Commit Graph

  • 36e9ca5611 Merge zizzer.eecs.umich.edu:/bk/newmem into ewok.(none):/home/gblack/m5/newmem Gabe Black 2006-07-26 03:48:48 -04:00
  • e803c8a912 Added alot of fp instructions, and some impdep instructions. Gabe Black 2006-07-26 03:42:16 -04:00
  • e081615cd9 Now ignore sigaction Gabe Black 2006-07-26 03:40:56 -04:00
  • 6e969c31c7 Merge zizzer.eecs.umich.edu:/z/m5/Bitkeeper/newmem into zizzer.eecs.umich.edu:/.automount/zooks/y/ksewell/research/m5-sim/newmem-o3 Korey Sewell 2006-07-23 13:41:53 -04:00
  • 19ca97af79 This changeset gets the MIPS ISA pretty much working in the O3CPU. It builds, runs, and gets very very close to completing the hello world succesfully but there are some minor quirks to iron out. Who would've known a DELAY SLOT introduces that much complexity?! arrgh! Korey Sewell 2006-07-23 13:39:42 -04:00
  • 887e00f18d Merge m5.eecs.umich.edu:/bk/newmem into ewok.(none):/home/gblack/m5/newmem Gabe Black 2006-07-23 03:06:12 -04:00
  • 8711078885 Added myself to the authors list. Gabe Black 2006-07-23 03:04:46 -04:00
  • f9729e999f Merge ktlim@zizzer:/bk/newmem into zamp.eecs.umich.edu:/z/ktlim2/clean/newmem-merge Kevin Lim 2006-07-23 00:10:52 -04:00
  • 6ef6e9b14d Fix up test.py Kevin Lim 2006-07-23 00:10:11 -04:00
  • c77cb31473 Reorganized SPARC binaries Gabe Black 2006-07-22 17:30:50 -04:00
  • 14b11a9734 Fixed subtract with carry, and started some work with floating point. Gabe Black 2006-07-22 15:50:40 -04:00
  • 7ccdb7accc Last minute check in. Very few functional changes other than some minor config updates. Also include some recently generated stats. Kevin Lim 2006-07-22 15:50:39 -04:00
  • db5f710a7b Merge ktlim@zizzer:/bk/newmem into zamp.eecs.umich.edu:/z/ktlim2/clean/newmem-merge Kevin Lim 2006-07-21 16:08:17 -04:00
  • a6c2e5cca0 Rearrange the FS configs to be more shared. Also check in the full-system tests. Reference stats coming soon. Kevin Lim 2006-07-21 15:56:35 -04:00
  • acf9cf21a2 Missed some files in a previous check-in. Also check-in the SMT tests. Reference stats will be coming soon. Kevin Lim 2006-07-21 15:53:07 -04:00
  • bf90e1dbde Minor functionality updates. Kevin Lim 2006-07-21 15:46:12 -04:00
  • bd33d8d4ac Some reorganization. Options are all handled at the user level script. Move createCpus function (now called connectCpu) to Util.py, where it can be used by other configs. Kevin Lim 2006-07-21 15:42:44 -04:00
  • 8bbe925192 Fixed a glitch in the disassembly output. Gabe Black 2006-07-20 21:01:57 -04:00
  • 419acd31cb Merge m5.eecs.umich.edu:/bk/newmem into ewok.(none):/home/gblack/m5/newmem Gabe Black 2006-07-20 19:04:09 -04:00
  • 0850c3dedb Merge zizzer:/bk/newmem into zeep.pool:/z/saidi/work/m5.newmem Ali Saidi 2006-07-20 19:04:08 -04:00
  • 851f91f245 Move PioPort timing code into Simple Timing Port object Make PioPort use it Make Physical memory use it as well Ali Saidi 2006-07-20 19:03:47 -04:00
  • e8a3295075 Enforce the timing cpu ticking at it's clock rate Add a max time option in seconds and a single system root clock be 1THz Ali Saidi 2006-07-20 19:00:40 -04:00
  • 15e5ce96c7 Merge zizzer:/bk/newmem into zeep.pool:/z/saidi/work/m5.newmem.head Ali Saidi 2006-07-19 17:59:04 -04:00
  • 6175f712b3 Merge zizzer:/bk/newmem into zeep.pool:/z/saidi/work/m5.newmem Ali Saidi 2006-07-19 17:24:45 -04:00
  • a50c5cc999 Change the device latency here to a latency rather than a Tick Ali Saidi 2006-07-19 17:24:20 -04:00
  • 87d4859458 Minor changes to reflect state used for regression stats. Kevin Lim 2006-07-19 16:09:34 -04:00
  • 4bd025742d Put regression tests back into m5. They are located in the "tests" directory. The directory output and reference outputs have changed slightly. Now the directory is ALPHA_SE/test/<test>/<cpu_model>/, and for the reference stats <test>/ref/<arch>/<cpu_model> Kevin Lim 2006-07-19 16:07:25 -04:00
  • 660ea2b176 Get the path to load the ini file from. I'm not sure if this fix is needed in other places as well. Kevin Lim 2006-07-19 15:28:53 -04:00
  • 85515c4976 O3CPU fixes. Kevin Lim 2006-07-19 15:28:02 -04:00
  • 0cedb23d3c Some minor compiling fixes. Kevin Lim 2006-07-19 15:26:48 -04:00
  • 1fdf14dd5a Update configs. Kevin Lim 2006-07-19 15:24:22 -04:00
  • 82b1a278ec Merge ktlim@zizzer:/bk/newmem into zamp.eecs.umich.edu:/z/ktlim2/clean/newmem-merge Kevin Lim 2006-07-19 14:46:05 -04:00
  • b7b603f9a7 Cleaned things up a little. Gabe Black 2006-07-19 02:07:00 -04:00
  • 44974a4462 Merge m5.eecs.umich.edu:/bk/newmem into ewok.(none):/home/gblack/m5/newmem Gabe Black 2006-07-18 18:23:23 -04:00
  • fe9e851e8c update test3 Korey Sewell 2006-07-17 16:50:20 -04:00
  • 31ac8e7337 Merge ktlim@zizzer:/bk/newmem into zamp.eecs.umich.edu:/z/ktlim2/clean/newmem-merge Kevin Lim 2006-07-14 17:54:43 -04:00
  • 138a4faf28 Minor updates. Kevin Lim 2006-07-14 17:53:16 -04:00
  • 40ebf0811a Fix the CheckerCPU being included via python. Kevin Lim 2006-07-14 17:51:29 -04:00
  • 07186de5a1 forgot tid Korey Sewell 2006-07-14 13:22:35 -04:00
  • b2c51d064b For now, halt context is the same as deallocating. suspend context will now take the thread off the activeThread list. Korey Sewell 2006-07-14 13:06:37 -04:00
  • f1a7e0d3b3 MIPS specific fixes ... the main thing is that SMT threads get their own stack space instead of all stacks start to space Korey Sewell 2006-07-14 04:52:08 -04:00
  • efc06d0545 Merge zizzer:/bk/newmem into zeep.pool:/z/saidi/work/m5.newmem Ali Saidi 2006-07-13 15:50:09 -04:00
  • e1b8e71500 fix help when no arguments are passed to m5 Ali Saidi 2006-07-13 15:48:41 -04:00
  • c368ff0bd8 add system.mem_mode = ['timing', 'atomic'] update scripts acordingly Ali Saidi 2006-07-13 15:48:17 -04:00
  • 1e4acb8e01 Move Dcache port creation from LSQUnit to LSQ in order to support Ron's recent changes, and using the O3CPU in SMT mode. Kevin Lim 2006-07-13 13:12:51 -04:00
  • 2af213022c Fix for bug when squashing and the fetching. Now fetch checks if the cache data is valid. Kevin Lim 2006-07-13 13:09:29 -04:00
  • a0a952d5ff Update for changes to draining. Kevin Lim 2006-07-13 13:08:58 -04:00
  • 63c999edef Fix help message printing. Might need to clean up the handling of the sys.exit() call, as right now it prints out "None" at the end (not sure why). Kevin Lim 2006-07-13 12:21:21 -04:00
  • 2bc9229ea7 memory mode information now contained in system object States are now running, draining, or drained. memory state information moved into system object system parameter is not fs only for cpus Implement drain() support in devices Update for drain() call that returns number of times drain_event->process() will be called Ali Saidi 2006-07-12 20:22:07 -04:00
  • 6dfaf06edf Merge ktlim@zizzer:/bk/newmem into zamp.eecs.umich.edu:/z/ktlim2/clean/newmem-merge Kevin Lim 2006-07-12 17:21:25 -04:00
  • 6f81ae5cad Be sure to include the EIO sources as well so we can run regression tests. Kevin Lim 2006-07-12 17:20:01 -04:00
  • e758c1fc04 Serialization changes to make O3CPU consistent with the other models. Kevin Lim 2006-07-12 17:18:34 -04:00
  • a9ca36639f Initial try of consolidating configuration files so they can be shared more easily, especially across regression tests and simple examples. Kevin Lim 2006-07-12 17:17:17 -04:00
  • 35ab2296d3 Push more default options to the Python object level as they are rarely changed. These are the changes that Steve was working on. Kevin Lim 2006-07-12 17:16:00 -04:00
  • 807afe8292 Updates for serialization. As long as the tickEvent doesn't need to be serialized (I don't believe it does because we drain all CPUs prior to checkpointing), it should be feasible to start up from other CPU's checkpoints. Kevin Lim 2006-07-12 17:11:57 -04:00
  • bbfe1db6b3 Merge ktlim@zizzer:/bk/newmem into zamp.eecs.umich.edu:/z/ktlim2/clean/newmem Kevin Lim 2006-07-12 15:25:34 -04:00
  • 6d120b7912 Track the PC of the cache data stored in fetch so it doesn't access memory multiple times if information is already in fetch. Kevin Lim 2006-07-12 15:24:27 -04:00
  • bf4fdbe25a Add --pdb Nathan Binkert 2006-07-12 15:21:23 -04:00
  • 10a8769d6d Merge m5.eecs.umich.edu:/bk/newmem into iceaxe.:/Volumes/work/research/m5/current Nathan Binkert 2006-07-12 15:19:08 -04:00
  • 3218538740 Fix __file__ for scripts Nathan Binkert 2006-07-12 15:18:49 -04:00
  • f7216daee0 Add a cache version of FS (should really make this an option in original) Ron Dreslinski 2006-07-11 16:03:42 -04:00
  • 2b7e685268 Merge zizzer:/z/m5/Bitkeeper/newmem into zizzer.eecs.umich.edu:/.automount/zazzer/z/rdreslin/m5bk/newmem Ron Dreslinski 2006-07-11 15:42:49 -04:00
  • 6bcc65c1f8 Fix ordering issue with squashed Icache Fetches and Static data in packet. Ron Dreslinski 2006-07-11 15:42:31 -04:00
  • 0b0cb2bca7 Merge ktlim@zizzer:/bk/newmem into zamp.eecs.umich.edu:/z/ktlim2/clean/newmem Kevin Lim 2006-07-11 13:43:30 -04:00
  • 7078d8d1b4 Fix option parsing. Nathan Binkert 2006-07-11 11:28:59 -04:00
  • 55ea050d48 Migrate most of main() and and all option parsing to python Nathan Binkert 2006-07-10 23:00:13 -04:00
  • ad4374e0cd Merge zizzer:/z/m5/Bitkeeper/newmem into zizzer.eecs.umich.edu:/.automount/zazzer/z/rdreslin/m5bk/newmem Ron Dreslinski 2006-07-10 17:19:54 -04:00
  • 6592045cbc Some fixes so that MSHR's are matched and we don't issue overlapping requests with detailed cpu Ron Dreslinski 2006-07-10 17:16:15 -04:00
  • 185a5502b7 Minor fixes. Kevin Lim 2006-07-10 16:31:42 -04:00
  • 5dbd7a3f76 Merge ktlim@zizzer:/bk/newmem into zamp.eecs.umich.edu:/z/ktlim2/clean/newmem-merge Kevin Lim 2006-07-10 15:41:35 -04:00
  • f60d8217e3 Some minor cleanups. Kevin Lim 2006-07-10 15:41:28 -04:00
  • fcaafdc48c Add parameters for backwards and forwards sizes for time buffers. Kevin Lim 2006-07-10 15:40:28 -04:00
  • 5584e2b26e Update config for a system with an L2 Ron Dreslinski 2006-07-10 12:42:35 -04:00
  • 39ffd24b64 Fix offset calculation. Now L2's work with timing&atomic. Ron Dreslinski 2006-07-10 12:35:18 -04:00
  • aef232a942 Update FS configs to use cpu connectors for ports Ron Dreslinski 2006-07-10 12:07:21 -04:00
  • 6c9bde608b Fix cpu in full system to match SE. Ron Dreslinski 2006-07-10 12:03:13 -04:00
  • cdf27a0a86 Merge zizzer.eecs.umich.edu:/z/m5/Bitkeeper/newmem into zizzer.eecs.umich.edu:/.automount/zooks/y/ksewell/research/m5-sim/newmem-o3 Korey Sewell 2006-07-07 19:02:12 -04:00
  • 43245d9c2f Support for recent port changes. Kevin Lim 2006-07-07 18:24:13 -04:00
  • 8ade33d324 Support Ron's changes for hooking up ports. Kevin Lim 2006-07-07 17:33:24 -04:00
  • 744e0055b7 Fix for bug when draining and a memory access is outstanding. Kevin Lim 2006-07-07 16:48:44 -04:00
  • b2a479cfc8 Merge ktlim@zizzer:/bk/newmem into zamp.eecs.umich.edu:/z/ktlim2/clean/newmem-merge Kevin Lim 2006-07-07 16:47:28 -04:00
  • 1faada9bd9 Take the name of the checkpoint directory in when calling checkpoint() or restoreCheckpoint(). Kevin Lim 2006-07-07 16:46:08 -04:00
  • e5aff1a7f3 Merge zizzer.eecs.umich.edu:/z/m5/Bitkeeper/newmem into zizzer.eecs.umich.edu:/.automount/zooks/y/ksewell/research/m5-sim/newmem-o3 Korey Sewell 2006-07-07 16:19:13 -04:00
  • 7811500eef Fix address range calculation. Still need bus to handle snoop ranges. On the way towards multi-level caches (L2) Ron Dreslinski 2006-07-07 16:02:22 -04:00
  • 743737c28b Merge zizzer.eecs.umich.edu:/z/m5/Bitkeeper/newmem into zizzer.eecs.umich.edu:/.automount/zooks/y/ksewell/research/m5-sim/newmem-o3 Korey Sewell 2006-07-07 15:58:22 -04:00
  • 74d4d67138 Minor fix for SMT Hello Worlds to finish correctly. Still, there is a problem with the LSQ and indexing out of range in the buffer. I havent nailed down the fix yet, but it's coming ... Korey Sewell 2006-07-07 15:58:03 -04:00
  • 018ba50f2c Switch out fixes for CPUs. Kevin Lim 2006-07-07 15:38:15 -04:00
  • 76c110d924 Remove hack now that ports work properly Ron Dreslinski 2006-07-07 15:16:41 -04:00
  • ea11c7bdbe Update cpus to use the getPort function to use a connector object to connect the I/D cache ports to memory Ron Dreslinski 2006-07-07 15:15:11 -04:00
  • 161e8bf874 Merge zizzer.eecs.umich.edu:/z/m5/Bitkeeper/newmem into zizzer.eecs.umich.edu:/.automount/zooks/y/ksewell/research/m5-sim/newmem-o3 Korey Sewell 2006-07-07 04:07:00 -04:00
  • c355df5bfe Fix so that O3CPU doesnt segfault on exit. Major thing was to not execute commit if there are no active threads in CPU. Korey Sewell 2006-07-07 04:06:26 -04:00
  • 55e59e26c1 Merge ktlim@zizzer:/bk/newmem into zamp.eecs.umich.edu:/z/ktlim2/clean/newmem-merge Kevin Lim 2006-07-06 23:20:44 -04:00
  • 6872b99c29 Be sure to call resume after restoring from a checkpoint. Kevin Lim 2006-07-06 23:16:22 -04:00
  • fff7531677 Support serializing and unserializing in the O3 CPU. Also a few small fixes for draining/switching CPUs. Kevin Lim 2006-07-06 23:13:38 -04:00
  • fbe3e22474 Fix the O3CPU to support the multi-pass method for checking if the system has fully drained. Kevin Lim 2006-07-06 17:57:20 -04:00
  • e7ccc94ea3 Various serialization changes to make it possible for the O3CPU to checkpoint. Kevin Lim 2006-07-06 17:53:26 -04:00
  • 1ccfdb442f Timing cache works for hello world test. Still need 1) detailed CPU (blocking ability in cache) 1a) Multiple outstanding requests (need to keep track of times for events) 2)Multi-level support 3)MP coherece support 4)LL/SC support 5)Functional path needs to be correctly implemented (temporarily works without multiple outstanding requests (simple cpu)) Ron Dreslinski 2006-07-06 16:52:05 -04:00
  • b55cda163e Merge ktlim@zizzer:/bk/newmem into zamp.eecs.umich.edu:/z/ktlim2/clean/newmem Kevin Lim 2006-07-06 16:51:50 -04:00
  • 8bf9709d91 Two minor FS compile fixes. Kevin Lim 2006-07-06 16:26:44 -04:00
  • 8ae4f45bc4 Fixes for draining. Kevin Lim 2006-07-06 16:06:00 -04:00