Commit graph

11356 commits

Author SHA1 Message Date
Mitch Hayenga
c0d19391d4 arm: Squash after returning from exceptions in v7
Properly done for the ERET instruction in v8, but not for v7.
Many control register changes are only visible after explicit
instruction synchronization barriers or exception entry/exit.
This means mode changing instructions should squash any
younger in-flight speculative instructions.
2016-02-29 19:13:13 -06:00
Curtis Dunham
aa674268e9 base: support gzip-compressed object files 2016-02-29 19:13:13 -06:00
Andreas Hansson
28289e5995 stats: Update stats to reflect forwarding of InvalidateReq 2016-02-24 04:16:59 -05:00
Andreas Hansson
7958f34797 mem: Ensure that InvalidateReq is not forwarded as ReadExReq
This patch fixes an issue where an InvalidationReq only traversed one
level of the cache hierarchy, and was subsequently turned into a
ReadExReq due to it needing writable, and the command not being
checked for explicitly.
2016-02-24 04:16:57 -05:00
Matteo Andreozzi
496a8c6c92 cpu: TraceGen fix for tick frequency check
Bug fix for check on protobuf file frequency being different than
global frequency.

The ASCII encoder script is also fixed, and the example trace used in
the regressions is updated.
2016-02-24 04:16:55 -05:00
Andreas Sandberg
e2cea54deb dev, arm: Implement the NoMali reset callback
Add a callback handler for the NoMali reset callback. This callback is
called whenever the GPU is reset using the register interface or the
NoMali API. The callback can be used to override ID registers using
the raw register API.
2016-02-23 11:49:35 +00:00
Andreas Sandberg
81a8ce3564 dev, arm: Refactor the NoMali GPU
Refactor and cleanup the NoMaliGpu class:

  * Use a std::map instead of a switch block to map the parameter enum
    describing the GPU type to a NoMali type.

  * Remove redundant NoMali handle from the interrupt callback.

  * Make callbacks and API wrappers protected instead of private to
    enable future extensions.

  * Wrap remaining NoMali API calls.
2016-02-23 11:49:34 +00:00
Andreas Sandberg
826e0047b0 arm: Ship Linux device trees with gem5
Ship aarch32 and aarch64 device trees with gem5. We currently ship
device trees as a part of the gem5 Linux kernel repository. This makes
tracking hard since device trees are supposed to be platform dependent
rather than kernel dependent (Linux considers device trees to be a
stable kernel ABI). It also makes code sharing between aarch32 and
aarch64 impossible.

This changeset implements a set of device trees for the new
VExpress_GEM5_V1 platform. The platform is described in a shared file
that is separate from the memory/CPU description. Due to differences
in how secondary CPUs are initialized, aarch32 and aarch64 use
different base files describing CPU nodes and the machine's
compatibility property.
2016-02-23 11:21:07 +00:00
Andreas Hansson
4619f0ee8b scons: Add missing override to appease clang
Make clang happy...again.
2016-02-23 03:27:20 -05:00
Tony Gutierrez
5a88f0931f ruby: move range change send from RubyPort to derived classes. 2016-02-18 10:50:16 -05:00
John Kalamatianos
a28a234069 gpu: fix bugs with MemFence, Flat Instrs and Resource utilization
Both Memory Fence is now flagged as Global Memory only to avoid resource
oversubscribing.
Flat instructions now check for Shared Memory resource busy to avoid
oversubscribing resources.
All WaitClass resources now use cycles (not ticks) to register the number
of pipe stages between Scoreboard and Execute to be consistent with
instruction scheduling logic which always used clock cycles.
2016-02-18 10:42:03 -05:00
Tony Gutierrez
9a0f1be21f gpu-compute: remove brig_object.hh from hsa_object.cc
brig_object.hh is specific to the HSAIL ISA, and hence should not be
included in ISA-agnostic code.
2016-02-17 11:46:02 -05:00
Tony Gutierrez
969babd26f ruby: send address ranges from RubyPort 2016-02-17 11:31:54 -05:00
Andreas Hansson
a34ff350da scons: Enable building with the gcc/clang Address Sanitizer
Allow the user to easily build gem5 with the Address Sanitizer, part
of both gcc and clang these days.
2016-02-17 03:56:20 -05:00
Andreas Hansson
0d50979888 misc: Add missing overrides to appease clang
Since the last round of fixes a few new issues have snuck in. We
should consider switching the regression runs to clang.
2016-02-15 03:40:32 -05:00
Andreas Hansson
407233f5d8 mem: Avoid using invalid iterator in cache lock list traversal
Fix up issue highlighted by Valgrind and the clang Address Sanitizer.
2016-02-15 03:40:04 -05:00
Michael LeBeane
b181cea364 ruby: make DMASequencer inherit from RubyPort
This patch essentially rolls back 10518:30e3715c9405 to make RubyPort the
parent class of DMASequencer.  It removes redundant code and restores some
features which were lost when directly inheriting from MemObject.  For
example,
DMASequencer can now communicate to other devices using PIO, which is useful
for memmory-mapped communication between multiple DMADevices.
2016-02-14 20:28:48 -05:00
Michael LeBeane
2ae4cce393 configs: add command-line option to stop debug output
This patch adds a --debug-end flag to main.py so that debug output can be
stoped at a specified tick, while allowing the simulation to continue. It is
useful in situations where you would like to produce a trace for a region of
interest while still collecting stats for the entire run. This is in contrast
to the currently existing --debug-break flag, which terminates the simulation
at the tick.
2016-02-13 12:36:43 -05:00
Michael LeBeane
8d923c7380 syscall_emul: Implement clock_getres() system call
This patch implements the clock_getres() system call for arm and x86 in linux
SE mode.
2016-02-13 12:33:07 -05:00
Andreas Hansson
c6cede244b stats: Update stats to reflect changes to cache and crossbar 2016-02-10 04:08:27 -05:00
Andreas Hansson
83a5977481 mem: Be less conservative in clearing load locks in the cache
Avoid being overly conservative in clearing load locks in the cache,
and allow writes to the line if they are from the same context. This
is in line with ALPHA and ARM.
2016-02-10 04:08:25 -05:00
Andreas Hansson
92f021cbbe mem: Move the point of coherency to the coherent crossbar
This patch introduces the ability of making the coherent crossbar the
point of coherency. If so, the crossbar does not forward packets where
a cache with ownership has already committed to responding, and also
does not forward any coherency-related packets that are not intended
for a downstream memory controller. Thus, invalidations and upgrades
are turned around in the crossbar, and the memory controller only sees
normal reads and writes.

In addition this patch moves the express snoop promotion of a packet
to the crossbar, thus allowing the downstream cache to check the
express snoop flag (as it should) for bypassing any blocking, rather
than relying on whether a cache is responding or not.
2016-02-10 04:08:25 -05:00
Andreas Hansson
f84ee031cc mem: Align cache behaviour in atomic when upstream is responding
Adopt the same flow as in timing mode, where the caches on the path to
memory get to keep the line (if present), and we use the
responderHadWritable flag to determine if we need to forward the
(invalidating) packet or not.
2016-02-10 04:08:24 -05:00
Andreas Hansson
986214f181 mem: Align how snoops are handled when hitting writebacks
This patch unifies the snoop handling in case of hitting writebacks
with how we handle snoops hitting in the tags. As a result, we end up
using the same optimisation as the normal snoops, where we inform the
downstream cache if we encounter a line in Modified (writable and
dirty) state, which enables us to avoid sending out express snoops to
invalidate any Shared copies of the line. A few regressions
consequently change, as some transactions are sunk higher up in the
cache hierarchy.
2016-02-10 04:08:24 -05:00
Andreas Hansson
fbdeb60316 mem: Deduce if cache should forward snoops
This patch changes how the cache determines if snoops should be
forwarded from the memory side to the CPU side. Instead of having a
parameter, the cache now looks at the port connected on the CPU side,
and if it is a snooping port, then snoops are forwarded. Less error
prone, and less parameters to worry about.

The patch also tidies up the CPU classes to ensure that their I-side
port is not snooping by removing overrides to the snoop request
handler, such that snoop requests will panic via the default
MasterPort implement
2016-02-10 04:08:24 -05:00
Curtis Dunham
bead7f249a scons: always generate sim/tags.cc
Due to insufficient build deps, the checkpoint tags might not get
updated; this commit solves this. Due to the uncommon nature of the
build target, regenerating tags.cc is a fairly clean solution. Since
SCons hashes file contents, it won't recompile anything unless a new
checkpoint upgrader is actually added.

--HG--
extra : amend_source : ed3879da7668554693f697076deaf5029cc9b954
2016-02-08 13:39:45 -06:00
Alexandru Dutu
0f27d70e90 x86: revamp cmpxchg8b/cmpxchg16b implementation
The previous implementation did a pair of nested RMW operations,
which isn't compatible with the way that locked RMW operations are
implemented in the cache models.  It was convenient though in that
it didn't require any new micro-ops, and supported cmpxchg16b using
64-bit memory ops.  It also worked in AtomicSimpleCPU where
atomicity was guaranteed by the core and not by the memory system.
It did not work with timing CPU models though.

This new implementation defines new 'split' load and store micro-ops
which allow a single memory operation to use a pair of registers as
the source or destination, then uses a single ldsplit/stsplit RMW
pair to implement cmpxchg.  This patch requires support for 128-bit
memory accesses in the ISA (added via a separate patch) to support
cmpxchg16b.
2016-02-06 17:21:20 -08:00
Steve Reinhardt
5200e04e92 arch, x86: add support for arrays as memory operands
Although the cache models support wider accesses, the ISA descriptions
assume that (for the most part) memory operands are integer types,
which makes it difficult to define instructions that do memory accesses
larger than 64 bits.

This patch adds some generic support for memory operands that are arrays
of uint64_t, and specifically a 'u2qw' operand type for x86 that is an
array of 2 uint64_ts (128 bits).  This support is unused at this point,
but will be needed shortly for cmpxchg16b.  Ideally the 128-bit SSE
memory accesses will also be rewritten to use this support.

Support for 128-bit accesses could also have been added using the gcc
__int128_t extension, which would have been less disruptive.  However,
although clang also supports __int128_t, it's still non-standard.
Also, more importantly, this approach creates a path to defining
256- and 512-byte operands as well, which will be useful for eventual
AVX support.
2016-02-06 17:21:20 -08:00
Steve Reinhardt
f5343df1e1 arch: get rid of dummy var init
MemOperand variables were being initialized to 0
"to avoid 'uninitialized variable' errors" but these
no longer seem to be a problem (with the exception of
one use case in POWER that is arguably broken and
easily fixed here).

Getting rid of the initialization is necessary to
set up a subsequent patch which extends memory
operands to possibly not be scalars, making the
'= 0' initialization no longer feasible.
2016-02-06 17:21:20 -08:00
Steve Reinhardt
92b750d5ef syscall_emul: fix bug in aux vector initialization
Writing 16 bytes from an 8-byte source value is a bad idea.
This doesn't appear to have broken anything, but showed up
as spurious differences when tracediffing runs.
2016-02-06 17:21:20 -08:00
Steve Reinhardt
f6b828d068 style: eliminate explicit boolean comparisons
Result of running 'hg m5style --skip-all --fix-control -a' to get
rid of '== true' comparisons, plus trivial manual edits to get
rid of '== false'/'== False' comparisons.

Left a couple of explicit comparisons in where they didn't seem
unreasonable:
invalid boolean comparison in src/arch/mips/interrupts.cc:155
>>        DPRINTF(Interrupt, "Interrupts OnCpuTimerINterrupt(tc) == true\n");<<
invalid boolean comparison in src/unittest/unittest.hh:110
>>            "EXPECT_FALSE(" #expr ")", (expr) == false)<<
2016-02-06 17:21:20 -08:00
Steve Reinhardt
2d91e741e8 x86: create function to check miscreg validity
In the process of trying to get rid of an '== false' comparison,
it became apparent that a slightly more involved solution was
needed.  Split this out into its own changeset since it's not
a totally trivial local change like the others.
2016-02-06 17:21:20 -08:00
Steve Reinhardt
4e65e25e25 util: fix apparent statetrace bug
In the process of trying to eliminate boolean comparisons,
I ran across this statement that appears to be a bug
(should have been an assignment).
2016-02-06 17:21:20 -08:00
Steve Reinhardt
2728f7bd81 util: added line length and boolean comparison style checkers
Added checkers for line length and boolean comparisons
(== true/== false) to the style script.
2016-02-06 17:21:20 -08:00
Steve Reinhardt
5592798865 style: fix missing spaces in control statements
Result of running 'hg m5style --skip-all --fix-control -a'.
2016-02-06 17:21:19 -08:00
Steve Reinhardt
dc8018a5c3 style: remove trailing whitespace
Result of running 'hg m5style --skip-all --fix-white -a'.
2016-02-06 17:21:18 -08:00
Steve Reinhardt
c8c82f09a2 util: clean up and extend style checker
Added a new Verifier object to check for and fix spacing
between if/while/for and following paren.

Restructured Verifier class to make it easier to add
new subclasses, particularly by using a global list of
verifiers to auto-generate command line options and
simplify the invocation loop.
2016-02-06 17:21:18 -08:00
Steve Reinhardt
9979355539 util: delete old unused style-checking scripts
The functions in these scripts were apparently folded into style.py but the
old scripts were orphaned without being deleted.  Get rid of them so their
existence is no longer confusing.
2016-02-06 17:21:18 -08:00
Mohammad Alian
24da47cf96 dist, dev: add an ethernet switch model 2016-02-06 13:33:34 -05:00
Steve Reinhardt
ce35c06c6e stats: update EIO stats for recent changes 2016-02-06 01:35:03 -05:00
Curtis Dunham
b4282af318 ext: fix SST connector
Should work with SST 5.1 and trunk as of right now.
2016-02-04 16:57:59 -06:00
Curtis Dunham
fe32b40269 ext: remove redundant parameter in example SST+gem5 test
The SST connector automatically adds --initialize-only to the gem5
"command line" (as it should); the config script doesn't need it.
2016-01-29 17:47:15 -06:00
Andreas Sandberg
b99fea78a6 ext: Update NoMali to external rev f08e0a5
Update NoMali from external revision 9adf9d6 to f08e0a5 and bring in
the following changes:

f08e0a5 Add support for tracking address space state
f11099e Fix job slot register handling when running new jobs
b28c98e api: Add a reset callback
29ac4c3 tests: Update gitignore to cover all future test cases
1c6b893 Propagate reset calls to all job slots
8f8ec15 Remove redundant reg vector in MMU
85d90d2 tests: Fix incorrect extern declaration
2016-01-29 12:14:21 +00:00
Tony Gutierrez
1285d639eb stats: update stats to after GPU checkin 2016-01-22 10:42:13 -05:00
Brad Beckmann
dcd8eeec3b ruby: removed Write_Only AccessPermission 2016-01-22 10:42:12 -05:00
Brad Beckmann
97a5e5b25e ruby: changed all references to numCPs to num-cp 2016-01-22 10:42:12 -05:00
David Hashe
698866d461 ruby: split CPU and GPU latency stats 2015-07-20 09:15:18 -05:00
Tony Gutierrez
1a7d3f9fcb gpu-compute: AMD's baseline GPU model 2016-01-19 14:28:22 -05:00
Tony Gutierrez
28e353e040 mem: write combining for ruby protocols
This patch adds support for write-combining in ruby.
2016-01-19 14:05:03 -05:00
Tony Gutierrez
d658b6e1cc * * *
mem: support for gpu-style RMWs in ruby

This patch adds support for GPU-style read-modify-write (RMW) operations in
ruby. Such atomic operations are traditionally executed at the memory controller
(instead of through an L1 cache using cache-line locking).

Currently, this patch works by propogating operation functors through the memory
system.
2016-01-19 13:57:50 -05:00