summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2017-12-14ext: Upgrade PyBind11 to version 2.2.1Jason Lowe-Power
This upgrade is necessary for pybind to build with GCC 7.2. We still need to add the patch for stl.h. MSC_FULL_VER change is no longer needed. See https://gem5-review.googlesource.com/c/public/gem5/+/2230 Change-Id: I806729217d022070583994c2dfcaa74476aef30f Signed-off-by: Jason Lowe-Power <jason@lowepower.com> Reviewed-on: https://gem5-review.googlesource.com/5801 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2017-12-14x86: Use operand size 4 when it would be 2 for cmpxchg8b.Gabe Black
This means the instruction is treated as cmpxchg8b when the effective operand size is 16 bits. Change-Id: I4d9bb295f96097e1746a9bbccb2c579d14738fab Reviewed-on: https://gem5-review.googlesource.com/6603 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Gabe Black <gabeblack@google.com>
2017-12-13scons, tests: Fix occasional linking errorAndreas Sandberg
There are some cases where scons incorrectly adds the same object multiple times to the linker command line. This seems to be caused by the test's source list being updated in place when determining test framework dependencies. Fix this by explicitly copying the source list and manipulate the copy. Without this change, the following command fails: scons ./build/ARM/unittests.opt/base/pixeltest.xml Whereas this command succeeds: scons ./build/ARM/base/pixeltest.opt Change-Id: I642efdf9d62a5478e49ba51efe1a3a5ba453e21f Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Jack Travaglini <giacomo.travaglini@arm.com> Reviewed-on: https://gem5-review.googlesource.com/6641 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Reviewed-by: Gabe Black <gabeblack@google.com>
2017-12-13scons, tests: Add support for GTest XML generationAndreas Sandberg
The GTest framework supports result generation in XML (JUnit). Enable this by creating unit test specific targets in the build/${BUILD_OPTS}/unittests.${VARIANT} directory. Targets in the directory use the following naming convention: ${SRC_PATH}/${TEST_NAME}.xml For example, the opt version of the bitunion test built for ARM would have this path: build/ARM/unittests.opt/base/bituniontest.xml Change-Id: I174dff16817734db05b08ce1d5bcf52e8697bbac Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Jack Travaglini <giacomo.travaglini@arm.com> Reviewed-on: https://gem5-review.googlesource.com/6624 Reviewed-by: Gabe Black <gabeblack@google.com>
2017-12-13scons: Make sure GTests have the right environment variablesAndreas Sandberg
SCons currently scrubs the environment variables used by GTests too aggressively. This breaks systems where libraries are installed in non-standard locations that need to be specified in LD_LIBRARY_PATH. Run said tests in the gtest_env SCons environment which white-lists the important environment variables. Change-Id: I5fc8fb5e51f09644dc976ee97b21c78ab349bf7d Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Jack Travaglini <giacomo.travaglini@arm.com> Reviewed-on: https://gem5-review.googlesource.com/6623 Reviewed-by: Gabe Black <gabeblack@google.com>
2017-12-13arm,sparc,x86,base,cpu,sim: Replace the Twin(32|64)_t types with.Gabe Black
Replace them with std::array<>s. Change-Id: I76624c87a1cd9b21c386a96147a18de92b8a8a34 Reviewed-on: https://gem5-review.googlesource.com/6602 Maintainer: Gabe Black <gabeblack@google.com> Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2017-12-13cpu,alpha,mips,power,riscv,sparc: Get rid of eaComp and memAccInst.Gabe Black
Neither of these were used, particularly memAccInst. Change-Id: I4ac9e44cf624e5de42519d586d7b699f08a2cdfc Reviewed-on: https://gem5-review.googlesource.com/6601 Maintainer: Gabe Black <gabeblack@google.com> Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
2017-12-13x86: Rework how "split" loads/stores are handled.Gabe Black
Explicitly separate the way the data is represented in the underlying representation from how it's represented in the instruction. In order to make the ISA parser happy, the Mem operand needs to have a single, particular type. To handle that with scalar types, we just used uint64_ts and then worked with values that were smaller than the maximum we could hold. To work with these new array values, we also use an underlying uint64_t for each element. To make accessing the underlying memory system more natural, when we go to actually read or write values, we translate the access into an array of the actual, correct underlying type. That way we don't have non-exact asserts which confuse gcc, or weird endianness conversion which assumes that the data should be flipped 8 bytes at a time. Because the functions involved are generally inline, the syntactic niceness should all boil off, and the final implementation in the binary should be simple and efficient for the given data types. Change-Id: I14ce7a2fe0dc2cbaf6ad4a0d19f743c45ee78e26 Reviewed-on: https://gem5-review.googlesource.com/6582 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2017-12-13base: Add endianness conversion functions for std::array types.Gabe Black
These swap the endianness of each element within the array individually. They probably obsolute the Twin(32|64)_t types which I believe were used for SPARC. Change-Id: Ic389eb24bdcdc0081068b0c5a37abdf416f6c924 Reviewed-on: https://gem5-review.googlesource.com/6581 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2017-12-12tests: Turn fbtest into a gtest and move it to src/base.Gabe Black
Change-Id: I9ca57e24f27e0eb747d1f27262972a8abcd10fc8 Reviewed-on: https://gem5-review.googlesource.com/6342 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2017-12-12tests: Move the cprintftest unit test into src/base.Gabe Black
That way it will live alongside the code it tests. Change-Id: I00baad2206870a4619b7cee792a1d4c303dad04d Reviewed-on: https://gem5-review.googlesource.com/6324 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2017-12-12tests: Convert the cprintf unit test into a gtest.Gabe Black
Change-Id: I0f78a202d1f5fd29cda94ca93b540618831fe898 Reviewed-on: https://gem5-review.googlesource.com/6323 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2017-12-12tests: Move the trietest unit test into base.Gabe Black
This puts it alongside trie.hh, the header file it tests. Change-Id: Id8ca0c1d68bdc01807c5ba4b51c0142b1221385d Reviewed-on: https://gem5-review.googlesource.com/6281 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Brandon Potter <Brandon.Potter@amd.com>
2017-12-12tests: Plumb dumps of the test trie into the gtest macros.Gabe Black
With this change, when one of the tests fails, it will output a dump of the trie data structure, making it a little easier to tell what happened. Change-Id: I0816ed727ef0b50fefd7ec485356b4fe8790bfe1 Reviewed-on: https://gem5-review.googlesource.com/6267 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
2017-12-12tests: Convert the trie unit test into a gtest.Gabe Black
Change-Id: Idcf60260d9bda1b8ef5b6f5d59b74ca218395f0c Reviewed-on: https://gem5-review.googlesource.com/6265 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-by: Brandon Potter <Brandon.Potter@amd.com>
2017-12-12tests: Add an implementation of the Logger interface for use gtests.Gabe Black
On exiting log types (panic and fatal), the message is set to an ADD_FAILURE_AT macro, and the test is exited by throwing an otherwise unexpected exception. On non-exiting log types, the message is sent to the SUCCEEDED macro which currently doesn't output anything. Change-Id: I1bb569e6cb8308dbc4c3e04eea7a962bd2b1ddd8 Reviewed-on: https://gem5-review.googlesource.com/6264 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
2017-12-12misc: Rework the logging functions.Gabe Black
Removed the "verbose" switch which wasn't used. Replaced the "get(LogLevel)" function with a get for each level. The parameter was always constant, so we can just call the right function at the right time. Made the "exit" behavior of panic/fatal a part of the logging implementation so that it can be overridden, and corrected a comment which said that both fatal and panic called ::abort(). Got rid of the printEpilogue function by reworking the print() methods. The subclasses of Logger can now override a "log" function which takes a composed message, letting the Logger class centralize how the message is put together and leaving the actual output mechanism to the subclass. Unfortunately there wasn't a way to tell gcc that the panic/fatal macros wouldn't return, so there needed to be an exit_helper wrapper function which calls the actual logger exit function. That can be marked as noreturn, unlike the virtual exit function. If the exit function does return, the wrapper will call ::abort(), placating gcc and ensuring that even if exit isn't implemented properly, exit_helper will still not return. That also provides a handy default implementation. Change-Id: I66d0cebd59f1127db980f3b565dbdf60687d8862 Reviewed-on: https://gem5-review.googlesource.com/6263 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2017-12-12config: Fix need to set ISA of switch cpus.Austin Harris
Since BaseCPU.createThreads() no longer overrides the BaseCPU.isa parameter, switch_cpus should have the ISA copied. This fixes a fatal error in BaseCPU when restoring from a checkpoint. Change-Id: I4fdcacb76da46bdbe1ce37dcf05c5a6a8a9e5237 Signed-off-by: Austin Harris <austinharris@utexas.edu> Reviewed-on: https://gem5-review.googlesource.com/6241 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2017-12-08arm: Change access permission in TPIDRURO and TPIDRURWGiacomo Travaglini
This patch corrects the TPIDRURO and TPIDRURW access flags: TPIDRURO is now readable in secure user mode, an TPIDRURW is readable and writable in secure user mode. Change-Id: I6293d9a3bcc7adc0f655bf98d29aca51eca5a002 Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com> Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Chuan Zhu <chuan.zhu@arm.com> Reviewed-on: https://gem5-review.googlesource.com/6381 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2017-12-08x86,misc: add additional info on faulting X86 instruction, fetched PCMatt Sinclair
Print faulting instruction for unmapped address panic in faults.cc and print extra info about corresponding fetched PC in base.cc. Change-Id: Id9e15d3e88df2ad6b809fb3cf9f6ae97e9e97e0f Reviewed-on: https://gem5-review.googlesource.com/6461 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2017-12-07tests: Accept SourceFilters as sources for GTest.Gabe Black
This change introduces the idea of a SourceFilter which is an object that can filter a SourceList and which can be composed with other SourceFilters using | and & operators. This means a filter can be constructed ahead of time, possibly before all sources have been discovered, and then later applied to any SourceList necessary. This change also modifies GTest so that it accepts SourceFilters in addition to normal source files. These filters will be applied to the final list of all sources, and the result included in the build for that test. By default, gtests will build in all sources tagged with 'gtest lib'. This change also introduces the keyword argument "skip_lib" which will exclude those files. They can then be left out entirely, or they can be re-included as part of a more elaborate filter. That would be useful if someone wanted to write a unit test for, for instance, the warn, etc. macros which rely on the gtest logging support. Those classes could be replaced by something under the control of the unit test, while still including the rest of the gtest library. Change-Id: I13a846dc884b86b9fdcaf809edefd57bb4168b8e Reviewed-on: https://gem5-review.googlesource.com/6262 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2017-12-07tests: Add a pseudo target to run all the unit tests for build/variant.Gabe Black
Telling scons to build build/${BUILD_OPTS}/unittests.${VARIANT} will get it to build and run all googletest based unit tests under the ${BUILD_OPTS} build options (ARM, ALPHA, X86, etc.), and compiled with the flags, etc., for the ${VARIANT} variant (ie. opt, debug, etc.). This will make it easy to run the unit tests without having to actually know where they are, what tests are available, etc. This target is called unittests* and not something based on gtest or googletest since it's my intention for all unit tests to be based on googletest, making the distinction unnecessary. Since the target is essentially part of the external interface for scons, I wanted to name it something general so it'll be less likely that we have to change it. Change-Id: I8fdec768d821974309c92a2ce4c96dce7df24fa5 Reviewed-on: https://gem5-review.googlesource.com/6282 Reviewed-by: Gabe Black <gabeblack@google.com> Reviewed-by: Brandon Potter <Brandon.Potter@amd.com> Maintainer: Gabe Black <gabeblack@google.com>
2017-12-07arch-riscv: Move compressed ops out of ISAAlec Roelke
This patch moves static portions of the compressed instruction definitions out of the ISA generated code. Change-Id: I61daae8b8c03a9e0f012790a132aa4d34a6ec296 Reviewed-on: https://gem5-review.googlesource.com/6026 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Alec Roelke <ar4jc@virginia.edu>
2017-12-06x86: Split apart x87's FSW and TOP, and add a missing break.Gabe Black
The FSW and TOP values are technically part of the same register, but they have very different behaviors. One of them can be renamed and float along without affecting global state, while the other requires serialization. They just need to *look* like the same register when read by the user. Also, there was a missing break in setMiscRegNoEffect. Change-Id: If58de0f566f65068208240f4001209fb9e1826d6 Reviewed-on: https://gem5-review.googlesource.com/6441 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Gabe Black <gabeblack@google.com>
2017-12-06misc: Update MAINTAINERS with learning-gem5 tagJason Lowe-Power
Change-Id: Ic91fb1d9b2c3c42946cb84c1ec52d9376d4e50b4 Signed-off-by: Jason Lowe-Power <jason@lowepower.com> Reviewed-on: https://gem5-review.googlesource.com/6422 Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
2017-12-06base: Split out the pixel class in framebuffer.(cc|hh).Gabe Black
These are really two separate things. Also, while it's realitively straightforward to write a unit test for the pixel conversion code, the framebuffer object is serializable and brings in more dependencies. Change-Id: If954caeb0bfedb1002cfb1a7a115a00c90d56d19 Reviewed-on: https://gem5-review.googlesource.com/6341 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2017-12-06base: Handle zero fill in cprintf when printing floats.Gabe Black
The fill_zero flag was being followed for ints, but not for floats. This makes the cprintf unit test pass. Change-Id: I4d17a3c9327aea05e0a3c81be1886c0c9256f03c Reviewed-on: https://gem5-review.googlesource.com/6322 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2017-12-06tests: Fix the source file for the cprintftime test.Gabe Black
It was using the source file for the cprintftest unit test. Change-Id: I534798e892ad55cef2f48be2ba9d732aa1993819 Reviewed-on: https://gem5-review.googlesource.com/6321 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
2017-12-06scons: Several fixes having to do with tags and sets.Gabe Black
There were a few places where tags weren't being converted to sets correctly which unfortunately only manifested when called in certain ways. This would be a pretty reasonable place to add some python unit tests... Change-Id: I87509369b4ec6f702b7521e52bf63701a87ec436 Reviewed-on: https://gem5-review.googlesource.com/6261 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2017-12-06scons: Track and reuse object nodes for a given source file.Gabe Black
scons gets upset if two different environments are used to set up a particular object file. This change adds two dicts to the SourceFile class, one for static and one for shared object files, which are keyed off of the appropriate suffix. If a suffix hasn't been set up yet, a new node of the appropriate type is set up and stored in the cache, and then whatever is in the cache (new or old) is returned. Change-Id: Ice4b4fc728b438a4d3316c3ff6667c0480d2a6d7 Reviewed-on: https://gem5-review.googlesource.com/6224 Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com> Maintainer: Gabe Black <gabeblack@google.com>
2017-12-05x86: LOOP's operand size defaults to 64 bits in 64 bit mode.Gabe Black
The microcode for those instructions needs a directive which overrides that setting in the instructions emulation environment. Reported-by: Matt Sinclair <mattdsinclair@gmail.com> Change-Id: I474d938c0b3cf01da92ec817a58b08de783f1967 Reviewed-on: https://gem5-review.googlesource.com/6301 Reviewed-by: Gabe Black <gabeblack@google.com> Maintainer: Gabe Black <gabeblack@google.com>
2017-12-05learning-gem5: Fix missing misc.hh in hello_object.ccHanhwi Jang
misc.hh has been renamed in commit 1088f0c4ac3999fc3c363cc51daef4cfb360a2bd Change-Id: Ic4f8c6423e6a5466f8d924e793a24f62bb4eca9c Reviewed-on: https://gem5-review.googlesource.com/6361 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2017-12-05arm: Add support for the dc {civac, cvac, cvau, ivac} instrNikos Nikoleris
This patch adds support for decoding and executing the following ARMv8 cache maintenance instructions by Virtual Address: * dc civac: Clean and Invalidate by Virtual Address to the Point of Coherency * dc cvac: Clean by Virtual Address to the Point of Coherency * dc cvau: Clean by Virtual Address to the Point of Unification * dc ivac: Invalidate by Virtual Addrsess to the Point of Coherency Change-Id: I58cabda37f9636105fda1b1e84a0a04965fb5670 Reviewed-by: Sudhanshu Jha <sudhanshu.jha@arm.com> Reviewed-by: Stephan Diestelhorst <stephan.diestelhorst@arm.com> Reviewed-on: https://gem5-review.googlesource.com/5060 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
2017-12-05arm: Add support for the mcr dc{ic,i,c}mvac, dccmvau instructionsNikos Nikoleris
This patch adds support for the ARMv7 cache maintenance intructions: * mcr dccmvac cleans a VA to the PoC * mcr dcimvac invalidates a VA to the PoC * mcr dccimvac cleans and invalidates a VA to the PoC * mcr dccmvau cleans a VA to the PoU Change-Id: I6511f203039ca145cc9128ddf61d09d6d7e40c10 Reviewed-by: Stephan Diestelhorst <stephan.diestelhorst@arm.com> Reviewed-by: Anouk Van Laer <anouk.vanlaer@arm.com> Reviewed-on: https://gem5-review.googlesource.com/5059 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2017-12-05mem-ruby: Prevent ruby from crashing on CMOsNikos Nikoleris
Ruby has no support for cache maintenace operations. As a workaround, after printing a warning, we treat them as no-ops in the memory system and respond immediately without handling them. There should be workarounds in the memory system already that allow execution to proceed without the requirement for cache maintenance operations. Change-Id: I125ee4fa37b674c636d87f2d9205bbc1a74da101 Reviewed-on: https://gem5-review.googlesource.com/5057 Reviewed-by: Jieming Yin <bjm419@gmail.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2017-12-05arm: Add CMO support for Non-Cacheable memoryNikos Nikoleris
Cache Maintainance operations to the point of coherence are treated as normal cahceable requests and clean and/or invalidate the caches of all PEs. Change-Id: Ia4a749c2318fe29c8601848b034b8315c4186c8a Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-on: https://gem5-review.googlesource.com/5056 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2017-12-05cpu: Add support for CMOs in the cpu modelsNikos Nikoleris
Cache maintenance operations go through the write channel of the cpu. This changes makes sure that the cpu does not try to fill in the packet with data. Change-Id: Ic83205bb1cda7967636d88f15adcb475eb38d158 Reviewed-by: Stephan Diestelhorst <stephan.diestelhorst@arm.com> Reviewed-on: https://gem5-review.googlesource.com/5055 Maintainer: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2017-12-05mem: Ignore clean requests in the abstract memoryNikos Nikoleris
Systems with atomic cores and the fastmem option enabled bypass the whole memory system and access the abstract memory directly. Cache maintenance operations which would be normally handled before the point of unification/coherence should be ignored by the abstract memory. Change-Id: I696cdd158222e5fd67f670cddbcf2efbbfd5eca4 Reviewed-by: Curtis Dunham <curtis.dunham@arm.com> Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-on: https://gem5-review.googlesource.com/5054 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Jason Lowe-Power <jason@lowepower.com>
2017-12-05mem: Handle CMO responses in the snoop filterNikos Nikoleris
Previously responses would either transfer the ownership of the line or the actual data to the cache that send out the original request. Cache clean operations are different since they bring neither data nor ownership. When they are also invalidating the cache that send out the original request will invalidate any existing copies. This patch makes the snoop filter handle the cache clean responses accordingly. Change-Id: I27165cb45b9dc57882526329c62db35f100d23df Reviewed-by: Stephan Diestelhorst <stephan.diestelhorst@arm.com> Reviewed-by: Sudhanshu Jha <sudhanshu.jha@arm.com> Reviewed-by: Anouk Van Laer <anouk.vanlaer@arm.com> Reviewed-on: https://gem5-review.googlesource.com/5053 Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2017-12-05mem: Allow CMOs as snooping requests in the snoop filterNikos Nikoleris
The snoop filter performs sanity checks of the type of packets that are expected to snoop caches above. Cache maintenace operations are expected to perform a clean and or invalidate on all caches down to the specified point of reference and therefore could also generate snoops. Change-Id: I7f8fef246a85faa87ccd289c28b49686ed7caa08 Reviewed-by: Stephan Diestelhorst <stephan.diestelhorst@arm.com> Reviewed-by: Anouk Van Laer <anouk.vanlaer@arm.com> Reviewed-on: https://gem5-review.googlesource.com/5052 Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2017-12-05mem: Co-ordination of CMOs in the xbarNikos Nikoleris
A clean packet request serving a cache maintenance operation (CMO) visits all memories down to the specified xbar. The visited caches invalidate their copy (if the CMO is invalidating) and if a dirty copy is found a write packet writes the dirty data to the memory level below the specified xbar. A response is send back when all the caches are clean and/or invalidated and the specified xbar has seen the write packet. This patch adds the following functionality in the xbar: 1) Accounts for the cache clean requests that go through the xbar 2) Generates the cache clean response when both the cache clean request and the corresponding writeclean packet has crossed the destination xbar. Previously transactions in the xbar were identified using the pointer of the original request. Cache clean transactions comprise of two different packets, the clean request and the writeclean, and therefore have different request pointers. This patch adds support for custom transaction IDs that by default take the value of the request pointer but can be overriden by the contructor. This allows the clean request and writeclean share the same id which the coherent xbar uses to co-ordinate them and send the response in a timely manner. Change-Id: I80db76386a1caded38dc66e6e18f930c3bb800ff Reviewed-by: Stephan Diestelhorst <stephan.diestelhorst@arm.com> Reviewed-on: https://gem5-review.googlesource.com/5051 Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2017-12-05mem: Add support for handling CMOs in the MSHRsNikos Nikoleris
To add support for cache maintenance operations (CMOs) in the MSHRs, this change adds the following functionality: - If a CMO request hits in the MSHRs, we deferred as we can't coalesce it with any other requests. - When we promote any deferred targets, we promote them in order and stop if we encounter a CMO request. If the CMO request is at the beginning of the deferred targets list it will be the only promoted target. Change-Id: I10d1f7e16bd6d522d917279c5d408a3f0cee4286 Reviewed-by: Stephan Diestelhorst <stephan.diestelhorst@arm.com> Reviewed-on: https://gem5-review.googlesource.com/5050 Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2017-12-05mem: Add support for CMOs in the cacheNikos Nikoleris
This change adds support for maintenance operations (CMOs) in the cache. The supported memory operations clean and/or invalidate a cache block as specified by its VA to the specified xbar (PoU, PoC). A cache maintenance packet visits all memories down to the specified xbar. Caches need to invalidate their copy if it is an invalidating CMO. If it is (additionally) a cleaning CMO and a dirty copy exists, the cache cleans it with a WriteClean request. Change-Id: Ibf31daa7213925898f3408738b11b1dd76c90b79 Reviewed-by: Stephan Diestelhorst <stephan.diestelhorst@arm.com> Reviewed-on: https://gem5-review.googlesource.com/5049 Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2017-12-05mem: Promote deferred targets only when the block is validNikos Nikoleris
When a response indicates that there are no other sharers of the block, the cache can promote its copy of the block to writable and potential service deferred targets even if the request didn't ask for a writable copy. Previously, a response would guarantee the presence of the block in the cache. A response could either be filling, upgrading or a response to an invalidation due to a pending whole line write. Responses to cache maintenance invalidations break this assumption. This change adds an extra check to make sure that the block was already valid or that the response is filling before promoting the block. Change-Id: I6839f683a05d4dad4205c23f365a925b7b05e366 Reviewed-by: Curtis Dunham <curtis.dunham@arm.com> Reviewed-by: Anouk Van Laer <anouk.vanlaer@arm.com> Reviewed-on: https://gem5-review.googlesource.com/5048 Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2017-12-05mem: Add support for cache maintenance operation requestsNikos Nikoleris
This change adds new packet cmds and request flags for cache maintenance operations. 1) A cache clean operation writes dirty data in the first memory below the specified xbar and updates any old copies in the memories above it. 2) A cache invalidate operation invalidates all copies of the specified block in the memories above the specified xbar 3) A clean and invalidate operation is a combination of the two operations above Change-Id: If45702848bdd568de532cd57cba58499e5e4354c Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Curtis Dunham <curtis.dunham@arm.com> Reviewed-by: Anouk Van Laer <anouk.vanlaer@arm.com> Reviewed-on: https://gem5-review.googlesource.com/5047 Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2017-12-05mem: Support for specifying the destination of a WriteCleanNikos Nikoleris
Previously, WriteClean packets would always write to the first memory below unless the memory was unable to allocate in which case it would be forwarded further below. This change adds support for specifying the destination of a WriteClean packet. The cache annotates the request with the specified destination and marks the packet as write-through upon its creation. The coherent xbar checks packets for their destination and resets the write-through flag when necessary e.g., the coherent xbar that is set as the PoC will reset the write-through flag for packets to the PoC. Change-Id: I84b653f5cb6e46e97e09508649a3725d72d94606 Reviewed-by: Curtis Dunham <curtis.dunham@arm.com> Reviewed-by: Anouk Van Laer <anouk.vanlaer@arm.com> Reviewed-on: https://gem5-review.googlesource.com/5046 Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2017-12-05mem: Add support for WriteClean packets in the memory systemNikos Nikoleris
This change adds support for creating and handling WriteClean packets. The WriteClean operation is almost identical to a WritebackDirty with the exception that the cache generating a WriteClean retains a copy of the block. Change-Id: I63c8de62919fad0f9547d412f8266aa4292ebecd Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Curtis Dunham <curtis.dunham@arm.com> Reviewed-by: Anouk Van Laer <anouk.vanlaer@arm.com> Reviewed-on: https://gem5-review.googlesource.com/5045 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com>
2017-12-05mem: Add a WriteClean command to the packet classNikos Nikoleris
A WriteClean packet allows a cache to write a block to a memory below without evicting its copy. A typical usecase for a WriteClean packet is a cache clean operation. Change-Id: If356cb067da5ddf3210c135f41ef0891fb811568 Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Reviewed-by: Curtis Dunham <curtis.dunham@arm.com> Reviewed-by: Anouk Van Laer <anouk.vanlaer@arm.com> Reviewed-on: https://gem5-review.googlesource.com/5044 Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
2017-12-05mem-cache: Add support for checking whether a cache is busyNikos Nikoleris
This changeset adds support for checking whether the cache is currently busy and a timing request would be rejected. Change-Id: I5e37b011b2387b1fa1c9e687b9be545f06ffb5f5 Reviewed-on: https://gem5-review.googlesource.com/5042 Reviewed-by: Jason Lowe-Power <jason@lowepower.com> Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com> Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com>
2017-12-05mem: Add function to check if the slave can receive a timing reqNikos Nikoleris
This changeset adds support for tryTiming, an interface that allows a master to check if the slave is busy or otherwise if it can accept a timing request. Change-Id: Idc7c2337ae9ccf5dec54f308e488660591419a63 Reviewed-on: https://gem5-review.googlesource.com/5041 Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com> Reviewed-by: Christian Menard <christian.menard@tu-dresden.de> Reviewed-by: Jason Lowe-Power <jason@lowepower.com>