Commit graph

1064 commits

Author SHA1 Message Date
Ali Saidi
eb15e8ba0b Merge zizzer:/bk/m5 into zeep.eecs.umich.edu:/tmp/m5
--HG--
extra : convert_revision : 8444a4f15530bef95350e124f9b840c74e236f7d
2004-07-21 16:26:51 -04:00
Ali Saidi
2497bb5990 updated config files to fix what was broken in paper rush
--HG--
extra : convert_revision : 3128d0edd9367cff6a24cd923068535b59de25fd
2004-07-21 16:26:43 -04:00
Nathan Binkert
5b1da825d8 Wrap tracing stuff in #if TRACING_ON to avoid compile error
in m5.fast

--HG--
extra : convert_revision : 329d810bfe282e88133ca35da62ee6dcc73a38f5
2004-07-21 15:44:57 -04:00
Steve Reinhardt
e76c98e6be Add python-file-to-c-string utility for compiling python into the simulator
(for later piping to a forked interpreter, unless we go crazy and embed it).

--HG--
extra : convert_revision : 70860b12780830a16c48aedd495fd79730dfce4c
2004-07-18 23:03:17 -07:00
Lisa Hsu
a7a7d6ff71 add some new stats and make one more change to the way the split lifo works.
--HG--
extra : convert_revision : f9154409014cb9edba6c33be1e1063f13c593ea3
2004-07-18 00:03:07 -04:00
Lisa Hsu
24821f7166 split_lifo.cc:
oops, also need to move a newly replaced blk to tail, if not there already.

--HG--
extra : convert_revision : 6e118cd135987c401f3663575d08f33c2c0c4ceb
2004-07-16 23:29:33 -04:00
Lisa Hsu
09067d4c29 Merge zizzer.eecs.umich.edu:/bk/m5
into shizzle.(none):/home/hsul/work/linux-clean

--HG--
extra : convert_revision : 6bdf1a161163ca1cec55b2a608b0e31eac8ef6b3
2004-07-16 21:04:25 -04:00
Lisa Hsu
f03793fe19 split_lifo.cc:
clear the isUsed flag for a block when it is replace in the 2Q scheme

--HG--
extra : convert_revision : 3284fe5ba4eb29dbd472237c09f95a8c92b2da5a
2004-07-16 21:01:49 -04:00
Nathan Binkert
0bbf9a4791 Merge zizzer.eecs.umich.edu:/bk/m5
into ziff.eecs.umich.edu:/z/binkertn/research/m5/latest

--HG--
extra : convert_revision : 87387b4f896ed945196b2090484c932c8b7e5abc
2004-07-14 23:02:11 -04:00
Nathan Binkert
bc5c52335c Add a clock multiplier for simple CPU so that it is possible
to do multiple instructions per cycle.

--HG--
extra : convert_revision : 5588ae38071f170792aad93899fef6842b7d818d
2004-07-14 23:01:54 -04:00
Erik Hallnor
a6939573ce Fix serialization when a tx interrupt isn't scheduled
dev/uart.cc:
    Fix serialization when a tx interrupt isn't scheduled.

--HG--
extra : convert_revision : 3029e284cbba417e3f453d83fdf2ef307d684916
2004-07-14 21:34:51 -04:00
Steve Reinhardt
374e12c2de Merge zizzer.eecs.umich.edu:/z/m5/Bitkeeper/m5
into zizzer.eecs.umich.edu:/z/stever/bk/m5

--HG--
extra : convert_revision : c79927401293a4496223cf190b9fc6cab6ea12de
2004-07-14 15:48:19 -04:00
Steve Reinhardt
842da6a9bb SConscript:
Make decoder.cc/hh depend on isa_parser.py.

SConscript:
    Make decoder.cc/hh depend on isa_parser.py.

--HG--
extra : convert_revision : ff575d7f819da951423b36275e960fed652f9b2b
2004-07-14 15:48:11 -04:00
Nathan Binkert
5c20353129 adjust latencies so that they more closely match numbers seen
in the real world.
re-work kernel ini files for more flexibility

--HG--
extra : convert_revision : 075c31c9daf251117cde7ef659e37e32d5e4b0c8
2004-07-14 09:44:31 -04:00
Nathan Binkert
7739f32758 fix paths for changes in disk image layout
--HG--
extra : convert_revision : 506ff845efc8d786639c29931eb4abd3a5860fcf
2004-07-14 09:06:56 -04:00
Nathan Binkert
6f43d2b8d8 Fix infinite loop in ns_gige stat machine
argh!

dev/ns_gige.cc:
    Exit the state machine so that we don't come right back where
    we started and enter an infinite loop.

--HG--
extra : convert_revision : a5f2b5b5a692de6c80e4b02d7f9bc5d27fe17252
2004-07-14 09:02:15 -04:00
Lisa Hsu
961805c73a remove a panic in split.cc:regernateBlkAddr that shouldnt' be there.
--HG--
extra : convert_revision : 59029318d714cd93e998e45429cf96694020cc3c
2004-07-13 11:05:43 -04:00
Nathan Binkert
195768d476 put the formatting fixes back into devtime
util/ccdrv/devtime.c:
    put the formatting fixes back.  no tabs.

--HG--
extra : convert_revision : 541e91294ee5383f3b321dd084876ae7c353cce9
2004-07-12 23:01:53 -04:00
Nathan Binkert
13f8dc981f make the cache access latency a parameter that is based on bus
ticks for the most commonly accessed devices.

dev/baddev.cc:
    Get rid of the constant cache access latency.
    For unimportant devices, don't add any latency.
dev/ide_ctrl.cc:
dev/ide_ctrl.hh:
dev/ns_gige.cc:
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/uart.cc:
dev/uart.hh:
    make the cache access latency a parameter that is based on bus
    ticks.
dev/io_device.cc:
dev/io_device.hh:
    add an io latency variable
dev/ns_gige.hh:
    this moved to io_device.hh

--HG--
extra : convert_revision : 4883130feeaef48abee492eddf0b8eb40eb94789
2004-07-12 22:58:22 -04:00
Ali Saidi
c2e5caf360 modified devtime to print out raw data
--HG--
extra : convert_revision : 5ad673d6f684acaffeb61db794f7e2ac099ba99d
2004-07-12 22:51:20 -04:00
Ali Saidi
3a8e5599b5 modified ccdrv to print out raw data.
util/ccdrv/devtime.c:
    modified to print out raw data.

--HG--
extra : convert_revision : 1f76ab8a022476e6a5ed787aa38547dd0ca99e36
2004-07-12 22:47:46 -04:00
Nathan Binkert
ef16b46484 formatting
--HG--
extra : convert_revision : cbd0c148a50643191372fdba2f771d7e145f1bff
2004-07-12 22:33:27 -04:00
Lisa Hsu
f857bd9dff make tx/rx fifo size a param, also fix the empty dma problem by adjusting the state machine.
dev/ns_gige.cc:
    make tx/rx fifo sizes a param. the default is 128K for each.  also, make the state machine not move onto txFragRead if there is no room in the txfifo for data.
dev/ns_gige.hh:
    make tx/rx fifo size a param

--HG--
extra : convert_revision : ed91eb31c2b21d4cdc6de87d8641df6197be5209
2004-07-12 16:09:52 -04:00
Nathan Binkert
537e840cc3 Fix a couple of bugs in lisa's cache partitioning code.
--HG--
extra : convert_revision : d3b6920609a59a33ac19bb1b395f5ae9694c2b53
2004-07-11 15:35:11 -04:00
Lisa Hsu
9dc9523a1c the changeset of partitioned caches. this changeset creates Split tagstore class that makes either an LRU/LRU cache or a LRU/LIFO cache, it is also able to make just a LIFO cache.
SConscript:
    add these files for partitioned caches
base/traceflags.py:
    add Split as a traceflag

--HG--
extra : convert_revision : 6aa8f7c389ee1145b33be006c820236f11f5397c
2004-07-10 21:57:59 -04:00
Andrew Schultz
4f4252fa89 Remove second pciToDma translation on current PRD address pointer b/c
second translation (which is superfluous) doesn't work properly when
system memory is higher than 128MB

--HG--
extra : convert_revision : 9cdf6925689d376953b1aa071bcd1e2f06419202
2004-07-09 22:32:27 -04:00
Nathan Binkert
539da68156 Defer registration only if we're sampling
--HG--
extra : convert_revision : 1bae80c02b39f91bb7867dad37cf170fd63ae838
2004-07-09 15:08:42 -04:00
Lisa Hsu
bbbfebd885 Merge zizzer.eecs.umich.edu:/bk/m5
into shizzle.(none):/home/hsul/work/m5clean

--HG--
extra : convert_revision : 78beb2735c6536ce3e639d59c99e755e9139f4f1
2004-07-09 11:50:40 -04:00
Lisa Hsu
7b2691d537 when you add caches, dma commands within the state machine are delayed and cause the state machine to exit until the dma comes back. thus, all relevant code must be executed BEFORE going to do the dma code.
dev/ns_gige.cc:
    rearrange code so nothing gets skipped when "doing dma."

--HG--
extra : convert_revision : cca66885f45e7df1831e2d8ccaddf5ece7600b13
2004-07-09 11:50:27 -04:00
Nathan Binkert
10446ec4ae fix compile problem in floss_reasons.cc
--HG--
extra : convert_revision : 86252dca4cd0bd6c1bcfc5bc14aaa3b941c3a8f8
2004-07-08 23:36:09 -04:00
Nathan Binkert
778c675079 - fix bad merge
- pass in sample name to mysql

--HG--
extra : convert_revision : 5a31085931037637d05d6acdf601dee582a46710
2004-07-08 18:16:30 -04:00
Nathan Binkert
6999918dd4 Clean up some mysql stuff to make things work again and
hopefully improve performance a tad.

base/stats/mysql.cc:
    - it's not called sample anymore, it's called tick
    - don't bother to cleanup deleted runs.  Doing this for each run
    is not necessary, it can be done all at once
    - don't query for a bin id every time, just do it once.
    - use locking in a few places to prevent two processes from
    stepping on eachother.
    - don't duplicate subdata ids.  use -1,y and x,-1
base/stats/mysql.hh:
    It's not called sample anymore, it's called tick

--HG--
extra : convert_revision : 95de8498b627c9175da28a66604ec7c719f7804c
2004-07-08 17:48:13 -04:00
Nathan Binkert
ec5dfbccff Clean up floss statistics names to avoid duplicate names
--HG--
extra : convert_revision : 57f34deaad00cc368e282a9f55ea8525b7c84563
2004-07-08 17:44:08 -04:00
Nathan Binkert
ea376e5c42 unify all system boot ini files into a single one. To run
different experiments, use --<foo>.system:readfile=<script>
to change the experiment

--HG--
extra : convert_revision : ff622a567bb10a14c5dd427a54814677cb205d7c
2004-07-08 17:37:22 -04:00
Nathan Binkert
d025147f3b FullCPU needs its registration deferred
--HG--
extra : convert_revision : 1bb84b34333540456c2735e11a8927262b4db4c8
2004-07-08 16:08:47 -04:00
Nathan Binkert
a8a7ffce6b Tick is signed, so make Event start be large and positive,
not negative

--HG--
extra : convert_revision : 96be42a30fc0bd59eaff1d19161d881cab3895ae
2004-07-08 16:02:37 -04:00
Nathan Binkert
989eb88576 Scripts that can be used with server.readfile and client.readfile
to run the netperf benchmark

--HG--
extra : convert_revision : a2ce490e0c44996d0876a6839ad333643ec252c8
2004-07-08 16:01:13 -04:00
Nathan Binkert
7c2084dcfe store the checkpoint to the proper directory
dev/disk_image.cc:
    store the disk checkpoint to the proper directory

--HG--
extra : convert_revision : f76347576691560f00c60b86ac2044960d5a193c
2004-07-08 15:59:46 -04:00
Andrew Schultz
7b2535d7bc Change mount command for mounting disk images
--HG--
extra : convert_revision : 450e8c7332c1e3162cdbef4177b7f54b47b1648a
2004-07-08 15:22:41 -04:00
Nathan Binkert
32df4eccb4 Fix the cache's address range.
**warning** this is still a hack.  The caches need to figure this
out on their own!

--HG--
extra : convert_revision : 4fbebf302bfbeaa35e5aad276934d8d1ce5c6baf
2004-07-08 14:23:49 -04:00
Lisa Hsu
9dba9e462f fix the failing regression. the sconscript ("makefile") was missing the etherbus.cc src file.
SConscript:
    THIS is why the regression keeps failing.  it's missing the the etherbus file in the sconscript.

--HG--
extra : convert_revision : adbcfb6036d8e8ecb51c4f484d0030059542684b
2004-07-08 09:05:26 -04:00
Ali Saidi
02dfbe92fc fixed a bad merge from linux<->tru64
--HG--
extra : convert_revision : d7a5acd89a2bdc013c8a2bd022cc7048a3920f01
2004-07-08 00:28:02 -04:00
Lisa Hsu
014f451d1f Merge zizzer.eecs.umich.edu:/bk/m5
into shizzle.(none):/home/hsul/work/linux-clean

--HG--
extra : convert_revision : e0192f56c189e595daf3c020a56e1c02010aefbc
2004-07-06 20:04:17 -04:00
Nathan Binkert
8d6fb1e9b5 disable the binned_fn stuff
cpu0 -> cpu since we're not all set up for MP in the
ini files just yet.

--HG--
extra : convert_revision : 69a5af791ea89fec7e647d55644cc4eeaf924651
2004-07-06 19:27:33 -04:00
Nathan Binkert
2fecc3c785 Make binning work with stuff other than FS_MEASURE
--HG--
extra : convert_revision : e3fa3cfbdf2f13dd3a8d2266dd64c2c335f998d6
2004-07-06 19:22:39 -04:00
Nathan Binkert
347166d908 Fix some ini stuff.
--HG--
extra : convert_revision : b870550f52901fe723eeab24a14ac4c1222ee1bd
2004-07-06 19:21:47 -04:00
Lisa Hsu
19fd3439c7 tag each mem_req as coming from the nic if it is DMA'd from the NIC. the NIC tells the DMA interface, which in turn sets a new nic_req flag in the MemReq it makes.
dev/ns_gige.cc:
    tell all outgoing dma events that this request is from the NIC

--HG--
extra : convert_revision : 62af17a2728a0ff729e7723dc29bd0d130ca5fe3
2004-07-06 17:55:16 -04:00
Lisa Hsu
151bb42173 Merge zizzer.eecs.umich.edu:/bk/m5
into shizzle.(none):/home/hsul/work/linux-clean

--HG--
extra : convert_revision : cd46aa31dab82e43c3c4f8bb6f0fe2e018acfd12
2004-07-06 17:42:58 -04:00
Erik Hallnor
4869ba881e Remove duplicate stat
cpu/trace/reader/m5_reader.cc:
    Wrap assert variable in NDEBUG

--HG--
extra : convert_revision : 8c79dc30eff2f2fa0110a04c30df17ec4417c28c
2004-07-06 11:29:23 -04:00
Ali Saidi
fe3a86e2ef updated to work on all platforms (actually map sure that the physical
address we are trying to access is mapped before accessing it.)

--HG--
extra : convert_revision : 104341334a3d2bb812a6b0b6277ab353f8f9b39e
2004-07-05 16:49:56 -04:00