The vector with all kernels is preallocated on the executable init.
Thus, reduce the scope of global lock to the binary creation only.
Change-Id: I73035013a6562175069137e895bba815f466ee35
[ROCm/clr commit: 0640d36019]
1) For Dynamic CO variables, free the device pointer in
DynCO destructor instead of DeviceVar destructor.
2) For Static CO Remove Fatbinary,
only call hipFree for valid device Vars instead of all devices.
Change-Id: I84291f5371b2c05d1d0bcdb4f9c6bd122e7c9b21
[ROCm/clr commit: a2d8199402]
Modified hipFuncSetAttribute to handle pointers to dynamic functions
returned by hipModuleGetFunction.
Change-Id: I54b98f9d31a79630dd7edcd363fad81f1d89219b
[ROCm/clr commit: 1a710dabcf]
- Remove newline from logging as log function internally inserts a new
line
Change-Id: I25eb2242a1f1e87cf811bcc373d1d485b2e027a8
[ROCm/clr commit: 058b2702db]
- Return error code instead of segfault during kernel launch when library isn't built with current gpu arch
- Fix string print bug
Change-Id: I91aaa50150fee220317cd2eb4e075d825cc448b2
[ROCm/clr commit: 76c8b3157b]
This reverts commit f870f04dd9.
Reason for revert: Doesn't match with CUDA behavior
Change-Id: I413f3d241f864c1e7c21681c5fff8f216ef54306
[ROCm/clr commit: 4f5d180ca7]
- Add helper function to validate whether a dynamic function is present or not
- If getStatFunc fails, then check whether the function passed is a valid dynamic function
Change-Id: I4141172f42dce7ff689709560c483e602b5dad4b
[ROCm/clr commit: b4d8586a11]
German Andryeyev
Rahul Manocha
Ioannis Assiouras
Jaydeep Patel
Ajay
kjayapra-amd
Saleel Kudchadker
Rakesh Roy
Satyanvesh Dittakavi
anusha GodavarthySurya
Julia Jiang
Rahul Garg