There were three different StaticInst flags for memory barriers,
IsMemBarrier, IsReadBarrier, and IsWriteBarrier. IsReadBarrier was never
used, and IsMemBarrier was for both loads and stores, so a composite of
IsReadBarrier and IsWriteBarrier.
This change gets rid of IsMemBarrier and replaces by setting
IsReadBarrier and IsWriteBarrier at the same time. An isMemBarrier
accessor is left, but is now implemented by checking if both of the
other flags are set, and renamed to isFullMemBarrier to make it clear
that it's checking both for both types of barrier, not one or the other.
Change-Id: I702633a047f4777be4b180b42d62438ca69f52ea
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/33743
Reviewed-by: Gabe Black <gabeblack@google.com>
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
According to page 57 in the RISC-V manual, CSR accesses "need to be
performed in program order with respect to those instructions whose
execution behavior is affected by the state of the accessed CSR".
Thus, we need to make them SerializeAfter to ensure that the following
instructions are executed with the potential changes to the CSR. In
theory, we could be smarter here by only considering write accesses to
CSRs and considering the following instructions, but for now we simply
serialize for every CSR access.
Change-Id: I69391fccaec31c34d944c55bac2f04d37947ebfe
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/25655
Tested-by: kokoro <noreply+kokoro@google.com>
Tested-by: Gem5 Cloud Project GCB service account <345032938727@cloudbuild.gserviceaccount.com>
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
The "shamt" in integer shift immediate instructions is an unsigned
immediate encoded in bits[25:20]. While the original Gem5 uses bits[31:20]
as an int64_t. This patch fixes the problem by:
- Adding a new parameter "imm_code" for format IOp and use the correct
bitfields SHAMT5 or SHAMT6 to assign "imm_code" for each instruction.
- Use uint64_t instead of default int64_t to assign parameter "imm_type"
of format IOp.
The instructions affected include:
- Shift Left Logical Immediate, slli
- Shift Right Logical Immediate, srli
- Shift Right Arithmetic Immediate, srai
- Shift Left Logical Word Immediate, slliw
- Shift Right Logical Word Immediate, srliw
- Shift Right Arithmetic Word Immediate, sraiw
Change-Id: Iad34ccd036c11630409f84f6de2b939224e100e6
Signed-off-by: Ian Jiang <ianjiang.ict@gmail.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/22563
Reviewed-by: Alec Roelke <alec.roelke@gmail.com>
Maintainer: Alec Roelke <alec.roelke@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
The original Gem5 does not give correct disassembly for instruction fence
and fence.i. This patch fixes the problem by adding two bitfields PRED and
SUCC and a new format FenceOp and a template FenceExecute, in which
operands are generated based on PRED and SUCC in the disassembling
function.
Change-Id: I78dbf125fef86ce40785c498a318ffb1569da46c
Signed-off-by: Ian Jiang <ianjiang.ict@gmail.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/22569
Reviewed-by: Alec Roelke <alec.roelke@gmail.com>
Maintainer: Alec Roelke <alec.roelke@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
In disassembling compressed instructions, the original Gem5 gives needless
operands, such as register or immediate. This patch fixes the problem.
- Existing formats fixed: CIOp, CJOp, CBOp and Jump.
- New formats added: CIAddi4spnOp (for c.addi4spn only) and CompressedROp (with
templates CBasicDeclare and CBasicExecute)
Change-Id: Ic293836983256a59d3a7aca091c8184b410516a4
Signed-off-by: Ian Jiang <ianjiang.ict@gmail.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/22566
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Reviewed-by: Alec Roelke <alec.roelke@gmail.com>
Maintainer: Alec Roelke <alec.roelke@gmail.com>
For U-type instructions auipc and lui, the 20-bit immediate is left-shifted
by 12 bits in decoding. While the original Gem5 gives the left-shifted
value directly in disassembly.
This patch fixes the problem by
- Assign the original 20-bit immediate to internal variable "imm".
- Output "imm" directly in disassembly, as how the original Gem5 does.
- Do the left-shift to "imm" later in the function defining of each
instruction, rather than in decoding.
Change-Id: I300e26fd9c79478783c39fcd6ff70ea06db88884
Signed-off-by: Ian Jiang <ianjiang.ict@gmail.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/22564
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Reviewed-by: Alec Roelke <alec.roelke@gmail.com>
Maintainer: Jason Lowe-Power <jason@lowepower.com>
c_j(al) has a special format, called CJ.
The jump offset format is instbits[12:2] --> offset[11|4|9:8|10|6|7|3:1|5]
Currently in decoder.isa, c_j format is JOp, the imm and branchTarget are incorrect
In the execute section (decoder.isa:228), the imm fields is ignored and the offset is calculated correctlly.
As a result, we get decoder flush for each c_j instance
I've added CJOp format in compressed.isa, and use it in execute section.
In addition, c_j is mappped to jal zero, cj_imm, and actually is neither indirect control nor a function call
I fixed the flags accordently.
I'll fix all IsRet, IsCall and IsIndirectControl flags for rest of (c_)jal(r) in my next commit.
I ran coremark -O0 before my fix and I got 37.7% branch miss-rate, after the fix the branch miss-rate is <13%
Change-Id: I608d5894a78a1ebefe36f21e21aaea68b42bccfc
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/17808
Maintainer: Jason Lowe-Power <jason@lowepower.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Reviewed-by: Alec Roelke <alec.roelke@gmail.com>
(1) Atomic Memory Operation (AMO)
This patch changes how RISC-V AMO instructions are implemented. For each
AMO, instead of issuing a locking load and an unlocking store request to
downstream memory system, this patch issues a single memory request that
contains a corresponding AtomicOpFunctor to the memory system. Once the
memory system receives the request, the atomic operation is executed in
one single step.
This patch also changes how AMO instructions handle acquire and release
flags in AMOs (e.g., amoadd.aq and amoadd.rl). If an AMO is associated
with an acquire flag, a memory fence is inserted after the AMO completes
as a micro-op. If an AMO is associated with a release flag, another
memory fence is inserted before the AMO executes. If both flags are
specified, the AMO is broken down into a sequence of 3 micro-ops:
mem fence -> atomic RMW -> mem fence. This change makes this AMO
implementation comply to the release consistency model.
(2) Load-Reserved (LR) and Store-Conditional (SC)
Addresses locked by LR instructions are tracked in a stack data
structure. LR instruction pushes its target address to the stack, and SC
instruction pops the top address from the stack. As specified by RISC-V
ISA, a SC fails if its target address does not match with the most recent
LR.
Previously, there was a single stack for all hardware thread contexts.
A shared stack between thread contexts can lead to a infinite sequence
of failed SCs if LRs from other threads keep pushing new addresses to
this stack.
This patch gives each context its private stack to address the problem.
This patch also adds extra memory fence micro-ops to lr/sc to guarantee
a correct execution order of memory instructions with respect to release
consistency model.
Change-Id: I1e95900367c89dd866ba872a5203f63359ac51ae
Reviewed-on: https://gem5-review.googlesource.com/c/8189
Reviewed-by: Alec Roelke <ar4jc@virginia.edu>
Maintainer: Alec Roelke <ar4jc@virginia.edu>
Fence instruction origially had two flags NonSpeculative and
MemBarrier. In O3 model, MemBarrier instructions are inserted
into the instruction queue by the InstructionQueue::insertBarrier (at
src/cpu/o3/iew_impl.hh:1083). Barrier instructions are implicitly
assumed to be non-speculative.
Adding NonSpeculative flag to fence instruction makes it inserted into
the instruction queue twice (at src/cpu/o3/iew_impl.hh:1083 and :1111).
This can lead to a deadlock if both pointers to the instruction are not
cleared from the queue when the instruction retires.
This patch removes NonSpeculative flag from the fence inst.
Change-Id: I26573d12a0b52f43b73c0e51158286dc98d05ea4
Reviewed-on: https://gem5-review.googlesource.com/c/8183
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Reviewed-by: Alec Roelke <ar4jc@virginia.edu>
Maintainer: Alec Roelke <ar4jc@virginia.edu>
This patch adds the uret, sret, and mret instructions for use with
returning from user-, supervisor-, and machine-level code, respectively.
These instructions read the STATUS register to determine the previous
privilege level and modify it to re-enable interrupts at the old
privilege level. These instructions can only be executed at the
corresponding privilege level or higher.
Change-Id: I6125c31cb2fdcc3f83eca86910519e81ffbbbfc9
Reviewed-on: https://gem5-review.googlesource.com/11136
Maintainer: Alec Roelke <alec.roelke@gmail.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Reviewed-by: Robert Scheffel <robert.scheffel1@tu-dresden.de>
RISC-V has a set of CSRs that contain information about a trap that was
taken into each privilegel level, such as illegal instruction bytes or
faulting address. This patch adds that register, modifies existing
faults to make use of it, and adds a new fault for future use with
handling page faults and bad addresses.
Change-Id: I3004bd7b907e7dc75e5f1a8452a1d74796a7a551
Reviewed-on: https://gem5-review.googlesource.com/11135
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Alec Roelke <alec.roelke@gmail.com>
This patch updates the CSRs to match the RISC-V privileged specification
version 1.10. As interrupts, faults, and privilege levels are not yet
supported, there are no meaninful side effects that are implemented.
Performance counters are also not yet implemented, as they do not have
specifications. Currently they act as cycle counters.
Note that this implementation trusts software to use the registers
properly. Access protection, readability, and writeability of registers
based on privilege will come in a future patch.
Change-Id: I1de89bdbe369b5027911b2e6bc0425d3acaa708a
Reviewed-on: https://gem5-review.googlesource.com/7441
Reviewed-by: Gabe Black <gabeblack@google.com>
Maintainer: Alec Roelke <ar4jc@virginia.edu>
If an instruction is invalid, some assertions may in the decoder may
fail the entire simulation. Instead, we want to raise an
IllegalInstFault instead of failing immediately in the decoder if the
invalid instruction is being speculatively executed.
Change-Id: I5cb72ba06f07f173922f86897ddfdf677e8c702f
Reviewed-on: https://gem5-review.googlesource.com/9261
Maintainer: Alec Roelke <ar4jc@virginia.edu>
Reviewed-by: Monir Zaman <monir.zaman.m@gmail.com>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
There is a bug in RISC-V's compressed branch instructions where the
offsets are not stored in ImmOp's immediate field, causing incorrect
branchTarget() return values. This patch adds a new compressed branch
op format, CBOp, which correctly stores the offset.
Change-Id: Iac6e9b091d63f3dce4717ee5a9ec31a7cbd6c377
Reviewed-on: https://gem5-review.googlesource.com/8441
Reviewed-by: Tuan Ta <qtt2@cornell.edu>
Maintainer: Alec Roelke <ar4jc@virginia.edu>
This patch applies correct miscellaneous or multiply-accumulate op
classes to floating point instructions which had previously been
incorrectly classed as add or multiply instructions.
Change-Id: I959dd8d3152aa341e0f060b003ce1da8c4d688fb
Reviewed-on: https://gem5-review.googlesource.com/6521
Reviewed-by: Alec Roelke <ar4jc@virginia.edu>
Maintainer: Alec Roelke <ar4jc@virginia.edu>
Using the fetestexcept function to check for specific types of floating
point exceptions is unreliable for some kinds of
floating-point-to-integer conversion operations. RISC-V code used to
make use of them to check for some exceptional cases like overflow and
underflow, which caused incorrect output when compiler optimization is
turned on. This patch changes the use of fetestexcept to explicit checks
for those exceptional cases.
Change-Id: Id983906ea0664dc246e115a9e470d9ab7733bde1
Reviewed-on: https://gem5-review.googlesource.com/6402
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Alec Roelke <ar4jc@virginia.edu>
This patch makes use of ImmOp's polymorphism to remove unnecessary
casting from the implementations of arithmetic instructions with
immediate operands and to remove the CUIOp format by combining it with
the CIOp format (compressed arithmetic instructions with immediate
operands). Interestingly, RISC-V specifies that instructions with
unsigned immediate operands still need to sign-extend the immediates
from 12 (or 20) bits to 64 bits, so that is left alone.
Change-Id: If20d70c1e90f379b9ed8a4155b2b9222b6defe16
Reviewed-on: https://gem5-review.googlesource.com/6401
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Reviewed-by: Tuan Ta <qtt2@cornell.edu>
Maintainer: Alec Roelke <ar4jc@virginia.edu>
This patch moves static portions of the memory instructions out of the
ISA generated code and puts them into arch/riscv/insts. It also
simplifies the definitions of load and store instructions by giving
them a common base class.
Change-Id: Ic6930cbfc6bb02e4b3477521e57b093eac0c8803
Reviewed-on: https://gem5-review.googlesource.com/6024
Reviewed-by: Gabe Black <gabeblack@google.com>
Maintainer: Alec Roelke <ar4jc@virginia.edu>
When both the C and C++ versions are visible, the compiler will complain that
it doesn't know which one to use. By specifying the std namespace, it will
know to use the C++ version.
Change-Id: Ie1bbe1d95eadbad9644b4915c21f924d7d5c0b22
Reviewed-on: https://gem5-review.googlesource.com/4060
Reviewed-by: Alec Roelke <ar4jc@virginia.edu>
Maintainer: Alec Roelke <ar4jc@virginia.edu>
This patch fixes some bugs that were missed with the changes to the
decoder that enabled compatibility with compressed instructions. In
order to accommodate speculation with variable instruction widths, a few
assertions in decoder had to be changed to returning faults as the
specification describes should normally happen. The rest of these
assertions will be changed in a later patch.
[Remove commented-out debugging line and add clarifying comment to
registerName in utility.hh.]
Change-Id: I3f333008430d4a905cb59547a3513f5149b43b95
Reviewed-on: https://gem5-review.googlesource.com/4041
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Maintainer: Alec Roelke <ar4jc@virginia.edu>
This patch adds compatibility with the 64-bit compressed extension to
the RISC-V ISA, RV64C. Current versions of the toolchain may use
compressed instructions in glibc by default, which can only be
overridden by recompiling the entire toolchain (simply adding
"-march=rv64g" or "-march=rv64imafd" when compiling a binary is not
sufficient to use uncompressed instructions in glibc functions in the
binary).
[Update diassembly generation for new RegId type.]
[Rebase onto master.]
Change-Id: Ifd5a5ea746704ce7e1b111442c3eb84c509a98b4
Reviewed-on: https://gem5-review.googlesource.com/3860
Reviewed-by: Alec Roelke <ar4jc@virginia.edu>
Maintainer: Alec Roelke <ar4jc@virginia.edu>
This patch restructures the RISC-V ISA description to use fewer classes
and improve its ability to be extended with nonstandard extensions in
the future. It also cleans up the disassembly for some of the CSR and
system instructions by removing source and destination registers for
instructions that don't have any.
[Fix class UImmOp to have an "imm" member rather than "uimm".]
[Update disassembly generation for new RegId class.]
Change-Id: Iec1c782020126e5e8e73460b84e31c7b5a5971d9
Reviewed-on: https://gem5-review.googlesource.com/3800
Maintainer: Alec Roelke <ar4jc@virginia.edu>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
Some of the functions in the Linux toolchain that allocate memory make
use of paired LR and SC instructions, which didn't work properly for
that toolchain. This patch fixes that so attempting to use those
functions doesn't cause an endless loop of failed SC instructions.
Change-Id: If27696323dd6229a0277818e3744fbdf7180fca7
Reviewed-on: https://gem5-review.googlesource.com/2340
Maintainer: Alec Roelke <ar4jc@virginia.edu>
Reviewed-by: Jason Lowe-Power <jason@lowepower.com>
RISC-V makes use of load-reserved and store-conditional instructions to
enable creation of lock-free concurrent data manipulation as well as
ACQUIRE and RELEASE semantics for memory ordering of LR, SC, and AMO
instructions (the latter of which do not follow LR/SC semantics). This
patch is a correction to patch 4, which added these instructions to the
implementation of RISC-V. It modifies locked_mem.hh and the
implementations of lr.w, sc.w, lr.d, and sc.d to apply the proper gem5
flags and return the proper values.
An important difference between gem5's LLSC semantics and RISC-V's LR/SC
ones, beyond the name, is that gem5 uses 0 to indicate failure and 1 to
indicate success, while RISC-V is the opposite. Strictly speaking, RISC-V
uses 0 to indicate success and nonzero to indicate failure where the
value would indicate the error, but currently only 1 is reserved as a
failure code by the ISA reference.
This is the seventh patch in the series which originally consisted of five
patches that added the RISC-V ISA to gem5. The original five patches added
all of the instructions and added support for more detailed CPU models and
the sixth patch corrected the implementations of Linux constants and
structs. There will be an eighth patch that adds some regression tests
for the instructions.
[Removed some commented-out code from locked_mem.hh.]
Signed-off by: Alec Roelke
Signed-off by: Jason Lowe-Power <jason@lowepower.com>
Fourth of five patches adding RISC-V to GEM5. This patch adds the RV64A
extension, which includes atomic memory instructions. These instructions
atomically read a value from memory, modify it with a value contained in a
source register, and store the original memory value in the destination
register and modified value back into memory. Because this requires two
memory accesses and GEM5 does not support two timing memory accesses in
a single instruction, each of these instructions is split into two micro-
ops: A "load" micro-op, which reads the memory, and a "store" micro-op,
which modifies and writes it back. Each atomic memory instruction also has
two bits that acquire and release a lock on its memory location.
Additionally, there are atomic load and store instructions that only either
load or store, but not both, and can acquire or release memory locks.
Note that because the current implementation of RISC-V only supports one
core and one thread, it doesn't make sense to make use of AMO instructions.
However, they do form a standard extension of the RISC-V ISA, so they are
included mostly as a placeholder for when multithreaded execution is
implemented. As a result, any tests for their correctness in a future
patch may be abbreviated.
Patch 1 introduced RISC-V and implemented the base instruction set, RV64I;
patch 2 implemented the integer multiply extension, RV64M; and patch 3
implemented the single- and double-precision floating point extensions,
RV64FD.
Patch 5 will add support for timing, minor, and detailed CPU models that
isn't present in patches 1-4.
[Added missing file amo.isa]
[Replaced information removed from initial patch that was missed during
division into multiple patches.]
[Fixed some minor formatting issues.]
[Fixed oversight where LR and SC didn't have both AQ and RL flags.]
Signed-off by: Alec Roelke
Signed-off by: Jason Lowe-Power <jason@lowepower.com>
Third of five patches adding RISC-V to GEM5. This patch adds the RV64FD
extensions, which include single- and double-precision floating point
instructions.
Patch 1 introduced RISC-V and implemented the base instruction set, RV64I
and patch 2 implemented the integer multiply extension, RV64M.
Patch 4 will implement the atomic memory instructions, RV64A, and patch
5 will add support for timing, minor, and detailed CPU models that is
missing from the first four patches.
[Fixed exception handling in floating-point instructions to conform better
to IEEE-754 2008 standard and behavior of the Chisel-generated RISC-V
simulator.]
[Fixed style errors in decoder.isa.]
[Fixed some fuzz caused by modifying a previous patch.]
Signed-off by: Alec Roelke
Signed-off by: Jason Lowe-Power <jason@lowepower.com>
Second of five patches adding RISC-V to GEM5. This patch adds the
RV64M extension, which includes integer multiply and divide instructions.
Patch 1 introduced RISC-V and implemented the base instruction set, RV64I.
Patch 3 will implement the floating point extensions, RV64FD; patch 4 will
implement the atomic memory instructions, RV64A; and patch 5 will add
support for timing, minor, and detailed CPU models that is missing from
the first four patches.
[Added mulw instruction that was missed when dividing changes among
patches.]
Signed-off by: Alec Roelke
Signed-off by: Jason Lowe-Power <jason@lowepower.com>
First of five patches adding RISC-V to GEM5. This patch introduces the
base 64-bit ISA (RV64I) in src/arch/riscv for use with syscall emulation.
The multiply, floating point, and atomic memory instructions will be added
in additional patches, as well as support for more detailed CPU models.
The loader is also modified to be able to parse RISC-V ELF files, and a
"Hello world\!" example for RISC-V is added to test-progs.
Patch 2 will implement the multiply extension, RV64M; patch 3 will implement
the floating point (single- and double-precision) extensions, RV64FD;
patch 4 will implement the atomic memory instructions, RV64A, and patch 5
will add support for timing, minor, and detailed CPU models that is missing
from the first four patches (such as handling locked memory).
[Removed several unused parameters and imports from RiscvInterrupts.py,
RiscvISA.py, and RiscvSystem.py.]
[Fixed copyright information in RISC-V files copied from elsewhere that had
ARM licenses attached.]
[Reorganized instruction definitions in decoder.isa so that they are sorted
by opcode in preparation for the addition of ISA extensions M, A, F, D.]
[Fixed formatting of several files, removed some variables and
instructions that were missed when moving them to other patches, fixed
RISC-V Foundation copyright attribution, and fixed history of files
copied from other architectures using hg copy.]
[Fixed indentation of switch cases in isa.cc.]
[Reorganized syscall descriptions in linux/process.cc to remove large
number of repeated unimplemented system calls and added implmementations
to functions that have received them since it process.cc was first
created.]
[Fixed spacing for some copyright attributions.]
[Replaced the rest of the file copies using hg copy.]
[Fixed style check errors and corrected unaligned memory accesses.]
[Fix some minor formatting mistakes.]
Signed-off by: Alec Roelke
Signed-off by: Jason Lowe-Power <jason@lowepower.com>
