The vALU instruction counters were previously 32 bits, but for some
workloads this value wraps around and triggers an assert failure
because the max vALU operations are reached. To resolve this, this
commit increases the counter size to 64 bits.
Change-Id: I90ed4514669485cfea7ccc37ba9d69665277bccb
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/29950
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Instruction s_setreg_b32 was unimplemented, but is used by hipified
rodinia 'srad'. The instruction sets values of hardware internal
registers. If the instruction is writing into MODE to control
single-precision FP round and denorm modes, a simple warn will be
printed; for all other cases (non-MODE hw register or other
precisions), panic will happen.
Change-Id: Idb1cd5f60548a146bc980f1a27faff30259e74ce
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/29949
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Xianwei Zhang <xianwei.zhang@amd.com>
Instructions that use the DPP field need to use the extra SRC0
register associated with the DPP instruction instead of the
"default" SRC0 register, since the default SRC0 register contains
the DPP information when DPP is being used. This commit fixes
2735c3bb88 to take this into account. Additionally, this commit
removes write of the src register from the DPP helper functions,
to avoid overwriting any changes made to the destination register.
Finally, this change modifies the instructions that use DPP to
simplify the flow through the execute() functions.
Change-Id: I80fd0af1f131f287f18ff73b3c1c9122d8c60823
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/29947
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
This change updates the constructors of the CU's pipe
stages/memory pipelines to accept a pointer to their
parent CU. Because the CU creates these objects, and
can pass a pointer to itself to these object via their
constructors, this is the safer way to initalize these
classes.
Change-Id: I0b3732ce7c03781ee15332dac7a21c097ad387a4
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/29945
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Barriers were not modeled properly. Firstly, barriers were
allocated to each WG that was launched, which is not
correct, and the CU would provide an infinite number
of barrier slots. There are a limited number of barrier slots
per CU in reality. In addition, the CU will not allocate
barrier slots to WGs with a single WF (nothing to sync if
only one WF).
Beyond modeling problems, there also the issue of deadlock.
The barrier could deadlock because not all WFs are freed
from the barrier once it has been satisfied. Instead, we
relied on the scoreboard stage to release them lazily,
one-by-one.
Under this implementation the scoreboard may not fully release
all WFs participating in a barrier; this happens because the
first WF to be freed from the barrier could reach an s_barrier
instruction again, forever causing the barrier counts across
WFs to be out-of-sync.
This change refactors the barrier logic to:
1) Create a proper barrier slot implementation
2) Enforce (via a parameter) the number of barrier
slots on the CU.
3) Simplify the logic and cleanup the code (i.e., we
no longer iterate through the entire WF list each
time we check if a barrier is satisfied).
4) Fix deadlock issues.
Change-Id: If53955b54931886baaae322640a7b9da7a1595e0
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/29943
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
Tested-by: kokoro <noreply+kokoro@google.com>
The LDS is capable of handling out-of-bounds accesses,
that is, accesses that are outside the bounds of the
chunk allocated to a WG. Currently, the simulator asserts
on these accesses. This patch changes the behavior of the
LDS to return 0 for reads and dropping writes that are
out-of-bounds.
Change-Id: I5f467d0f52113e8565e1a3029e82fb89cc6f07ea
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/29940
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Matt Sinclair <mattdsinclair@gmail.com>
Ubuntu 20.04 contains GCC verisons 9 and 10, which are not easily
obtainable (at least through APT) on Ubuntu 18.04. Therefore, a
Dockerfile for obtaining GCC versions in Ubuntu 20.04 has been added.
The orignal GCC version Dockerfile (Ubuntu 18.04) has been kept as
GCC versions 4.8, 5, and 6 are not obtainable, via APT, on Ubuntu
20.04. A complete migration to the 20.04 Dockerfile is not possible
until these earlier GCC versions are dropped.
The Docker images for GCC Versions 9 and 10 can be found here:
https://gcr.io/gem5-test/gcc-version-10https://gcr.io/gem5-test/gcc-version-9
The other Dockerfile directories have been renamed for consistency.
Change-Id: I569249331095ee62d1be5be479c7ba7da0077422
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/30514
Reviewed-by: Daniel Carvalho <odanrc@yahoo.com.br>
Reviewed-by: Bobby R. Bruce <bbruce@ucdavis.edu>
Maintainer: Bobby R. Bruce <bbruce@ucdavis.edu>
Tested-by: kokoro <noreply+kokoro@google.com>
This script runs a series of compilations on gem5. The following
compilers are tested:
clang-9
clang-8
clang-7
clang-6
clang-5
clang-4
clang-3.9
gcc-9
gcc-8
gcc-7
gcc-6
gcc-5
gcc-4.8 (to be dropped soon:
https://gem5.atlassian.net/browse/GEM5-218)
They are tested by building the following build targets:
ARM
ARM_MESI_Three_Level
Garnet_standalone
GCN3_X86
MIPS
NULL_MESI_Two_Level
NULL_MOESI_CMP_directory
NULL_MOESI_CMP_token
NULL_MOESI_hammer
POWER
RISCV
SPARC
X86
X86_MOESI_AMD_BASE
For each, ".opt" and ".fast" compiler build settings are tested.
clang-9 and gcc-9 are tested against all targets with each build
setting. For the remaining compilers, a random build target is
chosen. After the script has run, the output of the tests can be
found in "compile-test-out".
Docker is required to run this script.
Change-Id: Id3bf4c89b9d424c87e9409930ee2aceaef72cb29
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/30395
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Reviewed-by: Daniel Carvalho <odanrc@yahoo.com.br>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
The RubyPrefetcher uses makeNextStrideAddress() with
a negative stride to find prefetched address.
The type of this expression is:
uint64_t + uint32_t * int;
This gives wrong result due to implicit conversion.
Fix this with static cast and it works correctly:
uint64_t + int * int;
Change-Id: I36e17e00d5c66c3699fe1d5b287971225a162d04
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/31314
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Maintainer: Jason Lowe-Power <power.jg@gmail.com>
Tested-by: kokoro <noreply+kokoro@google.com>
Motivation:
An AddressSizeFault on AArch32 can only happen during a table walk
since the register used as a base by LD/ST is always 32 bit wide.
On AArch64 on the other hand, addresses can be 64bit wide;
when MMU is off (no virtual memory) an invalid physical address
can be specified
Change-Id: Id3ef170e99202c6b0b511fa7205c754956861720
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/31274
Reviewed-by: Nikos Nikoleris <nikos.nikoleris@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This will still be technically possible with the right converters, but
this removes the tags, ignore file, and style checking hooks related to
mercurial. We no longer maintain a mercurial mirror of the main git
repository, and this support adds clutter and could diverge from the git
style hooks, etc, over time.
Change-Id: Icf4833c4f0fda51ea98989d1d741432ae3ddc6dd
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/31174
Reviewed-by: Jason Lowe-Power <power.jg@gmail.com>
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Gabe Black <gabeblack@google.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
This changeset drops fetches if there is no entry reserved in the
fetch buffer for that instruction. This can happen due to a fetch
attempted to be issued in the same cycle where a branch instruction
flushed the fetch buffer, while an ITLB or I-cache request is still
pending.
Change-Id: I3b80dbd71af27ccf790b543bd5c034bb9b02624a
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/29932
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Anthony Gutierrez <anthony.gutierrez@amd.com>
Reviewed-by: Onur Kayıran <onur.kayiran@amd.com>
Maintainer: Anthony Gutierrez <anthony.gutierrez@amd.com>
These classes are all basically empty now that Alpha has been deleted,
except in cases where the arch versions had copied versions of the Alpha
code.
This change pulls all the generic logic out of the arch versions, making
the arch versions much simpler and making it clearer what the core
functionality of the class is, and what parts are architecture specific
details.
In the future, the way the StackTrace class is instantiated should be
delegated to the Workload class so that ISA agnostic code doesn't need
to know about a particular ISA's StackTrace class, and so that
StackTrace logic can, at least theoretically, be specialized for a
particular workload. The way a stack trace is collected could vary from
OS to OS, for example.
Change-Id: Id8108f94e9fe8baf9b4056f2b6404571e9fa52f1
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/30961
Reviewed-by: Gabe Black <gabeblack@google.com>
Maintainer: Gabe Black <gabeblack@google.com>
Tested-by: kokoro <noreply+kokoro@google.com>
KVM_CAP_ARM_USER_IRQ is a KVM extension introduced in newer versions of
Linux (>= 4.12). It supports delivering interrupt from the kernel-space
timer to the user-space GIC, which means that it will be unnecessary to
use the memory-mapped timer and emulate it in gem5 anymore.
Using the option provided by this change, Linux is able to boot with 1
CPU successfully, and the speed is slightly faster then the memory-
mapped timer option. However, multicore seems to hang during boot and
still needs more investigation to be enabled.
JIRA: https://gem5.atlassian.net/browse/GEM5-663
Change-Id: I146bbcce3cf66f8f5ebee04ea5f1b9f54868721a
Signed-off-by: Hsuan Hsu <hsuan.hsu@mediatek.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/30921
Tested-by: kokoro <noreply+kokoro@google.com>
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
The memory-mapped timer emulated by gem5 is driven by the underlying
gem5 tick, which means that we must align the tick with the host time
to make the timer interrupt fire at a nearly native rate.
In each KVM execution round, the number of ticks incremented is
directly calculated from the number of instructions executed. However,
when a guest CPU switches to idle state, KVM seems to stay in kernel-
space until the POSIX timer set up in user-space raises an expiration
signal, instead of trapping to user-space immediately; and somehow the
instruction count is just too low to match the elapsed host time. This
makes the gem5 tick increment very slowly when the guest is idle and
drastically slow down workloads being sensitive to the guest time which
is driven by timer interrupt.
Before switching to KVM to execute the guest code, gem5 programs the
POSIX timer to expire according to the remaining ticks before the next
event in the event queue. Based on this, we can come up with the
following solution: If KVM returns to user-space due to POSIX timer
expiration, it must be time to process the next gem5 event, so we just
fast-forward the tick (by scheduling the next CPU tick event) to that
event directly without calculating from the instruction count.
There is one more related issue needed to be solved. The KVM exit
reason, KVM_EXIT_INTR, was treated as the case where the KVM execution
was disturbed by POSIX timer expiration. However, there exists a case
where the exit reason is KVM_EXIT_INTR but the POSIX timer has not
expired. Its cause is still unknown, but it can be observed via the
"old_value" argument returned by timer_settime() when disarming the
POSIX timer. In addition, it seems to happen often when a guest CPU is
not in idle state. When this happens, the above tick event scheduling
incorrectly treats KVM_EXIT_INTR as POSIX timer expiration and fast-
forwards the tick to process the next event too early. This makes the
guest feel external events come too fast, and will sometimes cause
trouble. One example is the VSYNC interrupt from HDLCD. The guest seems
to get stuck in VSYNC handling if the KVM CPU is not given enough time
between each VSYNC interrupt to complete a service. (Honestly I did not
dig in to see how the guest handled the VSYNC interrupt and how the
above situation became trouble. I just observed from the debug trace of
GIC & HDLCD & timer, and made this conclusion.) This change also uses
a workaround to detect POSIX timer expiration correctly to make the
guest work with HDLCD.
JIRA: https://gem5.atlassian.net/browse/GEM5-663
Change-Id: I6159238a36fc18c0c881d177a742d8a7745a23ca
Signed-off-by: Hsuan Hsu <hsuan.hsu@mediatek.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/30919
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
We should also deal with change of the imask bit, or we will lose timer
interrupt if the timer expires before the guest kernel unmasks the bit.
More precisely, consider the following common pattern in timer interrupt
handling:
1. Set the interrupt mask bit (CNTV_CTL.IMASK)
2. Reprogram the downcounter (CNTV_TVAL) for the next interrupt
3. Clear the interrupt mask bit (CNTV_CTL.IMASK)
The timer can expires between step 2 & 3 if the value programmed in step
2 is small enough, and this seems very likely to happen in KVM mode. If
we don't check for timer expiration right after unmasking, we will miss
the only chance to inject the interrupt.
JIRA: https://gem5.atlassian.net/browse/GEM5-663
Change-Id: I75e8253bb78d15ae72cb985ed132f896d8e92ca6
Signed-off-by: Hsuan Hsu <hsuan.hsu@mediatek.com>
Reviewed-on: https://gem5-review.googlesource.com/c/public/gem5/+/30918
Reviewed-by: Andreas Sandberg <andreas.sandberg@arm.com>
Maintainer: Andreas Sandberg <andreas.sandberg@arm.com>
Tested-by: kokoro <noreply+kokoro@google.com>
