d346663f6ddff56d9a8a8ec
2004-11-15 15:42:15 -05:00
Nathan Binkert
f7323aebe8 fix udp checksums
base/inet.hh:
    tcp/udp/ip checksums are calculated/manipulated in NBO always,
    so there's no need to convert to host endian.

--HG--
extra : convert_revision : 13d6aff025a6c072d9d829c2e9fad4f13399b1f7
2004-11-15 15:40:35 -05:00
Steve Reinhardt
c250a16e05 Merge zizzer.eecs.umich.edu:/z/m5/Bitkeeper/m5
into zizzer.eecs.umich.edu:/z/stever/bk/m5

--HG--
extra : convert_revision : 033375b63b56104a5c1823985eeee7bb6cdcc7de
2004-11-15 02:44:23 -05:00
Nathan Binkert
bd4f5f843d mount surge image read-only
--HG--
extra : convert_revision : b91a93f9dc004c587d3ec763ecde98e5bd241159
2004-11-15 02:44:22 -05:00
Steve Reinhardt
fed64a3b36 Add support for sampled PC profiling to FullCPU.
Simple text list of symbol (or address) and count
will be dumped to m5prof.<cpu-name> if the cpu's
pc_sample_interval param is set.

SConscript:
    Add cpu/full_cpu/pc_sample_profile.cc
base/callback.hh:
    Add a comment about MakeCallback.
    Fix type in another comment.
base/loader/symtab.cc:
    Revamp findNearestSymbol() to provide addresses of both
    nearest symbols (preceding and following) as well as
    string for former.

    Move global definition of debugSymbolTable here too.
base/loader/symtab.hh:
    Revamp findNearestSymbol() to provide addresses of both
    nearest symbols (preceding and following) as well as
    string for former.

    Move global declaration of debugSymbolTable here too.
cpu/exetrace.cc:
    Use new findNearestSymbol() interface for trace symbols.
kern/linux/linux_system.cc:
sim/system.cc:
    Remove extern of debugSymbolTable (now in symtab.hh)
sim/process.cc:
    Initialize debugSymbolTable if binary has a symbol table.

--HG--
extra : convert_revision : 0b5393dc39c40ac88c953684708f1125da550671
2004-11-15 01:56:40 -05:00
Nathan Binkert
928656aecb make it so we can individually use the sampling cpu on
either the client or the server.

Make an ini file that makes our benchmarks easier to run.

--HG--
extra : convert_revision : c270cd1c027cb40d5c5bcfc2ed63f9db4b155116
2004-11-15 01:55:24 -05:00
Ali Saidi
0097ab1ecb Merge zizzer:/bk/m5 into zeep.eecs.umich.edu:/tmp/m5
--HG--
extra : convert_revision : 3c7b8a86b528a61f5cb14c86ec7c1eb9276824af
2004-11-15 01:36:28 -05:00
Nathan Binkert
10e88b43fd Fix more of the stats python stuff
util/stats/info.py:
    Make the binnings stuff work again.
util/stats/stats.py:
    small patch for graphing
    make it so we can print out bins for the stat command

--HG--
extra : convert_revision : c0279ac7030fd5146dd00801baa41e7baf97d1f4
2004-11-15 01:36:27 -05:00
Ali Saidi
e462e9627a commit ploticus code to repository
--HG--
extra : convert_revision : 861423a859d05e7915d9f83ef40c188fe21bcb31
2004-11-15 00:43:34 -05:00
Ali Saidi
ed1ff7fca9 updated to make memory a touch slower in the HT case and the
make sure the interrupt controller is on cpu die

--HG--
extra : convert_revision : 07b7eadbec7694859957279c81b9bdb4ebd044a3
2004-11-14 18:52:04 -05:00
Nathan Binkert
53b606683b New system configurations that work around the bus bridge
memory system bug

--HG--
extra : convert_revision : 29aa265126691eee386bc73a86e5d382e12b59a9
2004-11-14 17:41:56 -05:00
Ron Dreslinski
9eedd19cfb Merge zizzer:/z/m5/Bitkeeper/m5
into zazzer.eecs.umich.edu:/z/rdreslin/m5bk/clean

--HG--
extra : convert_revision : 19dce611da2e93f85af1eef1d590d820e8db3c5f
2004-11-14 17:31:59 -05:00
Ron Dreslinski
050ea6c522 Fix some small bugs with bus blocking and adress arbitration
Fix it so that when snoop response is called on other interfaces besides master, it just does nothing

--HG--
extra : convert_revision : d1e29b738c932642d53c011b50cfd915f67af51b
2004-11-14 17:31:43 -05:00
Nathan Binkert
9d80d88a71 update graph generating code
util/stats/stats.py:
    tweak the graphing stuff for the new configurations we have.
    add more graph types.
    nsgige -> etherdev
    deal with memory hierarchy change by using L2 instead of L3

--HG--
extra : convert_revision : 55362e79d9f8d0d68aa08129f5af944b378a9f4c
2004-11-14 17:22:33 -05:00
Ron Dreslinski
f7d1166e04 Merge zizzer:/z/m5/Bitkeeper/m5
into zazzer.eecs.umich.edu:/z/rdreslin/m5bk/clean

--HG--
extra : convert_revision : 170f5fd8891b02ad3cc04112c6f304ede3254dae
2004-11-14 16:19:11 -05:00
Ali Saidi
6650da85e7 by default bin interrupts seperately from kernel code.
--HG--
extra : convert_revision : fb890907402b7c1e873a2ac32e72cb6e0ed73f13
2004-11-14 15:26:48 -05:00
Ali Saidi
a89398e262 patched nfs boot scripts
configs/boot/nfs-server.rcS:
    made the partition size correct

--HG--
extra : convert_revision : 4e9bdbe3e465aea2a914fd928b47296031c2523b
2004-11-13 23:09:17 -05:00
Ali Saidi
f63dd31f4e fixed nfs scripts hopefully for the last time
configs/boot/nfs-client.rcS:
    lucky #3

--HG--
extra : convert_revision : a55f29e4c087e7ab95ea584316ad6fe4ebb5df9a
2004-11-13 23:08:08 -05:00
Steve Reinhardt
2e0695ec9a Get rid of obsolete sim/sim_stats.* files (looks like these
are replaced by sim/stats.hh and sim/stat_control.*)

dev/ns_gige.cc:
dev/sinic.cc:
    Include sim/stats.hh instead of sim/sim_stats.hh

--HG--
extra : convert_revision : 5e07932eab45ae4fb719baa4f94c5f62092a8446
2004-11-13 21:13:25 -05:00
Nathan Binkert
7e4229fb8f Add the Simple Integrated Network Interface Controller
--HG--
extra : convert_revision : 2bce25881a104e8282a5ed819769c6a7de414fb2
2004-11-13 17:10:48 -05:00
Nathan Binkert
9f8db6f446 don't handle device specific configuration stuff in the base
class.  While we're at it, very minor formatting.

--HG--
extra : convert_revision : b7c719f92ece7234838434294b33833289adbada
2004-11-13 17:05:13 -05:00
Nathan Binkert
4760ae46c6 Use the new PacketFifo class to avoid manual calculations
--HG--
extra : convert_revision : afa193904b7ed4d5e5c50e9dcb78e8e855b00ecc
2004-11-13 16:52:08 -05:00
Nathan Binkert
8922d69953 change the serialization of a packet so that we don't
do a nameOut.  This fixes a subtle bug in serialization that
can pop up.

--HG--
extra : convert_revision : b52df977dcbef1c9bd0d4405ba0b36dff3737cdf
2004-11-13 16:46:56 -05:00
Nathan Binkert
acb98fb0f6 we shouldn't ever pass around references to PacketPtrs,
const references are ok, or pass by value.

--HG--
extra : convert_revision : 7280a1c7d22b9294fddbe50f02f6f4c6ca9b2e5b
2004-11-13 16:33:16 -05:00
Nathan Binkert
bd3e3c0230 forgot a change in the previous commit. the ide controller
doesn't have its own interrupt functions

dev/ide_ctrl.hh:
    oops. we don't have our own interrupt functions anymore
    we get them from the base class.

--HG--
extra : convert_revision : 3eac228ec59f4fea0b0e49f961e8b21705dee27f
2004-11-13 16:24:47 -05:00
Nathan Binkert
b0fd676a25 Make it possible for PioDevices to have unknown pio
latencies.

--HG--
extra : convert_revision : 08f2250efc078406c497a6820ae30f5fdf687ea5
2004-11-13 16:04:33 -05:00
Nathan Binkert
b031888038 Use parameter structs for initialization so it's easier
to add new devices.  Abstract the Platform more so that
it is unnecessary to know know platform specifics for
interrupting or translating PCI DMA addresses.

dev/ide_ctrl.cc:
    convert to parameter struct for initialization
    use the interrupt functions in the PciDev base class
    convert from tsunami to using platform
    We don't need an interrupt controller here.
dev/ide_ctrl.hh:
    don't use Tsunami, use Platform
    make the IdeDisk a friend so that it can access my plaform
    convert to parameter struct for construction
dev/ide_disk.cc:
    don't use tsunami references, but platform references
dev/ns_gige.cc:
    Convert to parameter struct for initialzation. Use code in
    base class for interrupts so we don't need to know anything
    about the platform. Don't need an IntrControl *.
dev/ns_gige.hh:
    We don't need a Tsunami * anymore
    convert to a parameter struct for construction
dev/pcidev.cc:
    deal with new parameter struct
dev/pcidev.hh:
    - Move all of the configuration parameters into a param struct
    that we can pass into the constructor.
    - Add a Platform * for accessing new generic interrupt post/clear
    and dma address translation fuctions
    - Create functions for posting/clearing interrupts and translating
    dma addresses
dev/platform.cc:
    have default functions that panic on pci calls
dev/platform.hh:
    don't make the pci stuff pure virtual, but rather provide
    default implementations that panic.  Also, add dma address
    translation.
dev/tsunami.cc:
    this-> isn't necessary here.
    add pci address translation
dev/tsunami.hh:
    implement the pciToDma address translation

--HG--
extra : convert_revision : 7db27a2fa1f1bd84704921ec7ca0280b5653c43e
2004-11-13 15:45:22 -05:00
Nathan Binkert
e9f3279334 sort #includes
don't need to include ev5.hh anymore

--HG--
extra : convert_revision : f80d2767936d3fb07a9cb7fd4709cafe9ea93e63
2004-11-13 14:46:02 -05:00
Nathan Binkert
425dda00df Macros are nasty, so let's get rid of them. Convert all
all macros in ev5.hh to inline functions or constant typed
variables and make them follow our style while we're at it.

All of the stuff in this file actually belongs in the ISA
traits code, but this is a first step at getting things done
in the right manner.

arch/alpha/alpha_memory.cc:
arch/alpha/alpha_memory.hh:
arch/alpha/ev5.cc:
arch/alpha/isa_desc:
dev/ns_gige.cc:
kern/tru64/tru64_events.cc:
    deal with changes in ev5.hh
arch/alpha/ev5.hh:
    Macros are nasty, so let's get rid of them.  Convert all
    all macros to inline functions or constant typed variables.
    Make them follow our style while we're at it.

    All of the stuff in this file actually belongs in the ISA
    traits code, but this is a first step at getting things done
    in the right manner.
arch/alpha/isa_traits.hh:
    move some of the ev5 specific code into the isa
arch/alpha/vtophys.cc:
base/remote_gdb.cc:
    deal with isa addition
cpu/exec_context.hh:
    be less isa specific and use the isa traits to figure out
    what we can.
dev/alpha_console.cc:
dev/pciconfigall.cc:
dev/tsunami_cchip.cc:
dev/tsunami_io.cc:
dev/tsunami_pchip.cc:
dev/uart.cc:
    deal with changes in ev5.hh
    I don't believe this masking is actually necessary.  We should
    look at removing it later.
dev/ide_ctrl.cc:
    sort #includes
    deal with changes in ev5.hh

--HG--
extra : convert_revision : c8a3adf0a4b1d198aefe38fc38b295abf289b08a
2004-11-13 14:01:38 -05:00
Nathan Binkert
8f74f77cf2 remove the global static check_interrupts variable now that
it's not used.

--HG--
extra : convert_revision : a4f50c1eb030a132766c4018d17c463ad5a97a9b
2004-11-13 13:57:51 -05:00
Nathan Binkert
bde7f4d121 defining SYSTEM_EV5 isn't all that necessary
--HG--
extra : convert_revision : 2ed4866db9483820d550bad00fdbc8dd027f95ba
2004-11-13 11:33:43 -05:00
Nathan Binkert
c4e5ef64b3 defining SYSTEM_EV5 isn't all that necessary
--HG--
extra : convert_revision : 7d39dd9f814434cb95ec769204d7f2426b0290fd
2004-11-13 11:32:17 -05:00
Ali Saidi
2282b0aa61 Merge zizzer:/bk/m5 into zeep.eecs.umich.edu:/tmp/m5
--HG--
extra : convert_revision : 7696b915099bc847bbb9bd9f97fd30b1825c68d5
2004-11-12 18:47:14 -05:00
Ali Saidi
8fc76693d2 slightly different parameters for nfs script
--HG--
extra : convert_revision : 1576f875646567cc661a8e3f07c00c4b282e9f64
2004-11-12 18:47:07 -05:00
Ron Dreslinski
cf5eea0e45 Merge zizzer:/z/m5/Bitkeeper/m5
into zazzer.eecs.umich.edu:/z/rdreslin/m5bk/clean

--HG--
extra : convert_revision : 4b14db6c78ed6b5f792fa9df6391af2fa2cc6b81
2004-11-12 15:27:01 -05:00
Lisa Hsu
78118bffbb add surge and spec-surge readfiles. also make the naming system uniform, there were 3 different ways!!! i like ali's.
--HG--
rename : configs/boot/client.netperf.maerts => configs/boot/netperf-maerts-client.rcS
rename : configs/boot/client.netperf.rr => configs/boot/netperf-rr.rcS
rename : configs/boot/server.netperf => configs/boot/netperf-server.rcS
rename : configs/boot/client.netperf.stream => configs/boot/netperf-stream-client.rcS
extra : convert_revision : 8c841e18dac9634f0680b857f3d0676b100dd448
2004-11-12 15:03:28 -05:00
Ron Dreslinski
ad3e3217e3 Merge
--HG--
extra : convert_revision : d7a5cccd3472bc59b840a0e5285cd65dcc2484fe
2004-11-12 14:50:28 -05:00
Ron Dreslinski
501db90f2c Make changes so that coherence works on a timing bus for the top-level of caches.
This added a snoopResponse callback to the caches, and a NACK to requests.

cpu/memtest/memtest.cc:
    Modified to work with do_events:
    No multiple requests to the same block outstanding at the same time from the same tester
    Using false sharing, each tester does only 1 byte access using it's id as the blk offset
    Allow for cycles before signaling deadlock, with do events it can take time to complete (NACK/blocked bus)
cpu/memtest/memtest.hh:
    Updated to keep an id with each tester (used for address generation)
    Updated to keep a list of outstanding address to prevent multiple outstanding per tester
    //Should really look into doing store forwarding within the tester, then we can test more functionality

--HG--
extra : convert_revision : 05fbcf547e4ffab9d220aeb73126ed787ca82239
2004-11-12 14:40:07 -05:00
Ali Saidi
856fdbe5aa Added nfs-client.rcS and nfs-server.rcS
--HG--
extra : convert_revision : 48cf156ce397b68075c07ed072c95644affa3ef1
2004-11-12 14:21:52 -05:00
Steve Reinhardt
2db94b9186 Bus interface range fixes for caches (to fix bus brudge range bug).
Add "fast write" behavior on writeInvalidates.

--HG--
extra : convert_revision : 0c5c1e06d6cbb608610b99fb4e4fb04db4997c7b
2004-11-10 22:18:20 -05:00
Steve Reinhardt
1e4acc382a add DPRINTFs
--HG--
extra : convert_revision : 7d2fad2e4f9125d41c8d1972d8f19f38bb95c06e
2004-11-10 19:57:55 -05:00
Nathan Binkert
b64619d575 Make it so we can cast an EthAddr to a 64-bit integer
--HG--
extra : convert_revision : 2f470d122997fbc375f04e9c2682b6528adef50c
2004-11-10 18:49:55 -05:00
Nathan Binkert
338513c681 Use the inPalMode function instead of the PC_PAL macro
--HG--
extra : convert_revision : 58e0a19ba98777e5d2e2572ed02dee1914378ff7
2004-11-10 18:44:37 -05:00
Nathan Binkert
ab8409738e Merge zizzer.eecs.umich.edu:/bk/m5
into ziff.eecs.umich.edu:/z/binkertn/research/m5/latest

--HG--
extra : convert_revision : 64eca6a11aeae236e802f9ca8eb97ca05d394ec7
2004-11-10 18:37:33 -05:00
Nathan Binkert
4328480b56 Don't use the global check_interrupts variable. Add a per-cpu
checkInterrupts variable and use that to determine whether an interrupt
can occur on a given cycle.

arch/alpha/ev5.cc:
    XC -> CPU (and xc -> CPU) since we're really talking about a CPU here
    Don't use the global check_interrupts variable.  Add a per-cpu
    checkInterrupts variable and use that to determine whether an interrupt
    can occur on a given cycle.

--HG--
extra : convert_revision : be4c0247e5834005c60a45796a222cffd327b64e
2004-11-10 18:37:19 -05:00
Nathan Binkert
82638c2e98 don't print dprintk twice on the same line
--HG--
extra : convert_revision : f41fc47221b6fb83eb92c080de073a11cf04c353
2004-11-10 18:15:04 -05:00