Previously calling a function with an INT_MAX argument would result in
the following (incorrectly extended) trace:
```
lui a1, 524288 : IntAlu : D=0xffffffff80000000
c_addiw a1, -1 : IntAlu : D=0xffffffff7fffffff
```
I noticed this due to a kernel assertion that checked the second argument
was bigger than the first. Since INT_MAX was incorrectly being extended to
0xffffffff7fffffff, the generated slt comparison instruction was returning
1 instead of the expected zero (which would have happened with 0x7fffffff).
The problem in the current addiw logic is that the immediate value is an
int64_t, so the 32-bit Rs1/Rc1 values are promoted to 64-bit for the
aritmetic operation, thereby making the current cast redundant.
Fix this by placing parens around the whole expression and truncating
that to 32 bits.
Change-Id: I7b18a8101b1c2614b9f056004e6a7f87b66b64c9
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/56103
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
These registers used to be accessed with a two dimensional index, with
one dimension specifying the register, and the second index specifying
the element within that register. This change linearizes that index down
to one dimension, where the elements of each register are laid out one
after the other in sequence.
Change-Id: I41110f57b505679a327108369db61c826d24922e
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/49148
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Maintainer: Giacomo Travaglini <giacomo.travaglini@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Global instruction address calculation when using an SGPR or SGPR pair
as a base address was being calculated incorrectly when 64-bit addresses
were to be generated.
From the ISA documentation, the SGPR should be read as 32-bit or 64-bit
depending on "ADDRESS_MODE." The VGPR-offset (computed from the lower
32-bits of vaddr) should always be 32-bits and the offset is 12 bits
from the instruction. This means the 32-bit mask should only be applied
to vaddr to get the VGPU-offset rather than the final sum.
The SGPR base format is being seen in more recent clang/ROCm versions to
avoid unnecessary copies of SGPRs into VGPRs to use VGPRs as the base
address.
Change-Id: I48910611fcfac5b62bc63496bbaabd6f6e53fe0d
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/55643
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Maintainer: Matt Sinclair <mattdsinclair@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
The 7th parameter of breakpoint_set_code is dontStop. It seems the
fastmodel would prefetch something or do some evaluation ahead with the
flag set. This behavior prevents the instruction stepping feature of
gdb. The implementation of the feature is creating a breakpoint on the
next instruction and contining the simulation. Without stopping on the
breakpoint, it wouldn't invoke the breakpoint callback, since it may
evaulate the code we just want it to stop already. We should set the
dontStop to false to fix this issue.
Change-Id: Iaf8acd3235fa9625c1423ef34606e1fa5d0c531a
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/55484
Reviewed-by: Earl Ou <shunhsingou@google.com>
Reviewed-by: Gabe Black <gabe.black@gmail.com>
Maintainer: Gabe Black <gabe.black@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Ported from https://gem5-review.googlesource.com/c/public/gem5/+/48019:
Certain DS insts are classfied as Loads, but don't actually go through
the memory pipeline. However, any instruction classified as a load
marks its destination registers as free in the memory pipeline.
Because these instructions didn't use the memory pipeline, they
never freed their destination registers, which led to a deadlock.
This patch explicitly calls the function used to free the destination
registers in the execute() method of those Load instructions that
don't use the memory pipeline.
Change-Id: I8231217a79661ca6acc837b2ab4931b946049a1a
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/55463
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Maintainer: Matt Sinclair <mattdsinclair@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
These are mapped to instruction definitions like MOV_R_R, even though
one or more of the Rs might have come from a fixed value. Because
MOV_R_R (for instance) is only defined once, using a fixed text constant
there won't work because that can only have one value.
Instead, use a variable which will have the value of that constant so
that the same disassembly code will work no matter what fixed value was
used.
Change-Id: Ie45181c6becce80ad44fa30fc3323757ef713d7c
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/55444
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Maintainer: Matt Sinclair <mattdsinclair@gmail.com>
Reviewed-by: Gabe Black <gabe.black@gmail.com>
When performing a real mode far jump, we were computing the offset into
the segment more or less correctly, but then when we tried to actually
set the PC using it, we used the second of the two wrip microop
arguments. The first argument is an unsigned value and is intended to be
a base to work from when figuring out the new IP, and the second
argument is a signed offset which can be used to implement relative
jumps/branches. When we used the second operand for our new value,
setting the first operand to t0 (the zero register on x86), we would
inadvertantly sign extend it since the wrip instruction would treat it
as a signed value.
Instead, we can just switch the two operands so that the wrip microop
treats the desired value as the unsigned base, and then adds a signed t0
to it, which will still be 0 one way or the other.
Also, while researching this bug, I found that the size used for
computing the new IP is always the operand size, and never the address
size. This CL fixes that problem as well by removing the faulty
override.
Change-Id: I96ac9effd37b40161dd8d0b634c5869e767a8873
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/55243
Reviewed-by: Gabe Black <gabe.black@gmail.com>
Maintainer: Gabe Black <gabe.black@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
The BaseCPU type had been specializing itself based on the value of
TARGET_ISA, which is not compatible with building more than one ISA at a
time.
This change refactors the CPU models so that the BaseCPU is more
general, and the ISA specific components are added to the CPU when the
CPU types are fully specialized. For instance, The AtomicSimpleCPU has a
version called X86AtomicSimpleCPU which installs the X86 specific
aspects of the CPU.
This specialization is done in three ways.
1. The mmu parameter is assigned an instance of the architecture
specific MMU type. This provides a reasonable default, but also avoids
having having to use the ISA specific type when the parameter is
created.
2. The ISA specific types are made available as class attributes, and
the utility functions (including __init__!) in the BaseCPU class can
refer to them to get the types they need to set up the CPU at run time.
Because SimObjects have strange, unhelpful semantics as far as assigning
to their attributes, these types need to be set up in a non-SimObject
class, which is then brought in as a base of the actual SimObject type.
Because the metaclass of this other type is just "type", things work
like you would expect. The SimObject doesn't do any special processing
of base classes if they aren't also SimObjects, so these attributes
survive and are accessible using normal lookup in the BaseCPU class.
3. There are some methods like addCheckerCPU and properties like
needsTSO which have ISA specific values or behaviors. These are set in
the ISA specific subclass, where they are inherently specific to an ISA
and don't need to check TARGET_ISA.
Also, the DummyChecker which was set up for the BaseSimpleCPU which
doesn't actually do anything in either C++ or python was not carried
forward. The CPU type still exists, but it isn't installed in the
simple CPUs.
To provide backward compatibility, each ISA implements a .py file which
matches the original .py for a CPU, and the original is renamed with a
Base prefix. The ISA specific version creates an alias with the old CPU
name which maps to the ISA specific type. This way, old scripts which
refer to, for example, AtomicSimpleCPU, will get the X86AtomicSimpleCPU
if the x86 version was compiled in, the ArmAtomicSimpleCPU on arm, etc.
Unfortunately, because of how tags on PySource and by extension SimObjects
are implemented right now, if you set the tags on two SimObjects or
PySources which have the same module path, the later will overwrite the
former whether or not they both would be included. There are some
changes in review which would revamp this and make it work like you
would expect, without this central bookkeeping which has the conflict.
Since I can't use that here, I fell back to checking TARGET_ISA to
decide whether to tell SCons about those files at all.
In the long term, this mechanism should be revamped so that these
compatibility types are only available if there is exactly one ISA
compiled into gem5. After the configs have been updated and no longer
assume they can use AtomicSimpleCPU in all cases, then these types can
be deleted.
Also, because ISAs can now either provide subclasses for a CPU or not,
the CPU_MODELS variable has been removed, meaning the non-ISA
specialized versions of those CPU models will always be included in
gem5, except when building the NULL ISA.
In the future, a more granular config mechanism will hopefully be
implemented for *all* of gem5 and not just the CPUs, and these can be
conditional again in case you only need certain models, and want to
reduce build time or binary size by excluding the others.
Change-Id: I02fc3f645c551678ede46268bbea9f66c3f6c74b
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/52490
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Gabe Black <gabe.black@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Port large DS read/write instructions from
https://gem5-review.googlesource.com/c/public/gem5/+/48342.
This implements the 96 and 128b ds_read/write instructions in a similar
fashion to the 3 and 4 dword flat_load/store instructions.
These instructions are treated as reads/writes of 3 or 4 dwords, instead
of as a single 96b/128b memory transaction, due to the limitations of
the VecOperand class used in the amdgpu code.
In order to handle treating the memory transaction as multiple dwords,
the patch also adds in new initMemRead/initMemWrite functions for ds
instructions. These are similar to the functions used in flat
instructions for the same purpose.
Change-Id: Iee2de14eb7f32b6654799d53dc97d806288af98f
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/55344
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Maintainer: Matt Sinclair <mattdsinclair@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Port the fixes for scalar source checks from arch-gcn3 at
https://gem5-review.googlesource.com/c/public/gem5/+/48344.
Scalar sources can either be a general-purpose register or a constant
register that holds a single value.
If we don't check for if the register is a general-purpose register,
it's possible that we get a constant register, which then causes all of
the register mapping code to break, as the constant registers aren't
supposed to be mapped like the general-purpose registers are.
This fix adds an isScalarReg check to the instruction encodings that
were missing it.
Change-Id: I30dd2d082a5a1dcc3075843bcefd325113ed1df6
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/55343
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Maintainer: Matt Sinclair <mattdsinclair@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Add initrd_filename and initrd_addr parameters to specify that an
initrd/initramfs should be loaded into memory from a file, just like the
DTB blob. The user must specify the initrd file, and they can specify
the initrd load address as well. However, in practice, it's expected
that the dev/machine backend will derive the initrd load address from
the dtb load address, which is how a bootloader would typically do it.
Change-Id: I6378927c2984b7ccdd1471486dd7803500ef5883
Signed-off-by: Alistair Delva <adelva@google.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/54184
Reviewed-by: Richard Cooper <richard.cooper@arm.com>
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Maintainer: Giacomo Travaglini <giacomo.travaglini@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
There are many SE mode system calls that are implemented
in src/sim/syscall_emul.cc or src/sim/syscall_emul.hh.
And they work well under X86 and ARM platforms. However,
they are not supported in se_workload.cc under the RISC-V
platform. This patch adds support for all the system calls
already implemented in syscall_emul.hh/cc to the RISC-V
platform (in arch/riscv/linux/se_workload.cc).
Change-Id: Ia47c3c113767b50412b1c8ade3c1047c894376cf
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/54803
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Bobby Bruce <bbruce@ucdavis.edu>
Rather than make each ISA include boilerplate to ignore a
SyscallReturn's value when it's marked as suppressed or needing a retry,
put that code into the SyscallDesc::doSyscall method instead.
That has two benefits. First, it removes a decent amount of code
duplication which is nice from a maintenance perspective. Second, it
puts the SyscallDesc in charge of figuring out what to do once a system
call implementation finishes. That will let it schedule a retry of the
system call for instance, without worrying about what the ISA is doing
with the SyscallReturn behind its back.
Jira Issue: https://gem5.atlassian.net/browse/GEM5-1123
Change-Id: I76760cba75fd23e6e3357f6169c0140bee3f01b6
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/54204
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Maintainer: Giacomo Travaglini <giacomo.travaglini@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Rather than make each ISA include boilerplate to ignore a
SyscallReturn's value when it's marked as suppressed or needing a retry,
put that code into the SyscallDesc::doSyscall method instead.
That has two benefits. First, it removes a decent amount of code
duplication which is nice from a maintenance perspective. Second, it
puts the SyscallDesc in charge of figuring out what to do once a system
call implementation finishes. That will let it schedule a retry of the
system call for instance, without worrying about what the ISA is doing
with the SyscallReturn behind its back.
Jira Issue: https://gem5.atlassian.net/browse/GEM5-1123
Change-Id: I3732a98c8e0d0b2b94d61313960aa0782c0b971f
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/54023
Maintainer: Gabe Black <gabe.black@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
This patch is modifying both TableWalker and MMU to effectively
store/use partial translations
* TableWalker changes: If there is a TLB supporting partial
translations (implemented with previous patch), the TableWalker will
craft partial entries and forward them to the TLB as walks are performed
* MMU changes: We now instruct the table walker to start a page
table traversal even if we hit in the TLB, if the matching entry
holds a partial translation
JIRA: https://gem5.atlassian.net/browse/GEM5-1108
Change-Id: Id20aaf4ea02960d50d8345f3e174c698af21ad1c
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/52125
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
