f4168a708c
Merge zizzer:/bk/linux into zower.eecs.umich.edu:/z/hsul/work/bk/linux
Lisa Hsu
2004-03-23 17:11:19 -05:00
75cef1a801
Add initial support for IDE
Andrew Schultz
2004-03-23 17:10:07 -05:00
b45fd9c541
Merge zizzer:/bk/linux into zower.eecs.umich.edu:/z/hsul/work/bk/linux
Lisa Hsu
2004-03-22 17:11:36 -05:00
e8f7aca8cf
Automerge
Andrew Schultz
2004-03-22 16:59:43 -05:00
c8640365c2
Change member variable names to lower case
Andrew Schultz
2004-03-22 16:58:18 -05:00
8baccdc6d3
Changed to inherit from DmaDevice to avoid multiple inheritance (now assumes that all PCI devices are DmaDevices, and if not, they can just ignore the DMA stuff)
Andrew Schultz
2004-03-22 16:50:09 -05:00
b18a36a6f2
Fix to compile m5.fast
Erik Hallnor
2004-03-22 13:59:51 -05:00
ddeaa4d5e1
Merge zizzer:/bk/linux into zower.eecs.umich.edu:/z/hsul/work/bk/linux
Lisa Hsu
2004-03-22 13:21:18 -05:00
3bc851b7db
forgot to check this in earlier - no confusion between types and member names!
Lisa Hsu
2004-03-22 13:15:14 -05:00
ea604923aa
Update to store uncompressed data in the cache blocks at all times. This will allow us to simulate compression without paying the uncompress penalty on reads. This also had the side-effect of getting copies and compression working together (collecting of code in helper function made this easier.)
Erik Hallnor
2004-03-21 01:20:19 -05:00
b4ff8c555f
Make it so that alpha_console can read the frequency from the correct sim object(depends on which platform tsunami/tlaser is being used)
Ron Dreslinski
2004-03-19 14:51:02 -05:00
1eb0c9626a
Updated kern/* code so we can have one console code and the simulator changes the HWRPB to the correct system type/variation.
Ali Saidi
2004-03-17 16:12:13 -05:00
2d38766e09
Pass cache block pointer when we have it.
Erik Hallnor
2004-03-15 21:24:50 -05:00
25a07e7467
merge etherpkt and makefile
Lisa Hsu
2004-03-12 11:07:32 -05:00
48e6ed48e0
first pass at ns_ethernet device. more will come later as i merge in nate's new ether infrastructure.
Lisa Hsu
2004-03-12 11:04:58 -05:00
256a1ac478
merge from head of m5. note: i did not bother to merge the PIO stuff with scsi since we're not going that route anymore anyway. esp because it was giving me an error i couldn't seem to fix.
Lisa Hsu
2004-03-11 20:04:41 -05:00
3bc8cffc75
merge with m5 head
Lisa Hsu
2004-03-11 18:52:29 -05:00
1039028d40
traceflags merge
Lisa Hsu
2004-03-11 18:47:29 -05:00
c6c61870fa
It was annoying me that there was PCIConfigAll and PciDev and PciConfigData. so for consistency I changed it.
Lisa Hsu
2004-03-11 12:45:27 -05:00
7f3a9e0f95
Use the specific tag type where we can.
Erik Hallnor
2004-03-11 00:54:54 -05:00
2c60d7aa9e
Performance improvement for unaligned copy
Erik Hallnor
2004-03-07 01:26:16 -05:00
532374bf2b
change the naming of MainBins. didn't check this in before because i had random debug statements i didn't want to push along with it.
Lisa Hsu
2004-03-05 19:18:58 -05:00
6eebb270be
fix ALPHA
Lisa Hsu
2004-03-05 15:15:23 -05:00
34576de15a
changes that affect post checkpoint runs.
Lisa Hsu
2004-03-05 08:16:33 -05:00
d9689c58d3
bin user and kernel stats
Lisa Hsu
2004-03-05 06:15:30 -05:00
12662c0b6d
nother fix
Lisa Hsu
2004-03-05 06:14:33 -05:00
8434ae9b03
Merge zizzer:/bk/m5 into zower.eecs.umich.edu:/z/hsul/work/bk/bincleanup
Lisa Hsu
2004-03-05 05:54:54 -05:00
9c48327145
forgot to check this in
Lisa Hsu
2004-03-05 05:54:46 -05:00
01f493a28a
Merge zizzer.eecs.umich.edu:/bk/m5 into ziff.eecs.umich.edu:/z/binkertn/research/m5/latest
Nathan Binkert
2004-03-05 05:50:33 -05:00
a57bc93e82
Don't dump an ethertrace by default Update the system dir for the addition of parallel and dirty netperf
Nathan Binkert
2004-03-05 05:50:19 -05:00
66325a2a30
Make payload splitting accessible with a #define
Nathan Binkert
2004-03-05 05:49:49 -05:00
adc176d13f
add a mechanism for delayed interrupts (aka. interrupt coalescing) re-do fixed dma latency
Nathan Binkert
2004-03-05 05:48:56 -05:00
c74cb77fa4
Add support for a propagation delay on the link. The electrical delay is more or less folded into the packet time, but an additional delay is possible representing crossing a long haul link, or some switches, etc.
Nathan Binkert
2004-03-05 05:47:05 -05:00
ec6265b044
constructor option to make the wrapped event autodelete
Nathan Binkert
2004-03-05 05:45:45 -05:00
34d36c45bf
use %#x so we know that it is hex
Nathan Binkert
2004-03-05 05:44:46 -05:00
4992680ea2
add ticks per picosecond
Nathan Binkert
2004-03-05 05:44:11 -05:00
1a300ef60d
small bugfix, forgot to increment iter
Lisa Hsu
2004-03-05 05:38:04 -05:00
4fa703f2ec
serialization for binning. it is WAAAAAAAY past my bedtime.
Lisa Hsu
2004-03-05 05:09:05 -05:00
2ef27db62c
Merge zizzer:/bk/m5 into zower.eecs.umich.edu:/z/hsul/work/bk/bincleanup
Lisa Hsu
2004-03-04 21:57:17 -05:00
f3861d0cc7
Overall gist of this is to 'dynamicize' the tracking of functions so you can choose them at the .ini stage rather than at compile time.
Lisa Hsu
2004-03-04 21:57:09 -05:00
1d09cd71a4
Reenable functioning copies.
Erik Hallnor
2004-03-04 20:03:38 -05:00
8ad803058f
Automerged
Erik Hallnor
2004-03-04 15:06:34 -05:00
7c089b2001
Copy implementations
Erik Hallnor
2004-03-04 14:57:57 -05:00
b91ea433b9
Script and ini file for new specweb images (the threaded ones)
Andrew Schultz
2004-03-02 22:02:33 -05:00
5537f02d49
Add some stats to track latencies of small and large DMA transactions.
Erik Hallnor
2004-03-02 18:45:08 -05:00
430a1700c6
Reduce redundant code, fix so that if the dma request is only as big as the header, and fix so that it doesn't keep requesting the payload bus if the sendRequest was unsuccesful.
Ron Dreslinski
2004-03-02 15:13:01 -05:00
b05d7f99ac
Make sure to arbitrate for the payload bus in the header splitting case, and to free up the header bus by returning false. If there is no header splitting it will work normally.
Ron Dreslinski
2004-03-02 12:59:24 -05:00
594d7d028d
Add more statistics about dma requests, shorten names for easier readability of stats
Ron Dreslinski
2004-03-02 02:49:19 -05:00
c92a09e213
Add a name parameter so the statistics work properly
Ron Dreslinski
2004-03-02 02:21:15 -05:00
9cd698000c
Add statistics to calculate average round-trip delay for DMA
Ron Dreslinski
2004-03-02 01:04:12 -05:00
9be01eec9a
Change ini again because I messed up the resolve
Ron Dreslinski
2004-03-01 21:38:33 -05:00
e9b985f91a
Auto merged
Ron Dreslinski
2004-03-01 21:26:21 -05:00
3a0db71f02
Add support for two buses to the DMA, allows header splitting The two busses are labeled header/payload. The header_size designates the number of cache blocks to send on the header bus.
Ron Dreslinski
2004-03-01 21:17:28 -05:00
eabc59ca2d
Initial cut at adding support to make the state machine take time
Nathan Binkert
2004-03-01 16:22:51 -05:00
18d90037cb
Use #define CPU_10GHZ and CPU_4GHZ. This should get fixed when we switch over to making ticks represent time directly.
Nathan Binkert
2004-03-01 16:22:09 -05:00
ad2f3f3a63
Support several memory configurations for hooking up devices to memory in different configurations
Nathan Binkert
2004-03-01 16:19:12 -05:00
7a9cf40bbf
add support for averaging
Nathan Binkert
2004-03-01 16:18:02 -05:00
2aed8753dd
Only need an execution context if we're doing data
Nathan Binkert
2004-03-01 16:17:18 -05:00
e943bedeaa
Merge ehallnor@zizzer:/bk/m5 into zazzer.eecs.umich.edu:/z/ehallnor/m5
Erik Hallnor
2004-02-29 23:13:27 -05:00
19c25e7c21
Fix an assert, only need the ExecContext if we are doing data,
Erik Hallnor
2004-02-29 23:13:04 -05:00
123513ab0d
Merge zizzer.eecs.umich.edu:/bk/m5 into ziff.eecs.umich.edu:/z/binkertn/research/m5/latest
Nathan Binkert
2004-02-29 22:56:59 -05:00
31ccbde829
Fix the swichover code. It's only for FULL_SYSTEM
Nathan Binkert
2004-02-29 22:56:42 -05:00
cbc42f1d71
Remove copys from isa_desc, and implement a store and forward bus bridge
Erik Hallnor
2004-02-29 22:41:11 -05:00
7f688ba6a9
unused function name
Nathan Binkert
2004-02-29 21:35:12 -05:00
133375d5b1
make it so that we can prevent descriptor DMAs or data DMAs from touching the memory system. They can still have a latency though if configured.
Nathan Binkert
2004-02-29 21:34:25 -05:00
ff516e8362
mark the request satisfied since well, it is.
Nathan Binkert
2004-02-29 20:38:25 -05:00
27b926c431
Avoid more bogus addresses
Nathan Binkert
2004-02-29 20:35:55 -05:00
0a3948bcfd
Merge zizzer.eecs.umich.edu:/m5/Bitkeeper/m5 into zizzer.eecs.umich.edu:/.automount/ziff/z/binkertn/research/m5/latest
Nathan Binkert
2004-02-29 20:25:52 -05:00
ee96799519
Initial cleanup pass of lisa's function call tracking code.
Nathan Binkert
2004-02-29 20:22:32 -05:00
a574c260e3
Merge zizzer.eecs.umich.edu:/bk/m5 into ziff.eecs.umich.edu:/z/binkertn/research/m5/latest
Nathan Binkert
2004-02-29 19:31:58 -05:00
45f377bcd8
more debugging support
Nathan Binkert
2004-02-29 19:29:30 -05:00
31f82cef41
Make the progress event work even after restoring from a checkpoint
Nathan Binkert
2004-02-29 18:49:44 -05:00
2272164d66
Fix dumping to work on the opteron. struct timeval can vary in size, so we're explicit about the fields.
Nathan Binkert
2004-02-29 17:05:23 -05:00
b491bda241
Add support for multiple address ranges in memory interfaces.
Erik Hallnor
2004-02-29 16:18:49 -05:00
2399fb1196
The DMAInterface was never updated to the new blocking model. Need to hold the request locally until it is retransmitted.
Erik Hallnor
2004-02-26 20:36:29 -05:00
006fb9b421
Quick hack to allow rerequests for the future.
Erik Hallnor
2004-02-26 16:59:04 -05:00
40c350fcfe
Merge.
Steve Reinhardt
2004-02-26 07:08:45 -08:00
6f5e104fc5
Make SW prefetch flag a parameter again, and add code to make it actually do something on FullCPU. Still disabled, as it causes detailed-boot to hang when you turn it on.
Steve Reinhardt
2004-02-26 07:05:36 -08:00
f8ad539aa5
Save the status of faulting write-hints so that they won't accidentally be issued to the memory system.
Nathan Binkert
2004-02-26 01:18:13 -05:00
f93e9e063a
Make this compile under non full system.
Erik Hallnor
2004-02-25 21:32:37 -05:00
752b6cf7c6
delay execContext registration from create() to init() callback in the sampling cpu also.
Lisa Hsu
2004-02-25 18:25:20 -05:00
a253251ae0
prevent bogus addresses from being sent to the memory system
Nathan Binkert
2004-02-25 16:16:17 -05:00
087a2ed6a1
Merge zizzer.eecs.umich.edu:/bk/m5 into ziff.eecs.umich.edu:/z/binkertn/research/m5/latest
Nathan Binkert
2004-02-25 16:13:00 -05:00
b678152277
Fix stats debugging and always compile it in for the debug target
Nathan Binkert
2004-02-25 16:12:48 -05:00
6a306d4caf
add in an init() callback for CPU's so that no stats are accessed prior to the end of the build process. (Done by doing the registerExecContext() calling sequence in the init() process rather than the create() process).
Lisa Hsu
2004-02-24 14:59:25 -05:00
f9c91277f3
Quote args properly in tracediff.
Steve Reinhardt
2004-02-23 08:07:32 -08:00
6d574e9cff
Only one request may be on the dma interface at a time, so add some states to indicate that the dma read/write hasn't issued yet, and add some support to reissue the request when the interface becomes unblocked.
Nathan Binkert
2004-02-23 02:49:09 -05:00
cc863129b8
extend the hack by turning the address into a real physical address
Nathan Binkert
2004-02-23 02:46:47 -05:00
Nathan Binkert