gem5.git
20 years agoMerge zizzer:/bk/m5 into zeep.eecs.umich.edu:/z/saidi/work/m5
Ali Saidi [Wed, 30 Jun 2004 20:06:50 +0000 (16:06 -0400)]
Merge zizzer:/bk/m5 into zeep.eecs.umich.edu:/z/saidi/work/m5

--HG--
extra : convert_revision : eec2cf5ef2f8452bca955fd8aecb062d07502b0b

20 years agoUpdate MAX ASN in kernel to 127 since that's what the ev5 supports
Ali Saidi [Wed, 30 Jun 2004 20:06:47 +0000 (16:06 -0400)]
Update MAX ASN in kernel to 127 since that's what the ev5 supports

--HG--
extra : convert_revision : e1feca854301682a7ce58e8f4ff149b10d2beb4b

20 years agofix the -I flag stuff for CPP so it actually works right.
Nathan Binkert [Wed, 30 Jun 2004 18:53:26 +0000 (14:53 -0400)]
fix the -I flag stuff for CPP so it actually works right.
What was I smoking?

--HG--
extra : convert_revision : 24488707a35febe006ec50a1fa7c9fad4842eadc

20 years agoadd asn when tracing tlb stuff
Nathan Binkert [Wed, 30 Jun 2004 18:51:40 +0000 (14:51 -0400)]
add asn when tracing tlb stuff

--HG--
extra : convert_revision : abc774179b3d4c979efd98e32d1d071b142e7b48

20 years agoMerge zizzer:/bk/m5 into zizzer.eecs.umich.edu:/z/alschult/m5
Andrew Schultz [Wed, 30 Jun 2004 13:47:32 +0000 (09:47 -0400)]
Merge zizzer:/bk/m5 into zizzer.eecs.umich.edu:/z/alschult/m5

--HG--
extra : convert_revision : 0bd6506e79600775dd0f72996120321f249b3411

20 years agoChange assert to check on virtual addresses because the physical
Andrew Schultz [Wed, 30 Jun 2004 13:47:23 +0000 (09:47 -0400)]
Change assert to check on virtual addresses because the physical
address can be zero for a copy

--HG--
extra : convert_revision : c6089969396aee2d67fa695a31cb7c5aea794338

20 years agoMerge zizzer.eecs.umich.edu:/bk/m5
Lisa Hsu [Wed, 30 Jun 2004 04:51:11 +0000 (00:51 -0400)]
Merge zizzer.eecs.umich.edu:/bk/m5
into shizzle.(none):/home/hsul/work/linux-clean

--HG--
extra : convert_revision : a072c1560e1f8b5cac378ed5f047a8a9c86b97b9

20 years agofix transmit side checksum offloading to not generate a pseudo header.
Lisa Hsu [Wed, 30 Jun 2004 04:50:56 +0000 (00:50 -0400)]
fix transmit side checksum offloading to not generate a pseudo header.

dev/ns_gige.cc:
    transmit side checksum offloading doesn't need pseudo header generation, it just takes the pseudo header checksum and uses it.

--HG--
extra : convert_revision : 9741bd650415c18ed37b06a453b23610d028135b

20 years agoMerge zizzer:/bk/m5 into zeep.eecs.umich.edu:/z/saidi/work/m5
Ali Saidi [Tue, 29 Jun 2004 20:08:33 +0000 (16:08 -0400)]
Merge zizzer:/bk/m5 into zeep.eecs.umich.edu:/z/saidi/work/m5

--HG--
extra : convert_revision : 7e1d8e3723f47338f0b24071ee95c66a9561dd2f

20 years agoAnother fix for the too much work problem in 2.6. This should do it.
Ali Saidi [Tue, 29 Jun 2004 20:08:26 +0000 (16:08 -0400)]
Another fix for the too much work problem in 2.6. This should do it.
Both rx/tx interrupts are now scheduled for the future to give the
linux kernel time to get out of its loop.

--HG--
extra : convert_revision : 8fee0a25fde0ce0545c924f8547bed460602e006

20 years agoGet rid of old ini files that haven't been used in a long time.
Nathan Binkert [Tue, 29 Jun 2004 17:03:20 +0000 (13:03 -0400)]
Get rid of old ini files that haven't been used in a long time.

--HG--
extra : convert_revision : 43c21727952d46363a46ef3d8fe89546576edc08

20 years agoWith the new uart code 300 cycles isn't quite enough, 350 seems to
Ali Saidi [Tue, 29 Jun 2004 01:23:10 +0000 (21:23 -0400)]
With the new uart code 300 cycles isn't quite enough, 350 seems to
work. When everything gets changed to seconds this should be updated.

--HG--
extra : convert_revision : 9f1064ff6fec5deceb591904f4571c9129ecc998

20 years agofix up the recordEvent stuff to support ignoring events
Nathan Binkert [Mon, 28 Jun 2004 20:49:35 +0000 (16:49 -0400)]
fix up the recordEvent stuff to support ignoring events

arch/alpha/ev5.cc:
cpu/simple_cpu/simple_cpu.cc:
    update for new event interface
base/stats/events.cc:
    implement the ignore event function which matches sim objects from which
    to ignore events.

    Make insert event like insert data and make it able to insert many
    events in a single transaction with the database.
base/stats/events.hh:
    Make it possible to ignore events
sim/sim_object.cc:
    make recordEvent a member function of SimObject to implement
    the ignore function easily
sim/sim_object.hh:
    implement the ignore event stuff in the sim object.  This is a
    bit of a hack, but an easy place to put it.

--HG--
extra : convert_revision : ba3f25a14ad03662c53fb35514860d69be8cd4f0

20 years agoDon't serialize the absolute path so we can move checkpoints around.
Nathan Binkert [Mon, 28 Jun 2004 19:54:05 +0000 (15:54 -0400)]
Don't serialize the absolute path so we can move checkpoints around.

sim/serialize.cc:
sim/serialize.hh:
    export the checkpoint directory

--HG--
extra : convert_revision : ba5b796fd930ab0487a502beefcf76bd6bb7c366

20 years agoadd support to add a string to indicate a particular sample of a run
Nathan Binkert [Mon, 28 Jun 2004 19:53:06 +0000 (15:53 -0400)]
add support to add a string to indicate a particular sample of a run
name to make statistical sampling easier.

--HG--
extra : convert_revision : b24db4ef747b8ef00a0131db237e97de60c27dc3

20 years agoMerge zizzer:/bk/m5 into isabel.reinhardt.house:/z/stever/bk/m5
Steve Reinhardt [Sun, 27 Jun 2004 22:17:16 +0000 (15:17 -0700)]
Merge zizzer:/bk/m5 into isabel.reinhardt.house:/z/stever/bk/m5

--HG--
extra : convert_revision : ad13cd0c7d700ec75ed98f517cbde3147e698940

20 years agoallow the use of old console code and update elf_object not to rely on EM_ALPHA value.
Ali Saidi [Sun, 27 Jun 2004 18:33:55 +0000 (14:33 -0400)]
allow the use of old console code and update elf_object not to rely on EM_ALPHA value.

base/loader/elf_object.cc:
    EM_ALPHA value isn't official, so perhaps we shouldn't use it
dev/alpha_console.cc:
dev/alpha_console.hh:
    this change allows the use of old console code

--HG--
extra : convert_revision : cfacd64ae7fd2595158ca1a83ebcdb66ee7e119b

20 years agoremoved LINUX system dir, all we really need is one that both tru64 and linux can...
Ali Saidi [Sun, 27 Jun 2004 01:43:17 +0000 (21:43 -0400)]
removed LINUX system dir, all we really need is one that both tru64 and linux can share

--HG--
extra : convert_revision : 7417031db632255749d2a8201d761a001fb7de2e

20 years agorewrote uart and renamed console.cc to simconsole to reduce confusion
Ali Saidi [Sun, 27 Jun 2004 01:26:28 +0000 (21:26 -0400)]
rewrote uart and renamed console.cc to simconsole to reduce confusion

base/traceflags.py:
    removed TsunamiUart/TlaserUart and added a plain Uart
dev/alpha_console.cc:
    updated for new simconsole
dev/platform.hh:
    added a uart member to platform
dev/simconsole.cc:
dev/simconsole.hh:
    removed lots of legacy code, it should all be ours now.
    converted tabs to 8 spaces
    added our copyright
dev/tsunami.cc:
    uses simconsole.hh rather than console.hh
dev/tsunami_cchip.cc:
dev/tsunami_io.cc:
    never needed console.hh
dev/tsunami_io.hh:
    this does need eventq.hh and it just happend to be working whenn console.hh was
    included everywhere
dev/tsunamireg.h:
    added a couple more 8250/16550 uart defines
dev/uart.cc:
    new uart code, rewritten to support both tlaser and tsunami (both a 8250 and 8530
    uart).
dev/uart.hh:
    updated for new uart, legacy code removed

--HG--
rename : dev/console.cc => dev/simconsole.cc
rename : dev/console.hh => dev/simconsole.hh
rename : dev/tsunami_uart.cc => dev/uart.cc
rename : dev/tsunami_uart.hh => dev/uart.hh
extra : convert_revision : e663352d49d4c2d3c95643030cf73c0e85ba2f08

20 years agominor changes to endian code so that it runs on x86 linux
Ali Saidi [Wed, 23 Jun 2004 22:39:11 +0000 (18:39 -0400)]
minor changes to endian code so that it runs on x86 linux

base/intmath.hh:
    only need FloorLog2(size_t) on a mac, so ifdefed for this
dev/alpha_console.cc:
    Actually allocate the alphaAccess struct.

--HG--
extra : convert_revision : 1f50b1a025c8ee728a9f3d2c603ea38347234f54

20 years agoMerge zizzer:/bk/linux into zeep.eecs.umich.edu:/z/saidi/work/m5-endian
Ali Saidi [Wed, 23 Jun 2004 21:55:54 +0000 (17:55 -0400)]
Merge zizzer:/bk/linux into zeep.eecs.umich.edu:/z/saidi/work/m5-endian

--HG--
extra : convert_revision : d4938c6011173d3017f47fd592c4b5e4c8d543a3

20 years agoFix to properly shadow the DEV bit in the Drive/Head register so other
Andrew Schultz [Wed, 23 Jun 2004 19:37:05 +0000 (15:37 -0400)]
Fix to properly shadow the DEV bit in the Drive/Head register so other
disks are properly detected and handled

--HG--
extra : convert_revision : ffc3046deb68458ee2ef6fa5263dc471488abc45

20 years agomore modifications for cross-endian support. linux now gets to pciconfig
Ali Saidi [Wed, 23 Jun 2004 19:07:09 +0000 (15:07 -0400)]
more modifications for cross-endian support. linux now gets to pciconfig

dev/alpha_console.cc:
    rather than acessing a byte array for alpha access, access the members
    **this requires an updated console**
dev/pcidev.cc:
    correctly type all the pci data and store in in little endian no
    matter what system we are on
dev/tsunami_uart.cc:
    correct a bug with the data type.
kern/linux/linux_system.cc:
    system type in hwprb needs to be endian happy as well.

--HG--
extra : convert_revision : 8de9bb69365b5d30fceaf4fa342a1639f92d7a83

20 years agoMerge zizzer:/bk/linux into zeep.eecs.umich.edu:/z/saidi/work/m5-linux
Ali Saidi [Tue, 22 Jun 2004 21:20:34 +0000 (17:20 -0400)]
Merge zizzer:/bk/linux into zeep.eecs.umich.edu:/z/saidi/work/m5-linux

--HG--
extra : convert_revision : 448f6e7851c14dd2c9f6148122da873a19c20772

20 years agoifdefed ev5 vs. ev6 differences so Tlaser can work in the linux tree
Ali Saidi [Tue, 22 Jun 2004 21:20:19 +0000 (17:20 -0400)]
ifdefed ev5 vs. ev6 differences so Tlaser can work in the linux tree

arch/alpha/alpha_memory.cc:
arch/alpha/ev5.hh:
    Ifdefed TLASER code
arch/alpha/vtophys.cc:
    added back some code andrew removed and couldn't remember why.

--HG--
extra : convert_revision : f00d255f7a8a7bdb6e74f061dd014188e3b39e73

20 years agopulled from head before pushing linux tree
Ali Saidi [Tue, 22 Jun 2004 17:49:26 +0000 (13:49 -0400)]
pulled from head before pushing linux tree

--HG--
extra : convert_revision : 27868e84419a4bc2b9abf59a813b8c8acb8c6083

20 years agopull from head before pushing linux tree
Ali Saidi [Tue, 22 Jun 2004 17:48:49 +0000 (13:48 -0400)]
pull from head before pushing linux tree

--HG--
extra : convert_revision : 345f91c5c16c69db22035dc716e82fd77041380f

20 years agoInitial cut at Python-based descriptor files for param contexts.
Steve Reinhardt [Tue, 22 Jun 2004 05:42:17 +0000 (22:42 -0700)]
Initial cut at Python-based descriptor files for param contexts.
These are not used at all right now, and will certainly change
over time, but the tedious job of extracting them from .cc files
is done.

--HG--
extra : convert_revision : b1b30cdf523fe457449f7779f8c7951adab648b5

20 years agoHandle SIGABRT a little more nicely.
Steve Reinhardt [Tue, 22 Jun 2004 05:42:16 +0000 (22:42 -0700)]
Handle SIGABRT a little more nicely.

base/misc.cc:
    Don't dump trace in panic(), SIGABRT handler will do it now.
sim/main.cc:
    Add SIGABRT handler that prints curTick and dumps buffered trace (if any).
    This doesn't work as well as I would like since the buffered trace records
    often contain stale references to stack-resident temporary std::string objects.
    Someday we'll have to put in a fix for that.

--HG--
extra : convert_revision : 67576efbf5c10e63e255fc9a9ec520326fd3567b

20 years agoGet rid of DynInst 'valid' hack check... bug it was put in for is long gone.
Steve Reinhardt [Tue, 22 Jun 2004 05:39:59 +0000 (22:39 -0700)]
Get rid of DynInst 'valid' hack check... bug it was put in for is long gone.

--HG--
extra : convert_revision : ae2975894c9c4aeec65466ff9798a0420b4c7782

20 years agoDon't put store prefetches (like wh64) with invalid addresses
Steve Reinhardt [Tue, 22 Jun 2004 03:48:55 +0000 (20:48 -0700)]
Don't put store prefetches (like wh64) with invalid addresses
in the store buffer.

--HG--
extra : convert_revision : c70cba4d5976b21967a53c7c728f2cbda1dfc183

20 years agoRecord execution faults in DynInst fault field. Previously the
Steve Reinhardt [Mon, 21 Jun 2004 23:26:21 +0000 (16:26 -0700)]
Record execution faults in DynInst fault field.  Previously the
DynInst field was being set only on memory faults.  If an FP
memory op encountered an FP enable fault, then the bogus memory
op would still get processed since the pipeline relied on the
inst->fault field to tell if the op was bogus.

--HG--
extra : convert_revision : 7435d712a464ce71b518425e0580aa22822fb901

20 years agol
Lisa Hsu [Mon, 21 Jun 2004 21:25:18 +0000 (17:25 -0400)]
l

base/traceflags.py:
    added some more traceflags for ethernet to break it up better
dev/etherpkt.hh:
    since we are not network host order, must reverse bytes for these typechecks.

    also, overload isTcp/UdpPkt to take an argument so you don't have to reget the ip header if you've already got one.
dev/ns_gige.cc:
    1) add some functions that reverse Endianness so we can generate adn evaluate checksum adn dprintf data accurately/more understandably

    2) forget about the implementation of fifo fill/drain thresholds, it's not used by the driver much, nor does it matter with use sending/receiving in whole packets anyway.
    get rid of teh associated variables.

    3) get rid of txFifoCnt the variable, it's redundant and unnecessary, just use txFifoAvail.

    4) change io_enable to ioEnable, just to be picky.

    5) modify some DPRINTF's to be clearer, also added a lot, and spread them into better traceflag categories

    6) fix the device bug!  it's the intrTick = 0 at teh beginning of cpuInterrupt().

    7) clear some bools in regsReset() so they don't holdover wrong state

    8) fix pseudo header generation for Tcp checksumming to account for network order
dev/ns_gige.hh:
    change io_enable to ioEnable, get rid of fill/drain thresh related variables and txFifoCnt, which is redundant

--HG--
extra : convert_revision : c538b75731f3c9e04354f57e6df9a40aeca5096d

20 years agoMerge saidi@zizzer.eecs.umich.edu:/bk/linux
Ali Saidi [Mon, 21 Jun 2004 05:04:48 +0000 (01:04 -0400)]
Merge saidi@zizzer.eecs.umich.edu:/bk/linux
into ali-saidis-computer.local:/research/linux

--HG--
extra : convert_revision : 446dc5fad11a6b29954dc5b1e974430853aa2969

20 years agostart towards getting m5 endian compliant
Ali Saidi [Mon, 21 Jun 2004 04:58:30 +0000 (00:58 -0400)]
start towards getting m5 endian compliant

base/inifile.cc:
    Added mac os support and fixed a bug, on error we need to exit the
    child process not return
base/intmath.hh:
    gcc on macos wanted a seperate function for the size_t type
base/loader/elf_object.cc:
    I'm not sure why this works under linux because it seems to return
    the wrong value.
base/stats/text.cc:
    added define/include for mac os x
cpu/exec_context.hh:
cpu/simple_cpu/simple_cpu.cc:
    added endian conversion code
dev/alpha_console.cc:
    rather than accessing a charecter array of varying size depending on
    the access, lets actually do this properly.
dev/alpha_console.hh:
    get rid of now nolonger used consoleData
dev/disk_image.cc:
    We have to byte swap the data is some cases, added function to do that
dev/ethertap.cc:
    added preproc directive for mac os

--HG--
extra : convert_revision : 2b5685765cfa2844926d7397f363d2788e3d640a

20 years agoSerialized cpu interrupts
Ali Saidi [Thu, 17 Jun 2004 19:36:59 +0000 (15:36 -0400)]
Serialized cpu interrupts

cpu/simple_cpu/simple_cpu.cc:
    called basecpu serialization

--HG--
extra : convert_revision : 1a639b5e3c08e47a5d581c18b2b53fe87bd05b73

20 years agoFixes to IDE disk to fix serialization. Now passes simple CPU serialize
Andrew Schultz [Thu, 17 Jun 2004 15:24:14 +0000 (11:24 -0400)]
Fixes to IDE disk to fix serialization.  Now passes simple CPU serialize
tests

dev/ide_ctrl.cc:
    Formatting
dev/ide_disk.cc:
    Remove some junk, add an assert to serialize, and add missing serialize
    for command register.

--HG--
extra : convert_revision : 8f99857e32f278dd4e6f23deffc8047c6411d5b2

20 years agoFix serialize/unserialize of the timers and RTC events
Andrew Schultz [Wed, 16 Jun 2004 23:47:07 +0000 (19:47 -0400)]
Fix serialize/unserialize of the timers and RTC events

--HG--
extra : convert_revision : aecf09b3b13a23ffef852a1539e8d4eec32008ad

20 years agoUpdated serialization code and added #if tracing so that make fast
Ali Saidi [Wed, 16 Jun 2004 22:20:10 +0000 (18:20 -0400)]
Updated serialization code and added #if tracing so that make fast
builds

dev/ide_ctrl.cc:
    added #if to remove variables that are optimized out.
dev/tsunami_io.cc:
dev/tsunami_io.hh:
    Updated serialization code

--HG--
extra : convert_revision : b322a3299097cbd05b9b5bb8b0a80e9fa33bdc20

20 years agoEnable software prefetches in FullCPU by default.
Steve Reinhardt [Tue, 15 Jun 2004 21:08:34 +0000 (17:08 -0400)]
Enable software prefetches in FullCPU by default.

--HG--
extra : convert_revision : b492c3537b4ae443bfd2c4d8f17cebead389dfc6

20 years agoMerge isabel.reinhardt.house:/z/stever/bk/m5-head
Steve Reinhardt [Tue, 15 Jun 2004 17:48:31 +0000 (10:48 -0700)]
Merge isabel.reinhardt.house:/z/stever/bk/m5-head
into isabel.reinhardt.house:/z/stever/bk/m5

--HG--
extra : convert_revision : 1af377ff2dbe27f357410ab91a50b06c61f12e83

20 years agoGet software prefetching to work in full-system mode.
Steve Reinhardt [Tue, 15 Jun 2004 17:48:08 +0000 (10:48 -0700)]
Get software prefetching to work in full-system mode.
Mostly a matter of keeping prefetches to invalid addrs
from messing up VM IPRs.  Also discovered that wh64s were
not being treated as prefetches, when they really should be
(for the most part, anyway).

arch/alpha/alpha_memory.cc:
arch/alpha/alpha_memory.hh:
    - Get rid of intrlock flag for locking VM fault regs (a la EV5);
    instead, just don't update regs on VPTE loads (a la EV6).
    - Add NO_FAULT MemReq flag to indicate references that should not
    cause page faults (i.e., prefetches).
arch/alpha/ev5.cc:
    - Get rid of intrlock flag for locking VM fault regs (a la EV5);
    instead, just don't update regs on VPTE loads (a la EV6).
    - Add Fault trace flag.
arch/alpha/isa_desc:
    - Add NO_FAULT MemReq flag to indicate references that should not
    cause page faults (i.e., prefetches).
    - Mark wh64 as a "data prefetch" instruction so it gets controlled
    properly by the FullCPU data prefetch control switch.
    - Align wh64 EA in decoder so issue stage doesn't need to worry about it.
arch/alpha/isa_traits.hh:
    - Get rid of intrlock flag for locking VM fault regs (a la EV5);
    instead, just don't update regs on VPTE loads (a la EV6).
base/traceflags.py:
    - Add Fault trace flag.
cpu/simple_cpu/simple_cpu.hh:
    - Pass MemReq flags to writeHint() operation.
cpu/static_inst.hh:
    Update comment re: prefetches.

--HG--
extra : convert_revision : 62e466b0f4c0ff9961796270fa2e371ec24bcbb6

20 years agoReconfigure ini files to match the default system that I'd like m5_1.0_beta2
Nathan Binkert [Tue, 15 Jun 2004 14:58:41 +0000 (10:58 -0400)]
Reconfigure ini files to match the default system that I'd like
to use for future papers.  Notable changes include a 3 level
cache hierarchy, a bug fix in the main memory bandwidth and
narrowing of the machine.

--HG--
extra : convert_revision : db0da82b0adaa5fa3413354e6d36bff9e3bd37dc

20 years agoAdd fast writes (wh64) back to the cache.
Erik Hallnor [Mon, 14 Jun 2004 04:45:30 +0000 (00:45 -0400)]
Add fast writes (wh64) back to the cache.

--HG--
extra : convert_revision : 48081c86da3f7959957e35f8ec868da55c68b5ce

20 years agoMerge zizzer.eecs.umich.edu:/z/m5/Bitkeeper/m5
Erik Hallnor [Sun, 13 Jun 2004 09:52:59 +0000 (05:52 -0400)]
Merge zizzer.eecs.umich.edu:/z/m5/Bitkeeper/m5
into zizzer.eecs.umich.edu:/z/ehallnor/work/m5

--HG--
extra : convert_revision : 88c83150b7fd9260e0441334f2503b1e029d8c09

20 years agoAdd some rewritten trace readers.
Erik Hallnor [Sun, 13 Jun 2004 09:52:28 +0000 (05:52 -0400)]
Add some rewritten trace readers.

--HG--
extra : convert_revision : 4a085c5d8526d1bf3f7155cda40002281c0c3d1b

20 years agominor mods for mimicking NS83820 functionality
Lisa Hsu [Sat, 12 Jun 2004 18:24:20 +0000 (14:24 -0400)]
minor mods for mimicking NS83820 functionality

dev/ide_ctrl.cc:
    generalize these #defs
dev/ide_ctrl.hh:
    put these in pcireg.h
dev/ns_gige.cc:
    do i need io_enable?  and assert will fail if i actually need to implement it, which may give clue as to wehtehr i need to implmeent the mem_enable and bm_enable stuff.
dev/ns_gige.hh:
    implement this in case it's needed
dev/pcireg.h:
    put these defs in pcireg instead

--HG--
extra : convert_revision : 5e3581b5da17410f943907139bd479f15d2231e8

20 years agofix the SYSTEMDIR #def
Lisa Hsu [Sat, 12 Jun 2004 16:59:43 +0000 (12:59 -0400)]
fix the SYSTEMDIR #def

--HG--
extra : convert_revision : cf5fd4aeaf96787345e233ddcd58f5eae55e0c27

20 years agofix serialization
Lisa Hsu [Fri, 11 Jun 2004 19:26:21 +0000 (15:26 -0400)]
fix serialization

dev/ns_gige.cc:
    fix serialization and move regsReset into the cc file
dev/ns_gige.hh:
    put regsReset into cc instead of here in hh

--HG--
extra : convert_revision : 3a8796fa583e0765503104a9dbe28cc69f1a8fa9

20 years agofix ini messups with the reorg
Lisa Hsu [Fri, 11 Jun 2004 19:01:18 +0000 (15:01 -0400)]
fix ini messups with the reorg

--HG--
extra : convert_revision : 67a36f724800b74fd0fcec9727bb2ea98ddc73e9

20 years agoafter unserialization in the ide driver we need to reprogram the
Ali Saidi [Fri, 11 Jun 2004 05:55:20 +0000 (01:55 -0400)]
after unserialization in the ide driver we need to reprogram the
busbridges. Also small fix for gcc 3.3.3.

dev/etherpkt.hh:
    Gcc 3.3.3 wants assert.h to be include in this file.
dev/ide_ctrl.cc:
    after unserialization we need to tell the busbridges what addresses to
    respond to.

--HG--
extra : convert_revision : a421197a5be07761bdef571d0a9406d77788e270

20 years agomove ping_linux.ini into linux-kernel-tsunami dir
Ali Saidi [Thu, 10 Jun 2004 21:31:25 +0000 (17:31 -0400)]
move ping_linux.ini into linux-kernel-tsunami dir

--HG--
extra : convert_revision : da08795d48bd7b82fdfb7443e4ffa7e45c037a0c

20 years agoMerge zizzer.eecs.umich.edu:/bk/linux
Lisa Hsu [Thu, 10 Jun 2004 18:40:51 +0000 (14:40 -0400)]
Merge zizzer.eecs.umich.edu:/bk/linux
into lush.(none):/z/hsul/work/linux

--HG--
extra : convert_revision : 30b93e7a4548a00ba18e86aad85df4c790f51d43

20 years agofix some things on linux todo list. 1) the config.ini. 2) the etherpkt struct defs...
Lisa Hsu [Thu, 10 Jun 2004 18:27:44 +0000 (14:27 -0400)]
fix some things on linux todo list.  1) the config.ini.  2) the etherpkt struct defs into ined_hdrs.hh
3) add some new functions to etherpkt.  4) checkin the nice ping_linux.ini that i've been using.

dev/etherpkt.hh:
    remove the packet headers nate hated so much.
    also add some new functionality regarding packets, like isIpHdr(), etc.
dev/ns_gige.cc:
    improve the code given the nice new functions offered in etherpkt

--HG--
extra : convert_revision : 2e27f5a8dca5323c0fa22d3c51af44a35f6be1a2

20 years agoFixes for detailed boot, made cttz and ctlz instructions more compact,
Ali Saidi [Thu, 10 Jun 2004 17:30:58 +0000 (13:30 -0400)]
Fixes for detailed boot, made cttz and ctlz instructions more compact,
and started cleaning up config files.

arch/alpha/isa_desc:
    Made implementation of cttz and ctlz more compact
base/remote_gdb.cc:
    Added comment about PALcode debugger accesses
dev/baddev.cc:
dev/baddev.hh:
dev/ide_ctrl.cc:
dev/ide_ctrl.hh:
dev/pciconfigall.cc:
dev/pciconfigall.hh:
dev/tsunami_cchip.cc:
dev/tsunami_cchip.hh:
dev/tsunami_io.cc:
dev/tsunami_io.hh:
dev/tsunami_pchip.cc:
dev/tsunami_pchip.hh:
dev/tsunami_uart.cc:
dev/tsunami_uart.hh:
    Cleaned up includes and changed device from FunctionalMemory to
    PioDevice for detailed boot
dev/ns_gige.cc:
    The ethernet dev uses two BARs, and the first bars size was being set
    incorrectly.
dev/tsunamireg.h:
    I don't know why we were using the superpage as the PCI memory addr.
    Changed and works correctly with detailed boot.

--HG--
extra : convert_revision : b535e76612cb90b544305dc1aa8c5e0e774564bd

20 years agolift common code *kernel_stats.* into the kern directory.
Lisa Hsu [Thu, 10 Jun 2004 05:14:35 +0000 (01:14 -0400)]
lift common code *kernel_stats.* into the kern directory.

cpu/exec_context.hh:
    lift kernel_stats out in the linux tree.

--HG--
extra : convert_revision : 866c7f82d8b995b4dce295de1ab0980527d13f01

20 years agoMerge zizzer.eecs.umich.edu:/bk/m5
Lisa Hsu [Thu, 10 Jun 2004 05:05:38 +0000 (01:05 -0400)]
Merge zizzer.eecs.umich.edu:/bk/m5
into shizzle.(none):/home/hsul/work/m5clean

--HG--
extra : convert_revision : 97a6fbe1d48c2f8855bd1b0971130b991dd48a80

20 years agoforgot to remove these files
Lisa Hsu [Thu, 10 Jun 2004 05:05:03 +0000 (01:05 -0400)]
forgot to remove these files

--HG--
extra : convert_revision : 91723409e22f36622b99309a3af227ea0c382cc7

20 years agolif kernelt stats out of tru64 directory
Lisa Hsu [Thu, 10 Jun 2004 05:02:33 +0000 (01:02 -0400)]
lif kernelt stats out of tru64 directory

cpu/exec_context.hh:
    change this to reflect the lifted kernel stats file.

--HG--
extra : convert_revision : 0dda3babdf51ee7a57430af69c7e20322b4eb622

20 years agoChange SCSI controller back to original M5 controller and remove the
Andrew Schultz [Thu, 10 Jun 2004 05:00:47 +0000 (01:00 -0400)]
Change SCSI controller back to original M5 controller and remove the
Adaptec additions (will stay in BitKeeper revision history)

--HG--
extra : convert_revision : f2e073fd5a34332e6f4b780bde6b948b486e3f68

20 years agoRename '*.odesc' files to '*.od'.
Steve Reinhardt [Thu, 10 Jun 2004 04:01:20 +0000 (21:01 -0700)]
Rename '*.odesc' files to '*.od'.

--HG--
extra : convert_revision : 68b4f1cc414ab67e7bd19682854ef3f0b28986c0

20 years agoMerge zizzer:/bk/m5 into isabel.reinhardt.house:/z/stever/bk/m5
Steve Reinhardt [Thu, 10 Jun 2004 03:53:29 +0000 (20:53 -0700)]
Merge zizzer:/bk/m5 into isabel.reinhardt.house:/z/stever/bk/m5

--HG--
extra : convert_revision : 10205d5b7bd10a5894dde9481d0909e909760874

20 years agoMinor restructuring of Python config code, mostly to avoid walking
Steve Reinhardt [Thu, 10 Jun 2004 03:46:29 +0000 (20:46 -0700)]
Minor restructuring of Python config code, mostly to avoid walking
the source tree for *.odesc files every time we run the script.
This is now factored out into load_odesc.py, which should be used
to generate m5odescs.py, which is then used as the source of object
& parameter definitions.

util/config/m5configbase.py:
    - Move odesc loading code to separate load_odescs.py, so maybe someday
    that can be done once at build time.
    - Print out children of a node in the order they are added.
    - Automatically assign a parent-less node to the first node for which it
    is used as the value of a parameter.  (Easier demonstrated than explained.)
    - Calculate object paths dynamically when requested rather than trying
    to keep them up to date as objects get assigned to parents.

--HG--
rename : util/config/m5config.py => util/config/m5configbase.py
extra : convert_revision : 2183a09d32f3862ab377e0a929715f30505a03cb

20 years agoDon't print 0 stats for memory commands we don't see.
Erik Hallnor [Wed, 9 Jun 2004 18:41:06 +0000 (14:41 -0400)]
Don't print 0 stats for memory commands we don't see.

--HG--
extra : convert_revision : e72c8bec85c72a48b0b5997d184922faf46c456c

20 years agoMake TraceCPU only compile in non-fullsystem.
Erik Hallnor [Wed, 9 Jun 2004 00:11:40 +0000 (20:11 -0400)]
Make TraceCPU only compile in non-fullsystem.

--HG--
extra : convert_revision : 5df611ae03f41093257672c0ad46bfddc58e515c

20 years agoTracing now works for upto 4 threads. Easy change to get it to work for more, but...
Erik Hallnor [Tue, 8 Jun 2004 23:52:49 +0000 (19:52 -0400)]
Tracing now works for upto 4 threads. Easy change to get it to work for more, but I don't have any test handy to test it.

cpu/trace/reader/m5_reader.cc:
    Add thread num.
cpu/trace/trace_cpu.cc:
    Increase thread count to 4, might want to make this a parameter (but it only really costs us storage).

--HG--
extra : convert_revision : 97cd7843668a3ef85aad06e3180dc04d2ca30ac1

20 years agoAdd the capability to read and write memory trace files. Currently is cycle accurate...
Erik Hallnor [Tue, 8 Jun 2004 21:31:04 +0000 (17:31 -0400)]
Add the capability to read and write memory trace files. Currently is cycle accurate for a single thread FullCPU.

--HG--
extra : convert_revision : f8fe545313eb307cc6f5ff2c23894cc9870b1d5b

20 years agoExtra semicolon removed
Ali Saidi [Tue, 8 Jun 2004 17:47:45 +0000 (13:47 -0400)]
Extra semicolon removed

--HG--
extra : convert_revision : fca7af81cfa4256acc1f2fb667196ca7411d754f

20 years agoUpdated Copyright with information in bitkeeper changelogs
Ali Saidi [Tue, 8 Jun 2004 17:37:27 +0000 (13:37 -0400)]
Updated Copyright with information in bitkeeper changelogs

arch/alpha/alpha_linux_process.cc:
arch/alpha/alpha_linux_process.hh:
arch/alpha/alpha_memory.cc:
arch/alpha/alpha_memory.hh:
arch/alpha/alpha_tru64_process.cc:
arch/alpha/alpha_tru64_process.hh:
arch/alpha/aout_machdep.h:
arch/alpha/arguments.cc:
arch/alpha/arguments.hh:
arch/alpha/faults.cc:
arch/alpha/faults.hh:
arch/alpha/isa_traits.hh:
arch/alpha/osfpal.cc:
arch/alpha/osfpal.hh:
arch/alpha/pseudo_inst.cc:
arch/alpha/pseudo_inst.hh:
arch/alpha/vptr.hh:
arch/alpha/vtophys.cc:
arch/alpha/vtophys.hh:
base/bitfield.hh:
base/callback.hh:
base/circlebuf.cc:
base/circlebuf.hh:
base/compression/lzss_compression.cc:
base/compression/lzss_compression.hh:
base/compression/null_compression.hh:
base/cprintf.cc:
base/cprintf.hh:
base/cprintf_formats.hh:
base/date.cc:
base/dbl_list.hh:
base/endian.hh:
base/fast_alloc.cc:
base/fast_alloc.hh:
base/fifo_buffer.cc:
base/fifo_buffer.hh:
base/hashmap.hh:
base/hostinfo.cc:
base/hostinfo.hh:
base/hybrid_pred.cc:
base/hybrid_pred.hh:
base/inet.cc:
base/inet.hh:
base/inifile.cc:
base/inifile.hh:
base/intmath.cc:
base/intmath.hh:
base/loader/aout_object.cc:
base/loader/aout_object.hh:
base/loader/ecoff_object.cc:
base/loader/ecoff_object.hh:
base/loader/elf_object.cc:
base/loader/elf_object.hh:
base/loader/object_file.cc:
base/loader/object_file.hh:
base/loader/symtab.cc:
base/loader/symtab.hh:
base/misc.cc:
base/misc.hh:
base/mod_num.hh:
base/mysql.cc:
base/mysql.hh:
base/pollevent.cc:
base/pollevent.hh:
base/predictor.hh:
base/random.cc:
base/random.hh:
base/range.cc:
base/range.hh:
base/refcnt.hh:
base/remote_gdb.cc:
base/remote_gdb.hh:
base/res_list.hh:
base/sat_counter.cc:
base/sat_counter.hh:
base/sched_list.hh:
base/socket.cc:
base/socket.hh:
base/statistics.cc:
base/statistics.hh:
base/stats/events.cc:
base/stats/events.hh:
base/stats/flags.hh:
base/stats/mysql.cc:
base/stats/mysql.hh:
base/stats/mysql_run.hh:
base/stats/output.hh:
base/stats/statdb.cc:
base/stats/statdb.hh:
base/stats/text.cc:
base/stats/text.hh:
base/stats/types.hh:
base/stats/visit.cc:
base/stats/visit.hh:
base/str.cc:
base/str.hh:
base/time.cc:
base/time.hh:
base/trace.cc:
base/trace.hh:
base/userinfo.cc:
base/userinfo.hh:
cpu/base_cpu.cc:
cpu/base_cpu.hh:
cpu/exec_context.cc:
cpu/exec_context.hh:
cpu/exetrace.cc:
cpu/exetrace.hh:
cpu/full_cpu/op_class.hh:
cpu/full_cpu/smt.hh:
cpu/inst_seq.hh:
cpu/intr_control.cc:
cpu/intr_control.hh:
cpu/memtest/memtest.cc:
cpu/memtest/memtest.hh:
cpu/pc_event.cc:
cpu/pc_event.hh:
cpu/simple_cpu/simple_cpu.cc:
cpu/simple_cpu/simple_cpu.hh:
cpu/static_inst.cc:
cpu/static_inst.hh:
dev/alpha_console.cc:
dev/alpha_console.hh:
dev/baddev.cc:
dev/baddev.hh:
dev/disk_image.cc:
dev/disk_image.hh:
dev/etherbus.cc:
dev/etherbus.hh:
dev/etherdump.cc:
dev/etherdump.hh:
dev/etherint.cc:
dev/etherint.hh:
dev/etherlink.cc:
dev/etherlink.hh:
dev/etherpkt.cc:
dev/etherpkt.hh:
dev/ethertap.cc:
dev/ethertap.hh:
dev/ide_ctrl.cc:
dev/ide_ctrl.hh:
dev/ide_disk.cc:
dev/ide_disk.hh:
dev/io_device.cc:
dev/io_device.hh:
dev/ns_gige.cc:
dev/ns_gige.hh:
dev/ns_gige_reg.h:
dev/pciconfigall.cc:
dev/pciconfigall.hh:
dev/pcidev.cc:
dev/pcidev.hh:
dev/pcireg.h:
dev/platform.cc:
dev/platform.hh:
dev/simple_disk.cc:
dev/simple_disk.hh:
dev/tsunami.cc:
dev/tsunami.hh:
dev/tsunami_cchip.cc:
dev/tsunami_cchip.hh:
dev/tsunami_io.cc:
dev/tsunami_io.hh:
dev/tsunami_pchip.cc:
dev/tsunami_pchip.hh:
dev/tsunami_uart.hh:
dev/tsunamireg.h:
docs/stl.hh:
kern/linux/linux.hh:
kern/linux/linux_syscalls.cc:
kern/linux/linux_syscalls.hh:
kern/linux/linux_system.cc:
kern/linux/linux_system.hh:
kern/system_events.cc:
kern/system_events.hh:
kern/tru64/dump_mbuf.cc:
kern/tru64/dump_mbuf.hh:
kern/tru64/mbuf.hh:
kern/tru64/printf.cc:
kern/tru64/printf.hh:
kern/tru64/tru64.hh:
kern/tru64/tru64_events.cc:
kern/tru64/tru64_events.hh:
kern/tru64/tru64_syscalls.cc:
kern/tru64/tru64_syscalls.hh:
kern/tru64/tru64_system.cc:
kern/tru64/tru64_system.hh:
sim/async.hh:
sim/builder.cc:
sim/builder.hh:
sim/debug.cc:
sim/debug.hh:
sim/eventq.cc:
sim/eventq.hh:
sim/host.hh:
sim/main.cc:
sim/param.cc:
sim/param.hh:
sim/process.cc:
sim/process.hh:
sim/serialize.cc:
sim/serialize.hh:
sim/sim_events.cc:
sim/sim_events.hh:
sim/sim_exit.hh:
sim/sim_object.cc:
sim/sim_object.hh:
sim/stat_control.cc:
sim/stat_control.hh:
sim/stats.hh:
sim/syscall_emul.cc:
sim/syscall_emul.hh:
sim/system.cc:
sim/system.hh:
sim/universe.cc:
test/bitvectest.cc:
test/circletest.cc:
test/cprintftest.cc:
test/initest.cc:
test/lru_test.cc:
test/nmtest.cc:
test/offtest.cc:
test/paramtest.cc:
test/rangetest.cc:
test/sized_test.cc:
test/stattest.cc:
test/strnumtest.cc:
test/symtest.cc:
test/tokentest.cc:
test/tracetest.cc:
util/m5/m5.c:
util/m5/m5op.h:
util/tap/tap.cc:
    Updated Copyright
dev/console.cc:
dev/console.hh:
    This code isn't ours, and shouldn't have our copyright

--HG--
extra : convert_revision : 598f2e5eab5d5d3de2c1e862b389086e3212f7c4

20 years agoMerge saidi@zizzer:/z/m5/Bitkeeper/linux/
Ali Saidi [Sun, 6 Jun 2004 23:31:06 +0000 (19:31 -0400)]
Merge saidi@zizzer:/z/m5/Bitkeeper/linux/
into zeep.eecs.umich.edu:/z/saidi/work/m5

--HG--
extra : convert_revision : 9fa64783d4f5d03acff5d11cba61e46590d6a9dd

20 years agoUpdated to use new palcode with ctlz instruction
Ali Saidi [Sun, 6 Jun 2004 23:30:52 +0000 (19:30 -0400)]
Updated to use new palcode with ctlz instruction

--HG--
extra : convert_revision : fbea7f01c528ae472845ec2897c9405afb145417

20 years agoAdded ctlz and cttz instructions to isa_desc for use in the PAL code.
Ali Saidi [Fri, 4 Jun 2004 22:10:50 +0000 (18:10 -0400)]
Added ctlz and cttz instructions to isa_desc for use in the PAL code.

--HG--
extra : convert_revision : 74dec35113b795e792b7fc03947a05349a4ff669

20 years agoFix to TsunamiIO unserialize
Andrew Schultz [Fri, 4 Jun 2004 20:04:55 +0000 (16:04 -0400)]
Fix to TsunamiIO unserialize

dev/tsunami_io.cc:
    Timers don't need to be rescheduled cause they aren't scheduled by
    the default constructor

--HG--
extra : convert_revision : afb68e4f0c4e2a2c98f0037e061752690080a503

20 years agoMerge zizzer:/bk/linux
Andrew Schultz [Fri, 4 Jun 2004 19:13:04 +0000 (15:13 -0400)]
Merge zizzer:/bk/linux
into zower.eecs.umich.edu:/z/alschult/DiskModel/linux

--HG--
extra : convert_revision : b23a1d1a79ac5c593150d269d0523c474cf6a4a0

20 years agoChanges to support serialization with PCI devices
Andrew Schultz [Fri, 4 Jun 2004 19:12:27 +0000 (15:12 -0400)]
Changes to support serialization with PCI devices

dev/ide_ctrl.cc:
    Properly serialize/unserialize the PciDev base class to get it to remap
    the MMU
dev/ns_gige.cc:
dev/ns_gige.hh:
    Remove the "addr" paramter from the constructor and change the device
    to use PCI based MMU mappings only
dev/pciconfigall.cc:
    Change comments
dev/pcidev.cc:
    Properly setup the MMU after a serialize

--HG--
extra : convert_revision : 4b2e7ba58e3c24fac1ff6f80635e704d6ecc0eff

20 years agofixed serialization in tsunami_io and tsunami_uart and console
Ali Saidi [Fri, 4 Jun 2004 18:26:17 +0000 (14:26 -0400)]
fixed serialization in tsunami_io and tsunami_uart and console

dev/console.cc:
dev/tsunami_io.cc:
dev/tsunami_uart.cc:
    fixed serialization

--HG--
extra : convert_revision : 1608a116b00007922fa382ddb0c10442a8724f8d

20 years agoMerge
Ali Saidi [Fri, 4 Jun 2004 17:46:04 +0000 (13:46 -0400)]
Merge

dev/tsunami_pchip.cc:
    SCCS merged

--HG--
extra : convert_revision : 064e3bab82aacf813d1f049d72429fd990608044

20 years agoUpdated copyright on Tsunami and kern/linux files.
Ali Saidi [Fri, 4 Jun 2004 17:43:50 +0000 (13:43 -0400)]
Updated copyright on Tsunami and kern/linux files.

dev/baddev.cc:
dev/baddev.hh:
dev/ide_ctrl.cc:
dev/ide_ctrl.hh:
dev/ide_disk.cc:
dev/ide_disk.hh:
dev/ns_gige.cc:
dev/ns_gige.hh:
dev/pciconfigall.cc:
dev/pciconfigall.hh:
dev/pcidev.cc:
dev/pcidev.hh:
dev/pcireg.h:
dev/platform.cc:
dev/platform.hh:
dev/tsunami.cc:
dev/tsunami.hh:
dev/tsunami_cchip.cc:
dev/tsunami_cchip.hh:
dev/tsunami_io.cc:
dev/tsunami_io.hh:
dev/tsunami_pchip.cc:
dev/tsunami_pchip.hh:
kern/linux/linux.hh:
kern/linux/linux_syscalls.cc:
kern/linux/linux_syscalls.hh:
kern/linux/linux_system.cc:
kern/linux/linux_system.hh:
    Updated copyright
dev/tsunamireg.h:
    Updated copyright and fixed a ULL

--HG--
extra : convert_revision : 4800bd227c7064044ee98169d6a91f74c791956f

20 years agoFixes in translation in PCI to DMA address to fix boot problems with
Andrew Schultz [Thu, 3 Jun 2004 21:48:05 +0000 (17:48 -0400)]
Fixes in translation in PCI to DMA address to fix boot problems with
Linux 2.6 with DMA support

dev/ide_disk.cc:
    Add debug infomation for DMA transfers and fix handling of PRD pointer
    values
dev/ide_disk.hh:
    Reduce buffer (MAX_DMA) size to 64K
dev/tsunami_pchip.cc:
dev/tsunami_pchip.hh:
    Add handling of PCTL register and also fix the translate from PCI to DMA
    addresses which was incorrect

--HG--
extra : convert_revision : 562f55fa1c7099ad0f5a23f59dec2c8ec7601d43

20 years agoMerge zizzer:/bk/linux
Andrew Schultz [Wed, 2 Jun 2004 17:57:08 +0000 (13:57 -0400)]
Merge zizzer:/bk/linux
into zower.eecs.umich.edu:/z/alschult/DiskModel/linux

--HG--
extra : convert_revision : 2b4893331f15b07c7f83148f6271d8ced0fcd6be

20 years agoMerge zizzer.eecs.umich.edu:/bk/m5 into lush.(none):/z/hsul/work/linux
Lisa Hsu [Wed, 2 Jun 2004 14:03:31 +0000 (10:03 -0400)]
Merge zizzer.eecs.umich.edu:/bk/m5 into lush.(none):/z/hsul/work/linux

--HG--
extra : convert_revision : 795698a55e98f67008392725968d4ac0a29b5326

20 years agoRemoved dynamic cast to get interrupt frequency and replaced with a
Ali Saidi [Tue, 1 Jun 2004 21:36:38 +0000 (17:36 -0400)]
Removed dynamic cast to get interrupt frequency and replaced with a
virtual function in alphaaccess.cc

dev/alpha_console.cc:
dev/alpha_console.hh:
dev/platform.hh:
dev/tsunami.cc:
dev/tsunami.hh:
dev/tsunami_io.hh:
    Removed dynamic cast to get interrupt frequency and replaced with a
    virtual function

--HG--
extra : convert_revision : 01f514a33d8f76c6527ab25a713d5c86f9fd646e

20 years agoMerge zizzer:/bk/linux
Andrew Schultz [Tue, 1 Jun 2004 21:20:30 +0000 (17:20 -0400)]
Merge zizzer:/bk/linux
into zower.eecs.umich.edu:/z/alschult/DiskModel/linux

--HG--
extra : convert_revision : 44678cd6aa2fa9e381d5d719d227013f5eb2a45c

20 years agoFixes to the state machine to properly support software reset and to fix
Andrew Schultz [Tue, 1 Jun 2004 21:19:47 +0000 (17:19 -0400)]
Fixes to the state machine to properly support software reset and to fix
PIO writes.  This was mainly related to not shadowing the status register
properly, and also not setting some of the status bits expected by the
operating system for the PIO write protocol.

--HG--
extra : convert_revision : fcdfd588be6e4f237aa6057889f0b3bdf4ea7631

20 years agoMerge saidi@zizzer:/z/m5/Bitkeeper/m5/
Ali Saidi [Tue, 1 Jun 2004 20:03:16 +0000 (16:03 -0400)]
Merge saidi@zizzer:/z/m5/Bitkeeper/m5/
into zeep.eecs.umich.edu:/z/saidi/work/m5

--HG--
extra : convert_revision : 89e0bdd427b23a8f52b8ba53b18451df7be22f14

20 years agoAdded ability to specify system type/revision in config file. This
Ali Saidi [Tue, 1 Jun 2004 19:43:27 +0000 (15:43 -0400)]
Added ability to specify system type/revision in config file. This
way a Tru64 system can be either Tlaser (by default) or Tsunami.

Added debugSymbolTable to Tru64 for symbol printing in InstExec Trace

cpu/exetrace.cc:
    Fixed bug if debugSymbolTable doesn't exist, we shouldn't try to
    look something up in it.
kern/tru64/tru64_system.hh:
    Added ability to specify system type/revision in config file. This
    way a Tru64 system can be either Tlaser (by default) or Tsunami.

--HG--
extra : convert_revision : c69a7411f0aa118ca5a766e43b7ceb7a5bb04cdf

20 years agoRename OpClass enum in .odesc file too.
Steve Reinhardt [Tue, 1 Jun 2004 00:39:06 +0000 (17:39 -0700)]
Rename OpClass enum in .odesc file too.

--HG--
extra : convert_revision : cb1824a1b6747fabf44ffcf40b45b71b052b2253

20 years agoAdded and cleaned up some comments
Ali Saidi [Mon, 31 May 2004 23:47:17 +0000 (19:47 -0400)]
Added and cleaned up some comments

dev/tsunami_cchip.hh:
    clean up some comments
kern/linux/linux_syscalls.cc:
    Cleaned up spacing of syscall numbers
kern/linux/linux_system.hh:
    Added doxygen comments

--HG--
extra : convert_revision : 23ecdaa92a208458dd5c5d3c68ac9012ce2690da

20 years agoRenamed OpClass enum members: they all end in 'Op' now.
Steve Reinhardt [Mon, 31 May 2004 23:19:31 +0000 (16:19 -0700)]
Renamed OpClass enum members: they all end in 'Op' now.
Got rid of two inconsistent sets of strings that corresponded
to this enum, and replaced with a single set that clearly
matches the enum names.

arch/alpha/isa_desc:
arch/isa_parser.py:
cpu/full_cpu/op_class.hh:
    Renamed OpClass enum members.

--HG--
extra : convert_revision : bf596f7568a20b2e77c07ac349f253135141aef4

20 years agoMerged head into linux tree
Ali Saidi [Mon, 31 May 2004 22:58:54 +0000 (18:58 -0400)]
Merged head into linux tree

base/stats/mysql.hh:
    Rename of Statsistics namespace to stats... merge from head

--HG--
extra : convert_revision : a5a7f6268b35e75fba1b1800a74fcd6dbd09d974

20 years agoIgnore fast_cpu_exec.cc.
Steve Reinhardt [Mon, 31 May 2004 22:22:22 +0000 (15:22 -0700)]
Ignore fast_cpu_exec.cc.

--HG--
extra : convert_revision : 31495fbb3568a09e763978ea59e36cb239090fda

20 years agoAdd parameter defaults to .odesc files. There was a bug in
Steve Reinhardt [Mon, 31 May 2004 22:21:14 +0000 (15:21 -0700)]
Add parameter defaults to .odesc files.  There was a bug in
extract_odesc.py that caused these to get left out.

--HG--
extra : convert_revision : 32e76c8e699f3dd72937e92217f3ab43fb53b4de

20 years agoCleaned up and commented code. I think we are ready to merge with head.
Ali Saidi [Sun, 30 May 2004 21:45:46 +0000 (17:45 -0400)]
Cleaned up and commented code. I think we are ready to merge with head.

dev/baddev.hh:
dev/pcidev.hh:
dev/tsunami.hh:
dev/tsunami_cchip.hh:
dev/tsunami_io.hh:
dev/tsunami_pchip.hh:
    Added doxygen comments
dev/pciconfigall.hh:
    Added doxygen comments. Made the hlist of devices private and provided
    members to modify the data.
dev/pcidev.cc:
    updated for change in pciconfigall
dev/tsunami_pchip.cc:
    Deleted commented out code we don't need
kern/linux/linux_syscalls.cc:
    Simplified the number -> name conversion.
kern/linux/linux_syscalls.hh:
    Removed StandardNumber and replaced with Number.
kern/linux/linux_system.cc:
kern/linux/linux_system.hh:
    LinuxSkipIdeDelay50msEvent was simply the same as the SkipFunc event,
    so I removed it. Same with with LinuxSkipFuncEvent.

--HG--
extra : convert_revision : 1508c335f87d90373f5772f3a0407ea13e858d7e

20 years agoUpdated FastCPU model with all the recent changes.
Kevin Lim [Fri, 28 May 2004 18:42:59 +0000 (14:42 -0400)]
Updated FastCPU model with all the recent changes.

arch/alpha/ev5.cc:
    Updated to support new forms of setIntReg and setFloatRegDouble.  Will need to be cleaned up in the future.
arch/isa_parser.py:
    Added in FastCPU model.

--HG--
extra : convert_revision : 384a27efcb50729ea6c3cc11653f395c300e48db

20 years agoMerged in new FastCPU stuff with existing code.
Kevin Lim [Fri, 28 May 2004 15:41:52 +0000 (11:41 -0400)]
Merged in new FastCPU stuff with existing code.

arch/alpha/ev5.cc:
    Added templatized processInterrupts() function that can be used by all of the CPU models.
arch/alpha/isa_desc:
    Merged in changes to remove CPU dependence.
arch/isa_parser.py:
    Merged in changes.
cpu/static_inst.hh:
    Includes FastCPU execute methods.

--HG--
extra : convert_revision : fcaa1dca35a9b316c73982bec8680df564f50bd8

20 years agoFastCPU model added. It's very similar to the SimpleCPU, just without a lot of the...
Kevin Lim [Thu, 27 May 2004 21:46:16 +0000 (17:46 -0400)]
FastCPU model added.  It's very similar to the SimpleCPU, just without a lot of the stats tracking.
Also various changes to make the CPU model less ISA dependent, which includes moving the code that checks for interrupts up to the ISA level, moving code that zeroes the zero registers up to the ISA level, and removing opcode and ra from the regfile.

arch/alpha/alpha_memory.cc:
    The regfile has been changed so it no longer has the opcode and ra.  Instead the xc holds the actual instruction, and from there the opcode and ra can be obtained with OPCODE() and RA().
arch/alpha/ev5.cc:
    Moved code that once existed within simpleCPU to ev5, and templatized it.
    This way the CPU models can call processInterrupts and the ISA specific interrupt handling is left to the ISA's code.
    Also moved ISA specific zero registers from simpleCPU to here.
arch/alpha/ev5.hh:
    Added macros for obtaining the opcode and ra from the instruction itself, as there is no longer opcode or ra in the regfile.
arch/alpha/isa_desc:
    Added in declarations for the FastCPU model.
arch/alpha/isa_traits.hh:
    Removed opcode and ra from the regfile.  The xc now holds the actual instruction, and the opcode and ra can be obtained through it.
    Also added the declaration for the templated zeroRegisters() function, which will set the zero registers to 0.
arch/isa_parser.py:
    Added in FastCPUExecContext so it will generate code for the FastCPU model as well.
cpu/exec_context.cc:
    Added in a more generic trap function so "ev5_trap" doesn't need to be called.  It currently still calls the old method, with plans for making this ISA dependent in the future.
cpu/exec_context.hh:
    Exec context now has the instruction within it.  Also added methods for exec context to read an instruction from memory, return the current instruction, and set the instruction if needed.
    Also has declaration for more generic trap() function.
cpu/simple_cpu/simple_cpu.cc:
    Removed references to opcode and ra, and instead sets the xc's instruction with the fetched instruction.
cpu/static_inst.hh:
    Added declaration for execute() using FastCPUExecContext.

--HG--
extra : convert_revision : 0441ea3700ac50b733e485395d4dd4ac83666f92

20 years agotest to see if pushing to m5-tru64 triggers an e-mail
Ali Saidi [Thu, 27 May 2004 18:20:50 +0000 (14:20 -0400)]
test to see if pushing to m5-tru64 triggers an e-mail

--HG--
extra : convert_revision : df7aab669a8521d031c7202e36c3c086fa979854

20 years agopostint, postext:
Steve Reinhardt [Thu, 27 May 2004 18:12:49 +0000 (14:12 -0400)]
postint, postext:
Fix output path /m5 -> /z/m5

--HG--
extra : convert_revision : 11ebcc7e62f02be8b0aeb8857eebc33a27e29513

20 years agoMerge zizzer.eecs.umich.edu:/z/m5/Bitkeeper/m5
Steve Reinhardt [Thu, 27 May 2004 18:06:39 +0000 (14:06 -0400)]
Merge zizzer.eecs.umich.edu:/z/m5/Bitkeeper/m5
into zizzer.eecs.umich.edu:/z/stever/bk/m5

--HG--
extra : convert_revision : 38d50d3d3640f02622e5ace5ac12b67d63eda3f0

20 years agoconfig_files.doxygen:
Steve Reinhardt [Thu, 27 May 2004 18:06:29 +0000 (14:06 -0400)]
config_files.doxygen:
Apparently <> don't need backslashes if they're inside dbl-quoted strings.

--HG--
extra : convert_revision : dd5294781d0ef818d1a5b622f3415510baefb402