| Age | Commit message (Collapse) | Author |
|---|
|
This changeset adds functionality that allows system calls to retry without
affecting thread context state such as the program counter or register values
for the associated thread context (when system calls return with a retry
fault).
This functionality is needed to solve problems with blocking system calls
in multi-process or multi-threaded simulations where information is passed
between processes/threads. Blocking system calls can cause deadlock because
the simulator itself is single threaded. There is only a single thread
servicing the event queue which can cause deadlock if the thread hits a
blocking system call instruction.
To illustrate the problem, consider two processes using the producer/consumer
sharing model. The processes can use file descriptors and the read and write
calls to pass information to one another. If the consumer calls the blocking
read system call before the producer has produced anything, the call will
block the event queue (while executing the system call instruction) and
deadlock the simulation.
The solution implemented in this changeset is to recognize that the system
calls will block and then generate a special retry fault. The fault will
be sent back up through the function call chain until it is exposed to the
cpu model's pipeline where the fault becomes visible. The fault will trigger
the cpu model to replay the instruction at a future tick where the call has
a chance to succeed without actually going into a blocking state.
In subsequent patches, we recognize that a syscall will block by calling a
non-blocking poll (from inside the system call implementation) and checking
for events. When events show up during the poll, it signifies that the call
would not have blocked and the syscall is allowed to proceed (calling an
underlying host system call if necessary). If no events are returned from the
poll, we generate the fault and try the instruction for the thread context
at a distant tick. Note that retrying every tick is not efficient.
As an aside, the simulator has some multi-threading support for the event
queue, but it is not used by default and needs work. Even if the event queue
was completely multi-threaded, meaning that there is a hardware thread on
the host servicing a single simulator thread contexts with a 1:1 mapping
between them, it's still possible to run into deadlock due to the event queue
barriers on quantum boundaries. The solution of replaying at a later tick
is the simplest solution and solves the problem generally.
|
|
The EIOProcess class was removed recently and it was the only other class
which derived from Process. Since every Process invocation is also a
LiveProcess invocation, it makes sense to simplify the organization by
combining the fields from LiveProcess into Process.
|
|
Turns out that SPARC SE mode relied on M5_pid being "0" in
all cases. The entries in the SPARC TLBs are accessed with
M5_pid as their context. This is buggy in the sense that it
will never work with more than one process or any
initialization that doesn't have the M5_pid value passed in
as "0".
cd7f3a1dbf55 broke the SPARC build because it deletes M5_pid
and uses a _pid with a default of "100" instead. This caused
the SPARC TLB to never return any valid lookups for any
request; the program never moved past the first instruction
with SPARC SE in the regression tester.
The solution proposed in this changeset is to initialize
the address space identification register with the PID value
that is passed into the process class as a parameter from
Python. This should return the correct responses from the TLB
since the insertions and lookups into the page table will be
using the same PID.
Furthermore, there are corner cases in the code which elevate
privileges and revert to using context "0" as the context in
the TLB. I believe that these are related to kernel level
traps and hypervisor privilege escalations, but I'm not
completely sure. I've tried to address the corner cases
properly, but it would be beneficial to have someone who is
familiar with the SPARC architecture to take a look at this
fix.
|
|
|
|
|
|
This patch optimises the passing of StaticInstPtr by avoiding copying
the reference-counting pointer. This avoids first incrementing and
then decrementing the reference-counting pointer.
|
|
This patch adds a warning for missing field initializers for both gcc
and clang, and addresses the warnings that were generated.
|
|
This gets rid of cryptic bits of code with lots of bit manipulation, and makes
some comments redundant.
|
|
Try to decrease indentation, and remove some redundant FullSystem checks.
|
|
|
|
|
|
|
|
Do some minor cleanup of some recently added comments, a warning, and change
other instances of stack extension to be like what's now being done for x86.
|
|
|
|
This change is a low level and pervasive reorganization of how PCs are managed
in M5. Back when Alpha was the only ISA, there were only 2 PCs to worry about,
the PC and the NPC, and the lsb of the PC signaled whether or not you were in
PAL mode. As other ISAs were added, we had to add an NNPC, micro PC and next
micropc, x86 and ARM introduced variable length instruction sets, and ARM
started to keep track of mode bits in the PC. Each CPU model handled PCs in
its own custom way that needed to be updated individually to handle the new
dimensions of variability, or, in the case of ARMs mode-bit-in-the-pc hack,
the complexity could be hidden in the ISA at the ISA implementation's expense.
Areas like the branch predictor hadn't been updated to handle branch delay
slots or micropcs, and it turns out that had introduced a significant (10s of
percent) performance bug in SPARC and to a lesser extend MIPS. Rather than
perpetuate the problem by reworking O3 again to handle the PC features needed
by x86, this change was introduced to rework PC handling in a more modular,
transparent, and hopefully efficient way.
PC type:
Rather than having the superset of all possible elements of PC state declared
in each of the CPU models, each ISA defines its own PCState type which has
exactly the elements it needs. A cross product of canned PCState classes are
defined in the new "generic" ISA directory for ISAs with/without delay slots
and microcode. These are either typedef-ed or subclassed by each ISA. To read
or write this structure through a *Context, you use the new pcState() accessor
which reads or writes depending on whether it has an argument. If you just
want the address of the current or next instruction or the current micro PC,
you can get those through read-only accessors on either the PCState type or
the *Contexts. These are instAddr(), nextInstAddr(), and microPC(). Note the
move away from readPC. That name is ambiguous since it's not clear whether or
not it should be the actual address to fetch from, or if it should have extra
bits in it like the PAL mode bit. Each class is free to define its own
functions to get at whatever values it needs however it needs to to be used in
ISA specific code. Eventually Alpha's PAL mode bit could be moved out of the
PC and into a separate field like ARM.
These types can be reset to a particular pc (where npc = pc +
sizeof(MachInst), nnpc = npc + sizeof(MachInst), upc = 0, nupc = 1 as
appropriate), printed, serialized, and compared. There is a branching()
function which encapsulates code in the CPU models that checked if an
instruction branched or not. Exactly what that means in the context of branch
delay slots which can skip an instruction when not taken is ambiguous, and
ideally this function and its uses can be eliminated. PCStates also generally
know how to advance themselves in various ways depending on if they point at
an instruction, a microop, or the last microop of a macroop. More on that
later.
Ideally, accessing all the PCs at once when setting them will improve
performance of M5 even though more data needs to be moved around. This is
because often all the PCs need to be manipulated together, and by getting them
all at once you avoid multiple function calls. Also, the PCs of a particular
thread will have spatial locality in the cache. Previously they were grouped
by element in arrays which spread out accesses.
Advancing the PC:
The PCs were previously managed entirely by the CPU which had to know about PC
semantics, try to figure out which dimension to increment the PC in, what to
set NPC/NNPC, etc. These decisions are best left to the ISA in conjunction
with the PC type itself. Because most of the information about how to
increment the PC (mainly what type of instruction it refers to) is contained
in the instruction object, a new advancePC virtual function was added to the
StaticInst class. Subclasses provide an implementation that moves around the
right element of the PC with a minimal amount of decision making. In ISAs like
Alpha, the instructions always simply assign NPC to PC without having to worry
about micropcs, nnpcs, etc. The added cost of a virtual function call should
be outweighed by not having to figure out as much about what to do with the
PCs and mucking around with the extra elements.
One drawback of making the StaticInsts advance the PC is that you have to
actually have one to advance the PC. This would, superficially, seem to
require decoding an instruction before fetch could advance. This is, as far as
I can tell, realistic. fetch would advance through memory addresses, not PCs,
perhaps predicting new memory addresses using existing ones. More
sophisticated decisions about control flow would be made later on, after the
instruction was decoded, and handed back to fetch. If branching needs to
happen, some amount of decoding needs to happen to see that it's a branch,
what the target is, etc. This could get a little more complicated if that gets
done by the predecoder, but I'm choosing to ignore that for now.
Variable length instructions:
To handle variable length instructions in x86 and ARM, the predecoder now
takes in the current PC by reference to the getExtMachInst function. It can
modify the PC however it needs to (by setting NPC to be the PC + instruction
length, for instance). This could be improved since the CPU doesn't know if
the PC was modified and always has to write it back.
ISA parser:
To support the new API, all PC related operand types were removed from the
parser and replaced with a PCState type. There are two warts on this
implementation. First, as with all the other operand types, the PCState still
has to have a valid operand type even though it doesn't use it. Second, using
syntax like PCS.npc(target) doesn't work for two reasons, this looks like the
syntax for operand type overriding, and the parser can't figure out if you're
reading or writing. Instructions that use the PCS operand (which I've
consistently called it) need to first read it into a local variable,
manipulate it, and then write it back out.
Return address stack:
The return address stack needed a little extra help because, in the presence
of branch delay slots, it has to merge together elements of the return PC and
the call PC. To handle that, a buildRetPC utility function was added. There
are basically only two versions in all the ISAs, but it didn't seem short
enough to put into the generic ISA directory. Also, the branch predictor code
in O3 and InOrder were adjusted so that they always store the PC of the actual
call instruction in the RAS, not the next PC. If the call instruction is a
microop, the next PC refers to the next microop in the same macroop which is
probably not desirable. The buildRetPC function advances the PC intelligently
to the next macroop (in an ISA specific way) so that that case works.
Change in stats:
There were no change in stats except in MIPS and SPARC in the O3 model. MIPS
runs in about 9% fewer ticks. SPARC runs with 30%-50% fewer ticks, which could
likely be improved further by setting call/return instruction flags and taking
advantage of the RAS.
TODO:
Add != operators to the PCState classes, defined trivially to be !(a==b).
Smooth out places where PCs are split apart, passed around, and put back
together later. I think this might happen in SPARC's fault code. Add ISA
specific constructors that allow setting PC elements without calling a bunch
of accessors. Try to eliminate the need for the branching() function. Factor
out Alpha's PAL mode pc bit into a separate flag field, and eliminate places
where it's blindly masked out or tested in the PC.
|
|
Also move the "Fault" reference counted pointer type into a separate file,
sim/fault.hh. It would be better to name this less similarly to sim/faults.hh
to reduce confusion, but fault.hh matches the name of the type. We could change
Fault to FaultPtr to match other pointer types, and then changing the name of
the file would make more sense.
|
|
Even though we're not incorrect about operator precedence, let's add
some parens in some particularly confusing places to placate GCC 4.3
so that we don't have to turn the warning off. Agreed that this is a
bit of a pain for those users who get the order of operations correct,
but it is likely to prevent bugs in certain cases.
|
|
store the process, not the system.
--HG--
extra : convert_revision : 2421af11f62f60fb48faeee6bddadac2987df0e8
|
|
--HG--
extra : convert_revision : a04a30df0b6246e877a1cea35420dbac94b506b1
|
|
across all architectures.
--HG--
extra : convert_revision : 18d441eb7ac44df4df41771bfe3dec69f7fa70ec
|
|
--HG--
extra : convert_revision : f799b65f1b2a6bf43605e6870b0f39b473dc492b
|
|
into zeep.pool:/z/saidi/work/m5.newmem
--HG--
extra : convert_revision : fd6464c9883783c7c2cbefba317f4a0f20dd24cb
|
|
configs/common/FSConfig.py:
Use binaries we've compiled instead of the ones that come with Legion
src/arch/alpha/interrupts.hh:
get rid of post(int int_type) and add a get_vec function that gets the interrupt vector for an interrupt number
src/arch/sparc/asi.cc:
Add AsiIsInterrupt() to AsiIsMmu()
src/arch/sparc/faults.cc:
src/arch/sparc/faults.hh:
Add InterruptVector type
src/arch/sparc/interrupts.hh:
rework interrupts. They are no longer cleared when created... A I/O or ASI read/write needs to happen before they are cleared
src/arch/sparc/isa_traits.hh:
Add the "interrupt" trap types to isa traits
src/arch/sparc/miscregfile.cc:
add names for all the misc registers and possible post an interrupt when TL is changed.
src/arch/sparc/miscregfile.hh:
Add a helper function to post an interrupt when pil < some set softint
src/arch/sparc/regfile.cc:
src/arch/sparc/regfile.hh:
InterruptLevel shouldn't really live here, moved to interrupt.hh
src/arch/sparc/tlb.cc:
Add interrupt ASIs to TLB
src/arch/sparc/ua2005.cc:
Add checkSoftInt to check if a softint needs to be posted
Check that a tickCompare isn't scheduled before scheduling one
Post and clear interrupts on queue writes and what not
src/base/bitfield.hh:
Add an helper function to return the msb that is set
src/cpu/base.cc:
src/cpu/base.hh:
get rid of post_interrupt(type) since it's no longer needed.. Add a way to see what interrupts are pending
src/cpu/intr_control.cc:
src/cpu/intr_control.hh:
src/dev/alpha/tsunami_cchip.cc:
src/python/m5/objects/IntrControl.py:
Make IntrControl have a system pointer rather than using a cpu pointer to get one
src/dev/sparc/SConscript:
add iob to SConsscrip
tests/quick/10.linux-boot/ref/alpha/linux/tsunami-simple-atomic-dual/config.ini:
tests/quick/10.linux-boot/ref/alpha/linux/tsunami-simple-atomic-dual/config.out:
tests/quick/10.linux-boot/ref/alpha/linux/tsunami-simple-atomic/config.ini:
tests/quick/10.linux-boot/ref/alpha/linux/tsunami-simple-atomic/config.out:
tests/quick/10.linux-boot/ref/alpha/linux/tsunami-simple-timing-dual/config.ini:
tests/quick/10.linux-boot/ref/alpha/linux/tsunami-simple-timing-dual/config.out:
tests/quick/10.linux-boot/ref/alpha/linux/tsunami-simple-timing/config.ini:
tests/quick/10.linux-boot/ref/alpha/linux/tsunami-simple-timing/config.out:
tests/quick/80.netperf-stream/ref/alpha/linux/twosys-tsunami-simple-atomic/config.ini:
tests/quick/80.netperf-stream/ref/alpha/linux/twosys-tsunami-simple-atomic/config.out:
update config.ini/out for intrcntrl not having a cpu pointer anymore
--HG--
extra : convert_revision : 38614f6b9ffc8f3c93949a94ff04b7d2987168dd
|
|
call into the Process object to handle system calls. Refactored the Process objects, and move the handler code into it's own file, and add some syscalls which are used in a natively compiled hello world. Software traps with trap number 3 (not syscall number 3) are supposed to cause the register windows to be flushed but are ignored right now. Finally, made uname for SPARC report a 2.6.12 kernel which is what m22-018.pool happens to be running.
--HG--
extra : convert_revision : ea873f01c62234c0542f310cc143c6a7c76ade94
|
|
into zower.eecs.umich.edu:/eecshome/m5/newmem
--HG--
extra : convert_revision : 2d7ae62a59b91d735bbac093f8a4ab542ea75eee
|
|
check writability of tlb cache entry before using
update tagaccess in places I forgot to
move the tlb privileged test up since it is higher priority
src/arch/sparc/faults.cc:
save only 32 bits of PC/NPC if Pstate.am is set
src/arch/sparc/isa/decoder.isa:
return only 32 bits of PC/NPC if Pstate.am is set
increment cleanwin correctly
src/arch/sparc/tlb.cc:
check writability of cache entry
update tagaccess in a few more places
move the privileged test up since it is higher priority
src/cpu/exetrace.cc:
mask off upper bits of pc if pstate.am is set before comparing to legion
--HG--
extra : convert_revision : 02a51c141ee3f9a2600c28eac018ea7216f3655c
|
|
into ewok.(none):/home/gblack/m5/newmemo3
src/sim/byteswap.hh:
Hand Merge
--HG--
extra : convert_revision : 640d33ad0c416934e8a5107768e7f1dce6709ca8
|
|
simplify and make complete some asi checks
implement all the twin asis and remove panic checks on their use
soft int is supported, so we don't need to print writes to it
src/arch/sparc/asi.cc:
make AsiIsLittle() be all the little asis.
Speed up AsiIsTwin() a bit
src/arch/sparc/faults.cc:
clean up the do*Fault code.... Make it work like legion, in particular
pstate.priv is left alone, not set to 0 like the spec says
src/arch/sparc/isa/decoder.isa:
implement some more twin ASIs
src/arch/sparc/tlb.cc:
All the twin asis are implemented, no need to say their not supported anymore
src/arch/sparc/ua2005.cc:
softint is supported now, no more need to
--HG--
extra : convert_revision : aef2a1b93719235edff830a17a8ec52f23ec9f8b
|
|
PSTATE to avoid name confusion.
src/arch/sparc/faults.cc:
1) s/Resumeable/Resumable/gc
2) s/if(/if (/gc
3) keep variables lowercase
4) change the way fields are accessed - instead of hard coding bitvectors, use masks (like HPSTATE::hpriv).
src/arch/sparc/faults.hh:
s/Resumeable/Resumable/
src/arch/sparc/isa_traits.hh:
This is unused and unnecessary.
src/arch/sparc/miscregfile.hh:
add bitfield masks for some important ASRs (HPSTATE, PSTATE).
--HG--
extra : convert_revision : f0ffaf48de298758685266dfb90f43aff42e0a2c
|
|
there is no interrupt_level_0 interrupt, so start the list at 0x40 so the adding is done correctly
src/arch/sparc/faults.cc:
there is no interrupt_level_0 interrupt, so start the list at 0x40 so the adding is done correctly
src/arch/sparc/faults.hh:
correct protection defines
src/arch/sparc/ua2005.cc:
set the softint appropriately on an timer compare interrupt
--HG--
extra : convert_revision : f41c10ec78db973b3f856c70b58a17f83b60bbe2
|
|
into zower.eecs.umich.edu:/eecshome/m5/newmem
src/arch/isa_parser.py:
src/arch/sparc/isa/formats/mem/basicmem.isa:
src/arch/sparc/isa/formats/mem/blockmem.isa:
src/arch/sparc/isa/formats/mem/util.isa:
src/arch/sparc/miscregfile.cc:
src/arch/sparc/miscregfile.hh:
src/cpu/o3/iew_impl.hh:
Hand Merge
--HG--
extra : convert_revision : ae1b25cde85ab8ec275a09d554acd372887d4d47
|
|
Fix fault formating and code for traps
fix a couple of bugs in the decoder
Cleanup/fix page table entry code
Implement more mmaped iprs, fix numbered tlb insertion code, add function to dump tlb contents
Don't panic if we differ from legion on a tcc instruction because of where legion prints its data and where we print our data
src/arch/sparc/faults.cc:
Fix fault formating and code for traps
src/arch/sparc/intregfile.hh:
allocate the correct number of global registers
src/arch/sparc/isa/decoder.isa:
fix a couple of bugs in the decoder: wrasi should write asi not ccr, done/retry should get hpstate from htstate
src/arch/sparc/pagetable.hh:
cleanup/fix page table code
src/arch/sparc/tlb.cc:
implement more mmaped iprs, fix numbered insertion code, add function to dump tlb contents
src/arch/sparc/tlb.hh:
add functions to write TagAccess register on tlb miss and to dump all tlb entries for debugging
src/cpu/exetrace.cc:
dump tlb entries on error, don't consider differences the cycle we take a trap to be bad.
--HG--
extra : convert_revision : d7d771900f6f25219f3dc6a6e51986d342a32e03
|
|
src/arch/sparc/asi.cc:
src/arch/sparc/asi.hh:
add sparc error asi
src/arch/sparc/faults.cc:
put a panic in if TL == MaxTL
src/arch/sparc/isa/decoder.isa:
Hpstate needs to be updated on a done too
src/arch/sparc/miscregfile.cc:
warn istead of panicing of fprs/fsr accesses
src/arch/sparc/tlb.cc:
add sparc error register code that just does nothing
fix a couple of other tlb bugs
src/arch/sparc/ua2005.cc:
fix implementation of HPSTATE write
src/cpu/exetrace.cc:
let exectrate mess up a couple of times before dying
src/python/m5/objects/T1000.py:
add l2 error status register fake devices
--HG--
extra : convert_revision : ed5dfdfb28633bf36e5ae07d244f7510a02874ca
|
|
and panics instead of fatals. This isn't technically what it should do, but it makes gdb stop at the panic rather than letting m5 exit.
--HG--
extra : convert_revision : 3b14c99edaf649e0809977c9579afb2b7b0d72e9
|
|
miscregs into the integer register file so they get renamed.
src/arch/alpha/syscallreturn.hh:
src/arch/mips/syscallreturn.hh:
src/sim/syscallreturn.hh:
Move the SyscallReturn class into sim/syscallreturn.hh
src/arch/sparc/faults.cc:
src/arch/sparc/isa/operands.isa:
src/arch/sparc/isa_traits.hh:
src/arch/sparc/miscregfile.cc:
src/arch/sparc/miscregfile.hh:
src/arch/sparc/process.cc:
src/arch/sparc/sparc_traits.hh:
Move some miscregs into the integer register file so they get renamed.
--HG--
extra : convert_revision : df5b94fa1e7fdca34816084e0a423d6fdf86c79b
|
|
manual of what happens during a trap says it should be 0, and other places say it doesn't matter.
--HG--
extra : convert_revision : 9ecb6af06657e936a208cbeb8e4a18305869b949
|
|
--HG--
extra : convert_revision : 4504f08fd94792819bd4419bbd2e0ebd1d7f29e9
|
|
use memcpy instead of bcopy
s/u_int32_t/uint32_t/g
fixup endian code to work with solaris
hack to make sure htole() works... Nate, have a good idea to fix this?
src/arch/sparc/faults.cc:
set the reset address to be 40 bits. Makes PC printing easier at least for now.
src/arch/sparc/isa/base.isa:
fix endian issues with condition codes
src/arch/sparc/tlb.hh:
add implemented physical addres constants
src/arch/sparc/utility.hh:
add tlb.hh to utilities
src/base/loader/raw_object.cc:
add a symbol <filename>_start to the symbol table for binaries files
src/base/remote_gdb.cc:
use memcpy instead of bcopy
src/cpu/exetrace.cc:
clean up printing a bit more
src/cpu/m5legion_interface.h:
add tons to the shared interface
src/dev/ethertap.cc:
s/u_int32_t/uint32_t/g
src/dev/ide_atareg.h:
fixup endian code to work with solaris
src/dev/pcidev.cc:
src/sim/param.hh:
hack to make sure htole() works...
--HG--
extra : convert_revision : 4579392184b40bcc1062671a953c6595c685e9b2
|
|
--HG--
extra : convert_revision : 38951352edbfc423fb6767a9aac49a703578c0ac
|
|
--HG--
extra : convert_revision : d17d28a9520524e5f56bd79beb9b2be6ce76a22f
|
|
--HG--
extra : convert_revision : 8023db1479cb9bf99fc9edfeb521c4e5b581f895
|
|
faults for ua2005, and commented out ones which are apparently dropped.
--HG--
extra : convert_revision : 32bd0c3a75d7c036ad4a3cb0bc1c32e0b6cb3d87
|
|
into zeep.eecs.umich.edu:/home/gblack/m5/newmemmemops
src/arch/sparc/faults.hh:
Hand merged.
--HG--
extra : convert_revision : 1bcefe47fa98e878a0dfbcfa5869b5b171927911
|
|
--HG--
extra : convert_revision : eb7e016a127417cbb0e1e2c733b17f82469c2f24
|
|
--HG--
extra : convert_revision : 6339c82d3655694445c3eb43e467b9aa6b4c8224
|
|
because apparently you need an xc for that and not a tc. Cleaned up the TrapInstruction fault in light of this.
--HG--
extra : convert_revision : 1805c9244cfd62d0ee7862d8fd7c9983e00c5747
|
|
src/arch/sparc/faults.cc:
Moved some code here from miscregfile.cc
src/arch/sparc/miscregfile.cc:
Moved code from here to faults.cc, and merged (read|set)MiscRegWithEffect and it's FS version from ua2005.cc
src/arch/sparc/miscregfile.hh:
readFSRegWithEffect is no longer a seperate function, and is instead done in the main readRegWith Effect.
--HG--
extra : convert_revision : 0b45f0f78e83929b32ddd2f443c8b1dbf9bc04fb
|
|
--HG--
extra : convert_revision : 46e9f26917efab642b80ea9e4303ec95d43d935e
|
|
--HG--
extra : convert_revision : 6e98cf839dc92bde5f06f9b9bf11ca6ac661c907
|
|
src/arch/sparc/faults.cc:
src/arch/sparc/faults.hh:
Added a function to do normal SPARC trap processing, and implemented the spill and fill faults for SE
src/arch/sparc/process.cc:
src/arch/sparc/process.hh:
Added fill and spill handlers which are stuffed into the processes address space. The location of these handlers are stored in fillStart and spillStart.
--HG--
extra : convert_revision : 59adb96570cce86f373fbc2c3e4c05abe1742d3b
|
