e650870c15update for new hierarchy parameters
Erik Hallnor
2004-01-13 15:23:54 -0500
003d2d724cMerge ehallnor@zizzer:/bk/m5 into zazzer.eecs.umich.edu:/z/ehallnor/m5
Erik Hallnor
2004-01-13 02:10:48 -0500
413bfc68e6Change to a new centralized way to specify the memory hierarchy parameters (do_data and do_events). If you use the defaults (false and true respectively) you don't have to change your INI files.
Erik Hallnor
2004-01-13 02:10:35 -0500
a44248aab0Added code to print out the symbol if one exists for an address
Ali Saidi
2004-01-12 19:59:02 -0500
4cda5224f6- quick support for using a regular expression to select which stats to print - print out the subname
Nathan Binkert
2004-01-12 18:39:39 -0500
308153665dCommit the correct version of the code
Nathan Binkert
2004-01-12 15:45:51 -0500
b1295a7317make the python stuff executable
Nathan Binkert
2004-01-12 13:38:37 -0500
8384f25620Check in python script that reads the python stats output. Still needs a lot of work, but it's at least as good as the normal output, and has some features that normal ouput is missing such as bin totalling with correct formulas and totalling of 2d vectors and such. It will also filter out specific stats for printing. (basically grep, but no mistakes.)
Nathan Binkert
2004-01-12 13:31:59 -0500
60a1b35e4daddad kern/linux
Ali Saidi
2004-01-11 22:00:35 -0500
b4301c652cMerge zizzer.eecs.umich.edu:/bk/m5 into ziff.eecs.umich.edu:/z/binkertn/research/m5/latest
Nathan Binkert
2004-01-11 21:22:33 -0500
f8ed615bc6Minor cleanup to bk ignore and makefile libelf rule.
Steve Reinhardt
2004-01-11 15:26:48 -0800
510eef0fa0Modify handling of serialize:dir parameter to make it more useful.
Steve Reinhardt
2004-01-11 15:24:18 -0800
7e07aa9300Fix cprintf bug, plus minor cleanup & better error messages.
Steve Reinhardt
2004-01-11 15:09:27 -0800
54b6d8c5a8the build directory is now correctly created
Ali Saidi
2004-01-08 22:42:21 -0500
98613950a3Fixed another small libelf problem, everything should work now, on any machine.
Ali Saidi
2004-01-08 16:35:47 -0500
e7033f718fshould now build... fixed relative path problem
Ali Saidi
2004-01-08 14:13:31 -0500
a3691fe09eMerge zizzer.eecs.umich.edu:/bk/m5 into ziff.eecs.umich.edu:/z/binkertn/research/m5/latest
Nathan Binkert
2004-01-08 12:25:49 -0500
14b4119aafChanged elf_object.* to use libelf for reading elf files. The makefile and setup filed had to be modified to build libelf. The ext repository MUST exist in the directory below m5!
Ali Saidi
2004-01-07 19:46:45 -0500
9dcfdce732Improve python output support fix some bugs
Nathan Binkert
2003-12-28 12:48:13 -0500
85d75e6dfcAdd a uni-directional bus bridge
Erik Hallnor
2003-12-28 00:41:47 -0500
55d94ba2e8Add python output support to the statistics package!
Nathan Binkert
2003-12-24 03:25:36 -0500
3f5ca9e5e8initial basic hooks into the stats package to output python code python dumping takes a name for the output data
Nathan Binkert
2003-12-24 02:35:33 -0500
fad8b13addRandom cleanup to make things work a bit better
Nathan Binkert
2003-12-24 01:45:56 -0500
9908987f15never print the cdf or pdf or the overflow bucket if we're simplescalr
Nathan Binkert
2003-12-24 00:22:02 -0500
a95d8e95cb- change the FormatFlags into more general StatFlags - make init and print StatFlags - default_mode -> DefaultMode - the display mode is no longer needed as part of the stat pass it into the display functions so the mode can be determined from dump to dump - get rid of old simplescalar bug and make the distribution min_val work correctly - get rid of the DisplayMode parameter to distribution updates since it is no longer needed
Nathan Binkert
2003-12-23 23:53:43 -0500
8f912297ab- Allow the stats package user to determine in which mode to dump - Factor out the formatted output into a display function - Make the stats package support tracing - Clean up output of binned data
Nathan Binkert
2003-12-23 21:45:48 -0500
42621e497dMake it so you can override CC and CXX
Nathan Binkert
2003-12-23 20:59:30 -0500
ce0396c19eMake this compile with NDEBUG defined
Nathan Binkert
2003-12-23 14:54:12 -0500
318e0c93edadd support for simple character input via the system console
Nathan Binkert
2003-12-22 17:51:14 -0500
c3ba166e65Fix binary path in configs/kernel/base.ini.
Steve Reinhardt
2003-12-22 08:47:55 -0800
ecd3ecb4f5Merge ehallnor@zizzer:/bk/m5 into zazzer.eecs.umich.edu:/z/ehallnor/m5
Erik Hallnor
2003-12-21 22:47:33 -0500
e249a54e4cAdd fully functional snooping to the hierarchy.
Erik Hallnor
2003-12-21 22:43:48 -0500
918c5d27abFix last patch to RTC initialization. It's still the case that the default is to use a fixed time, which was the main point of the last changeset.
Steve Reinhardt
2003-12-20 23:59:13 -0800
3fdeebac24Default RTC behavior is to set to Jan 1 2010 (midnight GMT), not current wall-clock time.
Steve Reinhardt
2003-12-20 20:17:18 -0800
126d810d13Merge zizzer:/bk/m5 into isabel.reinhardt.house:/z/stever/bk/m5
Steve Reinhardt
2003-12-20 19:56:31 -0800
51286f22b1we don't actually want to refer to the stable symlink because we want to keep revision control of which system we're referring to
Nathan Binkert
2003-12-20 19:18:34 -0500
2a720ee99dupdate to reflect changes in the directory structure
Nathan Binkert
2003-12-20 19:14:17 -0500
520442b01aFix a bunch of memory leak type bugs, and add some better checking in places to make sure stuff is behaving properly.
Nathan Binkert
2003-12-20 16:25:48 -0500
f182f7a2eeMerge ehallnor@zizzer:/bk/m5 into zazzer.eecs.umich.edu:/z/ehallnor/m5
Erik Hallnor
2003-12-19 18:33:54 -0500
573236a7fbMerge ehallnor@zizzer:/bk/m5 into zazzer.eecs.umich.edu:/z/ehallnor/m5
Erik Hallnor
2003-12-19 12:10:06 -0500
7d8366ac49Fix non-determinism on misspeculated uncached store conditionals.
Steve Reinhardt
2003-12-19 08:10:53 -0800
4f8acd2709Minor cleanup of trace/output stuff (leftover from EINTR bug fix).
Steve Reinhardt
2003-12-19 08:04:40 -0800
255ac8372dMake stuff build on openbsd
Nathan Binkert
2003-12-19 00:02:20 -0500
1154c4de7bOkay, FULL_SYSTEM uses cpu_num
Erik Hallnor
2003-12-17 23:43:19 -0500
df9a435b8fMerge ehallnor@zizzer:/bk/m5 into zazzer.eecs.umich.edu:/z/ehallnor/m5
Erik Hallnor
2003-12-17 22:55:55 -0500
27cb54e92dInitial probe checkin. Still need to write the snoopProbe functions for coherence to work properly.
Erik Hallnor
2003-12-17 22:55:45 -0500
888ea0dfd9Re-add code to handle following multitple chains that got removed when chain stuff moved into dispatch
Steve Raasch
2003-12-17 15:31:51 -0500
96fddedb9cAdd parameters to Segmented IQ
Steve Raasch
2003-12-17 15:30:15 -0500
41f38e2646Be careful what you put inside an assert() !!!
Steve Raasch
2003-12-17 15:27:23 -0500
d5a187dd5aMiscellaneous minor fixes/improvements. Most notable is that the rundiff trace-output lossage is fixed!
Steve Reinhardt
2003-12-17 09:43:03 -0800
16021aa436Fix previously committed call_pal fix... the main problem was in the makefile, such that decoder.cc was not getting rebuilt. Also add -fno-strict-aliasing to fix all the bizarre problems I've been having with g++ 3.3.x.
Steve Reinhardt
2003-12-16 11:46:52 -0800
f47dcadfd8Merge zizzer:/bk/m5 into isabel.reinhardt.house:/z/stever/bk/m5
Steve Reinhardt
2003-12-15 21:07:04 -0800
2cd5e980d2Fixes for full-system call_pal instruction.
Steve Reinhardt
2003-12-15 21:06:09 -0800
d0feece6e0Minor enhancements to rundiff.
Steve Reinhardt
2003-12-15 20:58:18 -0800
eb1cfea1aaAnother fix for gcc 3.3.x.
Steve Reinhardt
2003-12-15 20:56:47 -0800
e537fb3fd6Add #define parameter and corresponding build option to disable FastAlloc. Gcc 3.3.x appears to have a bug that breaks FastAlloc when compiled with optimization.
Steve Reinhardt
2003-12-15 20:55:10 -0800
ce4aba3e54Add support for getting the hostname
Nathan Binkert
2003-12-15 18:24:02 -0500
ef659565ffClean up time handling code
Nathan Binkert
2003-12-15 18:23:48 -0500
3c6fc7c269- Add a new target m5.fast (or just fast for short) - make stuff compile with NDEBUG - remove the TRACE setup option since it isn't useful anymore
Nathan Binkert
2003-12-11 19:22:03 -0500
1cb7159966Make memtest compile again.
Steve Reinhardt
2003-12-11 08:46:21 -0800
45d0a46080Minor util/find-files fix.
Steve Reinhardt
2003-12-11 08:30:35 -0800
7c708c8d1b- Switch events to use a priority enum instead of integers. This lets us centralize priorities so we can see what's going on. - Shift serialize & cpu-switch events to happen before CPU ticks (to be consistent with starting new CPU on same cycle instead of next cycle). - Get rid of unnecessary bus stats reset callback.
Steve Reinhardt
2003-12-11 08:29:52 -0800
777c1ebfabStats & serialization tweaks & cleanup. Unserializing from a checkpoint now gives identical results to running from scratch and doing at switchover at the same cycle! - CPUs start at cycle 0 again, not cycle 1. - curTick is now serialized & unserialized. - Stats get reset in main (before event loop). Since this is done after curTick is unserialized, simTicks gets set correctly for running from a checkpoint. - Simplify serialization to happen in a single pass. - s/Serializeable/Serializable/
Steve Reinhardt
2003-12-11 00:16:46 -0800
7e6dcd812cRename RegResetCallback to registerResetCallback(). Add reset callback for bus (to fix idle cycles computation).
Steve Reinhardt
2003-12-10 17:49:41 -0800
4ce6118fdaFactor ExecContext::setStatus(), BaseCPU::execCtxStatusChange(), and SimpleCPU::setStatus() into separate functions. For example, setStatus(Active) is now activate().
Steve Reinhardt
2003-12-10 17:47:28 -0800
db6038937dMinor fixes. Somehow do-tests.pl suddenly started failing on test5 because of a wrong exit code; fixed that.
Steve Reinhardt
2003-12-09 20:25:52 -0800
3d8cd20467Merge zizzer:/bk/m5 into isabel.reinhardt.house:/z/stever/bk/m5
Steve Reinhardt
2003-12-09 16:43:54 -0800
06aafccce7Make the simulator compile on an Itanium.
Nathan Binkert
2003-12-09 17:41:08 -0500
a1dcdeb3d8Merge zizzer:/bk/m5 into isabel.reinhardt.house:/z/stever/bk/m5
Steve Reinhardt
2003-12-09 14:21:46 -0800
5e360051cfMinor tweaks to make a switchover at tick N match restarting from a checkpoint at tick N.
Steve Reinhardt
2003-12-09 14:21:38 -0800
f13509e968Generate a fault when a privileged PAL call is executed in non-privileged (non-PAL) mode.
Steve Reinhardt
2003-12-09 14:19:35 -0800
368de4109fSweeping change in the stats package....again...
Nathan Binkert
2003-12-09 13:05:43 -0500
f8b78a81b9Remove mostly-broken binning stuff from "get" Add parser debugging command-line option
Steve Raasch
2003-12-08 17:34:14 -0500
9511bc025dMerge zizzer.eecs.umich.edu:/m5/Bitkeeper/m5 into zizzer.eecs.umich.edu:/y/sraasch/m5
Steve Raasch
2003-12-08 17:13:18 -0500
ac51f420c7Fetch-loss now uses the stats package
Steve Raasch
2003-12-08 17:12:45 -0500
1d7c11af7dInstead of keeping track of the fraction of time that we're idle, keep track of the fraction of time we're not idle. This works better because the default processor state is idle, and the default stat value is 0. Keep the stat as idleFraction which is a formula that is equal to 1 - notIdleFraction
Nathan Binkert
2003-12-08 14:01:48 -0500
0ff2457bfamove setStatus into the .cc file
Nathan Binkert
2003-12-08 13:15:18 -0500
37a718e45eVarious cleanup. Mostly forcing initialization of some fields... I noticed that checkpoints dumped from m5.debug and m5.opt had differences, which is no longer the case (other than addresses encoded in names).
Steve Reinhardt
2003-12-06 09:57:59 -0800
b6c77fe6f8remote_gdb.cc: fix remote gdb
Lisa Hsu
2003-12-04 19:26:31 -0500
a229495400Minor fixes to the Segmented and Seznec IQ models
Steve Raasch
2003-12-03 12:48:57 -0500
9984b2bd6cFormatting & doxygen docs for new syscall emulation code.
Steve Reinhardt
2003-12-01 22:39:27 -0800
745f0044cdMerge zizzer:/bk/m5 into isabel.reinhardt.house:/z/stever/bk/m5
Steve Reinhardt
2003-12-01 19:34:51 -0800
7976794aadRestructuring of LiveProcess etc. to support multiple emulated OS syscall interfaces, and specific support for Alpha Linux. Split syscall emulation functions into several groups, based on whether they depend on the specific OS and/or architecture (and all combinations of above), including the use of template functions to support syscalls with slightly different constants or interface structs.
Steve Reinhardt
2003-12-01 19:34:38 -0800
5b24b5a5c5Change the way writebacks are handled to make probing easier. Minor fixes to get coherence-test to work again.
Erik Hallnor
2003-11-25 18:11:48 -0500
6f1fa02f51Extract documentation for all classes.
Erik Hallnor
2003-11-24 15:46:39 -0500
c02fc7e99fget rid of the errors from simple-test1 and test1-6
Lisa Hsu
2003-11-23 12:57:16 -0500
713e2b12e4Merge zizzer.eecs.umich.edu:/bk/m5 into zans.eecs.umich.edu:/z/binkertn/research/m5/stats
Nathan Binkert
2003-11-22 15:09:30 -0500
d1e12b57ceWe only need to choose that we do want binning, or we don't. no need for multiple bin classes. If multiple bins are needed, we can always do it with ini type config instead.
Nathan Binkert
2003-11-22 15:07:53 -0500
d760ca18abMerge ehallnor@zizzer:/bk/m5 into zazzer.eecs.umich.edu:/z/ehallnor/m5
Erik Hallnor
2003-11-20 23:03:42 -0500
e5eb961274full_cpu.cc: change name of numTicks to "numTicks" instead of "number of ticks" (which screws up scripts).
Lisa Hsu
2003-11-20 08:06:33 -0500
3e57b99ad6Pass all the related information (the memreq) to the findBlock call. Working towards not caring about block addresses anymore.
Erik Hallnor
2003-11-19 16:41:41 -0500
24061d5bffPass MemReqPtr by reference.
Erik Hallnor
2003-11-19 16:40:05 -0500
65ea19b6f5Move some MSHR typedefs into the class, transition from MSHRblah to MSHR::blah.
Erik Hallnor
2003-11-19 15:26:14 -0500
92348c503cmake the name consistent with what it does.
Nathan Binkert
2003-11-15 12:32:36 -0500
123d0294b5Make it so that we create a directory for each checkpoint so that there aren't so many files littering your directory
Nathan Binkert
2003-11-14 01:19:16 -0500
Erik Hallnor
Ali Saidi
Nathan Binkert
Steve Reinhardt
Steve Raasch
Lisa Hsu