The requirement to have an environment variable exported to run a program
is not common, and many new users trip up on it.
Before this commit, M5_PATH was a requirement to run those scripts, or
else simulation would fail with:
IOError: Can't find a path to system files.
After this patch, as long as users indicate all required files with
command line options, M5_PATH is not needed.
This patch changes the M5_PATH semantics slightly to more closely match
PATH and so be more intuitive to users: after this commit, if the
given path contains a slash /, then the path is not searched for inside
M5_PATH, which is exactly how PATH works. Users can then select images
in the CWD with a leading ./ just as done for executables.
This is backwards incompatible if users were already specifying their paths
as ./, but this interface feels saner, because otherwise writing on the CLI
e.g.:
--disk-image path/to/my.disk
would previously fail to find the disk, even if it existed, which is very
counter-intuitive. The following will still fail however:
--disk-image my.disk
which is not ideal, but for now is a comprise between backwards
compatibility of having an M5_PATH and what users expect from CLI
interfaces.
Change-Id: Ic91e1cc20557b35b69490b6dc420e7d324fae1fc
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/23672
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
Tested-by: kokoro <noreply+kokoro@google.com>
All ISAs except SPARC can now take multiple disk images by passing
the --disk-image option multiple times.
Before this patch, several ISAs automatically mounted a secondary disk
called "linux-bigswap2.img", which had to be in M5_PATH even if the end
user did not want more than one disk. This was the case for for example
for X86 but not ARM.
This change was done to:
* allow ARM to have a second disk image in fs.py, which was not possible,
and allow other ISAs like X86 and ARM to take any number of disk images
* provide a simpler, more intuitive CLI interface that does not require
magic disk images to be present in M5_PATH to work for ISAs such as X86.
Linux does not need that secondary image to boot correctly, so it is
more friendly to support a minimal setup that requires the least amount
of binaries to boot, and let supply the second image manually only if
they need it.
* make fs.py --disk-image work more similarly across all ISAs
SPARC was left with a single disk only because its setup was a bit more
complex and would require further testing.
Change-Id: I8b6e08ae6daf0a5b6cd1d57d285a9677f01eb7ad
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/23671
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Giacomo Travaglini <giacomo.travaglini@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
CNTFRQ_EL0 should be initialised to a uniform value in all cores present
in the system. Previously, this was only done if EL3 was present,
however architecture states CNTFRQ_EL0 may be written from the highest
EL implemented.
This patch moves this initilization outside of the EL3-only one.
Change-Id: Ibaa197de53d531ba898e5137ba4f46a8c9554699
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/24683
Reviewed-by: Gabe Black <gabeblack@google.com>
Maintainer: Giacomo Travaglini <giacomo.travaglini@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Some ABIs (including 32 bit ARM, 64 bit x86) allocate their argument
registers differently depending on their return value. For instance,
if the value needs to be returned in memory because it's too big,
the caller could pass a pointer to where the result should be stored
when the function returns. This pointer acts like an invisible first
argument, offsetting where all the normal arguments actually live.
This change adds a mechanism to handle that case. The Result templates
can now declare an allocate() static method which is given a
ThreadContext *, and a reference to the Position object. It can perform
any adjustment it needs to before the normal argument extraction
starts.
Change-Id: Ibda9095f0e8c9882742d24f5effe309ccb514188
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/23747
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
A recent-ish change modified ExeTraceRecord::traceInst to make it more
consistent with DPRINTF-s by using dprintf_flag to print the trace
string. The generated string was passed as the format however, and that
means that all % characters in the output (from register names, for
example) are interpreted as format characters, mangling the output and
making cprintf angry since there are no corresponding arguments.
This change sets the format to "%s" instead, and passes the trace
string as the first argument. The argument won't be parsed for format
specifiers, and so should no longer get mangled.
Change-Id: I8fa9c2c22179a5b55104a618a4af4080a3931c5f
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/24643
Reviewed-by: Ciro Santilli <ciro.santilli@arm.com>
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
It looks like this function is supposed to allow you to set up a PC
based event which will trigger when the simulator executes a particular
kernel function. That event doesn't actually do anything, but you can
set a breakpoint there with gdb when debugging gem5 itself.
There are a couple of problems with this function. First, it assumes
that you want to set the breakpoint based on the first system in your
simulation. Frequently simulations have only one system, but there
isn't any rule that says they must, or any way to pick a different
system.
Second, this function assumes that you're in FS mode, that there is a
kernel, and that there is a kernel symbol table to look symbols up in.
On top of that, this function is a bit redundant since you can just use
gdb to debug the kernel inside a simulated system.
Change-Id: I8dadbd42fc7d4ccba2a035a2a72e6ede4b872f3c
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/24644
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
Tested-by: kokoro <noreply+kokoro@google.com>
TheISA::initCPU is basically an ISA specific implementation of reset
logic on architectural state. As such, it only needs to be called if
we're not going to load a checkpoint, ie in initState.
Also, since the implementation was the same across all CPUs, this
change collapses all the individual implementations down into the base
CPU class.
Change-Id: Id68133fd7f31619c90bf7b3aad35ae20871acaa4
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/24189
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
The build was failing with:
/usr/bin/ld: unrecognized option '--as-needed -fuse-ld=gold'
and --verbose confirms that a single quoted CLI parameter was being
executed:
"-Wl,--as-needed -fuse-ld=gold"
This happened because at Ifb001786a66b0dd9b29865e39a5740313002f250
--as-needed was added, and because it is the second option to happen before
the following main.subst, it exposed the fact that the existing main.subst
was wrong, because it returns a string instead of the expected array.
Change-Id: I619d242d60fe9d27438638ac11c2b92512881f26
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/24624
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Reviewed-by: Gabe Black <gabeblack@google.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This patch adds the reference 32KHz clock to VExpress_GEM5_Base derived
platforms. This is in preparation for supporting the SP805 Watchdog.
I/O voltage domain and platform clock domain coupling is transferred
to the __init__ method for correctness.
Change-Id: Ic743fd986793f1e43b75fa60260c9b43b2737763
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/24204
Maintainer: Giacomo Travaglini <giacomo.travaglini@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
A bug has been introduced with the new test url.
The line break should have used a backslash or (this is the recommended
way by PEP8) the implied line continuation via parenthesis.
This error was preventing the test to be loaded with the error message:
Exception thrown while loading
"/tmpfs/src/git/jenkins-gem5-prod/tests/gem5/m5_util/test_exit.py"
Ignoring all tests in this file.
and was not producing a failure (the test was not run: it was jus
ignored).
Change-Id: I0afe252d66d2f6546caaf5e7be811f34f88df82c
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/+/24625
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
Tested-by: kokoro <noreply+kokoro@google.com>
All of the state being checkpointed would either be provided by the
config directly, or would be brought into the TLB through normal fill
operations. Having this state in the checkpoint complicates the
checkpoint and significantly decreases compatibility with other TLB
implementations, or even variations of the same TLB, for instance if
the size was changed.
Change-Id: I4ea079dd01ff18fbc458b3aaaf88519dbcfdd869
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/24389
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Maintainer: Giacomo Travaglini <giacomo.travaglini@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
The helper is meant to check if the local binary is younger than the
remote binary (on gem5.org). If the call fails it is giving up and
it is just using the local regression (producing a warning).
The code is not handling the blocking behaviour of the connection:
simulaton might stall indefinitely
The patch is addressing this by providing a 10 seconds timeout.
Change-Id: I8f9c2e555c9a55d850a66d02f8e55f56ceda2ca3
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/24531
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
So far lots of tests will download binaries inside the gem5 directory.
The path is also specific to the test being run.
This doesn't play well with an environment where gem5 is cloned from
scratch for every build, or if several gem5 are cloned in a single
machine.
Binaries will be automatically downloaded every time this happens.
This patch is adding a --bin-path option, so that it's possible to
setup a fixed directory with all pre-downloaded binaries.
By default it is set to None to preserve original behaviour.
Change-Id: I42fb25e3ce0a495c73672b15a097b1bd2607795c
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/+/24525
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Reviewed-by: Bobby R. Bruce <bbruce@ucdavis.edu>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Add an invalidation function to the AssociativeSet, so that entries
can be properly invalidated by also invalidating their replacement
data.
Both setInvalid and reset have been merged into invalidate to
indicate users that they are using an incorrect approach by
generating compilation errors, and to match CacheBlk's naming
convention.
Change-Id: I568076a3b5adda8b1311d9498b086c0dab457a14
Signed-off-by: Daniel R. Carvalho <odanrc@yahoo.com.br>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/24529
Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Maintainer: Nikos Nikoleris <nikos.nikoleris@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
First, remove a deprecated flag that gcc no longer recognizes.
Second, disable suffix based implicit makefile rules. These, in
combination with the %.o: boot.S rule, were tricking make into deleting
it's own makefile. How, you might ask?
make wants to update its makefile, since that's a thing it does
automatically. This is useful if you, for instance, have computed
header dependencies.
make decides it can make a file called "makefile" from a file called
"makefile.o" by doing a linking step.
make decides it can make makefile.o from boot.S from the %.o: boot.S
rule, which it does.
It then attempts to link makefile.o into makefile, but that fails
because it lacks a "main" function since it's using a built in rule
which doesn't know not to expect main. The makefile is clobbered in the
process.
make then deletes makefile.o because it was an implicit target,
eliminating all the evidence.
Change-Id: Ib0dfc333dc554caf5772dd8468dba6ba821f98ac
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/24329
Reviewed-by: Chun-Chen TK Hsu <chunchenhsu@google.com>
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Maintainer: Giacomo Travaglini <giacomo.travaglini@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
