Commit graph

1502 commits

Author SHA1 Message Date
Nathan Binkert
b46baf107f enable the Trace, Statistics, and Serialize param contexts.
objects/Root.mpy:
    Fake the param context stuff for now.
sim/param.cc:
    Make empty vector enums work
sim/serialize.cc:
    serialize_dir is always valid

--HG--
extra : convert_revision : c46373f0f4c70e6a2f01a81c0fa6bacab72d4c4f
2005-02-09 17:33:28 -05:00
Ron Dreslinski
118b374b84 Merge zizzer:/z/m5/Bitkeeper/m5
into zizzer.eecs.umich.edu:/.automount/zazzer/z/rdreslin/m5bk/timing_L1

--HG--
extra : convert_revision : c12f7ad9143bc69d25c39132d30889f22c73edf1
2005-02-09 16:23:56 -05:00
Nathan Binkert
061f40df08 Fixes to thes pbs send script
util/pbs/send.py:
    - add a -d to set the job root directory allowing one to run
    send.py from anywhere.
    - specify full paths to files instead of relative paths to make -d
    work and to allow ssh qsub to work again.
    - make the Link directory only copy links that point to regular files.

--HG--
extra : convert_revision : dd330cee08b97c5d72c3d58ef123f83ac7ccede7
2005-02-09 16:20:53 -05:00
Nathan Binkert
26ef1f56c8 Add the split_first and split_last functions on strings.
base/str.cc:
base/str.hh:
    Add a couple functions that allow you to split a string at
    the first or last instance of a delimiter.

--HG--
extra : convert_revision : 2af22639e1b67ac61577c00475a555841a56f902
2005-02-09 13:46:23 -05:00
Nathan Binkert
89ba024b98 Fix the panic message so that it looks more like M5's panic.
Make it so the same path is not added to the system path twice.

--HG--
extra : convert_revision : fe18db38cc4e335ad3525a364e9f8faf62b60e52
2005-02-09 13:41:53 -05:00
Nathan Binkert
6eaa4d3571 fix indent (so emacs and vi aren't screwed up.)
--HG--
extra : convert_revision : 589f37476fec14aa5e3c6e018631e291113d4e69
2005-02-09 13:40:02 -05:00
Ron Dreslinski
230a5a608d Merger
cpu/simple_cpu/simple_cpu.hh:
    Merge

--HG--
extra : convert_revision : 1b6003ac731051fefacb7d7a30c317553b4bf1bc
2005-02-09 12:56:24 -05:00
Ron Dreslinski
d9317dd348 Some more useful debugging info for kernel panic and die events
Increase the default number of CSHR's, we should really fix this or make it a parameter

Use a setBlocked call to tell the bus it should block

New technique for sampling and switchover:
1) Sampler switchover event happens
2) All cpus in the current phase of sampling associated with this sampler are signaled to switchover
3) Each cpu drains it's pipe of things being executed (stops fetching and waits for empty pipe)
4) Once the pipe is empty the cpu calls back to the sampler to signal it has finished, and moves into the switchedout state (continues not to fetch)
5) The sampler collects all the signals, once all cpus are drained it calls the new cpu's in the next phase to takeover from the correct cpu
6) The statistics are reset and the next switchover time is calculated from this point

cpu/base_cpu.cc:
cpu/base_cpu.hh:
cpu/simple_cpu/simple_cpu.cc:
cpu/simple_cpu/simple_cpu.hh:
    Reconfigure the way the sampling switchover works
cpu/pc_event.cc:
    More debugging information on kernel panic's
kern/linux/linux_system.cc:
    More debug info for Kernel Die events
kern/linux/linux_system.hh:
    More debug info for kernel die events

--HG--
extra : convert_revision : 61cc42e43ba738705aa1f1d167b65d4d6dee51ae
2005-02-09 10:27:00 -05:00
Nathan Binkert
c4089562d5 Tweak genini.
test/genini.py:
    Make it possible to run genini from a different directory.

--HG--
extra : convert_revision : 57cfb010d6114512040bf334ea21c9ed87234be0
2005-02-05 13:50:25 -05:00
Nathan Binkert
fc64ab8102 Make pbs submission scripts available to all.
Fix up configuration scrupts to have better support for
running on the simulation pool.

--HG--
extra : convert_revision : 0178c8600b193d6c0ca69163fb735a7fa0e70782
2005-02-05 13:49:17 -05:00
Ali Saidi
b78d7c2f16 Merge zizzer:/bk/m5 into zeep.eecs.umich.edu:/z/saidi/work/m5
--HG--
extra : convert_revision : 0151e501074988eedb52d91254870c265935f229
2005-02-03 23:51:02 -05:00
Ali Saidi
7e559f6c50 Add Monet configuration, update p4 parameters, couple of typo fixes
dev/tsunami_cchip.cc:
    add a fake register to tsunami that we can do 32bit reads to.
    Warn on access.

--HG--
extra : convert_revision : d87860f3b527528151c23431556039bca6e12945
2005-02-03 23:50:57 -05:00
Steve Reinhardt
6dd5509490 Merge zizzer.eecs.umich.edu:/z/m5/Bitkeeper/m5
into zizzer.eecs.umich.edu:/z/stever/bk/m5

--HG--
extra : convert_revision : aea0708fa6684e3203c03f17e8ae6ae87e893f04
2005-02-03 20:50:07 -05:00
Steve Reinhardt
06bee6397d Minor bug fix/update to tracediff util.
util/tracediff:
    Fix bug (used += instead of .= for string concatenation in Perl...
    wrong language!).

    Also updated for new config (s/Universe/root/).

--HG--
extra : convert_revision : 0db3f22794037dc51cc29f78a75bd22012a8ecd9
2005-02-03 20:49:14 -05:00
Steve Reinhardt
0aaf8ec6b8 Add support for CPU models to execute the effective
address calculation and memory access portions separately.
Not currently used by any CPU models, but Kevin says he needs this.

Also clean up handling of execution tracing for memory accesses
(move it all into isa_desc and out of CPU models).

Got rid of some ancient unused code too.

arch/alpha/isa_desc:
    Add execute() methods to EAComp and MemAcc portions of memory
    access instructions, to allow CPU models to execute the effective
    address calculation and memory access portions separately.

    Requires the execution context to remember the effective address
    across the two invocations.  Added setEA() and getEA() methods to
    execution context to support this.  A model that does not use the
    split execution model can panic if these methods are called.

    Also added hook to call traceData->setAddr() after EA computation
    on any load or store operation.
arch/isa_parser.py:
    Call traceData->setData() on memory writes (stores).
cpu/simple_cpu/simple_cpu.cc:
    Get rid of unused code.
cpu/simple_cpu/simple_cpu.hh:
    Add (non-functional) setEA() and getEA() methods for new
    split memory access execution support.

--HG--
extra : convert_revision : bc2d2c758c4ca753812b9fa81f21038e55929ff0
2005-02-03 20:47:11 -05:00
Nathan Binkert
52403a5ddf get rid of defined and just access the env dict directly
get rid of the alias for true to True and false to False to keep
consistent python syntax.

util/stats/info.py:
    Fix typo

--HG--
extra : convert_revision : e69588a8de52424e043315e70008ca3a3ede7d5b
2005-02-03 17:04:54 -05:00
Nathan Binkert
936ad58402 small python config related fixes.
dev/simconsole.cc:
sim/universe.cc:
    isValid isn't compatible with new python stuff, so whack it.

--HG--
extra : convert_revision : 0c50038769a558650479c51122a8be5d92e7d9c4
2005-02-03 14:33:02 -05:00
Steve Reinhardt
15e1ad8f6b Fix minor doxygen problem.
--HG--
extra : convert_revision : c15b1c5ab1c87b8d1cea87ffa383d1f4d45f107c
2005-02-03 13:10:35 -05:00
Nathan Binkert
a736a8fab6 Update config file language to take simobj and no longer use siminst
objects/AlphaConsole.mpy:
objects/AlphaTLB.mpy:
objects/BadDevice.mpy:
objects/BaseCPU.mpy:
objects/BaseCache.mpy:
objects/BaseSystem.mpy:
objects/Bus.mpy:
objects/CoherenceProtocol.mpy:
objects/Device.mpy:
objects/DiskImage.mpy:
objects/Ethernet.mpy:
objects/Ide.mpy:
objects/IntrControl.mpy:
objects/MemTest.mpy:
objects/Pci.mpy:
objects/PhysicalMemory.mpy:
objects/Platform.mpy:
objects/Process.mpy:
objects/Repl.mpy:
objects/Root.mpy:
objects/SimConsole.mpy:
objects/SimpleDisk.mpy:
objects/Tsunami.mpy:
objects/Uart.mpy:
    simobj now requires a type= line if it is actually intended
    to be a type
sim/pyconfig/SConscript:
    keep track of the filename of embedded files for better
    error messages.
sim/pyconfig/m5config.py:
    Add support for the trickery done with the compiler to get the
    simobj language feature added to the importer.

    fix the bug that gave objects the wrong name in error messages.
test/genini.py:
    Globals have been fixed and use execfile

--HG--
extra : convert_revision : b74495fd6f3479a87ecea7f1234ebb6731279b2b
2005-02-02 21:13:01 -05:00
Lisa Hsu
45bb2bf14d some changes for a split lifo in the new python config.
objects/BaseCache.mpy:
    add all the Split parameters to the BaseCache simobj.

--HG--
extra : convert_revision : 4fcba3ce730f730ca8628ac7f4aa0fb9476474ab
2005-02-01 17:35:01 -05:00
Lisa Hsu
38e6d1f782 checking in outstanding changes for partitioning that never made it to the head in my mad prelim rush.
--HG--
extra : convert_revision : 59ca9cca2f2090d234708a695228bf4a91fc5f52
2005-02-01 15:40:02 -05:00
Ali Saidi
da269014da removed all tsunami dependence on tlaserreg.h (RTC defines) and
all but tlaser_node.cc dependence on tlaserreg.h

dev/tsunami_io.cc:
dev/tsunamireg.h:
    removed tlaserreg.h

--HG--
extra : convert_revision : 148a5d79530e5ed721a49279f684a48041deed2b
2005-01-30 16:58:39 -05:00
Lisa Hsu
c3aaf43c88 make nat runs used vegas congestion avoidance implementation.
--HG--
extra : convert_revision : 29d2c4b70c56f13642466bd88c82f36ba849ed9d
2005-01-28 15:57:40 -05:00
Ali Saidi
2aae3636ee Merge zizzer:/bk/m5 into zeep.eecs.umich.edu:/z/saidi/work/m5
--HG--
extra : convert_revision : ed089f6062639ae5be930fbaea3dd7f7622653cc
2005-01-27 16:01:32 -05:00
Ali Saidi
7ec713edc8 added support for outputing Dot if pydot is installed
--HG--
extra : convert_revision : bdb1032cddb2478e999399647f893d320260ef7e
2005-01-27 16:01:25 -05:00
Kevin Lim
12ce227f0b Merge zizzer.eecs.umich.edu:/z/m5/Bitkeeper/m5
into zizzer.eecs.umich.edu:/.automount/zamp/z/ktlim2/m5-patched/m5-new

--HG--
extra : convert_revision : 2848966f3daf7aba741d4d38db9c87145d72ee26
2005-01-25 16:27:23 -05:00
Nathan Binkert
534ad6bc7e We need more cshrs so we don't segfault.
--HG--
extra : convert_revision : 54b1987d86e3f307e13de0396cf149653f4f6e6a
2005-01-25 11:15:54 -05:00
Nathan Binkert
e8d70bcd45 Fix the stats ParamContext
--HG--
extra : convert_revision : 23e49ca8bd9f50be32a4c28ba435a358d7093f83
2005-01-25 02:02:42 -05:00
Nathan Binkert
8b62d41f11 Fix pc_sample_interval default
--HG--
extra : convert_revision : 553b0489e8b8a83a4e8bbf9601eb82902436c761
2005-01-25 02:00:01 -05:00
Kevin Lim
e6b99b0768 Merge zizzer.eecs.umich.edu:/bk/m5
into zamp.eecs.umich.edu:/z/ktlim2/m5-patched/m5-new

--HG--
extra : convert_revision : e802c800a478c297d3aa780a9ea3c6701453d91d
2005-01-21 18:31:30 -05:00
Nathan Binkert
fa7fa5a909 send output to the JOBDIR
--HG--
extra : convert_revision : d53864c55f20aa00754106390878ce04a4ff468e
2005-01-21 15:49:46 -05:00
Ron Dreslinski
bc0661a3da Merge zizzer:/z/m5/Bitkeeper/m5
into zazzer.eecs.umich.edu:/z/rdreslin/m5bk/timing_L1

--HG--
extra : convert_revision : 20c65e224a5fbd396a98baa2111a0e22a1534586
2005-01-21 15:24:12 -05:00
Nathan Binkert
2b78a9c132 fix bug in python code
--HG--
extra : convert_revision : 67a931326e5a8e6b814ea6d3bcd384126b0cbba4
2005-01-21 15:13:27 -05:00
Nathan Binkert
f3ff5fe28d Fixup checkpointing with the new config stuff.
objects/Root.mpy:
    add checkpointing as a parameter to the root object.

--HG--
extra : convert_revision : 3b809ebd776c8a9256a4ad6f8783cd96ab5cb1b3
2005-01-21 15:09:42 -05:00
Ron Dreslinski
a40bc2ba77 Merge zizzer:/z/m5/Bitkeeper/m5
into zizzer.eecs.umich.edu:/.automount/zazzer/z/rdreslin/m5bk/timing_L1

--HG--
extra : convert_revision : dcbdcd39ef8920e05a69cd5bc7fa7c9164268125
2005-01-21 12:30:40 -05:00
Ron Dreslinski
77ae088345 Make it so that with dual CPU's we create more writebuffers by default, so that we can handle more uncached access (we will get some from multiple processors)
--HG--
extra : convert_revision : 4bb913340cda47925321da5f95aaa2ff95dbe9dd
2005-01-21 12:23:03 -05:00
Ron Dreslinski
f3aac53458 Merge zizzer:/z/m5/Bitkeeper/m5
into zizzer.eecs.umich.edu:/.automount/zazzer/z/rdreslin/m5bk/timing_L1

--HG--
extra : convert_revision : 193cb1585489c51f274f978f0adde7c8fbec050f
2005-01-21 11:43:27 -05:00
Nathan Binkert
2b57a1b8c1 Fix some mysql code issues.
base/stats/events.cc:
    cast this to an unsigned long long to make the compiler happy
    on 64-bit platforms.

--HG--
extra : convert_revision : e893d21bb228cbfb0c23adb7c4eeb1f62209ca1c
2005-01-21 11:29:19 -05:00
Nathan Binkert
be12fb379f Add more options that we may want to tweak on a per-job basis.
Fix the nat config.

--HG--
extra : convert_revision : 22bd4f20408371b644b51fd8d6b79486e1e5a59b
2005-01-21 11:27:14 -05:00
Nathan Binkert
1c31e247e7 Merge zizzer.eecs.umich.edu:/bk/m5
into ziff.eecs.umich.edu:/z/binkertn/research/m5/latest

--HG--
extra : convert_revision : 47fe06540907f6810cd1bccfc96a8705fd38c4ba
2005-01-21 11:19:26 -05:00
Ron Dreslinski
40a2fa3e12 Merge zizzer:/z/m5/Bitkeeper/m5
into zizzer.eecs.umich.edu:/.automount/zazzer/z/rdreslin/m5bk/timing_L1

--HG--
extra : convert_revision : cf9baafa40dbca1a89581484fd4c9d8336a26d66
2005-01-21 04:57:12 -05:00
Ron Dreslinski
7f10e948a5 Merge zizzer:/z/m5/Bitkeeper/m5
into zazzer.eecs.umich.edu:/z/rdreslin/m5bk/new_config

--HG--
extra : convert_revision : 8c87cd948d501b1fc58d57f1f505d1f72725ecde
2005-01-21 04:55:55 -05:00
Ron Dreslinski
867d80d55d Fix annoying bug that lead to dropped packets in the P4 system
dev/pktfifo.cc:
    Make it so that we don't unserialize the size of the fifo, clobering the fact that we were trying to make it bigger, and leading to a misleading config.out that states the un-overwritten max_size.
    Perhaps this should instead be a panic if the size (amount that was serialized) is bigger than the maxsize that was assigned by the configuration file.

--HG--
extra : convert_revision : d4b0527bfd7a584554ddc87c9b2103f7a3a72332
2005-01-21 04:55:43 -05:00
Nathan Binkert
6cbce7631a Add a couple statistics contexts.
arch/alpha/pseudo_inst.cc:
    rename the context for consistency.
sim/pyconfig/m5config.py:
    Add a ParamContext class so that param contexts work with
    the new config stuff.

--HG--
extra : convert_revision : 3a6b583a25c86237baca7a2b4eccc9d12f86a384
2005-01-21 04:34:01 -05:00
Ron Dreslinski
9bb66e7766 Merge zizzer:/z/m5/Bitkeeper/m5
into zazzer.eecs.umich.edu:/z/rdreslin/m5bk/timing_L1

--HG--
extra : convert_revision : 1892825b1e7ba47353e7a79ca4dcb84b64a73f79
2005-01-20 17:21:42 -05:00
Ali Saidi
29789443ba Merge zizzer:/bk/m5 into zeep.eecs.umich.edu:/z/saidi/work/m5
--HG--
extra : convert_revision : d5e6f6e696d7607f89894856d940afc89ae8e37b
2005-01-19 18:40:09 -05:00
Ali Saidi
886f905785 added total bytes/bandwidth/packets formulas to nics
cleaned up stability code and wrote some better help for stats.py
fixed sample bug in info.py

dev/ns_gige.cc:
dev/ns_gige.hh:
dev/sinic.cc:
dev/sinic.hh:
    add total bandwidth/packets/bytes stats
util/stats/info.py:
    fixed samples bug
util/stats/stats.py:
    cleaned up stability code and wrote a bit better help

--HG--
extra : convert_revision : cae06f4fac744d7a51ee0909f21f03509151ea8f
2005-01-19 18:40:02 -05:00
Lisa Hsu
db00ac3bc6 Add a new aggregate trace flag that doesn't print gobs of ethernet raw data, but prints everything else.
--HG--
extra : convert_revision : ddb0636f5d5f0c746bd121eaafc0c685b422b006
2005-01-19 18:18:15 -05:00
Kevin Lim
6ac7954242 Merge zizzer.eecs.umich.edu:/bk/m5
into zamp.eecs.umich.edu:/z/ktlim2/m5-patched/m5-new

--HG--
extra : convert_revision : 3b9413a6861eaddb8de753d76e217106d41cc347
2005-01-19 16:34:35 -05:00
Ron Dreslinski
10371be698 Fix serialization of txPacket (need to properly calculate the length before serialization)
--HG--
extra : convert_revision : e93321495d19afb86949daf2c14102ffb6bcef41
2005-01-19 16:26:19 -05:00