The "optimized" version of memcpy is outdated and
was used in win32 only.
Change-Id: I7f2e0e9051e37cec95438266824b5b0025c324c6
[ROCm/clr commit: 7448113cfc]
Avoid a deadlock on the host call buffer creation. Since the buffer will be
allocated in the queue thread, then use direct device memory allocation
skipping the global context lock.
Change-Id: I09b55ee03bb42ab5d320c152b52a8c842c5fdcc1
[ROCm/clr commit: 62559a6e5a]
- Create a vector to allow multiple TS to be stored in Command.
- This would mean we dont wait for entire batch in Accumulate command
to finish when we exhaust signals.
- Reduce the number of signals created at init to 64. This min value
may still need to be tuned but the KFD allows max of 4094 interrupt
signals per device.
- Store kernel names whenever they are available and not just when
profiling. If we dynamically enable profiling like for Torch, a crash
can happen if hipGraphInstantiate wasnt included in Torch profile scope
beacuse we previously entered kernel names only when profiler is
attached.
Change-Id: I34e7881a25bbc763f82fdeb3408a8ea58e1ec006
[ROCm/clr commit: c157bfb202]
PAL optimized the logic for the barriers, which caused failures with CP DMA on Navi4x.
Change barrier's code to match the most recent PAL optimizations.
Change-Id: I55eeab20f51eb8e920bcbb4b55fbe3c7f77fd3fa
[ROCm/clr commit: 1239309c90]
__amd_streamOpsWrite blitkernel in device-libs has only 3 args.
so getting rid of the 4th unused arg (sizeBytes)
Change-Id: I81cc1107f8b424bf58558c93a2495a1b878aef91
[ROCm/clr commit: e643406caa]
Extension allows to execute the kernels without a wait barrier and L1
invalidation.
Change-Id: I96c485204303f54a0240b93134f4560673e4bd17
[ROCm/clr commit: 13c6f56ca9]
Luxmark still uses HSAIL path and one subtest can benefit from the wave limit.
Change-Id: I16c94e09cd6e2afd6341cb76bf2e9ab7b7713214
[ROCm/clr commit: dec1158d04]
Add GPU_DEBUG_ENABLE to control ttpm behavior. If enabled,
then HW will collect more debug info at some perf cost
Change-Id: Icee0686b903a7b1bd483710b9d611877cd43c6aa
[ROCm/clr commit: 7d661bc7df]
Extra CPU read back will be performed before every submission to make sure
previous writes over PCIE reached GPU. HDP flush is done by CP.
Change-Id: I402d28ca26c8cee4a3920feb3599af8c285d0889
[ROCm/clr commit: cfc07c88ee]
- Add the new fillBuffer kernel, which allows to launch a limited
number of workgroups for memory fill operation
- Switch fill memory to 16 bytes write by default
- Allow to limit the workgroups with DEBUG_CLR_LIMIT_BLIT_WG
Change-Id: Ibad1822f2d42b2fc71bcfc1917c31409c0623e8e
[ROCm/clr commit: f1dc81f427]
Add support of HIP_FORCE_DEV_KERNARG under PAL.
Fix persistent memory detection for a resource view.
Change-Id: Ifb7db2db14e0c2205a9661cfa53887ec61ab26a4
[ROCm/clr commit: 5f297d75d9]
- Track all captured commands under a new AccumulateCommand
- Add begin() and end() methods to capture commands
- Explicit TS object now passed to certain methods because
profilingBegin() and profilingEnd() now happen separately and thus can
run into threading issues
Change-Id: I171106bdcad72b057836cb2f3fc398db3533119f
[ROCm/clr commit: 40f41f4d0b]
Restore PAL platform destruction.
Update CmdAllocatorCreateInfo::AllocInfo for the new interface.
Change-Id: Iea418eed7ee26166039a4a9cc1999438856e9097
[ROCm/clr commit: bd00826446]
This reverts commit cab71e6e00.
Implement the right way to make ExternalSemaphores be signalled
only after prior works on the stream have been finished.
Change-Id: I9d5974e05d5f229170b928db4566c14e40e3cbaa
[ROCm/clr commit: d433df4761]
- Program unique AQL index for debugger. The logic manages AQL array of packets per HW queue.
- Provide debug state to PAL
Change-Id: I38fa1f5435fa711fd1d44dc391f2e61eb2a25efa
[ROCm/clr commit: d97cc0abbd]
The change enables VM support in graphs on Windows. That allows
to avoid caching of all allocations at the cost of map/unmap
overhead during memory create/destroy.
Change-Id: I792be00fba099e5e5d3cd44a963e1dfd6976a86d
[ROCm/clr commit: 04b696abee]
- The implementation in mempool graphs requires refcounting VA object.
That requires release() to update the map only on the actual destruction.
- Add GPU event tracking for paging operation. Otherwise, runtime
may not always flush IB.
Change-Id: Idf99ffb894321a38e04b490116a7ca435635918d
[ROCm/clr commit: 7ef2da5aba]
This reverts commit dfa7790030.
Reason for revert: Deferred to a future release.
Change-Id: Ia66c37f0ab9734dee73c930d10d7469d5fd57254
[ROCm/clr commit: 5dc104b3ea]
Windows kills threads on exit without any notification. However,
runtime can still destroy VirtualGPU object from the host thread with
HostQueue destruction.
This change also forces RGP trace transfer on the last capture without
any delays.
Change-Id: I768e87e99e1d23a021e63c12f36e450817743759
[ROCm/clr commit: ad33a021cb]
OCL runtime uses WGP mode and total CU count reported in WGP.
Realtime values are still in CUs. That can mislead in the test results.
Report realtime in WGP values and convert to CUs for KMD.
Change-Id: I90b82615640734dd655be2b613ccac3cb8483239
[ROCm/clr commit: 0703b8380b]
HIP can't rely on the resource tracking, used in OCL and requires different explicit sync.
Make sure ROCCLR syncs compute only when SDMA is used and vise versa.
The new logic will allow to enable CPDMA without unnecessary waits.
Change-Id: Ib9d1788cfd5afa5ea2fec4c96a37d8b9c4d0059d
[ROCm/clr commit: ff6b4db70b]
- In case of HMM, use blit kernel instead of CPU memcpy for hipMemset
Change-Id: I89bfc96ff01a2375ed8df1b1c6bc05357dea84f7
[ROCm/clr commit: f097cda948]
Introducing a VirtualMemObj map as it is needed to differentiate
between virtual address ranges and actual physical memory
This is because a whole VA range can have several physical memories
as chunks.
Change-Id: Ie2a972b4faf3f7d552cfa53e77898f80ad75740a
[ROCm/clr commit: 905088e4e7]
Implement map/unmap for PAL backend
Create commands since PAL uses the IQueue to map/unmap
Change-Id: I97e26a7d28ae5e10774c9ca65307153100945621
[ROCm/clr commit: 67657d6099]
This code change is to improve error handling.
This code change does not fix issue itself.
Before this code change, hostcallBuffer_ point is initialized in the end of
create() function. If create function fails and returns early,
hostcallBuffer_ point is not initialized. This non-initialized point can
cause access violation when object is destructed.
This code change put the initialization of the pointer in the constructor.
Change-Id: I7fb6e764eb0547196dca03db237e49d3ff0fd06a
[ROCm/clr commit: 5528812aa9]
Pass active queue for transfers in the cache coherency layer.
That will allow to use device transfer queue only for
cases when active queue isn't available, because using device
transfer queue from another active queue may cause a deadlock
Change-Id: Ifbe7e0303b77dbf6eeda3939ffbc25a3df7472de
[ROCm/clr commit: 95d55fdfa8]
Metadata in Codeobject version 5 is the extension of CO3 and CO4.
Add the detection of the new fields and program them in
the setup of the kernel arguments.
Change-Id: I27e58df77320ad00f4f16d35912668db803826af
[ROCm/clr commit: be6a06384e]
kjayapra-amd
Rahul Manocha
Ioannis Assiouras
Payam Ghafari
German Andryeyev
Saleel Kudchadker
Ajay
Tao Sang
Jaydeep Patel
Maneesh Gupta
Xie,AlexBin
Todd tiantuo Li
Rakesh Roy
Christophe Paquot