Use HSA_AMD_SVM_ATTRIB_AGENT_ACCESSIBLE flag for the initial
allocation instead of HSA_AMD_SVM_ATTRIB_AGENT_ACCESSIBLE_IN_PLACE.
Change-Id: Ia52fe205563df1ea916dc2dc81e749e11c16f83d
This addresses the rocprof hang seen with direct dispatch. The
workaround queues the handler back if any of the signal value in the batch
is not decremented. To rememmber the last position in the list, we save
the parsed command in the current timestamp struct.
Change-Id: I02959e463cfe3cee83c54808ffd6e6f48f43b4e8
hipIpcOpenMemHandle should return the device pointer which is
similar to the base ptr of the original allocation even if the offset
to the original pointer is passed to hipIpcGetMemHandle
Change-Id: I99c0553e8c67c15b5fed880b6a4c74bce39c3aee
MT doesn't use GPU waits, but CPU for sync between engines.
Change the threshold values for CPU waits for direct dispatch.
That will bring behavior closer to MT.
Change-Id: Ia41c3cb812614962aff2746b6cf858f1bf77dda2
In adition to removing the HSAIL logic from the ROCm backend, guard all
of the HSAIL includes in the common layer behind the WITH_COMPILER_LIB
define. This is to avoid including HSAIL headers when building with
no support for it.
In common logic replace the use of the aclType enum with the new
Program::file_type_t enum. This is essentially a local copy of the HSAIL
enum to avoid including any HSAIL headers.
Change-Id: Ica0651d1b29dfccc255cc584eb82a5cb35e1b520
The check has to be performed inside the signal loop, because
active signals need to be processed to avoid a stale timestamp
class.
Change-Id: I26af8287aae18eb19c096d9358cd0b86cfd1c561
- With direct disaptch profiling state is enabled to trigger the
callback on HSA signal. However ROCr has very low peformance on
the first call to get the profiling info. That impacts some tiny
performance tests.
Change-Id: Idacd1b10a473fcfb5feef3074b7191d35743f769
The copy image workaround could be recursively used by ROCclr blit kernel.
Avoid such situation by using stack variable.
Change-Id: Iadaa8cad9216220194760dd461a9533bb236aea0
Since the majority of the Hostcall implementation now sits in the
commmon layer, the PAL backend simply just needs to invoke it. One thing
that is missing though is HSA signal support.
The newly added pal::Signal class is a light emulaion of what HSA
signals provide. The current implementation is just enough to get
Hostcall working, but it can be expanded in the future if needed to
fully emulate HSA signals.
The major difference for now between PAL and ROCm hostcall
implemenations is that PAL doesn't support blocking signals. This will
be enabled in the near future. For now use active wait for PAL.
Change-Id: I746557354ab9d71a7d4a31f9320fcc2fee5aee7f
HIP requires to return AccessedBy query for all device, but ROCr
can process one per query. Hence send the queries for all
available devices and then accumulate the results in runtime.
Change-Id: I082f9adb8e31c775a8ad1bf7a5af37440ef4bd16
HIP tests require HIP callbacks to be processed in another thread.
This change will use a thread from HSA signal callbacks to make
sure a HIP callback was done asynchronously.
Also process the callback before changing the status of command
Change-Id: Icef85d0e0f808663882cf6881ff1be3e5eca29ac
- Don't notify if the batch is empty, because that means
the current command was processed already.
- Disable pinning optimization to avoid a race condition on stall.
- TS marker submition requires extra AQL barrier
to track the status.
Change-Id: I17eff4ad12ac66cfe1bb44048bebb1891805279d
GPU waits have noticeable overheads on compute with extra
AQL barrier packet and on SDMA with power saving features. This
change introduces a wait on CPU for 30 us in case the app has tiny
operations.
Change-Id: I761ba3af595f3f48544980058a9077dda15aa5f9
This change unifies the hostcall implementation for all the backends,
by pushing the common logic to the device layer. This is done by
replacing the use of hsa_signal_t with device::Signal (a light wrapper
around it).
Change-Id: I7b6fca7930b5a0b199da5d85e2e048354cc04e7b
- Avoid GPU wait on the marker submission and update the command
batch after HSA signal callback upon HSA barrier completion.
Change-Id: I5c1c97212aefc2ae4b99aa9e2a81627ee9a38c1c
In pitch workaround, we need to copy the image to copy buffer
when application wants to read image to buffer. After this
patch, we unconditionally copy the image data to the copy buffer.
Change-Id: I71b0d19459542dfbb3ca51a2c8a3a81367fa2fb5
The existing workgroup calculation logic for GWS initialization is
incorrect. It tries to add together workgroups across dimensions,
leading to major under-count in 2D and 3D kernels. An (x,y,z) kernel
uses x * y * z blocks, not x + y + z.
In addition, the previous logic was incorrect for the case of launching
a single-threaded kernel. It calculated 0 workgroups, leading to
initializing GWS to -1.
Change-Id: I1bb20a0d5b6e0cc10ac55901c28d8f93aac61c09
kjayapra-amd
Jason Tang
German Andryeyev
Ravi C Akkenapally
Saleel Kudchadker
Brian Sumner
pghafari
Satyanvesh Dittakavi
Vladislav Sytchenko
Sarbojit Sarkar
Alex Xie
Tao Sang
Payam
Rahul Garg
Joseph Greathouse