Commit graph

3387 commits

Author SHA1 Message Date
Lisa Hsu
3922b2e076 warmup of 1B cpu cycles.
configs/example/fs.py:
configs/example/se.py:
    warm up of 1B CPU cycles

--HG--
extra : convert_revision : 0f3263f466fde4cd86e0663930e83617a6b3faad
2006-10-23 19:32:57 -04:00
Lisa Hsu
764f27a0c9 Merge zizzer:/bk/newmem
into  zed.eecs.umich.edu:/z/hsul/work/m5/newmem

--HG--
extra : convert_revision : bb58679e101570d50c040519fb08ffbabfee7416
2006-10-23 18:46:05 -04:00
Lisa Hsu
4da3938ed9 get rid of the "resume" step at the end of changeToTiming/Atomic because this will cause an assertion when you do the CPU switch. instead, push the responsibility of the resume upwards towards the user - documented in se.py and fs.py so it should be ok.
--HG--
extra : convert_revision : 7530cf140844e18cc26df80057f8760f29ec952b
2006-10-23 18:45:30 -04:00
Lisa Hsu
0a2387f38c make this parallel to the other cpu types so that resume works correctly.
--HG--
extra : convert_revision : 3c165af27ea0e6c7f2a17819c1717d8900f54cc1
2006-10-23 18:43:56 -04:00
Lisa Hsu
049f8d53a9 make a lot of the same changes as to fs.py for checkpointing.
1) rearrange the options to be in a nice logical order
2) add an option for what i call "standard switch", which is from simple->timing->detailed
3) make checkpointing code such that checkpoints taken from the command line override checkpoint instructions compiled into binaries.
4) add an option for maximum number of checkpoints - simulation will stop at max or maxtick, whichever is first

--HG--
extra : convert_revision : 8d905e1b297ae664d60f8c8ba48b2aac25437fc6
2006-10-23 18:42:46 -04:00
Lisa Hsu
40a04f2f40 changes regarding fs.py
1) rearrange the options to be in a nice logical order
2) add an option for what i call "standard switch", which is from simple->timing->detailed
3) change the client/server naming system to testsys/drivesys
4) make checkpointing code such that checkpoints taken from the command line override checkpoint instructions compiled into binaries.
5) add an option for maximum number of checkpoints - simulation will stop at max or maxtick, whichever is first

doesn't fully work because of a caching issue, but the python side of things i think should work - the counterpart of se.py does work.
i think i should factor out a lot of the common code in both, but i'll do that after this checkin, just to get this in the tree.

configs/example/fs.py:
    1) rearrange the options to be in a nice logical order
    2) add an option for what i call "standard switch", which is from simple->timing->detailed
    3) change the client/server naming system to testsys/drivesys
    4) make checkpointing code such that checkpoints taken from the command line override checkpoint instructions compiled into binaries.
    5) add an option for maximum number of checkpoints - simulation will stop at max or maxtick, whichever is first

--HG--
extra : convert_revision : 078e22800ff83f6e950bf5cc6fb16a98320e7c51
2006-10-23 18:07:51 -04:00
Gabe Black
ef8b7713ca Minor compile fix. Not sure why this is broken.
--HG--
extra : convert_revision : 6f181b15f37114ca0a3965cabcb2036bd2f97916
2006-10-23 11:17:59 -04:00
Gabe Black
18b2d94b8c Move around more SPARC memory code, and make block memory operations work with the timing cpu
--HG--
extra : convert_revision : 37358504c4d05d78d08c19ba3d0c99d38c4babf5
2006-10-23 11:17:15 -04:00
Gabe Black
466c387318 Merge zizzer.eecs.umich.edu:/bk/newmem
into  zeep.eecs.umich.edu:/home/gblack/m5/newmem

--HG--
extra : convert_revision : cb15101d24ef2969e1819d6bdeeb2dd1f23f02d1
2006-10-23 09:44:58 -04:00
Gabe Black
274d2670a1 Add reference outputs for SPARC on the atomic timing cpu model
--HG--
extra : convert_revision : b64ff7c05504da6112631baaae8f0d927469e16f
2006-10-23 07:57:16 -04:00
Gabe Black
20208d00e6 Broke Load/Store instructions into microcode, and partially refactored memory operations in the SPARC ISA description.
--HG--
rename : src/arch/sparc/isa/formats.isa => src/arch/sparc/isa/formats/formats.isa
rename : src/arch/sparc/isa/formats/mem.isa => src/arch/sparc/isa/formats/mem/basicmem.isa
rename : src/arch/sparc/isa/formats/blockmem.isa => src/arch/sparc/isa/formats/mem/blockmem.isa
rename : src/arch/sparc/isa/formats/mem.isa => src/arch/sparc/isa/formats/mem/mem.isa
extra : convert_revision : dbbb00f997a102871b084b209b9fa08c5e1853ee
2006-10-23 07:55:52 -04:00
Gabe Black
e9908e3c85 Don't let interupts interupt microcode at undesired points.
--HG--
extra : convert_revision : a8ddc6b213b1a1b0d9c5cd194b88ac0c6bfb2a21
2006-10-23 02:39:02 -04:00
Gabe Black
a8973c6054 Files in base shouldn't depend on things in sim. Changed "sim/host.hh" to <inttypes.hh>
--HG--
extra : convert_revision : c1e46c012a26cdb0603416f8e8a99e0ecb1c09bc
2006-10-23 02:37:54 -04:00
Gabe Black
f31d73a433 Start making memory ops work with InitiateAcc and CompleteAcc, and some minor cleanups
--HG--
extra : convert_revision : 178a8c5d0506c75ad7a7e8d691c8863235ed7e95
2006-10-23 02:36:46 -04:00
Gabe Black
3d6ff82552 Change the default constructors to take ExtMachInsts rather than regular MachInsts
--HG--
extra : convert_revision : 8fa34f82e0cbf5ce81775d572b182826c578581f
2006-10-23 02:32:58 -04:00
Steve Reinhardt
d2856c2fde Add mutex test to Benchmarks.py.
--HG--
extra : convert_revision : 9b4f1ce9a181ac5a01e5b6a68067079969dfe9ce
2006-10-22 12:52:58 -04:00
Steve Reinhardt
968311d096 Another missing case in a switch (like Nate's earlier fix).
--HG--
extra : convert_revision : b2f195c29861a09e9dd99aefcf4a173be2f8c97c
2006-10-22 12:51:49 -04:00
Steve Reinhardt
810dee6e98 Have tracediff print warning if no traceflags are set.
Elaborate on description a bit.

--HG--
extra : convert_revision : 2649961b53d6fb2774ddfb60219415ae4251db2d
2006-10-22 12:51:00 -04:00
Steve Reinhardt
5e2263fc52 Small bug fixes for timing LL/SC. Better now but
not necessarily 100% there yet.

src/mem/cache/cache_impl.hh:
    Generate response packet on failed store conditional.
src/mem/packet.hh:
    Clear packet flags when reinitializing.
    (SATISFIED in particular is one we don't want to leave set.)

--HG--
extra : convert_revision : 29207c8a09afcbce43f41c480ad0c1b21d47454f
2006-10-21 23:35:00 -07:00
Steve Reinhardt
0159529343 Add Quiesce trace flag to track CPU quiesce/wakeup events.
--HG--
extra : convert_revision : 23be99d0fe6e2184523efe5d9e0a1ac7bf19d087
2006-10-21 23:32:14 -07:00
Steve Reinhardt
883ed108e4 Just give up if a store conditional misses completely
in the cache (don't treat as normal write miss).

--HG--
extra : convert_revision : c030eb6ba25318cae422e4da31e3b802049c8c74
2006-10-21 17:19:33 -07:00
Steve Reinhardt
82e90bf5e0 Fix formatting that got screwed up when tabs were removed.
--HG--
extra : convert_revision : 98596542a5774fe010e25632836ce92b66779f53
2006-10-21 13:54:48 -07:00
Steve Reinhardt
1e6aa0d0d0 Refactor coherence state table initialization.
--HG--
extra : convert_revision : eb36dd2cc1463e5076f4758a59cf68cc6b2bafc5
2006-10-21 13:43:14 -07:00
Steve Reinhardt
fdad936f79 Merge vm1.(none):/home/stever/bk/newmem-head
into  vm1.(none):/home/stever/bk/newmem-llsc

--HG--
extra : convert_revision : 157d07cc56e8ea68741d1b8536a9856488cb4a69
2006-10-21 11:41:53 -07:00
Steve Reinhardt
e70f5507e2 Get rid of unused handleTargets() function.
--HG--
extra : convert_revision : 90032c3831d10e98c6453cd6144f9c00b9f97219
2006-10-21 11:38:23 -07:00
Steve Reinhardt
3ac1ca8ff6 Tweak a few things for better page fault debugging.
src/sim/faults.cc:
    Fix fault message.
src/kern/tru64/tru64.hh:
    Add DPRINTF to see where new thread stacks are allocated.
src/arch/alpha/faults.cc:
    Add print statement so we know what the faulting address is in SE mode.

--HG--
extra : convert_revision : 6eb2b513c339496a0d013b7e914953a0a066c12d
2006-10-21 05:28:05 -04:00
Steve Reinhardt
0e121bc14f Updated to work with new command line argument ordering.
Note that command line syntax has totally changed as a result.
See comments for more details.

--HG--
extra : convert_revision : bdb6e27abd2da83c7468dfe2a95e8bf54757ac6c
2006-10-21 02:24:27 -07:00
Nathan Binkert
5e34c3fe13 Merge zizzer.eecs.umich.edu:/bk/newmem
into  iceaxe.:/Volumes/work/research/m5/incoming

--HG--
extra : convert_revision : c9153e5dca1d1f46a34770c645761d7b0419e8ce
2006-10-21 00:32:09 -07:00
Nathan Binkert
dff1a022b8 Missing case
--HG--
extra : convert_revision : 128896dd1a654fe9a02e2c07ef6ce6799b62f21f
2006-10-21 00:31:46 -07:00
Ron Dreslinski
30cd2298df Add some default options, point it to the /dist version of the splash benchmarks
--HG--
extra : convert_revision : cd3b4f395b360d646b8b60464768eaad0fd110a4
2006-10-20 21:13:10 -04:00
Ron Dreslinski
e855a7e6f2 Merge zizzer:/bk/newmem
into  zazzer.eecs.umich.edu:/z/rdreslin/m5bk/newmemcleanest

--HG--
extra : convert_revision : 4678ce5fb0dc29a28d9cd21e687f9cee967d21fa
2006-10-20 20:04:45 -04:00
Ron Dreslinski
e198e58e1e Clean up splash2 so it works in v2.0
configs/splash2/run.py:
    Update the splash2 file

--HG--
extra : convert_revision : b57ef1ab4b8fd1eaf281358db623b7581b96546b
2006-10-20 19:53:52 -04:00
Gabe Black
0b5cf4ba6e Merge zizzer.eecs.umich.edu:/bk/newmem
into  zeep.eecs.umich.edu:/home/gblack/m5/newmem

--HG--
extra : convert_revision : 2711fec2bf72801999b060e65f0bf744c18734fb
2006-10-20 16:39:47 -04:00
Nathan Binkert
6c6b78126a Construct a correct value of PYTHONHOME from the interpreter
running SCons, make it into a sticky option that can be
overridden at build time, and set it up before the interpreter
is started.  Also, fix the code that turns sticky options into
config/*.hh so that it works with types other than bool.

--HG--
extra : convert_revision : 602398b35d4da4e813f78865678ed348fdea7270
2006-10-20 11:37:59 -07:00
Ron Dreslinski
ad783962c5 Give physical memory some latency to stress the system
--HG--
extra : convert_revision : 3ca32ff9140770d0774cac5e82807a0574db09dd
2006-10-20 13:36:26 -04:00
Ron Dreslinski
316e0fa879 Add a config file in the example with the memtester and some parser options.
--HG--
extra : convert_revision : e70ccc3de4f7a3ae20ff9ec672853ee1555ed41b
2006-10-20 13:32:24 -04:00
Ron Dreslinski
ba24ce6bb6 Get rid of a variable put back by merge.
--HG--
extra : convert_revision : 5ddb6ae5d5412f062c07c16a27b79483430b5f22
2006-10-20 13:05:39 -04:00
Ron Dreslinski
54ed57cc4c Merge zizzer:/bk/newmem
into  zazzer.eecs.umich.edu:/z/rdreslin/m5bk/newmemcleanest

src/mem/tport.cc:
    Merge PacketPtr changes

--HG--
extra : convert_revision : 0329c5803a3df67af3dda89bd9d4753fd1a286d1
2006-10-20 13:04:59 -04:00
Ron Dreslinski
28e9641c2c Use fixPacket function everywhere.
Fix fixPacket assert function.
Stop timing port from forwarding the request if a response was found in its queue on a read.

src/cpu/memtest/memtest.cc:
src/cpu/memtest/memtest.hh:
src/python/m5/objects/MemTest.py:
    Add parameter to configure what percentage of mem accesses are functional
src/mem/cache/base_cache.cc:
src/mem/cache/cache_impl.hh:
    Use fix Packet function
src/mem/packet.cc:
    Fix an assert that was checking the wrong thing
src/mem/tport.cc:
    Properly detect if we need to do the access to the functional device

--HG--
extra : convert_revision : 447cc1a9a65ddd2a41e937fb09dc0e7c74e9c75e
2006-10-20 13:01:21 -04:00
Nathan Binkert
a4c6f0d69e Use PacketPtr everywhere
--HG--
extra : convert_revision : d9eb83ab77ffd2d725961f295b1733137e187711
2006-10-20 00:10:12 -07:00
Nathan Binkert
7245d4530d refactor code for the packet, get rid of packet_impl.hh
and call it packet_access.hh and fix the #includes so
things compile right.

--HG--
extra : convert_revision : d3626c9715b9f7e51bb3ab8d97e971fad4e0b724
2006-10-19 23:38:45 -07:00
Nathan Binkert
5b246a0567 initialize end, clean up loop
--HG--
extra : convert_revision : e1c107f0c0fd5d535acd2d6c43571a5df57c9ed3
2006-10-19 23:35:59 -07:00
Nathan Binkert
b937d54510 Fix compile of m5.fast
--HG--
extra : convert_revision : a8a37c318e55e48e697e4aaba339328f000b3f60
2006-10-19 23:34:59 -07:00
Steve Reinhardt
0d2908a584 Delete unused file src/mem/cache.hh
--HG--
extra : convert_revision : 11bd043bb72eef0239fa60155e1f5a5e02de7cbc
2006-10-19 22:59:38 -07:00
Steve Reinhardt
3772e4fc97 m5term: assume localhost if host name not provided.
util/term/term.c:
    Reindent.
util/term/term.c:
    Assume localhost if only port number is given on command line.

--HG--
extra : convert_revision : 768e61a56339a0795ca258cca788e9a2c20cbaae
2006-10-19 21:42:30 -07:00
Ron Dreslinski
780aa0a0eb Fix corner case on assertion.
I need to move over to using the fixPacket function so I don't have to make the same changes everywhere.
Still a functional access bug someplace I need to track down in timing mode.

src/mem/cache/base_cache.cc:
src/mem/cache/cache_impl.hh:
    Fix corner case on assertion
tests/configs/memtest.py:
    Updated memtester with uncacheable addresses and functional accesses

--HG--
extra : convert_revision : e6fa851621700ff9227b83cc5cac20af4fc8444f
2006-10-19 21:26:46 -04:00
Ron Dreslinski
cc1feb9f6d Fix memtester to use functional access, fix cache to work functionally now that we could test it.
src/cpu/memtest/memtest.cc:
    Fix memtest to do functional accesses
src/mem/cache/cache_impl.hh:
    Fix cache to handle functional accesses properly based on memtester changes
    Still need to fix functional accesses in timing mode now that the memtester can test it.

--HG--
extra : convert_revision : a6dbca4dc23763ca13560fbf5d41a23ddf021113
2006-10-19 21:07:53 -04:00
Ron Dreslinski
210e73f2a2 Small changes:
?? doesn't compile in warn statements
Should have been false, where I had a true.

src/cpu/o3/lsq_impl.hh:
    Apparently you can't have ?? in a warn statement (Something about trigraphs)
src/mem/cache/cache_impl.hh:
    Forgot to signal atomic mode in snoopProbe

--HG--
extra : convert_revision : c75cb76e193e852284564993440c8ea39e6de426
2006-10-19 20:18:17 -04:00
Ron Dreslinski
e34e564f79 Fixes to get single level uni-coherence to work.
Now to try L2 caches in FS.

src/mem/cache/base_cache.hh:
    Fix uni-coherence for atomic accesses in coherence protocol access to port
src/mem/cache/cache_impl.hh:
    Properly handle uni-coherence
src/mem/cache/coherence/simple_coherence.hh:
    Properly forward invalidates (not done for MSI+ protocols (assumed top level for now)
src/mem/cache/coherence/uni_coherence.cc:
src/mem/cache/coherence/uni_coherence.hh:
    Properly forward invalidates in atomic/timing uni-coherence

--HG--
extra : convert_revision : f0f11315e8e7f32c19d92287f6f9c27b079c96f7
2006-10-19 20:02:57 -04:00
Ron Dreslinski
9cf063eb8e Merge zizzer:/bk/newmem
into  zazzer.eecs.umich.edu:/z/rdreslin/m5bk/newmemcleanest

--HG--
extra : convert_revision : c6611b32537918f5bf183788227ddf69a9a9a069
2006-10-19 19:00:43 -04:00