Adobe AE allocates/destroys system memory per each frame
in some effects, which is a very expensive operation.
Disable pinning and force sysmem allocation in VidMM.
That allows to trigger resource cache in runtime and
avoid paging operations, caused by pinning.
The actual benchmark doesn't have significant improvement
from the change, but GPU view traces are much cleaner.
Change-Id: I92cdd87d37c64611788f05250f71d6fe489b4d7e
Move hidden heap creation to the kernel launch to make sure it's
allocated on the actual first usage.
Change-Id: I1b65a82fc06d9129ed45a69765bf14ea3d945b04
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
The heap must be cleared once per device, but ROCclr doesn't
create a queue per device in HIP. Hence, the clear operation will
be performed during the first queue creation.
Change-Id: I52ceb06d67d11cde6d019c5ab510059f426a9bfb
Adding virtual memory management APIs to rocclr.
The HIP layer will handle virtual allocs on devices.
Change-Id: Ia978f105c2c3fed3959c77580ba228e845105754
OCL2.2 requires SPIR-V and runtime doesn't support it.
Make sure PAL backend doesn't report any SPIR-V support.
Change-Id: I8d179069674205b54f7d20d149bcb675bee5cdb0
Use HSA_AMD_AGENT_INFO_COOPERATIVE_COMPUTE_UNIT_COUNT to get compute
units. This is needed to work around assymentric CU harvesting bug on
gfx90a. Add a new device property to get the max available CUs on the
device.
Change-Id: I878f38f14f16c1af01fc0a77157aea1e816a63b8
Report proper target id for xnack in HSAIL path. Runtime
will use ISA table and report hsailName().
Fix offline compilation path for PAL.
Change-Id: Ic0250bf6b9c193d867aec9800a319da1bf00c3ee
- Create hash values for binaries
- Add the binaries into RGP trace
- Add corresponding hash value for every dispatch
Change-Id: I2c3ce004d69f37d0d46bc4744e12f24273517f5e
Redshift sees around a 3x performance uplift this change.
Turning this on for OpenCL might cause unwanted behaviour, due to
apps like RSX running in the background all the time.
Change-Id: I9f32d5f2e05b6697a8aaa9ddf74474b5531bb7e1
Below logic allocates the host buffer whenever a subbuffer is created
from a SVM allocation. This is only needed for multi-device contexts.
HIP does not support multi-device contexts, hence this logic just ends
up performing unnecessary system allocations.
Change-Id: I8eae635f7c5289c52ef73434218c1658b788a456
Revert back to using the Raven (gfx902) target ID for Raven 2 (gfx909).
This is due to the HSAIL compiler not supporting gfx909.
In theory there should be no issue with running Raven isa on Raven 2.
Change-Id: I425edebc99075799eda5522fad231b8fb3184873
Anusha GodavarthySurya
German
Jason Tang
Todd tiantuo Li
kjayapra-amd
Ajay
jie1zhan
Sarbojit Sarkar
sdashmiz
Tao Sang
Christophe Paquot
Saleel Kudchadker
Satyanvesh Dittakavi
Julia Jiang
Vladislav Sytchenko
Brian Sumner
pghafari