The GPU device keeps a local copy of each ring buffers read pointer
(rptr) to avoid constant DMAs to/from host memory. This means it needs
to be periodically updated on the host side as the driver uses this to
determine how much space is left in the queue and may hang if it believe
the queue is full. For user-mode queues, this already happens when
queues are unmapped. For kernel mode queues (e.g., HIQ, KIQ) the rptr is
never updated leading to a hang.
In this patch the rptr for *all* queues is reported back to the kernel
whenever the queue reaches an empty state (rptr == wptr). Additionally
to handle PM4 queue wrap-around, the queue processing function checks if
the queue is not empty instead of rptr < wptr. This is state because the
driver fills PM4 queues with NOP packets on initialization and when wrap
around occurs.
Change-Id: Ie13a4354f82999208a75bb1eaec70513039ff30f
#525 Updated DRAMSys to v5.0. This PR further improves v5.0
inforporation into gem5 by better managing its new dependencies and
updating the DRAMSys tests to use v5.0.
This PR:
1. Adds a check which throws warning if DRAMSys cannot be build due to a
missing `cmake` instead of failing with a build error. `cmake` is not a
hard gem5 requirement. It is only required to build DRAMSys in the cases
it is required. It is therefore prudent to not fail a build in cases
`cmake` is not present on the host system.
2. Updates the "all-dependency" Docker images to include the optional
dependencies `git-lfs` (needed to clone the DRAMSys repo when running
the command outlined in ext/dramsys/README -- introduced in #525) and
`cmake` (needed to build DRAMSys).
3. Updates the Weekly workflow's `dramsys-tests`' `Checkout DRAMSys` job
to clone DRAMSys in the same manner as outlined in ext/dramsys/README.
This ensures the `dram-systests` test the instructions we give users.
4. `.gitignore` is added to ext/dramsys to ignore the
ext/dramsys/DRAMSys directory when cloned for building and integration
into gem5.
(2.) Should fix our failing weekly tests:
https://github.com/gem5/gem5/actions/runs/6912511984/job/18808339821 and
(3.) will ensure the changes introduced in #525 are tested.
The V_PERM_B32 instruction is selecting the correct byte, but is
shifting into place moving by bits instead of bytes. The V_OR3_B32
instruction is calling the wrong instruction implementation in the
decoder.
This patch fixes both issues plus a bonus fix for GCN3's V_PERM_B32.
(GCN3 does not have V_OR3_B32).
Change-Id: Ied66c43981bc4236f680db42a9868f760becc284
This PR is fixing remaining issues in the ArmISA::Interrupt class; more
specifically it is enabling
virtual interrupts in secure mode (when FEAT_SEL2 is present). Previous
version was assuming no
virtual interrupt was possible in secure mode. We fix this assumption by
replacing the security check
with the EL2Enabled helper which closely matches the Arm pseudocode
This clone is updated to reflect the new advice given in
ext/dramasys/README that was introduced in PR
https://github.com/gem5/gem5/pull/525 to upgrade DRAMSysm to v5.0.
Change-Id: I868619ecc1a44298dd3885e5719979bdaa24e9c2
'cmake' is required to build DRAMSysm.
This is an optional dependency for compiling DRAMSys. It is therefore
not required. It is included in the "all-dependencies" Docker images
as they may be needed if DRAMSys is desired.
Change-Id: I1a3e1a6fa2da4d0116d423e9267d4d3095000d4e
CMake is not required to build gem5. It is only required to build
and link the optional DRAMSysm library. Therefore, if the DRAMSys repo
has been cloned but CMake is not present this patch ensures no attempt
at building or linking DRAMSysm is made. A warning is thrown
inform the user of the missing CMake.
Change-Id: I4d22e3a16655fd90f6b109b4e75859628f7d532d
The AtomicWait event was not being woken up properly due to the
numPending count in the TBE not being decremented. This patch decrements
the count when Data is returned. Since that moves to a base state, the
TBE should no longer be needed.
Additionally added a transition which stalls and wait when an AtomicWait
occurs while in WI state so that it retries.
Change-Id: Ic8bfc700f9df3f95bea0799121898926a23d8163
When restoring checkpoints for certain applications, gem5 tries to
create new doorbells with a pre-existing queue ID and simulation crashes
shortly after. This commit adds existing IDs to the GPU device's used
VMID map so that new doorbells are aware of existing queue IDs and use a
new ID. This ensures that queue IDs are unique after checkpoint
restoration
The CPU should not sleep with a pending virtual interrupt
if secure mode EL2 is supported (FEAT_SEL2)
Change-Id: Ib71c4a09d76a790331cf6750da45f83694946aee
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Similarly to the physical version [1], we rewrite the
masking logic to account for FEAT_SEL2.
The interrupt table is taken from the Arm architecture reference
manual (version DDI 0487H.a, section D1.3.6, table R_BKHXL)
[1]: https://github.com/gem5/gem5/pull/430
Change-Id: Icb6eb1944d8241293b3ef3c349b20f3981bcc558
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
There is no need to call the methods for every kind
of interrupt. A pending one should short-circuit the
remaining checks
Change-Id: I2c9eb680a7baa4644745b8cbe48183ff6f8e3102
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
With this patch we align virtual interrupts with respect to
the physical ones by introducing a matching takeVirtualInt
method.
Change-Id: Ib7835a21b85e4330ba9f051bc8fed691d6e1382e
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Vitual interrupts are enabled in secure mode as well
after the introduction of FEAT_SEL2. Replacing the
secure mode check with the EL2Enabled one
Change-Id: Id685a05d5adfa87b2a366f6be42bf344168927d4
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
https://github.com/gem5/gem5/pull/386 included two cases in
"src/dev/reg_bank.hh" where `std:: min` was used to compare a an integer
of type `size_t` and another of type `Addr`. This causes an error on my
Apple Silicon Mac as the comparison between an "unsigned long" and an
"unsigned long long" is not permitted. To fix this issue this patch
changes `reg_size` from `size_t` to `Addr`, as well as it the types of
the values it was derived from and the variable used to hold the return
from the `std::min` calls. While not completely correct typing from a
labelling perspective (`reg_bytes` is not an address), functions in
"src/dev/reg_bank.hh" already abuse `Addr` in this way frequently (for
example, `bytes` in the `write` function).
When restoring checkpoints for certain applications, gem5 tries to
create new doorbells with a pre-existing queue ID and simulation crashes
shortly after. This commit checkpoints the existing VMID map so that any
new doorbells after restoration use a unique queue ID
Change-Id: I9bf89a2769db26ceab4441634ff2da936eea6d6f
https://github.com/gem5/gem5/pull/386 included two cases in
"src/dev/reg_bank.hh" where `std:: min` was used to compare a an integer
of type `size_t` and another of type `Addr`. This cause an error on my
Apple Silicon Mac as this is a comparison between an "unsigned long"
and an "unsigned long long" which (at least on my setup) was not
permitted. To fix this issue the `reg_size` was changed from `size_t` to
`Addr`, as well as it the types of the values it was derived from and
the variable used to hold the return from the `std::min` calls.
Change-Id: I31e9c04a8e0327d4f6f5390bc5a743c629db4746
This dockerfile is used to *build* applications (e.g., from
gem5-resources) which can be run using full system mode in a GPU build.
The next releases disk image will use ROCm 5.4.2, therefore bump the
version from 4.2 to that version.
Again this is used to *build* input applications only and is not needed
to run or compile gem5 with GPUFS. For example:
$ docker build -t rocm54-build .
/some/gem5-resources/src/gpu/lulesh$ docker run --rm -u $UID:$GID -v \
${PWD}:${PWD} -w ${PWD} rocm54-build make
Change-Id: If169c8d433afb3044f9b88e883ff3bb2f4bc70d2
When restoring checkpoints for certain applications, gem5 tries to
create new doorbells with a pre-existing queue ID and simulation crashes
shortly after. This commit adds existing IDs to the GPU device's used
VMID map so that new doorbells are aware of existing queue IDs and use a
new ID. This ensures that queue IDs are unique after checkpoint
restoration
Change-Id: I9bf89a2769db26ceab4441634ff2da936eea6d6f
This pull request contains a set of small patches which fix some bugs in
the gem5 prefetchers, and aligns out-of-the box prefetcher performance
more closely with that which a typical user would expect.
The performance patches have been tested with an out-of-the-box
(untuned) Stride prefetcher configuration against a set of SPEC 2017
SimPoints, and show a modest IPC uplift across the board, with no IPC
degradation.
The new defaults were identified as part of work on gem5 prefetchers
undertaken by Nikolaos Kyparissas while on internship at Arm.
The comp_anr parameter is currently unused. Both parameters (comp_wu and
comp_anr) are set to false by default
Change-Id: If09567504540dbee082191d46fcd53f1363d819f
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Currently, the GPU SQC (L1I$) and TCP (L1D$) have a performance bug
where they do not behave correctly when multiple requests to the same
cache line overlap one another. The intended behavior is that if the
first request that arrives at the Ruby code for the SQC/TCP misses, it
should send a request to the GPU TCC (L2$). If any requests to the
same cache line occur while this first request is pending, they should
wait locally at the L1 in the MSHRs (TBEs) until the first request has
returned. At that point they can be serviced, and assuming the line
has not been evicted, they should hit.
For example, in the following test (on 1 GPU thread, in 1 WG):
load Arr[0]
load Arr[1]
load Arr[2]
The expected behavior (confirmed via profiling on real GPUs) is that
we should get 1 miss (Arr[0]) and 2 hits (Arr[1], Arr[2]) for such a
program.
However, the current support in the VIPER SQC/TCP code does not model
this correctly. Instead it lets all 3 concurrent requests go straight
through to the TCC instead of stopping the Arr[1] and Arr[2] requests
locally while Arr[0] is serviced. This causes all 3 requests to be
classified as misses.
To resolve this, this patch adds support into the SQC/TCP code to
prevent subsequent, concurrent requests to a pending cache line from
being
sent in parallel with the original one. To do this, we add an
additional transient state (IV) to indicate that a load is pending to
this cache line. If a subsequent request of any kind to the same cache
line occurs while this load is pending, the requests are put on the
local wait buffer and woken up when the first request returns to the
SQC/TCP. Likewise, when the first load is returned to the SQC/TCP, it
transitions from IV --> V.
As part of this support, additional transitions were also added to
account for corner cases such as what happens when the line is evicted
by another request that maps to the same set index while the first load
is pending (the line is immediately given to the new request, and when
the load returns it completes, wakes up any pending requests to the same
line, but does not attempt to change the state of the line) and how GPU
bypassing loads and stores should interact with the pending requests
(they are forced to wait if they reach the L1 after the pending,
non-bypassing load; but if they reach the L1 before the non-bypassing
load then they make sure not to change the state of the line from IV if
they return before the non-bypassing load).
As part of this change, we also move the MSHR behavior from internally
in the GPUCoalescer for loads to the Ruby code (like all other
requests). This is important to get correct hits and misses in stats
and other prints, since the GPUCoalescer MSHR behavior assumes all
requests serviced out of its MSHR also miss if the original request to
that line missed.
Although the SQC does not support stores, the TCP does. Thus,
we could have applied a similar change to the GPU stores at the TCP.
However, since the TCP support assumes write-through caches and does not
attempt to allocate space in the TCP, we elected not to add this support
since it seems to run contrary to the intended behavior (i.e., the
intended behavior seems to be that writes just bypass the TCP and thus
should not need to wait for another write to the same cache line to
complete).
Additionally, making these changes introduced issues with deadlocks at
the TCC. Specifically, some Pannotia applications have accesses to the
same cache line where some of the accesses are GLC (i.e., they bypass
the GPU L1 cache) and others are non-GLC (i.e., they want to be cached
in the GPU L1 cache). We have support already per CU in the above code.
However, the problem here is that these requests are coming from
different CUs and happening concurrently (seemingly because different
WGs are at different points in the kernel around the same time).
This causes a problem because our support at the TCC for the TBEs
overwrites the information about the GPU bypassing bits (SLC, GLC) every
time. The problem is when the second (non-GLC) load reaches the TCC, it
overwrites the SLC/GLC information for the first (GLC) load. Thus, when
the the first load returns from the directory/memory, it no longer has
the GLC bit set, which causes an assert failure at the TCP.
After talking with other developers, it was decided the best way handle
this and attempt to model real hardware more closely was to move the
point at which requests are put to sleep on the wakeup buffer from the
TCC to the directory. Accordingly, this patch includes support for that
-- now when multiple loads (bypassing or non-bypassing) from different
CUs reach the directory, all but the first one will be forced to wait
there until the first one completes, then will be woken up and
performed. This required updating the WTRequestor information at the
TCC to pass the information about what CU performed the original request
for loads as well (otherwise since the TBE can be updated by multiple
pending loads, we can't tell where to send the final result to). Thus,
I changed the field to be named CURequestor instead of WTRequestor since
it is now used for more than stores. Moreover, I also updated the
directory to take this new field and the GLC information from incoming
TCC requests and then pass that information back to the TCC on the
response -- without doing this, because the TBE can be updated by
multiple pending, concurrent requests we cannot determine if this memory
request was a bypassing or non-bypassing request. Finally, these
changes introduced a lot of additional contention and protocol stalls at
the directory, so this patch converted all directory uses of z_stall to
instead put requests on the wakeup buffer (and wake them up when the
current request completes) instead. Without this, protocol stalls cause
many applications to deadlock at the directory.
However, this exposed another issue at the TCC: other applications
(e.g., HACC) have a mix of atomics and non-atomics to the same cache
line in the same kernel. Since the TCC transitions to the A state when
an atomic arrives. For example, after the first pending load returns to
the TCC from the directory, which causes the TCC state to become V, but
when there are still other pending loads at the TCC. This causes invalid
transition errors at the TCC when those pending loads return, because
the A state thinks they are atomics and decrements the pending atomic
count (plus the loads are never sent to the TCP as returning loads).
This patch fixes this by changing the TCC TBEs to model the number of
pending requests, and not allowing atomics to be issued from the TCC
until all prior, pending non-atomic requests have returned.
Change-Id: I37f8bda9f8277f2355bca5ef3610f6b63ce93563
- resources field in workload now supports a dict with resources id and
version.
- Older workload JSON are still supported but added a deprecation waring
This commit optimizes the address generation logic in the strided
prefetcher by introducing the following changes
(d is the degree of the prefetcher)
* Evaluate the fixed prefetch_stride only once (and not d-times)
* Replace 2d multiplications (d * prefetch_stride and distance *
prefetch_stride) with additions by updating the new base prefetch
address while looping
Change-Id: I49c52333fc4c7071ac3d73443f2ae07bfcd5b8e4
Signed-off-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Reviewed-by: Richard Cooper <richard.cooper@arm.com>
Reviewed-by: Tiberiu Bucur <tiberiu.bucur@arm.com>
The Stride Prefetcher will skip this number of strides ahead of the
first identified prefetch, then generate `degree` prefetches at
`stride` intervals. A value of zero indicates no skip (i.e. start
prefetching from the next identified prefetch address).
This parameter can be used to increase the timeliness of prefetches by
starting to prefetch far enough ahead of the demand stream to cover
the memory system latency.
[Richard Cooper <richard.cooper@arm.com>:
- Added detail to commit comment and `distance` Param documentation.
- Changed `distance` Param from `Param.Int` to `Param.Unsigned`.
]
Change-Id: I6c4e744079b53a7b804d8eab93b0f07b566f0c08
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Signed-off-by: Richard Cooper <richard.cooper@arm.com>
Currently, the GPU SQC (L1I$) and TCP (L1D$) have a performance bug
where they do not behave correctly when multiple requests to the same
cache line overlap one another. The intended behavior is that if the
first request that arrives at the Ruby code for the SQC/TCP misses, it
should send a request to the GPU TCC (L2$). If any requests to the
same cache line occur while this first request is pending, they should
wait locally at the L1 in the MSHRs (TBEs) until the first request has
returned. At that point they can be serviced, and assuming the line
has not been evicted, they should hit.
For example, in the following test (on 1 GPU thread, in 1 WG):
load Arr[0]
load Arr[1]
load Arr[2]
The expected behavior (confirmed via profiling on real GPUs) is that
we should get 1 miss (Arr[0]) and 2 hits (Arr[1], Arr[2]) for such a
program.
However, the current support in the VIPER SQC/TCP code does not model
this correctly. Instead it lets all 3 concurrent requests go straight
through to the TCC instead of stopping the Arr[1] and Arr[2] requests
locally while Arr[0] is serviced. This causes all 3 requests to be
classified as misses.
To resolve this, this patch adds support into the SQC/TCP code to
prevent subsequent, concurrent requests to a pending cache line from being
sent in parallel with the original one. To do this, we add an
additional transient state (IV) to indicate that a load is pending to
this cache line. If a subsequent request of any kind to the same cache
line occurs while this load is pending, the requests are put on the
local wait buffer and woken up when the first request returns to the
SQC/TCP. Likewise, when the first load is returned to the SQC/TCP, it
transitions from IV --> V.
As part of this support, additional transitions were also added to
account for corner cases such as what happens when the line is evicted
by another request that maps to the same set index while the first load
is pending (the line is immediately given to the new request, and when
the load returns it completes, wakes up any pending requests to the same
line, but does not attempt to change the state of the line) and how GPU
bypassing loads and stores should interact with the pending requests
(they are forced to wait if they reach the L1 after the pending,
non-bypassing load; but if they reach the L1 before the non-bypassing
load then they make sure not to change the state of the line from IV if
they return before the non-bypassing load).
As part of this change, we also move the MSHR behavior from internally
in the GPUCoalescer for loads to the Ruby code (like all other
requests). This is important to get correct hits and misses in stats
and other prints, since the GPUCoalescer MSHR behavior assumes all
requests serviced out of its MSHR also miss if the original request to
that line missed.
Although the SQC does not support stores, the TCP does. Thus,
we could have applied a similar change to the GPU stores at the TCP.
However, since the TCP support assumes write-through caches and does not
attempt to allocate space in the TCP, we elected not to add this support
since it seems to run contrary to the intended behavior (i.e., the
intended behavior seems to be that writes just bypass the TCP and thus
should not need to wait for another write to the same cache line to
complete).
Additionally, making these changes introduced issues with deadlocks at
the TCC. Specifically, some Pannotia applications have accesses to the
same cache line where some of the accesses are GLC (i.e., they bypass
the GPU L1 cache) and others are non-GLC (i.e., they want to be cached
in the GPU L1 cache). We have support already per CU in the above code.
However, the problem here is that these requests are coming from
different CUs and happening concurrently (seemingly because different
WGs are at different points in the kernel around the same time).
This causes a problem because our support at the TCC for the TBEs
overwrites the information about the GPU bypassing bits (SLC, GLC) every
time. The problem is when the second (non-GLC) load reaches the TCC, it
overwrites the SLC/GLC information for the first (GLC) load. Thus, when
the the first load returns from the directory/memory, it no longer has
the GLC bit set, which causes an assert failure at the TCP.
After talking with other developers, it was decided the best way handle
this and attempt to model real hardware more closely was to move the
point at which requests are put to sleep on the wakeup buffer from the
TCC to the directory. Accordingly, this patch includes support for that
-- now when multiple loads (bypassing or non-bypassing) from different
CUs reach the directory, all but the first one will be forced to wait
there until the first one completes, then will be woken up and
performed. This required updating the WTRequestor information at the
TCC to pass the information about what CU performed the original request
for loads as well (otherwise since the TBE can be updated by multiple
pending loads, we can't tell where to send the final result to). Thus,
I changed the field to be named CURequestor instead of WTRequestor since
it is now used for more than stores. Moreover, I also updated the
directory to take this new field and the GLC information from incoming
TCC requests and then pass that information back to the TCC on the
response -- without doing this, because the TBE can be updated by
multiple pending, concurrent requests we cannot determine if this memory
request was a bypassing or non-bypassing request. Finally, these
changes introduced a lot of additional contention and protocol stalls at
the directory, so this patch converted all directory uses of z_stall to
instead put requests on the wakeup buffer (and wake them up when the
current request completes) instead. Without this, protocol stalls cause
many applications to deadlock at the directory.
However, this exposed another issue at the TCC: other applications
(e.g., HACC) have a mix of atomics and non-atomics to the same cache
line in the same kernel. Since the TCC transitions to the A state when
an atomic arrives. For example, after the first pending load returns to
the TCC from the directory, which causes the TCC state to become V, but
when there are still other pending loads at the TCC. This causes invalid
transition errors at the TCC when those pending loads return, because
the A state thinks they are atomics and decrements the pending atomic
count (plus the loads are never sent to the TCP as returning loads).
This patch fixes this by changing the TCC TBEs to model the number of
pending requests, and not allowing atomics to be issued from the TCC
until all prior, pending non-atomic requests have returned.
Change-Id: I37f8bda9f8277f2355bca5ef3610f6b63ce93563
The current GPU TCP (L1D$) Ruby SLICC code had a bug where a GPU
load that wants to bypass the L1D$ (e.g., GLC or SLC bit was set)
but the line is in Invalid when that request arrives, results in
a non-bypassing load being sent to the GPU TCC (L2$) instead of
a bypassing load.
This issue was not caught by currently nightly or weekly tests,
because the tests do not test for correctness in terms of hits
and misses in the caches. However, tests for these corner cases
expose this issue.
To fix, this, this patch removes the check that the entry is valid
when deciding what to do with a bypassing GPU load -- since the
TCP Ruby code has transitions for bypassing loads in both I and V,
we can simply call the LoadBypassEvict event in both cases and the
appropriate transition will handle the bypassing load given the
cache line's current state in the TCP.
Change-Id: Ia224cefdf56b4318b2bcbd0bed995fc8d3b62a14
Print extra logs for the full/partial read/write access to the registers
through the register bank. The debug flag is empty by default and would
not print anything.
Test: run unittest of dev/reg_bank.test.xml to check the behavior would
not affect the original functionality.
run gem5 with debug flags and use m5term to poke on registers.
mtvec.mode is extended in the new riscv proposal, like fast interrupt.
This change moves that part from Fault class to ISA class for
extendable.
Ref: https://github.com/riscv/riscv-fast-interrupt
Since returned data is not needed for AtomicNoReturn and Store memory
requests, the coalescer need not spend time writing in dummy data for
packets of these types.
Change-Id: Ie669e8c2a3bf44b5b0c290f62c49c5d4876a9a6a
I believe the weekly test failures (example:
https://github.com/gem5/gem5/actions/runs/6832805510/job/18592876184)
are due to a container running out of memory when running the very-long
x86 boot tests. I found that the `-t $(nproc)` flag meant, on our
runners, 4 x86 full system gem5 simulations were being pawned. Locally I
found these gem5 x86 boot sims can reach 4GB in size so I suspect they
eventually grew big enough exceed the 16GB memory of the VM.
I have removed `-t $(nproc)` meaning each execution to see if this fixes
the issue (we may want to use `-t 2` later if the Weeklies take too long
running single-threaded).
Recent breaking changes in the DRAMSys API require user code to be
updated. These updates have been applied to the gem5 integration.
Furthermore, as DRAMSys started to use CMake dependency management,
it is no longer sensible to maintain two separate build systems for
DRAMSys. The use of the DRAMSys integration in gem5 will therefore
from now on require that CMake is installed on the target machine.
Additionally, support for snapshots have been implemented into DRAMSys
and coupled with gem5's checkpointing API.
This commit fixes a violation of the TLM2.0 protocol as well as a
bug regarding back-pressure:
- In the BEGIN_REQ phase, transaction objects are required to set
their response status to TLM_INCOMPLETE_RESPONSE. This was not
the case in the packet2payload function that converts gem5 packets
to TLM2.0 generic payloads.
- When the target applies back-pressure to the initiator, an assert
condition was triggered as soon as the response is retried. The
cause of this was an unintentional nullptr-access into a map.
Augmenting Datablock and WriteMask to support optional arg to
distinguish between return and no return. In the case of atomic no
return requests, log entries should not be created when performing the
atomic.
Change-Id: Ic3112834742f4058a7aa155d25ccc4c014b60199a
Added a resource constraint, AtomicALUOperation, to GLC atomics
performed in the TCC.
The resource constraint uses a new class, ALUFreeList array. The class
assumes the following:
- There are a fixed number of atomic ALU pipelines
- While a new cache line can be processed in each pipeline each cycle,
if a cache line is currently going through a pipeline, it can't be
processed again until it's finished
Two configuration parameters have been used to tune this behavior:
- tcc-num-atomic-alus corresponds to the number of atomic ALU pipelines
- atomic-alu-latency corresponds to the latency of atomic ALU pipelines
Change-Id: I25bdde7dafc3877590bb6536efdf57b8c540a939
pkt->req->isCacheMaintenance() would not include a check
for clean eviction before notifying the prefetcher,
causing gem5 to crash.
Change-Id: I1d082a87a3908b1ed46c5d632d45d8b09950b382
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Update the default prefetch options to achieve out-of-the box
prefetcher performance closer to that which a typical user would
expect. Configurations that set these parameters explicitly will be
unaffected.
The new defaults were identified as part of work on gem5 prefetchers
undertaken by Nikolaos Kyparissas while on internship at Arm.
Change-Id: Id63868c7c8f00ee15a0b09a6550780a45ae67e55
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
Prefetch queue entries were being squashed by comparing the address
of each queued prefetch against the block address of the demand
access. Only prefetches that happen to fall on a cache-line block
boundary would be squashed.
This patch converts the prefetch addresses to block addresses before
comparison.
Change-Id: I55ecb4919e94ad314b91c7795bba257c550b1528
Reviewed-by: Giacomo Travaglini <giacomo.travaglini@arm.com>
While we do run compiler tests weekly, 9/10 the issue is a strict check
in clang we did not check before incorporating code into the codebase.
Therefore, running a clang compilation as part of our CI would help us
catch errors quicker.
This reverts gem5#133, the temporary work-around for gem5#131, allowing
both SLC and GLC atomic requests to be made in the GPU tester.
The underlying issues behind gem5#131 have been resolved by gem5#367 and
gem5#397.
The decode_inst_dep_trace.py opens the trace file in read mode, and
subsequently reads the magic number from the trace file. Once the number
is read, it is compared against the string 'gem5' without decoding it
first. This causes the comparison to fail.
The fix addresses this by calling the decode() routine on the output of
the read() call. Please find the details in the associated issue #543
