derek/gem5
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Commit Graph

  • e47f1667b6 Don't mask the pc because the Alpha predecoder needs it to set the PAL mode bit in the ExtMachInst. Gabe Black 2007-06-02 03:41:47 +00:00
  • aba2eeaf8f Fix typo so m5.fast will compile Nathan Binkert 2007-06-01 20:41:46 -07:00
  • 66ee27078e Merge zizzer:/bk/newmem into zeep.pool:/z/saidi/work/m5.newmem Ali Saidi 2007-06-01 14:55:17 -04:00
  • be0aef9819 Merge zizzer:/bk/newmem into zeep.pool:/z/saidi/work/m5.newmem Ali Saidi 2007-06-01 14:18:45 -04:00
  • d8f6769962 cast sizeof(MachInst) to Addr before generating a mask Ali Saidi 2007-06-01 14:16:58 -04:00
  • d8c487c401 don't generate trace data unless tracing is on Ali Saidi 2007-06-01 13:44:24 -04:00
  • 6e8a06b237 Clean things up Gabe Black 2007-06-01 16:24:51 +00:00
  • 85caab4e8c Merge zizzer.eecs.umich.edu:/bk/newmem into ahchoo.blinky.homelinux.org:/home/gblack/m5/newmem-x86 Gabe Black 2007-06-01 16:24:50 +00:00
  • a7fe9345ee Minor error. Forgotten to remove brackets for threadPC. Vincentius Robby 2007-06-01 12:24:49 -04:00
  • a703fdfcf9 Add a second section to make sure the ROM is extended properly. Gabe Black 2007-05-31 22:21:21 +00:00
  • 2bdd4eda12 Add rom based macroops into the macroop dict instead of dropping them on the floor Gabe Black 2007-05-31 22:21:20 +00:00
  • 287446396c Do something with ROM based macroops Gabe Black 2007-05-31 22:21:19 +00:00
  • fbcc35450d Merge zizzer.eecs.umich.edu:/bk/newmem into ahchoo.blinky.homelinux.org:/home/gblack/m5/newmem-x86 Gabe Black 2007-05-31 22:21:18 +00:00
  • 8ab0ffa3dd Merge zizzer:/bk/newmem into zeep.pool:/z/saidi/work/m5.newmem Ali Saidi 2007-05-31 18:21:17 -04:00
  • 1da901d60f This is probably a more scons like way to do this Ali Saidi 2007-05-31 18:01:07 -04:00
  • d24a9c7d21 Make directives take parameters and use the directive function and not it's name Gabe Black 2007-05-31 20:45:06 +00:00
  • ace2890f9f Handle comments Gabe Black 2007-05-31 20:45:05 +00:00
  • c432588981 Merge zizzer.eecs.umich.edu:/bk/newmem into ahchoo.blinky.homelinux.org:/home/gblack/m5/newmem-x86 Gabe Black 2007-05-31 20:45:04 +00:00
  • 6b6de8aaae obey the m5 style Nathan Binkert 2007-05-31 13:45:03 -07:00
  • 83aa742d26 Merge zizzer:/bk/newmem into zamp.eecs.umich.edu:/.automount/greenville/w/acolyte/newmem Vincentius Robby 2007-05-31 16:02:31 -04:00
  • ecf1eb7248 Assign traceData to be NULL at BaseSimpleCPU constructor. Initialize a temporary variable for thread->readPC() at setupFetchRequest() to reduce function calls. exec tracing isn't needed for m5.fast binaries Moved MISCREG_GL, MISCREG_CWP, and MISCREG_TLB_DATA out of switch statement and use if blocks instead. Vincentius Robby 2007-05-31 16:01:41 -04:00
  • 4d6296a841 Merge zizzer:/bk/newmem into zeep.pool:/z/saidi/work/m5.newmem Ali Saidi 2007-05-31 15:33:30 -04:00
  • 473bf1a66a check that m4 is available before trying to use it Ali Saidi 2007-05-31 15:33:17 -04:00
  • 62fde97bb2 Early micro assembler Gabe Black 2007-05-31 13:52:48 +00:00
  • 7860c045e2 x86 work that hadn't been checked in. Gabe Black 2007-05-31 13:50:35 +00:00
  • 7797a239cc Fix cut-n-pasto to make the path correct Nathan Binkert 2007-05-30 17:19:20 -07:00
  • 0193476ea7 Fix compiling on Solaris since Nate's libelf change Ali Saidi 2007-05-30 17:08:12 -04:00
  • a9b7c558fd Merge vm1.(none):/home/stever/bk/newmem-head into vm1.(none):/home/stever/bk/newmem-cache2 Steve Reinhardt 2007-05-29 22:59:20 -07:00
  • 4e65d2678d tport.cc: Oops... forgot to update call site after changing function argument semantics. Steve Reinhardt 2007-05-30 01:53:28 -04:00
  • 94c19ad37d Merge vm1.(none):/home/stever/bk/newmem-head into vm1.(none):/home/stever/bk/newmem-cache2 Steve Reinhardt 2007-05-29 22:25:57 -07:00
  • 365e4ac374 A little more cleanup & refactoring of SimpleTimingPort. Make it a better base class for cache ports. Steve Reinhardt 2007-05-29 22:23:41 -07:00
  • 1d18fec5f3 Merge vm1.(none):/home/stever/bk/newmem-head into vm1.(none):/home/stever/bk/newmem-cache2 Steve Reinhardt 2007-05-28 21:00:42 -07:00
  • cd423aa9dd Merge zizzer.eecs.umich.edu:/bk/newmem into vm1.(none):/home/stever/bk/newmem-head Steve Reinhardt 2007-05-28 15:41:05 -07:00
  • 8dfd7f9804 Fix M4 command line... wasn't working on zizzer. A little more concise now. Steve Reinhardt 2007-05-28 18:39:35 -04:00
  • 05915ed6f7 Merge vm1.(none):/home/stever/bk/newmem-head into vm1.(none):/home/stever/bk/newmem-cache2 Steve Reinhardt 2007-05-28 08:13:40 -07:00
  • 41f6cbce9a Restructure SimpleTimingPort a bit: - factor out checkFunctional() code so it can be called from derived classes - use EventWrapper for sendEvent, move event handling code from event to port where it belongs - make sendEvent a pointer so derived classes can override it - replace std::pair with new class for readability Steve Reinhardt 2007-05-28 08:11:43 -07:00
  • 04ac944920 Reformat comments to meet line length restriction. Steve Reinhardt 2007-05-28 08:04:33 -07:00
  • 07bda077f2 Remove unnecessary include of physical.hh. Steve Reinhardt 2007-05-28 08:03:13 -07:00
  • 075f4b108a Merge vm1.(none):/home/stever/bk/newmem-head into vm1.(none):/home/stever/bk/newmem-cache2 Steve Reinhardt 2007-05-27 21:34:37 -07:00
  • 35147170f9 Move SimObject python files alongside the C++ and fix the SConscript files so that only the objects that are actually available in a given build are compiled in. Remove a bunch of files that aren't used anymore. Nathan Binkert 2007-05-27 19:21:17 -07:00
  • 6a48f6b67d Merge vm1.(none):/home/stever/bk/newmem-head into vm1.(none):/home/stever/bk/newmem-cache2 Steve Reinhardt 2007-05-26 20:15:35 -07:00
  • 4f0f217c1b Get rid of GNU libelf and its autoconf nastiness and replace it with FreeBSD's implementation Nathan Binkert 2007-05-26 18:15:22 -07:00
  • 87bb4c3792 Merge vm1.(none):/home/stever/bk/newmem-head into vm1.(none):/home/stever/bk/newmem-cache2 Steve Reinhardt 2007-05-26 17:10:35 -07:00
  • a3ae9486d5 Merge zizzer.eecs.umich.edu:/bk/newmem into doughnut.mwconnections.com:/home/gblack/m5/newmem-x86 Gabe Black 2007-05-25 19:29:32 -07:00
  • ad02a59f89 Make the lexer and parser use objects and not the last lexer and parser generated. Gabe Black 2007-05-25 19:26:26 -07:00
  • 44ebb8d3e2 Update to ply 2.3 Nathan Binkert 2007-05-24 21:54:51 -07:00
  • da46364b18 Fix getDeviceAddressRanges() to get snooping right. Steve Reinhardt 2007-05-22 07:30:55 -07:00
  • 0484867d85 Merge vm1.(none):/home/stever/bk/newmem-head into vm1.(none):/home/stever/bk/newmem-cache2 Steve Reinhardt 2007-05-22 06:36:01 -07:00
  • 41dde5f6fd memtest.hh: Fix description string. Minor whitespace cleanup. Steve Reinhardt 2007-05-22 06:32:24 -07:00
  • 9048c695a0 Another pass of minor changes in preparation for new protocol. Steve Reinhardt 2007-05-22 06:29:48 -07:00
  • 9f1c104ccd memtest.py: Make clocks more reasonable. Fix bug in sense of options.timing flag. Steve Reinhardt 2007-05-22 06:22:27 -07:00
  • 41241799ae Change getDeviceAddressRanges to use bool for snoop arg. Steve Reinhardt 2007-05-21 23:36:09 -07:00
  • 0a02e3a764 Merge vm1.(none):/home/stever/bk/newmem-head into vm1.(none):/home/stever/bk/newmem-cache2 Steve Reinhardt 2007-05-20 23:05:02 -07:00
  • 05d14cf3e2 Add new EventWrapper constructor that takes a Tick value and schedules the event immediately. Steve Reinhardt 2007-05-20 21:43:01 -07:00
  • 87adc37e91 Insist that PhysicalMemory object have at least one connection. Steve Reinhardt 2007-05-20 18:23:05 -07:00
  • dcce351eaa Merge vm1.(none):/home/stever/bk/newmem-head into vm1.(none):/home/stever/bk/newmem-cache2 Steve Reinhardt 2007-05-18 22:37:32 -07:00
  • 792d5b9e5e First set of changes for reorganized cache coherence support. Compiles but doesn't work... committing just so I can merge (stupid bk!). Steve Reinhardt 2007-05-18 22:35:04 -07:00
  • aa5b595f39 Oops... some places in C++ explicitly ask for a "functional" port. It would be better to move this to python IMO but for now I'll stick in a compatibility hack. Steve Reinhardt 2007-05-19 01:20:58 -04:00
  • 0305159abf PhysicalMemory has vector of uniform ports instead of one special one. Steve Reinhardt 2007-05-19 00:24:34 -04:00
  • a13d5af274 Merge zizzer.eecs.umich.edu:/bk/newmem into doughnut.mwconnections.com:/home/gblack/m5/newmem-x86 Gabe Black 2007-05-18 13:36:47 -07:00
  • 6a6e62014e Changes to make simple cpu handle pcs appropriately for x86 Gabe Black 2007-05-18 10:42:50 -07:00
  • a8278c3bde Update the release notes for the 2.0 beta 3 release Nathan Binkert 2007-05-16 19:09:18 -07:00
  • b85690e239 update all the regresstion tests for release Ali Saidi 2007-05-15 19:25:35 -04:00
  • c30e615689 Merge zizzer.eecs.umich.edu:/z/m5/Bitkeeper/newmem into zizzer.eecs.umich.edu:/tmp/newmem Ali Saidi 2007-05-15 19:04:34 -04:00
  • f487edf146 Merge zizzer:/bk/newmem into zeep.pool:/z/saidi/work/m5.newmem Ali Saidi 2007-05-15 18:06:52 -04:00
  • 0934f259d6 add an l2 cache option to se example config Ali Saidi 2007-05-15 18:06:35 -04:00
  • f317227b4e hopefully the final hacky change to make the bus bridge work ok cache blocks that get dmaed ARE NOT marked invalid in the caches so it's a performance issue here Ali Saidi 2007-05-15 17:39:50 -04:00
  • 224ae7813d Merge vm1.(none):/home/stever/bk/newmem-head into vm1.(none):/home/stever/bk/newmem-cache2 Steve Reinhardt 2007-05-14 13:54:22 -07:00
  • bda25c1d5e Merge zizzer.eecs.umich.edu:/z/m5/Bitkeeper/newmem into zizzer.eecs.umich.edu:/tmp/newmem Ali Saidi 2007-05-14 16:37:23 -04:00
  • fcf85725b5 Merge zizzer:/bk/newmem into zeep.pool:/z/saidi/work/m5.newmem Ali Saidi 2007-05-14 16:37:22 -04:00
  • 57104ea5f9 couple more bug fixes for intel nic Ali Saidi 2007-05-14 16:37:00 -04:00
  • ea4e6f2e3d add uglyiness to fix dmas Ali Saidi 2007-05-14 16:14:59 -04:00
  • fecae03a0b Eliminate unused PacketPtr from BaseCache's RequestEvent and ResponseEvent. Compiles but not tested. Steve Reinhardt 2007-05-13 23:09:10 -07:00
  • df3fc36fa9 Split BaseCache::CacheEvent into RequestEvent and ResponseEvent. Compiles but not tested. Steve Reinhardt 2007-05-13 22:58:06 -07:00
  • 404a91265e Merge zizzer.eecs.umich.edu:/z/m5/Bitkeeper/newmem into zizzer.eecs.umich.edu:/tmp/newmem Ali Saidi 2007-05-13 04:48:42 -04:00
  • af26532bbd fix handling of atomic packets fix up code for counting requests and responses Ali Saidi 2007-05-13 01:44:42 -04:00
  • debf04aef1 Make sure all addresses used in syscalls are truncated to 32 bits. Actually -all- arguements are truncated to 32 bits, but we should be able to get away with it. Gabe Black 2007-05-12 15:11:44 -07:00
  • 011db5c851 Move full CPU sim object stuff into the encumbered directory Nathan Binkert 2007-05-11 15:01:44 -07:00
  • 113319a7da Float should have a c++ param type Nathan Binkert 2007-05-11 11:48:58 -07:00
  • d667ce01b4 total should be the sum of the vector result of an operation, not sum the operands and then apply the operation. Nathan Binkert 2007-05-11 11:47:18 -07:00
  • 634d2e9d83 remove hit_latency and make latency do the right thing set the latency parameter in terms of a latency add caches to tsunami-simple configs Ali Saidi 2007-05-10 18:24:48 -04:00
  • 6d199f0b25 Merge zizzer.eecs.umich.edu:/bk/newmem into doughnut.mwconnections.com:/home/gblack/newmem-o3-micro Gabe Black 2007-05-09 22:04:58 -07:00
  • e08a5c6052 Merge zizzer:/bk/newmem into pb15.local:/Users/ali/work/m5.newmem.zeep Ali Saidi 2007-05-10 00:36:47 -04:00
  • 77fd734b99 update for bus bridge updates Ali Saidi 2007-05-10 00:30:53 -04:00
  • 4a37c48e8e add/update parameters for bus bridge Ali Saidi 2007-05-10 00:08:22 -04:00
  • 4ad1b58fdd Merge zizzer.eecs.umich.edu:/bk/newmem into doughnut.mwconnections.com:/home/gblack/newmem-o3-micro Gabe Black 2007-05-09 20:50:46 -07:00
  • 69ea50c163 couple of updates in the intel nic Ali Saidi 2007-05-09 22:39:43 -04:00
  • 9dfbea68a2 update for new reschedule semantics Ali Saidi 2007-05-09 22:34:54 -04:00
  • ff55888575 undo my previous bus change, it can make the bus deadlock.. so it still constantly reschedules itself Ali Saidi 2007-05-09 22:23:01 -04:00
  • 376cff64bd Merge zeep:/z/saidi/work/m5.newmem into zizzer.eecs.umich.edu:/tmp/newmem Ali Saidi 2007-05-09 18:20:25 -04:00
  • 3c608bf765 add a backoff algorithm when nacks are received by devices add seperate response buffers and request queue sizes in bus bridge add delay to respond to a nack in the bus bridge Ali Saidi 2007-05-09 18:20:24 -04:00
  • 37b45e3c8c fix the translating ports so it can add a page on a fault Ali Saidi 2007-05-09 15:37:46 -04:00
  • 939cbd8201 Merge zizzer:/bk/newmem into udhcp-macvpn-703.public.engin.umich.edu:/Users/ali/work/m5.newmem Ali Saidi 2007-05-09 12:02:36 -04:00
  • ee70d8cfc4 bit_val was being used directly in the statement in return. If type B had fewer bits than last, bit_val << last would get the wrong answer. Ali Saidi 2007-05-09 12:01:31 -04:00
  • c2ac0fd89b Fix insertBits so it doesn't shift things into oblivion Gabe Black 2007-05-08 17:19:33 +00:00
  • dc1c9e0300 Add a hack to truncate addresses to 32 bits in SE. Paging should be changed to use the architecture's TLB, at which point this can be removed. Gabe Black 2007-05-08 13:02:19 +00:00
  • aa095b63cd update for partial write fix changes Ali Saidi 2007-05-07 19:19:00 -04:00
  • a38c79ec22 the bridge never returns false when recvTiming() is called on its ports now, it always returns true and nacks the packet if there isn't sufficient buffer space fix the timing cpu to handle receiving a nacked packet Ali Saidi 2007-05-07 18:58:38 -04:00
  • 0dfc29a023 fix partial writes with a functional memory hack figure out the block size from devices attached to the bus otherwise use a default block size when no devices that care are attached Ali Saidi 2007-05-07 14:42:03 -04:00
  • b7292a1713 Merge zizzer:/bk/newmem into zeep.pool:/z/saidi/work/m5.newmem Ali Saidi 2007-05-01 18:14:45 -04:00