- Use getBuffer/releaseBuffer in BlitManager
- Cleanup XferBuffer as we use ManagedBuffer for both reads/writes
Change-Id: I2661b85dd012763b17a38a743fec1b1d79125f67
1) Create getBuffer method to return pinned host memory or staging buffer
2) for D2H path use managed buffer instead of static buffer
3) use staging buffer copy for 16KB < size < 1MB
4) use pinned memory copy for size > 1MB
Change-Id: I13d4d6ab60691bc6c7724239db1e11e23f0f3dc2
- Fix regression for D2H pinned copies which adds systemscope release.
- Skip cpu wait for D2H unpinned copies as we can pass the signal of the
barrier to rocr copy.
- Fix an old bug in sdmaEngineRetainCount_ logic
- Improve logging
Change-Id: If074bddb05564b15949b0d5f9bf12acd3692174e
BatchMemop should be positioned before the image support kernels
because the total number of kernels is determined by BlitLinearTotal,
when there is no image support on the device.
Change-Id: I8e53caf744ba54259ac04bad1762eef21806f3f2
We are passing this arg as an address, and memcpy complains about
overreading (8 bytes instead of 4).
Change-Id: Ica9207f6c5f6056a4bfc968280c76e779ded13ae
- Refactor blit code and clean ASAN instrumentation
- Use unified function for rocr copy
- Enable shader copy path for unpinned writeBuffer/readBuffer paths
- Set GPU_FORCE_BLIT_COPY_SIZE=16 which means we will use BLIT copy for
pinned copies or unpinned H2D/D2H copies < 16KB
Change-Id: I42045cca79234b340dbf53dafb93044199736ae4
Although unpinned copies require synchronizations
in HIP, runtime can avoid syncs for H2D copies with
a staging buffer
Change-Id: If2203c6bc0cbd89742823688dc8e89e9acd873b2
- Gfx12 TCC cacheline size is 256B, Increase to have alignment
compatible. Eventually this needs to be replaced with what the query
returns.
Change-Id: I545929446c4faa3f26872a6290b3a89657888596
If we are using the mask returned by getLastUsedSdmaEngine() then we
need to apply the SDMA Read/Write mask to it before using with HSA
copy_on_engine API.
Change-Id: I6e5dc6c187eeb3c61ee159e9d2a0fa7b4737c06e
The copies can get blocked if the last SDMA engine is used by another
copy and this can lead to perf drop in some of the tests like Gromacs.
Resetting the last engine by checking the engine status and fetching the
new mask after few copies can avoid this.
Change-Id: I8fe8ea678db508d291c6242f3741fa9215e99921
Add hipMemcpyDeviceToDeviceNoCU to force a non blit copy path. This
helps in cases where an app may determine that CU may be busy and copies
with SDMA may be quicker.
Change-Id: I59b415dd8f6022c244e8d75f265464d5c635df1e
- Remove large bar memcpy path. Since we end up waiting for a barrier,
its defeating the true intent of the copy, Also memcpy over PCIE\XGMI is
introducing variability in perf for HPC apps like GROMACS
Change-Id: I3b5c9d9ce93333959c39023bf4f703e2ccb6e3af
- Maintain a map of SDMA engine# to stream allocated following a greedy
approach
- Anything past that will query SDMA engine status always and go with a
SDMA or Blit copy path
Change-Id: Ibfaed7f951ab84d80cb0430596a4d11b5aec9202
- Leverage managed buffer that would use chunks for fill pattern. Use a
different chunk for the next fill to avoid wait
Change-Id: I254483c867e112f66564ffd8f55e0a605d8896c9
For the fillBuffer shader, if there are two 32bit writes to a MMIO
register, it can get dropped. It has to be a single 64bit write.
Add optimization to fillBuffer to write 64bit and 16bit writes.
Change-Id: I3aa78e027898f8ae01e9c8f09004615673720c2b
- The logic will trace compute, sdma read/write operations and
apply signals when necessary
- ROC_CPU_WAIT_FOR_SIGNAL, ROC_SYSTEM_SCOPE_SIGNAL
and ROC_SKIP_COPY_SYNC were added to control the tracking
Change-Id: I9e8e6174c63bf7784f7ab00964e2918c8667d364
Implement the global class for signals tracking per device queue.
Switch to the new tracking mechanism.
Change-Id: I3c4dda04b34e6d18d6a95510d84102909633b415
When HIP_ENABLE_DEFERRED_LOADING=0, many global variables will be
referenced but they are not initialized in that early time. The patch
will use constexpr to initialze global constant varables in compile
time.
Change-Id: I9d538b7abc6a0ce700ec3332b97fc144db5fc1ef
Optimization for the fence release removed a sync for mem fill.
Add simple const buffer management forr the filled pattern to avoid
pattern overwriting with the async fills.
Change-Id: I63773ac09ceec31d5396d24570e4647ff096326b
[ROCm][TCT][HIP] cooperative stream test case is failing.
Make sure lockXfer() in the blit manager returns a valid value.
Port the latest PAL backend logic into the ROCr backend.
This change doesn't fix the issue, reported in the ticket.
Change-Id: I54101a824f49a2dcfbbf5414cb5b3af41745306d
Saleel Kudchadker
Rahul Manocha
kjayapra-amd
Ioannis Assiouras
Jimbo Xie
Jatin Jaikishan Chaudhary
Sourabh Betigeri
German Andryeyev
Satyanvesh Dittakavi
Anusha GodavarthySurya
Ajay
Ravi C Akkenapally
Jason Tang
Tao Sang
Michael LIAO
Alex Xie
jujiang
Laurent Morichetti