summaryrefslogtreecommitdiff
path: root/src/mem/bus.cc
AgeCommit message (Collapse)Author
2007-07-14Fix & tweak DPRINTFs for tracediff w/new cache code.Steve Reinhardt
Note that we should *not* print pointer values in DPRINTFs as these needlessly clutter tracediff output. --HG-- extra : convert_revision : 25a448f1b3ac8d453a717a104ad6dc0112fb30bb
2007-06-20Minor error.Vincentius Robby
--HG-- extra : convert_revision : 514032e21c8861f20fcbcae7204e132088cc7dbc
2007-06-20Removed "adding instead of dividing" trick.Vincentius Robby
Caused slowdown in performance instead of speeding up. src/cpu/base.cc: Removed "adding instead of dividing" trick. src/mem/bus.cc: Fixed spelling in comments. Removed "adding instead of dividing" trick. --HG-- extra : convert_revision : 65a736f4f09a64e737dc7aeee53b117976330488
2007-06-09Add a startup function that will fast forward to the right clock edgeNathan Binkert
using a divide in order to not loop forever after resuming from a checkpoint --HG-- extra : convert_revision : 4bbc70b1be4e5c4ed99d4f88418ab620d5ce475a
2007-05-21Change getDeviceAddressRanges to use bool for snoop arg.Steve Reinhardt
--HG-- extra : convert_revision : 832e52ba80cbab2f5bb6d5b5977a499d41b4d638
2007-05-09undo my previous bus change, it can make the bus deadlock.. so it still ↵Ali Saidi
constantly reschedules itself --HG-- extra : convert_revision : b5ef1aa0a6a2e32bd775d2dbcad9cd9505ad9b78
2007-05-07fix partial writes with a functional memory hackAli Saidi
figure out the block size from devices attached to the bus otherwise use a default block size when no devices that care are attached configs/common/FSConfig.py: src/mem/bridge.cc: src/mem/bridge.hh: src/python/m5/objects/Bridge.py: fix partial writes with a functional memory hack src/mem/bus.cc: src/mem/bus.hh: src/python/m5/objects/Bus.py: figure out the block size from devices attached to the bus otherwise use a default block size when no devices that care are attached src/mem/packet.cc: fix WriteInvalidateResp to not be a request that needs a response since it isn't src/mem/port.hh: by default return 0 for deviceBlockSize instead of panicing. This makes finding the block size the bus should use easier --HG-- extra : convert_revision : 3fcfe95f9f392ef76f324ee8bd1d7f6de95c1a64
2007-04-04The MemoryObject tha owns a port should delete it if it so chooses when ↵Ali Saidi
deletePortRefs() is called on it with that port as a parameter. In this way a MemoryObject can keep a functional port around and give it to anyone who wants to do functional accesses rather than creating a new one each time. src/mem/bus.cc: src/mem/bus.hh: src/mem/cache/cache_impl.hh: only keep around one func port we give to anyone who wants it. Otherwise we can run out of port ids reasonably quickly if a lot of functional accesses are happening (e.g. remote debugging, dprintk, etc) --HG-- extra : convert_revision : 6a9e3e96f51cedaab6de1b36cf317203899a3716
2007-03-23Merge ktlim@zizzer:/bk/newmemKevin Lim
into zamp.eecs.umich.edu:/z/ktlim2/clean/tmp/clean2 src/cpu/base_dyn_inst.hh: Hand merge. Line is no longer needed because it's handled in the ISA. --HG-- extra : convert_revision : 0be4067aa38759a5631c6940f0167d48fde2b680
2007-03-233 memory system fixes:Kevin Lim
1. Update packet's flags properly when a snoop happens 2. Don't allow accesses to read a block's data if the block has outstanding MSHRs. This avoids a RAW hazard in MP systems that the memory system was not detecting properly earlier (a write required a block to upgrade, and while the upgrade was outstanding, a read came along and read old data). 3. Update MSHR's request upon a response being handled. If the MSHR has more targets than it can respond to in one cycle, then its request must be properly updated to the new head of the targets list. src/mem/bus.cc: Update packet's flags properly upon snoop. src/mem/cache/cache_impl.hh: Be sure to not allow accesses to a block with outstanding MSHRs. src/mem/cache/miss/miss_queue.cc: Update MSHR's request upon a response being handled. --HG-- extra : convert_revision : 76a9abc610ca3f1904f075ad21637148a41982d6
2007-03-08stop m5 from leaking like a sieveAli Saidi
don't create a new physPort/virtPort every time activateContext() is called add the ability to tell a memory object to delete it's reference to a port and a method to have a port call deletePortRefs() on the port owner as well as delete it's peer still need to stop calling connectMemoPorts() every time activateContext() is called or we'll overflow the bus id and panic src/cpu/thread_state.cc: if we hav ea (phys|virt)Port don't create a new on, have it delete it's peer and then reuse it src/mem/bus.cc: src/mem/bus.hh: add ability to delete a port by usig a hash_map instead of an array to store port ids add a function to do deleting src/mem/cache/cache.hh: src/mem/cache/cache_impl.hh: src/mem/mem_object.cc: src/mem/mem_object.hh: adda function to delete port references from a memory object src/mem/port.cc: src/mem/port.hh: add a removeConn function that tell the owener to delete any references to the port and then deletes its peer --HG-- extra : convert_revision : 272f0c8f80e1cf1ab1750d8be5a6c9aa110b06a4
2007-02-06More DPRINTF cleanup.Steve Reinhardt
--HG-- extra : convert_revision : db89cea42b46476d19333038522a6c144eafdab1
2007-02-06Minor DPRINTF fixes.Steve Reinhardt
--HG-- extra : convert_revision : 41956c9a480163ecac7807982215027e8ff1a4a9
2006-12-15Use my range_map to speed up findPort() in the bus. The snoop code could ↵Ali Saidi
still use some work. --HG-- extra : convert_revision : ba0a68bd378d68e4ebd80a101b965d36c8be1db9
2006-11-22Do a functional access to levels above on a read as a temporary solution for ↵Ron Dreslinski
L2's in FS Fix a small writeback bug when missing in the L2 in atomic mode src/mem/bus.cc: Fix a comment to make sense src/mem/cache/cache_impl.hh: Do a functional access to levels above on a read as a temporary solution for L2's in FS Also fix a small writeback miss in L2 issue src/mem/cache/coherence/simple_coherence.hh: src/mem/cache/coherence/uni_coherence.cc: src/mem/cache/coherence/uni_coherence.hh: Do a functional access to levels above on a read as a temporary solution for L2's in FS tests/quick/00.hello/ref/alpha/linux/o3-timing/m5stats.txt: tests/quick/00.hello/ref/alpha/linux/simple-timing/m5stats.txt: tests/quick/01.hello-2T-smt/ref/alpha/linux/o3-timing/m5stats.txt: Update ref's for writeback changes --HG-- extra : convert_revision : 937febd577b16b7fd97a5a68acaf53541828a251
2006-11-14Fix bugs around uni-coherence invalidates being propogated properly.Ron Dreslinski
src/mem/bus.cc: Make it so that invalidates being sent from the responder up don't call the responder but they should also not Panic. src/mem/packet.hh: If we don't have data in the packet, don't call deleteData: Example: InvalidateRequests never have data. --HG-- extra : convert_revision : 18766bc9f3bb4d852ac651d094254d347abd1634
2006-11-14Update atomic and functional paths for snoops as wellRon Dreslinski
--HG-- extra : convert_revision : 566d73438efb87ca683e4dee23454d880db3dfc7
2006-11-14Update bus bridges now that snoop ranges are passed properlyRon Dreslinski
src/mem/bridge.cc: Update brdiges, now that snoop addresses are properly forwarded. Bus bridge should only handle snoops on the second phase (SNOOP_COMMIT) src/mem/bus.cc: src/mem/bus.hh: Make sure if a busBridge has access to both things that snoop and things that respond it only takes the request once --HG-- extra : convert_revision : 26cc9ee4429be45d4476fa435e0e9a54843c2509
2006-11-13Fix some errors related to snooping and functional access in the busRon Dreslinski
src/mem/bus.cc: Only call snoop once per port, need to fix it so snoop ranges that overlap aren't added to list Functional accesses that call snoop and it goes to a higher bus may change the src, reset it after each snoop. --HG-- extra : convert_revision : 7276059c798a85cb9d138ccc5531298ecd055c13
2006-11-13Fix problems with snoop ranges not working properly on functional accessesRon Dreslinski
src/mem/bus.cc: Actually return the snoop list when asked for it. Don't get stuck in infinite functional loops --HG-- extra : convert_revision : 8e6dafbd10b30d48d28b6b5d4b464e8e8f6a3ddc
2006-11-09Be sure to populate the packet's finishTime field in the atomic timing case.Kevin Lim
--HG-- extra : convert_revision : ef34818eb2dea5b3a8e754bf56745a7cd2497bf0
2006-11-09Draining fixes.Kevin Lim
src/cpu/o3/cpu.cc: Handle draining properly when CPU isn't actually being used. src/cpu/simple/atomic.cc: Be sure to set status properly when draining. src/mem/bus.cc: Fix for draining. --HG-- extra : convert_revision : d9796e6693e974f022159029fc9743c49a970c8f
2006-11-07Fix up bus draining and add draining to the caches.Kevin Lim
src/mem/bus.cc: Fix up draining to work properly. src/mem/bus.hh: Initialize drainEvent to NULL. src/mem/cache/base_cache.cc: src/mem/cache/base_cache.hh: Add draining to the caches. --HG-- extra : convert_revision : 3082220a75d50876f10909f9f99bec535889f818
2006-11-02Merge ktlim@zizzer:/bk/newmemKevin Lim
into zamp.eecs.umich.edu:/z/ktlim2/clean/newmem-busfix --HG-- extra : convert_revision : a9a41e2c292bd95aa148e1cf4d9a77c0622a462b
2006-11-02Have bus use the BadAddress device to handle bad addresses. The O3 CPU ↵Kevin Lim
should be able to boot into Linux with caches on after this change. src/mem/bus.cc: src/mem/bus.hh: Bus now will be setup with a default responder, unless the user overrides it. This default responder should return BadAddress if no matching port is found. src/python/m5/objects/Bus.py: Bus now has a default responder for FS mode if the user doesn't override it. It returns BadAddress if no matching port is found. src/python/m5/objects/Tsunami.py: Add bad address device. Also record when the user has specified their own default responder. --HG-- extra : convert_revision : 59070477ae313ee711b2d59baa2369c9a91c5b85
2006-11-01Added code to handle draining.Gabe Black
--HG-- extra : convert_revision : 3861f553bde5865cd21a8a58a4c410896726f0a3
2006-10-20Use PacketPtr everywhereNathan Binkert
--HG-- extra : convert_revision : d9eb83ab77ffd2d725961f295b1733137e187711
2006-10-19Fix compile of m5.fastNathan Binkert
--HG-- extra : convert_revision : a8a37c318e55e48e697e4aaba339328f000b3f60
2006-10-12small bus updates for functional accessesAli Saidi
--HG-- extra : convert_revision : c7a6b199c74ed4b4ffab14bbffb51e72d75b7742
2006-10-11Make default ID unique (not broadcast)Ron Dreslinski
Fix a segfault associated with DefaultId src/mem/bus.cc: Handle a segfault in the bus when DefaultPort was being used src/mem/bus.hh: Make the Default ID more unique (it overlapped with Broadcast ID) --HG-- extra : convert_revision : 9182805c5cf4d9fe004e6c5be8547a8f41ed7bfe
2006-10-11Fix bus in FS mode.Ron Dreslinski
src/mem/bus.cc: Add debugging statement src/mem/bus.hh: Fix implementation of bus for subsequent recvTimings while handling a retry request. src/mem/tport.cc: Rework timing port to retry properly --HG-- extra : convert_revision : fbfb5e8b4a625e49c6cd764da1df46a4f336b1b2
2006-10-11More cache fixes. Atomic coherence now works as well.Ron Dreslinski
src/cpu/memtest/memtest.cc: src/cpu/memtest/memtest.hh: Make Memtester able to test atomic as well src/mem/bus.cc: src/mem/bus.hh: Handle atomic snoops properly for cache->cache transfers src/mem/cache/cache_impl.hh: Debug output. Clean up memleak in atomic mode. Set hitLatency. Still need to send back reasonable number for atomic return value. src/mem/packet.cc: Add command strings for new commands src/python/m5/objects/MemTest.py: Add param to test atomic memory. --HG-- extra : convert_revision : 43f880e29215776167c16ea90793ebf8122c785b
2006-10-11Don't call recvRetry if the bus is busy anyway. This takes care of a corner ↵Gabe Black
case as well when dealing with grants that aren't used. --HG-- extra : convert_revision : 38f7ef1b41477fb2a2438387ef3a81cccd3e7a8a
2006-10-11Make the bus work if the other sides recvRetry doesn't call sendTiming for ↵Gabe Black
some reason. --HG-- extra : convert_revision : e722ddb0354a5c021dc7c44a3e2f0a64e962442b
2006-10-10Merge zizzer:/n/wexford/x/gblack/m5/newmem_busRon Dreslinski
into zazzer.eecs.umich.edu:/z/rdreslin/m5bk/newmemcleanest src/mem/bus.cc: SCCS merged --HG-- extra : convert_revision : 18608114350c466a56ab499ae523b01fcb2f6ef2
2006-10-10Make the bus is occupied for none broadcast packets as well.Gabe Black
--HG-- extra : convert_revision : aef3c625172e92be8f29c4c57077fefee43046bb
2006-10-10Merge zizzer:/n/wexford/x/gblack/m5/newmem_busRon Dreslinski
into zazzer.eecs.umich.edu:/z/rdreslin/m5bk/newmemcleanest src/mem/bus.cc: SCCS merged --HG-- extra : convert_revision : eaae105025635c37af06cf72bb061ce82def9dc9
2006-10-10Debugging infoRon Dreslinski
src/base/traceflags.py: Add new flags for cacheport src/mem/bus.cc: Add debugging info src/mem/cache/base_cache.cc: Add debuggin info --HG-- extra : convert_revision : a6c4b452466a8e0b50a86e886833cb6e29edc748
2006-10-10Put in an accounting mechanism and an assert to make sure something doesn't ↵Gabe Black
try to send another packet while it's still waiting for the bus. --HG-- extra : convert_revision : 4a2b83111e49f71ca27e05c98b55bc3bac8d9f53
2006-10-10Changed the bus to use a bool to keep track of retries rather than a pointerGabe Black
src/mem/tport.cc: minor formatting tweak --HG-- extra : convert_revision : 7391d142815c5876fcc0f991bd053e6a1781c101
2006-10-09Fixes to the bus, and added fields to the packet.Gabe Black
src/mem/bus.cc: Put back the check to see if the bus is busy. Also, populate the fields in the packet to indicate when the first word and the entire packet will be delivered. src/mem/bus.hh: Remove the occupyBus function. src/mem/packet.hh: Added fields to the packet to indicate when the first chunk of a packet arrives, and when the entire packet arrives. --HG-- extra : convert_revision : cfc7670a33913d48a04d02c6d2448290a51f2d3c
2006-10-09Merge zizzer.eecs.umich.edu:/bk/newmemGabe Black
into zeep.eecs.umich.edu:/home/gblack/m5/newmem_bus --HG-- extra : convert_revision : 8267487b935eaf11665841ace3a5c664751b53b0
2006-10-09Potentially functional partially timed bandwidth limitted bus model.Gabe Black
src/mem/bus.cc: Fixes to the previous hand merging, and put the snooping back into recvTiming and out of it's own function. src/mem/bus.hh: Put snooping back into recvTiming and not in it's own function. --HG-- extra : convert_revision : fd031b7e6051a5be07ed6926454fde73b1739dc6
2006-10-08Fixes for functional path.Ron Dreslinski
If the cpu needs to update any state when it gets a functional write (LSQ??) then that code needs to be written. src/cpu/o3/fetch_impl.hh: src/cpu/o3/lsq_impl.hh: src/cpu/ozone/front_end_impl.hh: src/cpu/ozone/lw_lsq_impl.hh: src/cpu/simple/atomic.cc: src/cpu/simple/timing.cc: CPU's can recieve functional accesses, they need to determine if they need to do anything with them. src/mem/bus.cc: src/mem/bus.hh: Make the fuctional path do the correct tye of snoop --HG-- extra : convert_revision : 70d09f954b907a8aa9b8137579cd2b06e02ae2ff
2006-10-08Merge zizzer.eecs.umich.edu:/bk/newmemGabe Black
into zeep.eecs.umich.edu:/home/gblack/m5/newmem_bus src/mem/bus.cc: Hand merged. Needs to be fixed --HG-- extra : convert_revision : df03219ccfd18431cc726a063bd29d30554944a1
2006-10-08Move away from using the statusChange function on snoops. Clean up snooping ↵Ron Dreslinski
code in general. --HG-- extra : convert_revision : 5a57bfd7742a212047fc32e8cae0dc602fdc915c
2006-10-08bus changesGabe Black
src/mem/bus.cc: src/mem/bus.hh: minor fix and some formatting changes src/python/m5/objects/Bus.py: changed bits to bytes --HG-- extra : convert_revision : dcd22205604b7a2727eaf2094084c4858f3589c5
2006-10-08Merge zizzer.eecs.umich.edu:/bk/newmemGabe Black
into zeep.eecs.umich.edu:/home/gblack/m5/newmem_bus --HG-- extra : convert_revision : 1749397443ccb320d32f8dd23c71ed0431d30cb7
2006-10-08A possible implementation of a multiplexed bus.Gabe Black
--HG-- extra : convert_revision : 3c560eda12ffd8ca539c91024baf2770b963ede8
2006-10-05First pass at snooping stuff that compiles and doesn't break.Ron Dreslinski
Still need: -Handle NACK's on the recieve side -Distinguish top level caches -Handle repsonses from caches failing the fast path -Handle BusError and propogate it -Fix the invalidate packet associated with snooping in the cache src/mem/bus.cc: Make sure to snoop on functional accesses src/mem/cache/base_cache.cc: Wait to make a request into a response until it is ready to be issued src/mem/cache/base_cache.hh: Support range changes for snoops Set up snoop responses for cache->cache transfers src/mem/cache/cache_impl.hh: Only access the cache if it wasn't satisfied by cache->cache transfer Handle snoop phases (detect block, then snoop) Fix functional access to work properly (still need to fix snoop path for functional accesses) --HG-- extra : convert_revision : 4c25f11d7a996c1f56f4f7b55dde87a344e5fdf8
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-17 23:35:57 +0000