Gem5Fixture is used to define a fixture for building the gem5
binary. Most tests are expected to define their own Gem5Fixture,
however, as some might depend on the same binary (e.g.,
./build/ARM/gem5.opt), they will try to re-define a fixture for the
same target. This patchset changes Gem5Fixture to derive from
UniqueFixture.
In addition, this patchset changes the way global fixtures are
discovered to work with the new Gem5Fixture class. Instead of
enumerating them when test definitions are loaded, we do so after the
tests have been filtered according to specified tags (e.g., include
opt variant, exclude fast, debug variants).
Change-Id: Ie868a7e18ef6c3271f3c8a658229657cd43997cb
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/19251
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Maintainer: Giacomo Travaglini <giacomo.travaglini@arm.com>
This changeset adds a test to check the redirection features
added in faux-filesystem changeset. The test contains a
"chdir" system call to "/proc" which should be redirected to
"$(gem5-dir)/m5out/fs/proc" (as specified by the config files).
After "chdir", the test subsequently outputs the "/proc/cpuinfo"
file which should output a configuration of a fake cpu with
values set by a Python configuration file.
Note, the test will call "clone" once. To avoid a runtime error,
make sure that you run this test with "-n2" supplied to the
"config/example/se.py" script.
Change-Id: I505b046b7a4feddfa93a6ef0f0773ac43078cc94
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/17112
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
Fix problem with O3 and AMO instructions. At initial stages amo
instruction is considered a type of non-speculative store. After
the instruction has been commited and during the squash step,
acquire_release version of the AMO operation is considered speculative,
that differents results in an assert fault. This fix ensures that AMO
instructions are always considered non-speculative, during early stages
and during squas/removal of the instruction.
Change-Id: Ia0c5fbb9dc44a9991337b57eb759b1ed08e4149e
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/19815
Maintainer: Jason Lowe-Power <jason@lowepower.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Tested-by: kokoro <noreply+kokoro@google.com>
In the current SMMUv3 model, multiple micro/mainTLB are present at the
device interface (SMMUv3SlaveInterface), caching translations specific
to a device.
Those distributed TLBs are checked for a translation before checking for
centralized TLBs (shared by devices), like the configuration cache, walk
cache etc. This means that if a hit in these TLBs occurs, there won't
be a need to enter configuration stage (which is where the STE and CD
are retrieved). So if we invalidate a cached configuration (in
ConfigCache), we need to invalidate those interface TLB entries as well,
otherwise in theory we will keep the same translation even after a
change in configuration tables.
Change-Id: I4aa36ba8392a530267517bef7562318b282bee25
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-by: Michiel van Tol <michiel.vantol@arm.com>
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/19813
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
ARMv8.1-PAN adds a new bit to PSTATE. When the value of this PAN state
bit is 1, any privileged data access from EL1 or EL2 to a virtual memory
address that is accessible at EL0 generates a Permission fault.
This feature is mandatory in ARMv8.1 implementations.
This feature is supported in AArch64 and AArch32 states.
The ID_AA64MMFR1_EL1.PAN, ID_MMFR3_EL1.PAN, and ID_MMFR3.PAN fields
identify the support for ARMv8.1-PAN.
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Change-Id: I94a76311711739dd2394c72944d88ba9321fd159
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/19729
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
MSR <pstatefield>, #imm is used for setting a PSTATE field using an
immediate. Current implementation has the following flaws:
* There is no base MSR immediate definition: all the existing
PSTATE fields have a different class definition
* Those implementation make use of a generic data64 base class
which results in a wrong disassembly (pstate register is printed as an
integer register).
This patch is fixing this by defining a new base class (MiscRegImmOp64)
and new related templates. In this way, we aim to ease addition of new
PSTATE fields (in ARMv8.x)
Change-Id: I71b630ff32abe1b105bbb3ab5781c6589b67d419
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/19728
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
The fstat64 system call does an upcast on entries in the file
descriptor array to check if the file descriptor has a backing
host-filesystem file opened. It does so because it needs to pass
the host fd into the fstat call (since we rely on the host
filesystem to service filesystem system calls).
The upcast was overly specific. This changeset alters the system
call to use the most general base class of the file descriptor
entries that can satisfy the code.
Change-Id: I10daf820257cea4d678ee6917e01e9cc9cd1cf5e
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/17110
Tested-by: kokoro <noreply+kokoro@google.com>
Maintainer: Brandon Potter <Brandon.Potter@amd.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
This addresses the issue described in
64687ee mem-cache: Mark block as dirty after a SWPrefetchEXResp.
Previous patch misses cases when the prefetch response is ReadExResp or
UpgradeResp. Also, marking the block as dirty in serviceMSHRTargets
instead of in handleFill covers cases when the prefetch is coalesced with
other requests.
Change-Id: I2b377fdd240eb0f09e720b6bb284dee6545925ce
Signed-off-by: Tiago Mück <tiago.muck@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/19688
Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Reviewed-by: Daniel Carvalho <odanrc@yahoo.com.br>
Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This patch is rewriting the SMMUv3::processCommand method for the
following reasons:
* Command names were not matching spec
* Command encoding/opcode was wrong
The patch is not adding any new command: there is still a subset of
unimplemented commands; those are:
* CMD_TLBI_EL3_ALL
* CMD_TLBI_EL3_VA
* CMD_TLBI_EL2_ALL
* CMD_TLBI_EL2_VA
* CMD_TLBI_EL2_VAA
* CMD_TLBI_EL2_ASID
which require StreamWorld support, and
* CMD_ATC_INV
* CMD_PRI_RESP
* CMD_RESUME
* CMD_STALL_TERM
which require in sequence: ATS, PRI, Stall Model support
Change-Id: Ia2dd47b5588738402d9584a00cfc88c94c253ad0
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-by: Michiel van Tol <michiel.vantol@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/19668
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Add this check because Gicv3 does not have the cpu_addr attribute.
Test: Change VExpress_GEM5_V1() to VExpress_GEM5_V2() and run the
following command to boot Debian.
M5_PATH=$PWD/fs_files ./build/ARM/gem5.opt ./configs/example/arm/fs_bigLITTLE.py \
--dtb $PWD/fs_files/binaries/armv8_gem5_v2_1cpu.dtb \
--kernel $PWD/fs_files/binaries/vmlinux \
--disk $PWD/fs_files/disks/disk.img \
--cpu-type atomic --big-cpus 1 --little-cpus 0
Change-Id: I23595ae5238dc7cc915ab09300f91aa5e8c24fdc
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/19648
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Giacomo Travaglini <giacomo.travaglini@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
The assert() in the LSQ writeback() only allowed ReExec faults.
However, a SplitRequest which completed the translation in
PartialFault state (i.e. any but the very first cacheline
translation failed) may end up here. The assert() condition is
extended accordingly.
The patch also removes the superfluous/unused Complete/Squashed
states from the LSQ request. (The completion of the request is
recorded in the flags still.)
Change-Id: Ie575f4d3b4d5295585828ad8c7d3f4c7c1fe15d0
Signed-off-by: Gabor Dozsa <gabor.dozsa@arm.com>
Reviewed-by: Giacomo Gabrielli <giacomo.gabrielli@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/19174
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
SMMU circular queues have a wrap bit which is used in order to
distinguish between an empty queue and a full queue.
According to SMMUv3 spec:
Each index has a wrap flag, represented by the next higher bit adjacent
to the index value contained in PROD and CONS. This bit must toggle each
time the index wraps off the high end and back onto the low end of the
buffer. It is the responsibility of the owner of each index, producer or
consumer, to toggle this bit when the owner updates the index after
wrapping. It is intended that software reads the register, increments or
wraps the index (toggling wrap when required) and writes back both wrap
and index fields at the same time.
Change-Id: Idfeb397141f3627c2878caaeaa2625fadf671d2a
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-by: Michiel Van Tol <michiel.vantol@arm.com>
Reviewed-by: Adrian Herrera <adrian.herrera@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/19311
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Ciro Santilli <ciro.santilli@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
The patch is aiming to be spec compliant when it comes to setup
the SMMU command queue (while CR0.CMDQEN = 0), in the following ways:
* Writes to CMDQ_CONS (read index) are allowed during initialization
* Writes to CMDQ_BASE (cmdq pointer) are allowed during initialization
According to spec,
If they happen when the command queue is in fuction (CR0.CMDQEN = 1),
behaviour is constrained unpredictable, with the following options
1) The write is ignored
2) The register takes the value and it is unpredictable whether it
affects the SMMU command queue internal state.
In the model/patch we go for option 1.
Change-Id: I1c55bc571a8b3a1c0b0a525e429ab7b1480544ff
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-by: Michiel Van Tol <michiel.vantol@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/19633
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Ruby caches block incoming ports with messages on a locked
address to make sure the line would not be replaced by others.
But they do not check the lock upon capacity/conflict misses.
This change adds a new slicc statement "check_on_cache_probe" which takes
two arguments (mandatoryQueue for the controller, and the line subject
to eviction - i.e. address returned by cacheProbe).
If the line is locked, incoming message is delayed for 1 cycle and the
controller skips this request (i.e. does not trigger an event).
Coherence protocols should be updated accordingly. One use case for MESI
Two Level will be added in a separate change.
Signed-off-by: Pouya Fotouhi <pfotouhi@ucdavis.edu>
Change-Id: I79ca2d45518de7a4e382b520a11f8e221b0cb803
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/16808
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Srikant Bharadwaj <srikant.bharadwaj@amd.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
If a page table walk is squashed, the walker state is being deleted
in the squash code. If there are in flight requests, the deleted
walker state values may be clobbered, leading to undefined behavior.
This adds a squashed boolean to the walker state which is set if a
walk is squashed while requests are still in flight. When packets
for the in flight request return, we check if the walk was squashed
and return that the walk is complete once the number of in flight
requests reaches zero. The walker state is then freed by the PTW.
Change-Id: I57a64b1548b83a8a9e8441fc9d6f33e9842df2b3
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/19568
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
Tested-by: kokoro <noreply+kokoro@google.com>
According to the armarm:
ARMv8.1-HPD introduces the facility to disable the hierarchical
attributes, APTable, PXNTable, and UXNTable, in the translation tables.
This disable has no effect on the NSTable bit. This feature is
mandatory in ARMv8.1 implementations.
This feature is added only to the VMSAv8-64 translation regimes. ARMv8.2
extends this to the AArch32 translation regimes, see ARMv8.2-AA32HPD.
The ID_AA64MMFR1_EL1.HPDS field identifies the support for ARMv8.1-HPD.
Change-Id: Ibbf589b82f2c1e4437b43252f8f633e8f6fb0b80
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-by: Ciro Santilli <ciro.santilli@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/19610
Tested-by: kokoro <noreply+kokoro@google.com>
