sanchayanmaity/gem5
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
6202 commits 2 branches 0 tags 204 MiB
Commit graph

2934 commits

Author SHA1 Message Date
Gabe Black
b6bfe8af26 X86: Split out the internal memory space from the regular translate() and precompute mode. 2009-04-26 16:48:44 -07:00
Gabe Black
4ee34dfb4e X86: Centralize updates to the handy M5 reg. 2009-04-26 16:47:48 -07:00
Gabe Black
06b3e3c303 X86: Implement lowest priority interrupts more correctly. 
Lowest priority interrupts are now delivered based on a rotating offset into
the list of potential recipients. There could be parasitic cases were a
processor gets picked on and ends up at that rotating offset all the time, but
it's much more likely that the group will stay consistent and the pain will be
distributed evenly.
 2009-04-26 02:09:54 -07:00
Gabe Black
2f34a7eaeb X86: Tell the function that sends int messages who to send to instead of figuring it out itself. 2009-04-26 02:09:27 -07:00
Gabe Black
88ab4bb257 X86: Make the local APICs register themselves with the IO APIC. 
This is a hack so that the IO APIC can figure out information about the local
APICs. The local APICs still have no way to find out about each other.
Ideally, when the local APICs update state that's relevant to somebody else,
they'd send an update to everyone. Without being able to do a broadcast, that
would still require knowing who else there is to notify. Other broadcasts are
implemented using assumptions that may not always be true.
 2009-04-26 02:09:13 -07:00
Gabe Black
c5e2cf841d X86: Record the initial APIC ID which identifies an APIC in M5. 
The ID as exposed to software can be changed. Tracking those changes in M5
would be cumbersome, especially since there's no guarantee the IDs will remain
unique.
 2009-04-26 02:06:21 -07:00
Gabe Black
8d84f81e70 X86, Config: Make makeX86System consider the number of CPUs, and clean up interrupt assignment. 2009-04-26 02:04:32 -07:00
Gabe Black
9d0fa27d09 SPARC: Tighten up the clone system call and SPARCs copyRegs. 2009-04-24 23:11:21 -07:00
Steve Reinhardt
7c056e44e5 request: reorganize flags to group related flags together. 2009-04-23 06:44:32 -07:00
Gabe Black
ee7055c289 X86: Put the StoreCheck flag with the others, and don't collide with other flags. 2009-04-23 01:43:00 -07:00
Nathan Binkert
b4816037ba stats: expose statistics to python 2009-04-22 13:38:01 -07:00
Nathan Binkert
aa9b4e6a68 stats: Move flags into info.hh and use base/flags.hh to manage the flags 2009-04-22 13:38:01 -07:00
Nathan Binkert
8c3eb1a192 stats: Shuffle around info stuff so it can be accessed separately 2009-04-22 13:38:00 -07:00
Nathan Binkert
4d9f25b75c stats: Rename the info classes to hopefully make things a bit clearer 
FooInfoBase became FooInfo
FooInfo became FooInfoProxy
 2009-04-22 13:38:00 -07:00
Nathan Binkert
ca3d82b38a stats: remove simplescalar compatibility for printing 2009-04-22 10:25:14 -07:00
Nathan Binkert
61a68371be stats: fix initialization bug in distribution text output 2009-04-22 06:44:29 -07:00
Steve Reinhardt
e7fa4f2f8e i8254xGBe: major style overhaul. 
Moved DescCache template functions from .hh to .cc file.
Also fixed lots of line-wrapping problems, and some irregular indentation.
 2009-04-22 01:58:53 -04:00
Steve Reinhardt
6629d9b2bc mem: use single BadAddr responder per system. 
Previously there was one per bus, which caused some coherence problems
when more than one decided to respond.  Now there is just one on
the main memory bus.  The default bus responder on all other buses
is now the downstream cache's cpu_side port.  Caches no longer need
to do address range filtering; instead, we just have a simple flag
to prevent snoops from propagating to the I/O bus.
 2008-07-16 11:10:33 -07:00
Nathan Binkert
4d001e43da Automated merge with ssh://m5sim.org//repo/m5 2009-04-21 16:04:55 -07:00
Nathan Binkert
fcc142463d pseudo: only include kernel stats if FULL_SYSTEM. 2009-04-21 15:40:26 -07:00
Nathan Binkert
43c7698f49 arm: include missing file for arm 2009-04-21 15:40:26 -07:00
Nathan Binkert
50f1570352 arm: Unify the ARM tlb. We forgot about this when we did the rest. 
This code compiles, but there are no tests still
 2009-04-21 15:40:25 -07:00
Steve Reinhardt
03b3925e58 syscall_emul: style fixes (mostly wrapping overly long lines) 2009-04-21 08:17:36 -07:00
Steve Reinhardt
52b6764f31 syscall: Resolve conflicts between m5threads and Gabe's recent SE changes. 2009-04-21 08:17:36 -07:00
Daniel Sanchez
b0e9654f86 Commit m5threads package. 
This patch adds limited multithreading support in syscall-emulation
mode, by using the clone system call.  The clone system call works
for Alpha, SPARC and x86, and multithreaded applications run
correctly in Alpha and SPARC.
 2009-04-21 08:17:36 -07:00
Nathan Binkert
b0489d18ed SCons: Export export_vars so SConsopts files can add to them 2009-04-21 08:17:36 -07:00
Steve Reinhardt
97b6947eb7 Minor tweaks for future Ruby compatibility. 2009-04-21 08:17:36 -07:00
Steve Reinhardt
eb3b6935d3 request: add PREFETCH flag. 2009-04-21 08:17:10 -07:00
Steve Reinhardt
3083268d60 request: rename INST_READ to INST_FETCH. 2009-04-20 18:54:02 -07:00
Steve Reinhardt
7f8ea68a30 request: split public and private flags into separate fields. 
This frees up needed space for more public flags.  Also:
- remove unused Request accessor methods
- make Packet use public Request accessors, so it need not be a friend
 2009-04-20 18:40:00 -07:00
Gabe Black
9e9a34fed1 Mem: Fill out the comment that describes the LOCKED request flag. 2009-04-19 22:00:24 -07:00
Gabe Black
bd6f2bb538 Mem: Change isLlsc to isLLSC. 2009-04-19 21:44:15 -07:00
Gabe Black
089b384086 X86: Fix the functions that manipulate large bit arrays in the local APIC. 2009-04-19 13:47:15 -07:00
Gabe Black
eee74ba427 X86: Fix up a copyright. 2009-04-19 13:17:35 -07:00
Gabe Black
6910baa015 X86: Fix how the TLB handles the storecheck flag. 2009-04-19 04:57:51 -07:00
Gabe Black
0a6ff60caa X86: Recognize and handle the lock legacy prefix. 2009-04-19 04:57:28 -07:00
Gabe Black
61edc9ba66 X86: Implement a locking version of XADD. 2009-04-19 04:56:49 -07:00
Gabe Black
209cfc89fd X86: Implement a locking version of BTC. 2009-04-19 04:56:45 -07:00
Gabe Black
e475cf85f0 X86: Implement a locking version of BTR. 2009-04-19 04:56:43 -07:00
Gabe Black
43f58927d6 X86: Implement a locking version of CMPXCHG. 2009-04-19 04:56:40 -07:00
Gabe Black
b493906eb9 X86: Implement a locking version of BTS. 2009-04-19 04:56:36 -07:00
Gabe Black
985d959ea6 X86: Implement a locking version of DEC. 2009-04-19 04:56:34 -07:00
Gabe Black
4f2d4f466a X86: Implement a locking version of INC. 2009-04-19 04:56:31 -07:00
Gabe Black
2394f73f90 X86: Implement a locking version of NEG. 2009-04-19 04:56:28 -07:00
Gabe Black
9b9b7a412c X86: Implement a locking version of NOT. 2009-04-19 04:56:25 -07:00
Gabe Black
b8f81c62a2 X86: Implement a locking version of XCHG. 2009-04-19 04:56:22 -07:00
Gabe Black
750f5a0a67 X86: Implement a locking version of XOR. 2009-04-19 04:56:20 -07:00
Gabe Black
cfb289ebeb X86: Implement a locking version of SUB. 2009-04-19 04:56:16 -07:00
Gabe Black
789b3191b9 X86: Implement a locking version of AND. 2009-04-19 04:56:14 -07:00
Gabe Black
e742cad6f4 X86: Implement a locking version of SBB. 2009-04-19 04:56:11 -07:00
Gabe Black
193265c6e5 X86: Implement a locking version of ADC. 2009-04-19 04:56:08 -07:00
Gabe Black
2f607b882c X86: Implement a locking version of OR. 2009-04-19 04:56:06 -07:00
Gabe Black
a7f79c9049 X86: Implement a locking version of ADD. 2009-04-19 04:56:02 -07:00
Gabe Black
d90456a486 X86: Implement the stul microop. 
This microop does a store and unlocks the requested address. The RISC86
microop ISA doesn't seem to have an equivalent to this, so I'm guessing that
the store following an ldstl is automatically unlocking. We don't do it this
way for performance reasons since the behavior is the same.
 2009-04-19 04:55:58 -07:00
Gabe Black
d2554ff030 X86: Implement the ldstl microop. 
This microop does a load, checks that a store would succeed, and locks the
requested address.
 2009-04-19 04:55:43 -07:00
Gabe Black
1a8a765a5c CPUs: Make the atomic CPU support locked memory accesses. 2009-04-19 04:50:07 -07:00
Gabe Black
742c3f045e Memory: Add a LOCKED flag back in for x86 style locking. 2009-04-19 04:39:25 -07:00
Gabe Black
3e5f487663 Memory: Rename LOCKED for load locked store conditional to LLSC. 2009-04-19 04:25:01 -07:00
Gabe Black
ca85981478 SE mode: Make keeping track of the number of syscalls less hacky. 2009-04-19 04:15:32 -07:00
Gabe Black
e174239bd8 X86: Mask the PIC at startup to avoid a glitch which causes an NMI. 2009-04-19 04:15:06 -07:00
Gabe Black
5f164ba720 X86: Actually handle 16 bit mode modrm. 2009-04-19 04:14:31 -07:00
Gabe Black
93cccf7d19 X86: Make the TEST instruction set all the flags it's supposed to. 2009-04-19 04:14:16 -07:00
Gabe Black
f82c123242 X86: Implement broadcast IPIs. 2009-04-19 04:14:01 -07:00
Gabe Black
829e424353 X86: Fix the ordering of the vendor string reported by CPUID. 2009-04-19 04:13:45 -07:00
Gabe Black
8b2ac20753 X86: Keep track of what the initial count value was in the LAPIC timer. 2009-04-19 03:56:57 -07:00
Gabe Black
18b3863127 X86: Only recognize the first startup IPI after INIT or reset. 2009-04-19 03:56:36 -07:00
Gabe Black
4d32cd10ce X86: Use recvResponse to implement the idle bit in the Local APIC ICR. 2009-04-19 03:56:24 -07:00
Gabe Black
bdda224d41 X86: Add a function which gets called when an interrupt message has been delivered. 2009-04-19 03:54:11 -07:00
Gabe Black
3031af21c7 X86: Fix the flags for interrupt response messages. 2009-04-19 03:53:29 -07:00
Gabe Black
3eed59768c X86: Explicitly use the right width in a few places that need a 64 bit value. 2009-04-19 03:47:59 -07:00
Gabe Black
8761057c78 X86: Keep track of the pioAddr for the local APIC. 2009-04-19 03:47:12 -07:00
Gabe Black
038225a6ca X86: Implement far jmp. 2009-04-19 03:42:41 -07:00
Gabe Black
3b1b21cb15 X86: Some segment selectors can be used when "NULL". 2009-04-19 03:41:10 -07:00
Gabe Black
a0cc081997 X86: Fix a bug in the chks microop where it ignored that it found a fault. 2009-04-19 03:40:08 -07:00
Gabe Black
f2ff5b9249 X86: Make the interrupt entering microcode record the value to use, not actually use it. 2009-04-19 03:36:57 -07:00
Gabe Black
35eea4191b X86: LEA calculates an address before segmentation. 2009-04-19 03:24:51 -07:00
Gabe Black
bdd55ec8b6 X86: Implement the save machine status word instruction (SMSW). 2009-04-19 03:22:38 -07:00
Gabe Black
d86cd1d2a0 X86: Implement the load machine status word instruction (LMSW). 2009-04-19 03:17:14 -07:00
Gabe Black
eba640c963 X86: Only use %eax to select a function and look like we support sse2. 2009-04-19 03:11:24 -07:00
Gabe Black
27e54982b4 X86: Fix the mov to segment selector in real mode instruction microcode. 2009-04-19 03:08:40 -07:00
Gabe Black
633c96bd85 X86: The startup IPI delivery mode is not reserved. 2009-04-19 03:01:46 -07:00
Gabe Black
08f021aad0 X86: Implement the STARTUP IPI. 2009-04-19 02:56:03 -07:00
Gabe Black
d277feb925 X86: Implement the INIT IPI. 2009-04-19 02:53:00 -07:00
Gabe Black
a340b214cf X86: Fix the halt microop. 2009-04-19 02:51:09 -07:00
Gabe Black
641513fe08 X86: Start implementing the interrupt command register in the local APIC. 2009-04-19 02:43:22 -07:00
Gabe Black
9549694ecd X86: Make code that sends an interrupt from the IO APIC available for IPIs. 2009-04-19 02:42:19 -07:00
Gabe Black
d10195b1a4 CPU: If the simple CPU is already idle, just return from suspendContext, don't assert. 2009-04-19 02:23:29 -07:00
Gabe Black
05b5861419 X86: Condense the startupCPU code. 2009-04-19 02:20:57 -07:00
Gabe Black
f668340f2c X86: Set the local APIC ID to something meaningful. 2009-04-19 02:16:49 -07:00
Gabe Black
79a3a6aecb X86: Don't pretend to be an AMD CPU any more. We're not good enough at it. 2009-04-19 02:06:51 -07:00
Korey Sewell
d8a34a9745 mips-tlb-fix: check for alignment faults.\nMIPS was never updated to use TLBS correcty in SE mode. The error was forwarding translations directly to pageTable. The TLB should check for alignment faults at bare minimum here but in the long run we should be using TLBs in SE mode for MIPS. 2009-04-18 10:42:29 -04:00
Korey Sewell
e501e1af54 mips-syscall: mark with correct flag. \nMIPS was using wrong serialization flag on syscall instructions allowing O3 to handle SE mode syscalls incorrectly and speculate on instructions after a syscall 2009-04-18 10:42:29 -04:00
Korey Sewell
5c1742b822 o3-delay-slot-bpred: fix decode stage handling of uncdtl. branches.\n decode stage was not setting the predicted PC correctly or passing that information back to fetch correctly 2009-04-18 10:42:29 -04:00
Korey Sewell
cc9e834e93 mips-shadowsets: fix calcuations. \n Remove Shadowsets from Int & Arch. Reg Calculations 2009-04-18 10:42:28 -04:00
Steve Reinhardt
14808ecac9 o3, inorder: fix FS bug due to initializing ThreadState to Halted. 
For some reason o3 FS init() only called initCPU if the thread state
was Suspended, which was no longer the case.  There's no apparent
reason to check, so I whacked the test completely rather than
changing the check to Halted.
The inorder init() was also updated to be symmetric, though the
previous code was just a fancy no-op.
 2009-04-17 16:54:58 -07:00
Steve Reinhardt
b146131d18 o3: handle fetch with no active threads correctly. 
This situation can arise now on the first fetch cycle after
the last active thread is halted.  It seems easy enough to
deal with when it happens rather than trying to avoid it.
 2009-04-15 23:12:00 -07:00
Steve Reinhardt
bb974d5a47 o3: fix {read,set}ArchFloatReg* functions. 
Register indices were not being calculated properly.
 2009-04-15 23:10:43 -07:00
Steve Reinhardt
7617dcf736 ThreadState: initialize status to Halted in constructor. 
This provides a common initial status for all threads independent
of CPU model (unlike the prior situation where CPUs initialized
threads to inconsistent states).
This mostly matters for SE mode; in FS mode, ISA-specific startupCPU()
methods generally handle boot-time initialization of thread contexts
(since the right thing to do is ISA-dependent).
 2009-04-15 13:18:24 -07:00
Steve Reinhardt
8882dc1283 Get rid of the Unallocated thread context state. 
Basically merge it in with Halted.
Also had to get rid of a few other functions that
called ThreadContext::deallocate(), including:
 - InOrderCPU's setThreadRescheduleCondition.
 - ThreadContext::exit().  This function was there to avoid terminating
   simulation when one thread out of a multi-thread workload exits, but we
   need to find a better (non-cpu-centric) way.
 2009-04-15 13:13:47 -07:00
Gabe Black
5c79191603 X86: Fix minor bug in the page table walker from TLB shuffling. 2009-04-13 04:14:15 -07:00