Add initial implementation of virtual memory heap with
dynamic virtual memory mapping support for memory pools.
DEBUG_HIP_MEM_POOL_VMHEAP controls the new method.
Change-Id: I8dc5be2e0f34ab472f1800f43bb6243639a5e500
[ROCm/clr commit: 296dce5570]
There are 2 functional changes to this patch:
* Use GPU timing for internal markers for HIP.
* Measure CPU time closer to GPU timer, to reduce delta between GPU/CPU timestamp measurements.
There are some smaller non-functional updates:
* waifForFence -> waitForFence typo
* Remove unused drmProfiling
Change-Id: I4c5fa600a842ab60e454888779edcac8449a902a
[ROCm/clr commit: 179801a750]
- Resolve signal dependencies for barrier value packet if there are > 1
depenent signals. Barrier Value packet accounts for only 1 dep signal
- Better log
Change-Id: Ia506ad5d80b91d598f92e7b539f41756e9b4b64b
[ROCm/clr commit: 2d450e8b06]
wait() is redesigned with two pathes:
fast path: Use spinlock to wait for notify signal. If the
signal hasn't been received for some loops, go to slow path.
slow path: Use condition_variable's wait().
Improve monitor wrapper for better performance.
Fix some bugs left from name removing patch.
Change-Id: I893a8353121a25d11e37c8e631caf31cc1fc1f24
[ROCm/clr commit: f2ff56af9c]
Support gfx9-4-generic target to cover mi3XX.
Support features sramecc and xnack in generic target.
Improve some code formats.
Add more log on compiler.
Change-Id: I6b3c6af55c60cffd43ce6f17b75998f751b75713
[ROCm/clr commit: 3ad8f1b811]
This reverts commit 84fb57e7f9.
Reason for revert: Even though this change is valid, this would break backward compatibility.
Change-Id: I9c7cab83198c8d5c8485b11194099162e3e7a874
[ROCm/clr commit: ec6f83b544]
Support zero width and height for hipTextObjectCreate to align
with cuda.
Change-Id: I5d4c48625faf5f060ed2a7e634ec65e4ecac9da5
[ROCm/clr commit: efce2f77c4]
Add an atomic counter to track the outstanding HSA handlers.
Wait on CPU for the callbacks if the number exceeds the value
in DEBUG_HIP_BLOCK_SYNC env variable.
Change-Id: I95dc8c4bf0258c7e59411b7504220709ed6898c5
[ROCm/clr commit: 403f624bf8]
Windows alings fields to 8 bytes even with 32bit builds.
Add BUG_CLR_SYSMEM_POOL to cotnrol sysmempool.
Change-Id: I8622aabc9f7391ed7dd8583b252ce9eb41d62293
[ROCm/clr commit: 6bb7d1afdc]
- Remove the list of all chunks and use embedded chunk
information in each allocation. That simplifies Free() logic,
avoiding expensive loop if for some reason the number of
outstanding allocations significantly grew.
Change-Id: I9ea84d314320ce356ed24dd3180f262e2116c59b
[ROCm/clr commit: ad18146d8f]
Applications may submit commands withoout waits
for GPU. That causes a growth of SW unreleased commands.
Make sure runtime flushes SW queue, if it grows over some
threshold, controlled by DEBUG_CLR_MAX_BATCH_SIZE.
Change-Id: Ia4d85c24210ef91c394f638ab6b53b14323a0396
[ROCm/clr commit: 8657a77029]
- Don't generate callbacks for HIP events
- Don't process profiling info in the callback for HIP events
- Wait for CPU status update of the submitted commands
every 50 calls. That will allow to drain the commands and
destroy HSA signals.
Change-Id: Ib601a350e7e7c2b6c6209a172385389baccf73a9
[ROCm/clr commit: 364dfb0ed1]
* In a scenario where kernel is launched with hipExtLaunchKernelGGL and stop event is used, hipGraphInstantiate leaks. Since stop event is used, profiling is enabled and Timestamp (ReferencedCountedObject) is created, but it doesn't get released.
* The idea behind this solution is that profiling should be disabled when command is captured, hence the timestamp should not be created. Because information about capturing isn't available when kernel command is created, packet capturing state is used to determine whether to create a timestamp or not.
Change-Id: Ia23adac4592ded4fb5e236acf99e12e729f63692
[ROCm/clr commit: da5f1a6146]
The new set tracks only the queues that have a command
submitted to them. This allows for fast iteration
in waitActiveStreams.
Change-Id: I2c832eefa01280d9a87a5f57874d36d2e9441de7
[ROCm/clr commit: bcc545e6b8]
Ensure the member function Alloc() and Free() of command_pool_ will not be
accessed after command_pool_ be destructed.
Signed-off-by: Chong Li <chongli2@amd.com>
Change-Id: Ic2d36423302518a030bd61fa399290ebe2ed8194
[ROCm/clr commit: e6a5c81221]
Fix random language string that leads to compiling failure
of trap handler and TDR of hipMemset() on VM in release
mode of hip-rt
Change-Id: Ie1d874742b804f62ceda68064fa54f5d39c092b8
[ROCm/clr commit: 857d0d60b9]
- Added the optimized multi stream path in graph execution. It uses a fixed number of async streams in the execution
- Optimize the launch latency, where commands
creation and execution is done at the same time
- Optimize the scheduling to use less barriers and waiting signals if
the same queue can be detected
- The new path is controlled by DEBUG_HIP_FORCE_GRAPH_QUEUES
environment variable, where 0 will use the original path and any other
value will force the number of asynchronous queues for execution
- DEBUG_HIP_FORCE_ASYNC_QUEUE can force single queue async
execution in graphs(applicable for Navi families only)
Change-Id: I7eb40bc15c45f508d6911868a6f6d4c3598d380e
[ROCm/clr commit: 9db52f9a46]
=> Added support to capture multiple AQL Packets.
=> Added Interface to callback to hip runtime from rocclr to allocate
kernel args from the graph kernel arg pool.
=> Enabled Support to capture memset node.
Change-Id: I7e1c2ba06927459e024653058af142bd82192c43
[ROCm/clr commit: bd3a35bde1]
hipDeviceSynchronize called from __hipUnregisterFatBinary
accesses static maps and monitors. This change ensures these ojects
are not destroyed before __hipUnregisterFatBinary is called.
Additionally it disables the teardown process for static build.
Change-Id: I46b58641d60efcf6637a8e99cdd786ffe9e2c77d
[ROCm/clr commit: 9b33db9b24]
Capture AQL packets during GraphInstantiation and enqueue AQL packets during graph launch.
Added support to capture single graph memset node.
Capture support for memset node is currently disabled.
Memset capture will be enabled when capture for multiple packets are supported..
Change-Id: I14dfbc41731025cc3a548a730558915def3fa384
[ROCm/clr commit: 346da4bb40]
Implement std::mutex based monitor that has much
simpler logics than legacy monitor.
Create DEBUG_CLR_USE_STDMUTEX_IN_AMD_MONITOR to
toggle them.
If DEBUG_CLR_USE_STDMUTEX_IN_AMD_MONITOR = false
(by default), use legacy monitor;
If DEBUG_CLR_USE_STDMUTEX_IN_AMD_MONITOR = true,
use std::mutex based monitor.
If no perf drop of stl::mutex based monitor,
legacy one will be removed later.
Change-Id: I1d21368ff462477d3238d71e4e2a1a7d6b9167ad
[ROCm/clr commit: 73c02041e1]
- awaitCompletion code may do a endless spin wait for cases where we
dont submit a handler. One such case can be the hipExt*Launch API which
takes a stop event. In that case we optimize the stop event by attaching
a signal to the dispatch packet but dont submit a handler when we attach
the signal. That means if awaitCompletion() is called after that, we
would keep on waiting on command status on the host rather than simply
checking signal value.
Change-Id: Ie8bf175aeefa3f9e4299b1ae7ae9108dad67e283
[ROCm/clr commit: 561fb8a459]
We must be in protected way to get last command when calling
awaitCompletion() where lastCommand will be released and
possibly destroyed.
This can solve scope lock(notify_lock_) crash in
Event::notifyCmdQueue() with AMD_DIRECT_DISPATCH = true.
Change-Id: I4297166f912a71112f4a8945d993160ba9afdc34
[ROCm/clr commit: 749385155a]
1.Move global amd::monitor listenerLock before global
class runtime_tear_down as it will be referenced in
~RuntimeTearDown() after main(). It should be freed
later than runtime_tear_down.
2.Update Device::~Device() to SVM free coopHostcallBuffer_
before context_ is released and freed.
Change-Id: I1d21378ff463477d3238d71e5e2a1a7d6b9147ad
[ROCm/clr commit: 544c45364f]
Adding a safety check prevents an invalid memory access
if timestamps and kernelNames vectors are of different size.
The patch also moves the addKernelNames for the accumulate command
into dispatchAqlPacket function.
Change-Id: Iea0927e1253800403a1ae3f3d72de1e7d96476c3
[ROCm/clr commit: d44f44a5b1]
Fork() duplicates all system memory resources, but runtime can't duplicate
GPU resources. Thus, avoid tearDown() calls for the child process(s).
Change-Id: Id6b12bacd5112b9ad3747c218e09cba98ea1b42c
[ROCm/clr commit: ae2992ea43]
German Andryeyev
Aidan Belton-Schure
Saleel Kudchadker
Tao Sang
Pengda Xie
Anusha GodavarthySurya
Sourabh Betigeri
Jatin Chaudhary
Maneesh Gupta
Rahul Manocha
Julia Jiang
Todd tiantuo Li
kjayapra-amd
Branislav Brzak
Ioannis Assiouras
Vladana Stojiljkovic
Chong Li