c82113d022
Load local kernel symbols, and set aic7xxx_no_reset to prevent an initial (very painful) bus reset from occuring
Andrew Schultz
2004-02-21 20:31:08 -05:00
43787ad863
Added initial (hackish) support for translating between a PCI bus address and a physical memory address for DMA
Andrew Schultz
2004-02-21 20:29:38 -05:00
7f069e7bdb
Merge zizzer:/bk/linux into zizzer.eecs.umich.edu:/z/alschult/DiskModel/m5
Andrew Schultz
2004-02-21 14:30:06 -05:00
6ff8c35e02
Rewrite of AdaptecController
Andrew Schultz
2004-02-21 14:29:59 -05:00
e067bc751b
Change order of serialization
Nathan Binkert
2004-02-21 10:46:31 -05:00
610081079c
New Ethernet device. Descriptors are now done via DMA instead of faking it. Renamed stuff to follow our style. Lots of general cleanup.
Nathan Binkert
2004-02-21 10:43:37 -05:00
4362b42cce
Create a mp test for linux-boot, make the other ini files more modular for other mp testbenches
Ron Dreslinski
2004-02-20 16:53:54 -05:00
4768c72964
Fix the RTC code so it is in the cchip, only interrupt processors that are present
Ron Dreslinski
2004-02-20 16:51:19 -05:00
a1259a4fcf
Add a simple event wrapper class that takes a class pointer and member function and will schedule it for the future.
Nathan Binkert
2004-02-20 15:24:21 -05:00
ec06c63cc7
make the dma interface useable. make it so that pio devices must respond with some delay.
Nathan Binkert
2004-02-20 15:23:27 -05:00
695d51e513
make uncacheable stuff happen again
Nathan Binkert
2004-02-20 15:22:41 -05:00
38e8cb8df1
make the devices default to being attached to the memory bus but provide a #define that can be used to put them on a separate IO bus
Nathan Binkert
2004-02-20 15:22:02 -05:00
aa483a0026
Don't need to specify two etherdump devices
Nathan Binkert
2004-02-20 15:20:37 -05:00
a51cec6669
Make it so dump takes a void *
Nathan Binkert
2004-02-20 15:19:55 -05:00
ee8c8cbcd8
make etherdump work again
Nathan Binkert
2004-02-20 15:19:19 -05:00
7ad5b6ee22
Merge zizzer.eecs.umich.edu:/bk/m5 into ziff.eecs.umich.edu:/z/binkertn/research/m5/memory
Nathan Binkert
2004-02-20 15:08:02 -05:00
db940dd0b0
Add support for IPI's and extend RTC to interrupt all Processors
Ron Dreslinski
2004-02-20 14:28:59 -05:00
ee799f47d5
Merge zizzer:/bk/m5 into zamp.eecs.umich.edu:/z/rdreslin/m5bk/clean
Ron Dreslinski
2004-02-20 12:45:11 -05:00
373d70980e
Add serialization for packets on the ethernet link, and for link events
Ron Dreslinski
2004-02-20 12:44:41 -05:00
5e494f0c59
put back $SETUPDIR = $ENV{PWD} for amd purposes.
Lisa Hsu
2004-02-20 12:37:52 -05:00
cb928f0391
Merge zizzer.eecs.umich.edu:/bk/m5 into ziff.eecs.umich.edu:/z/binkertn/research/m5/memory
Nathan Binkert
2004-02-20 11:29:39 -05:00
4ec55bef11
interestingly, when setup is called from a perl script, when you chdir in the perl script, the ENV{PWD} does not change.
Lisa Hsu
2004-02-19 17:58:17 -05:00
d4637757f8
Remote an old hack that is now unnecessary
Andrew Schultz
2004-02-19 16:32:13 -05:00
e3fb3d1ad0
Misspeculation fix and (more importantly) change to allow for the ev6 style physical addressing. This has the uncacheable bit as bit 40 as opposed to bit 39. Additionally, we now support (at least superficially) a 44-bit physical address. To deal with superpage access in this scheme, any super page access with either bit 39 or 40 set is sign extended.
Andrew Schultz
2004-02-19 16:30:06 -05:00
12747d3084
Change the physical memory logic, and also add misspeculation fix to tlb index calls that are called from ExecContext::readIpr
Andrew Schultz
2004-02-18 21:38:55 -05:00
26d0cf70ea
Merge zizzer.eecs.umich.edu:/bk/m5 into ziff.eecs.umich.edu:/z/binkertn/research/m5/memory
Nathan Binkert
2004-02-18 14:17:44 -05:00
6610699987
Add COW support to the IIC.
Erik Hallnor
2004-02-17 22:53:15 -05:00
4bb902328a
Merge ehallnor@zizzer:/bk/m5 into zazzer.eecs.umich.edu:/z/ehallnor/m5
Erik Hallnor
2004-02-17 03:14:21 -05:00
5c77989c9b
Change the way data is handled in the cache blocks.
Erik Hallnor
2004-02-17 03:05:19 -05:00
59a6e9d705
Merge
Andrew Schultz
2004-02-16 22:04:55 -05:00
a73e263f94
Change logic for translating a memory addresses, extra checks for invalid physical addresses.
Andrew Schultz
2004-02-16 21:56:38 -05:00
772f95975d
Changed physical device addresses to fit within the 40-bit implemented physical address
Andrew Schultz
2004-02-16 21:20:18 -05:00
b797f99213
Rework BusBridge to have a latency and handle bus IDs correctly.
Erik Hallnor
2004-02-16 16:28:11 -05:00
706d6bea52
1 << x != (uint64_t)1 << x
Ali Saidi
2004-02-16 16:04:34 -05:00
9c9d1cb5bf
Changed device IRQ back to 1
Andrew Schultz
2004-02-16 01:33:43 -05:00
484a4de65a
Added new aic register definition, fix some interrupt related calls
Andrew Schultz
2004-02-16 01:07:16 -05:00
b82bd8fd78
changed interrupt index to be 64 bits long and fixed a bad include in adaptec_ctrl.hh
Ali Saidi
2004-02-16 00:21:42 -05:00
03e8793f89
Merge
Ali Saidi
2004-02-16 00:00:23 -05:00
de5fa66714
Rewrote interrupt code to handle masking correctly and changed every interrupt to use a different subnumber since both devices could interrupt at the same time and we don't want to loose one.
Ali Saidi
2004-02-15 23:56:44 -05:00
44b277a62f
Interrupt changes
Andrew Schultz
2004-02-15 22:36:23 -05:00
1978851fc7
Fix some debug printing errors
Andrew Schultz
2004-02-15 18:24:14 -05:00
55a7ad9977
Scsi controller operation changes
Andrew Schultz
2004-02-15 18:23:42 -05:00
9984412671
Fix bug with physical address translation
Andrew Schultz
2004-02-15 16:33:16 -05:00
7fd91f7c4c
Add full copy support.
Erik Hallnor
2004-02-14 03:25:39 -05:00
092ec1978d
Fix to remote debugger while in PAL
Andrew Schultz
2004-02-13 18:15:04 -05:00
6391bfac54
Merge zizzer.eecs.umich.edu:/m5/Bitkeeper/m5 into zizzer.eecs.umich.edu:/z/alschult/linux
Andrew Schultz
2004-02-13 16:42:42 -05:00
8c89212fa3
Merge zizzer.eecs.umich.edu:/m5/Bitkeeper/m5 into zizzer.eecs.umich.edu:/.automount/ziff/z/binkertn/research/m5/memory
Nathan Binkert
2004-02-13 15:44:46 -05:00
8232c9743d
fix up vtophys a bit
Nathan Binkert
2004-02-13 15:36:21 -05:00
cc53465cd5
Add support for all devices to get requests from a timing memory bus. In the future, this can be used for actual data, but for now, it's so that devices can respond to timing accesses properly. This way, an uncached access on a bus further away will take longer to respond.
Nathan Binkert
2004-02-11 16:07:55 -05:00
29eae76153
Merge
Ali Saidi
2004-02-11 15:36:36 -05:00
0f34a00703
added serializeation code
Ali Saidi
2004-02-11 15:32:30 -05:00
5e82f8d84c
Merge zizzer:/bk/m5 into isabel.reinhardt.house:/z/stever/bk/m5
Steve Reinhardt
2004-02-10 21:44:08 -08:00
d38f995aee
Fixes for Linux syscall emulation.
Steve Reinhardt
2004-02-10 21:43:57 -08:00
e72e8b28c8
Fix masking of read/write address to get read/write offset Changed base_linux.ini file to use physical addresses
Andrew Schultz
2004-02-10 22:35:18 -05:00
f7c4d57260
Removed mask from devices
Andrew Schultz
2004-02-10 20:45:48 -05:00
14c11155be
Add support for multiple outstanding aligned copies
Erik Hallnor
2004-02-10 20:37:38 -05:00
81d5ffe7de
Changed new linux stuff to work with new FunctionalMemory interface and some sundry problems with new interface
Andrew Schultz
2004-02-10 00:19:43 -05:00
eac2d6a668
Merge linux tree with head
Andrew Schultz
2004-02-09 17:50:47 -05:00
4478bcb7c7
Merge ehallnor@zizzer:/bk/m5 into zazzer.eecs.umich.edu:/z/ehallnor/m5
Erik Hallnor
2004-02-09 17:38:41 -05:00
f0abedf769
Add LRU aligned copies to the hierarchy, with only one outstanding copy. Aligned copies now fully work in LRU (just need to write the IIC doCopy call). At the moment they are slow since a stalled copy stalls the entire cache.
Erik Hallnor
2004-02-09 17:37:27 -05:00
48bb27be72
Added platform.hh
Andrew Schultz
2004-02-09 14:52:41 -05:00
581a6ea0dd
Fixes for updating mmu mappings
Andrew Schultz
2004-02-09 14:47:39 -05:00
b93cab4e85
Merge zizzer.eecs.umich.edu:/m5/Bitkeeper/linux into zizzer.eecs.umich.edu:/y/saidi/work/m5
Ali Saidi
2004-02-09 13:46:38 -05:00
1ded394fc3
Some changes to for linux 2.6.2
Ali Saidi
2004-02-09 13:40:58 -05:00
da8a702212
- Whack unused code - Make the MemoryController use address ranges (via Range) instead of an address and a mask
Nathan Binkert
2004-02-09 10:49:48 -05:00
087334bfe5
Merge zizzer.eecs.umich.edu:/m5/Bitkeeper/m5 into zizzer.eecs.umich.edu:/.automount/ziff/z/binkertn/research/m5/memory
Nathan Binkert
2004-02-09 10:36:32 -05:00
411d5497fa
Add that one IPR memory space address that we keep seeing
Nathan Binkert
2004-02-09 09:06:20 -05:00
6ece3a7e59
random fixes
Nathan Binkert
2004-02-09 09:04:52 -05:00
f1aca5e6c5
Commit regenerated trace_flags.* based on merged traceflags.pl. (bk auto merge messed these up...)
Steve Reinhardt
2004-02-09 00:44:56 -08:00
7f04b6ce59
bk resolve is making me commit these even though they're broken.
Steve Reinhardt
2004-02-09 00:32:21 -08:00
730296f7f9
Results of automatic (yet incomplete) merge.
Steve Reinhardt
2004-02-09 00:30:16 -08:00
d7b7363444
Add support for memory barriers.
Steve Reinhardt
2004-02-09 00:22:43 -08:00
88c5bd695f
Merge ehallnor@zizzer:/bk/m5 into zazzer.eecs.umich.edu:/z/ehallnor/m5
Erik Hallnor
2004-02-08 17:27:08 -05:00
9c7bec74b1
Separate template definitions and instantiations into *_impl.hh and *.cc. Eventually the *.cc fils should be generate by our super-duper python script.
Erik Hallnor
2004-02-08 17:26:27 -05:00
de285ff2a0
Totally rework the Range class. Now the range is from [start, end) to be consistent with the way that the stl works. It also makes lots of other stuff easier. (Maybe those guys were smart?) Overload the various comparison operators so that you can test for overlapping of ranges, etc.
Nathan Binkert
2004-02-07 01:16:50 -05:00
22f78a6d03
Merge zizzer.eecs.umich.edu:/bk/m5 into ziff.eecs.umich.edu:/z/binkertn/research/m5/kernel
Nathan Binkert
2004-02-06 19:00:12 -05:00
8a89f8c807
Small cleanup
Nathan Binkert
2004-02-06 18:59:12 -05:00
bdfb5de4b2
Merge ehallnor@zizzer:/bk/m5 into zazzer.eecs.umich.edu:/z/ehallnor/m5
Erik Hallnor
2004-02-06 00:13:44 -05:00
8f0060b4ae
Various extraction of common code into separate functions to make life easier.
Erik Hallnor
2004-02-06 00:13:26 -05:00
c76675be58
Now skips the calibrate_delay loop in the kernel boot
Andrew Schultz
2004-02-05 18:23:16 -05:00
77a30ed48d
fixed mapping of device memory and i/o space
Ali Saidi
2004-02-05 15:12:33 -05:00
09c40bfe43
added some comments
Ali Saidi
2004-02-05 13:22:02 -05:00
f4ef613ffd
mereged some comments with some of andrews changes
Ali Saidi
2004-02-05 13:17:44 -05:00
7a6a435983
added some comments
Ali Saidi
2004-02-05 13:05:20 -05:00
95c248c213
Modify the emulated system calls to support running the SPEC Int benchmarks for alpha-linux.
David Oehmke
2004-02-05 12:16:17 -05:00
4bbe8c1058
Merge zizzer.eecs.umich.edu:/bk/m5 into ziff.eecs.umich.edu:/z/binkertn/research/m5/kernel
Nathan Binkert
2004-02-05 02:53:03 -05:00
daa579fc0f
Fix PCI code so it builds properly now
Andrew Schultz
2004-02-05 02:25:45 -05:00
b6ff600bca
Add support for "serializing" instructions that flush execution pipeline (Alpha trapb & excb).
Steve Reinhardt
2004-02-04 21:42:00 -08:00
69e1e10f5d
Added code so that BAR writes will result in an updated memory mapping
Ali Saidi
2004-02-04 19:56:24 -05:00
f2af9be29e
Merge ehallnor@zizzer:/bk/m5 into zazzer.eecs.umich.edu:/z/ehallnor/m5
Erik Hallnor
2004-02-04 16:55:17 -05:00
2d6db0359e
Change IIC to use param class. Tweak parameters in CacheTags.
Erik Hallnor
2004-02-04 16:55:05 -05:00
aaaa782165
Removed the tsunami_pci* files
Andrew Schultz
2004-02-04 15:04:36 -05:00
f2b8df2420
Changes to support PciDev as a new base class for devices which use PCI
Andrew Schultz
2004-02-04 15:03:50 -05:00
98948b2e57
More reformatting of reference parameter declarations. The last change only caught the ones with types that started with capitals. This pass catches the rest (mostly STL and uint*_t types).
Steve Reinhardt
2004-02-04 11:16:30 -08:00
579c5b59e1
Removed pci_dev from AdaptecController
Andrew Schultz
2004-02-03 18:56:39 -05:00
9b18af0887
Forgot to fixup the boot_linux.ini file to use AdaptecController
Andrew Schultz
2004-02-03 17:09:49 -05:00
Andrew Schultz