Add ROCR memory detection and enable arena mem object for possible
access in HIP
Change-Id: Icf86ac789176bfee4ea8d36b0970a817d4c6a2f7
[ROCm/clr commit: 28597ec5b5]
Tensor flow build console shows "hip/lib/.hipInfo: No such file or directory"
Fixed by adding soft link for .hipInfo and this will maintain backward compatibility for file reorganization
Change-Id: I91c17115b56bbcabeb86cbcd12d3fb9aaa8f847e
[ROCm/clr commit: 1a4fd9b7ea]
hipErrorHostMemoryNotRegistered should be returned when an
unregistered pointer is passed to hipHostUnregister
Change-Id: Ie76086b602bde2cd7a7a2b049cd03d29454ff306
[ROCm/clr commit: ed4014e519]
- Fix a crash with AMD_CPU_AFFINITY=1 as numa_bitmask_alloc isnt the
right api to allocate bitmask
- Do not set affinity for ROCr thread. It worsens performance rather
than any improvement.
- Fix regression from my previous change for event handler.
Change-Id: I3ea75adc2a6333f29752283eddd5b555e9b58cc5
[ROCm/clr commit: 802c2c8a9f]
- Queue handler for hipEventRecord(aka marker_ts_) only if there is a
callback associated with it.
Change-Id: I8a9877ae0e342556053abbaacc9510744a8e772a
[ROCm/clr commit: 3c3c0ca4c5]
This reverts commit eff4583118.
Reason for revert: Catch2 test fails
Change-Id: If0ea0eab16b50f27c9acbc471530266917773ff1
[ROCm/clr commit: d888c0a9bf]
Flat folder structure requires hip to be installed in /opt/rocm and this need a change in build script
This patch is a workaround, till build scripts changes for flat folder structure are merged and will be removed once build script changes are merged to amd-master
Change-Id: Ib6f51b9d4e89afcc954a5d96595ce471cf7bd67a
[ROCm/clr commit: ef32ae6ada]
Update timeout for hostcall wait for signal. If the timeout is small it
checks frequent enough to affect performance for certain applications
which may be CPU bound.
Change-Id: I0a879559e4ad111b09a994a5b82a6faf6e4fea3f
[ROCm/clr commit: 9292abb2d8]
- the pointer attribute is fetched for might not be the current device
Signed-off-by: sdashmiz <shadi.dashmiz@amd.com>
Change-Id: Ibcea92b52da31adba5f62baae25b55671704c005
[ROCm/clr commit: ae506b40c0]
Queue creation and release fails during hipFree causing seg fault
if the null queue was not created until then and there is no
device memory left.
Change-Id: I118b86b58e777dac5415bc9ba0c4e23c986efa7d
[ROCm/clr commit: 4a90c3ccaf]
It can be too early to allocate memory at the begining of
Device::create() under PAL
Change-Id: I4bd76db7be3f6fb246243ea68022d8b0f860471d
[ROCm/clr commit: 3af3fe10de]
HIP installed in /opt/rocm
Wrapper header file in /opt/rocm/hip/include
Soft link for cmake and library files in /opt/rocm/hip
Depends-On: I30adccf7e6241dae12270156602d4b033ea99d93
Change-Id: If70b61020fa3ee85bf49171c842837da05191a48
[ROCm/clr commit: 4bcccb60c9]
This reverts commit bc59991924.
Reason for revert: want to add one more check
Change-Id: I02b621d0c98b9000f75ea214cf5b4967cfaee95d
[ROCm/clr commit: f8f8b0fd95]
Addresses the negative scenarios and assigns default NULL values to
the queried attributes when the pointer encountered is not valid.
This is to match the corresponding API's behavior in CUDA.
Change-Id: I4ce400b02416d455ec9649f731af2e327dccfe71
[ROCm/clr commit: a554d03b9a]