Some parts of clone are architecture dependent. In some cases, we are
able to use architecture-specific helper functions or register
aliases. However, there is still some architecture-specific that is
protected by ifdefs in the common clone implementation.
Move these architecture-specific bits to the architecture-specific OS
class instead to avoid these ifdefs and make the code a bit more
readable.
Change-Id: Ia0903d738d0ba890863bddfa77e3b717db7f45de
Signed-off-by: Andreas Sandberg <andreas.sandberg@arm.com>
Cc: Giacomo Travaglini <giacomo.travaglini@arm.com>
Cc: Javier Setoain <javier.setoain@arm.com>
Cc: Brandon Potter <Brandon.Potter@amd.com>
Reviewed-on: https://gem5-review.googlesource.com/c/15435
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Brandon Potter <Brandon.Potter@amd.com>
The __to_number helper function defined in base/str.hh is used by
unserializing code. Its purpose is to convert a string into an
integral/floating point number. Since enums underlying type can only be
an integer type, it makes sense to extend the helper function for enums
as well. In this way it will be possible to unserialize Enums and
containers of Enums without the need of casting.
Change-Id: I74069cc4c04ec8b5eb80939acea7ab18fb366dd4
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-by: Ciro Santilli <ciro.santilli@arm.com>
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/15336
Reviewed-by: Gabe Black <gabeblack@google.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
The sc_event_finder class in Accellera's implementation has a
non-standard report_error function that it uses internally. The TLM
headers were calling that function in their own event finder subclass.
This change replaces that call with what should be an equivalent bit of
code which is based on the report_error implementation.
Change-Id: Id57d26791df01403a77e70d5f4a00f650dc33599
Reviewed-on: https://gem5-review.googlesource.com/c/15063
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
The former is either the same as the later, or a custom wrapper if
C++11 isn't supported. Since we generally expect C++11 support, we can
skip the indirection.
Change-Id: I9a45e3854bb7cc56d094e3fe2773fe4b5c94403b
Reviewed-on: https://gem5-review.googlesource.com/c/15297
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
This makes those files more consistent with other headers in gem5. The
top level headers (tlm and tlm.h) and the tlm_utils headers were left
alone since the user might reasonably expect those to have a well known
name. The tlm_core headers by comparison are an implementation detail,
and users shouldn't include them directly by name.
Change-Id: Ibc82e8159e47717c79d76f50ed96fdd619294a06
Reviewed-on: https://gem5-review.googlesource.com/c/15067
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
One TLM test will complain if argc isn't 1 or 2, assuming that that
must mean that argc > 2. If it's 0 then the test will also complain and
fail. We therefore need to pass it at least a dummy value in argv/argc.
Change-Id: I5c64856f46d1459d7238e88ad8ba06933c7c38b8
Reviewed-on: https://gem5-review.googlesource.com/c/15065
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
A lot of TLM implementation assumes the string functions (memcpy,
memset, etc) are available, but don't include cstring. Lets do that
for them to avoid having to fix up a lot of places.
Change-Id: I1f26d3a227fe2cefc2613e24cbb0333216e46ee9
Reviewed-on: https://gem5-review.googlesource.com/c/15060
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
We already know those files belong to tlm because of the directory
they're in. Removing the prefix makes the paths of the headers less
enormously long.
Change-Id: I869e58fae904162f353bb31f4c0919fba08dffa6
Reviewed-on: https://gem5-review.googlesource.com/c/15059
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
The TLM event finder class was calling implementation specific
functions from the Accellera version of systemc. This change replaces those
calls with equivalent calls which match the gem5 implementation.
Change-Id: I0ecdb0a4bf09aeb1aad823a01105fbd88edb4601
Reviewed-on: https://gem5-review.googlesource.com/c/15075
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
Both basic systemc and tlm tests were present, but scons only looked
in the systemc directory when populating the test json. This change
makes it also look in the tlm directory so that those tests can be run.
Change-Id: Id65b744664350f6105fb3a4f28cbc7ab91d8c82e
Reviewed-on: https://gem5-review.googlesource.com/c/15056
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
These headers will need to be cleaned up and have some Accellera
specific quirks ironed out of them, but I'll do that in a later change
to make it clear what those changes are.
Change-Id: Ia4e08633ab552b4c616c66c9b7e2bbd78ebfe7b9
Reviewed-on: https://gem5-review.googlesource.com/c/15055
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
Up until Apple LLVM version 10.0.0 (clang-1000.11.45.5), partial
linked objects do not expose symbols that are marked with the hidden
visibility and consequently building gem5 on Mac OS fails. As a
workaround, we disable partial linking, however, we may want to
revisit in the future.
Change-Id: I0a26dae082bf723c2bd49d90e4497e44ecab9c41
Signed-off-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/15235
Reviewed-by: Andrea Mondelli <andrea.mondelli@ucf.edu>
Reviewed-by: Gabe Black <gabeblack@google.com>
Maintainer: Gabe Black <gabeblack@google.com>
The PL011 spec specifies a DMACR register at offset 0x48, which isn't
implemented in the model. Currently any attempt to access the register
results in a panic.
This change swaps the panic for a warning only when software writes into
DMACR to disable DMA, keeping the panic otherwise.
Change-Id: I04586b52df8d5d174536276fd7ae19e77ff4681a
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/15279
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
Implementation registers for the GICv2 model currently hold values
referring to a GIC-400 implementation. This patch is making them
parametrizable so that it is possible to instantiate a GIC-400 model.
The patch is also modifying Realview platform to use new GIC-400 model
in lieau of GICv2.
Change-Id: I446db8c796ee3c2708af91e9139f0a6e7947321b
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/15277
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
sc_in::bind and sc_core::sc_object::print shadow their respective base
class method. This change adds the necessary declarations to inform
the compiler that this is intentional.
Change-Id: I8c6bc0f05aa568630783db63d52be90f4ea4748c
Signed-off-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/15135
Reviewed-by: Gabe Black <gabeblack@google.com>
Maintainer: Gabe Black <gabeblack@google.com>
LLVM on Mac OS up until version 10.0.0 doesn't seem to support the
attribute for weak linking and fails to compile. Additionally, even
the basic example tests fail. Therefore, this change disables systemc
by default on Mac OS until we address these issues.
Change-Id: Iaa6360ec61317e01f4e645af8b509961b3144282
Signed-off-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/14818
Reviewed-by: Gabe Black <gabeblack@google.com>
Maintainer: Gabe Black <gabeblack@google.com>
Any operands which use read_code or write_code would need to start
using the floatToBits and bitsToFloat, but only ARM is using that
mechanism, and not on float operands.
Also I notice that the "predicate" mechanism for operands ignores the
read_code and write_code mechanism, and using both will not work
correctly. This change makes no attempt to fix that problem, but
shouldn't contribute to it either.
Change-Id: I0e3a7f78ed28f40cb66958ef12c32e862950fde0
Reviewed-on: https://gem5-review.googlesource.com/c/14456
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-by: Brandon Potter <Brandon.Potter@amd.com>
Maintainer: Gabe Black <gabeblack@google.com>
Boolean sc_buffers (either pure bool or sc_dt::sc_logic) should signal
positive and negative edges and resets even when their value doesn't
change, unlike sc_signals. The spec doesn't actually say that and just
mentions the value changed event, but it may have been implied that the
other types of events also happen, they just made special mention of
the value change event.
This change moves some code around a bit so that when _signalChange()
is called, if the underlying type is a boolean signal, it will
automatically notify the appropriate edge event and signal any reset.
Putting the functionality in _signalChange instead of delegating it to
the sc_buffer lets us have a single template for sc_buffer and makes
the base class template specialization handle whether the edge events
exist, and if so which should be notified.
Change-Id: Ic4ca86afc3fde6a9df5c15a0a7386e24ac89a9e2
Reviewed-on: https://gem5-review.googlesource.com/c/14916
Reviewed-by: Matthias Jung <jungma@eit.uni-kl.de>
Reviewed-by: Gabe Black <gabeblack@google.com>
Maintainer: Gabe Black <gabeblack@google.com>
Change 9af1214 added a new ctor to the LSQUnit, however
there is a typo/bug because it sizes the SQEntries
member variable to lqEntries + 1, as opposed to
sqEntries + 1. This change corrects the issue by
using sqEntries.
Change-Id: I19dfaa5c0e335bd7b84343a92034147d7c5d914e
Reviewed-on: https://gem5-review.googlesource.com/c/15015
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
In the systemc spec, process handles can be invalidated if they have no
children and are terminated. The implementation is not required to do
so however, and for the sake of simplicity gem5 currently does not. To
quote:
"When the underlying process instance terminates, if the process
instance has no surviving children, an implementation may choose to
invalidate any associated process handles, but it is not obliged to do
so."
Two tests have reference output which is affected by this (legal)
difference in behavior. In one case, the test creates new processes
which reuse the names of processes that have been terminated. Since
gem5 doesn't invalidate the old processes, a standard mechanism is
activated which renames the processes to something which is unique.
The other test has this same problem, and also prints the hierarchy
of processes several times during the test. In that hierarchy, the
terminated tests with no live children are still present, where in the
old reference output they've been removed.
This change updates the reference output to match gem5's behavior.
Change-Id: I363448de10080bdce01a4df92f991c67b31a2401
Reviewed-on: https://gem5-review.googlesource.com/c/14919
Reviewed-by: Matthias Jung <jungma@eit.uni-kl.de>
Maintainer: Gabe Black <gabeblack@google.com>
There are a number of cases where systemc leaves the decision of what
order things happen up to the kernel, and any ordering is legal and
conformant with the spec as long as it's repeatable. The reference
outputs reflect a particular choice of ordering which is implementation
specific, and while considerable effort has been made to make the gem5
ordering match the Accellera ordering, at a certain point that's no
longer practical or desirable.
This change manually updates the reference output for the tests that
haven't been excluded for other reasons which have this sort of ordering
problem. They have been individually examined, and as best as I can
tell changing the output this way does not mask any underlying error.
One or two real problems were discovered in addition to the ordering
issues, and those were fixed in earlier changes.
Change-Id: I915269998de3f33d5ab5d1fd754a830fd620184d
Reviewed-on: https://gem5-review.googlesource.com/c/14918
Reviewed-by: Matthias Jung <jungma@eit.uni-kl.de>
Maintainer: Gabe Black <gabeblack@google.com>
