Unit_hipMemPoolApi_BasicAlloc expects to work on device 0, but other
tests will set not-0 devices in mgpu. This leads to hang of
Unit_hipMemPoolApi_BasicAlloc. Fix by set device 0 in head code
of Unit_hipMemPoolApi_BasicAlloc.
SWDEV-508872 - Fix Perf_hipPerfMemFill_test
When mem size is 2G, the test is so slow that it looks like stuckness.
Set top mem size to 1G can make the test pass in an acceptiable time.
Change-Id: Ie26dbf597e5ba8cb898d1aae5ed5ecf0267c3228
[ROCm/hip-tests commit: 94eea4db59]
- Removed unnecessary hipEventRecord and fixed time calculation in
hipPerfDispatchSpeed.cc where it was off by a factor of 1,000.
Change-Id: If538e1d236cf0e6d3c69caf7af53c9095d812ad6
[ROCm/hip-tests commit: b1f9f86543]
Having same target name causes same includes to be called twice
Change-Id: I53469a07e6dee375ea4a4700ccac3c9487b79e4a
[ROCm/hip-tests commit: c03ad253fd]
- Do an hipEventRecord on null stream, that creates the streams and
avoids stream creation overhead when we time the core functionality
Change-Id: I117dccc42c92836fa113214d31bf14da49deba77
[ROCm/hip-tests commit: fb5e1d33d9]
Correct the size of allocated buffers.
Extend the number of executed tests
Make sure warm-up finishes, before starting the test
Use a non-blocking stream for Async tests
Align up the output with results
Change-Id: Ie107fd83c0a95dacb537d8bca0b534cf6a6d5032
[ROCm/hip-tests commit: 9971540ac8]