Commit Graph

8 Commits

Author SHA1 Message Date
Maneesh Gupta c4a93adb74 Replace hipLaunchKernel -> hipLaunchKernelGGL
Change-Id: I4d99009e1199811d417becf1e1b934ec4d4e30be


[ROCm/hip commit: 52e320f396]
2018-10-17 14:32:25 +05:30
Maneesh Gupta 4f42ee762d Apply .clangformat to all repo source files
Change-Id: I7e79c6058f0303f9a98911e3b7dd2e8596079344


[ROCm/hip commit: 1ba06f63c4]
2018-03-12 11:29:03 +05:30
Alex Voicu 70a41e7dac This switches HIP from its currently convoluted macro + pfe based dispatch mechanism to a more natural one partially based on the existing module API. The basic idea is that HCC will always correctly emit __global__ functions: as empty-bodied stubs, on host, and as kernels, on device. It then becomes trivial to obtain the mangled name on host, at dispatch, from the function's address, and then to use the mangled name to retrieve the kernel. This should address all problems stemming from serialisation, dubious mismatches due to the manufactured functor, macro-isms et al. It also immediately enables support for generalised globals as a consequence of that being available in the module API. Finally, it will make debug much easier, since the actual names of the __global__ functions will automatically be used in traces etc. One detail is that due to how dispatch works now (hipLaunchKernel and hipLaunchKernelGGL are themselves variadic function templates which deduce the function type of the callee), in certain cases it may be necesssary to insert explicit casts to ensure that the variadic argument list selects a viable overload - this can be observed in some unit tests. Eventually we may be able to remove this limitation, but for now it does not appear terribly onerous. The code is not extremely HIPpie, nor is it fully optimised, but rather is intended as a starting point for the HIP team to make its own.
[ROCm/hip commit: c2482d1255]
2017-11-01 15:09:59 +00:00
Ben Sander bd7a374f20 Make hipMultiThreadStreams1 test a little harsher.
Fail faster if synchronization rules are violated.
Run vectorAddRevers to read last elements of array first - if the
vector add kernel starts before preceding copy finishes we
will read stale data and flag the error.

Increase default array sizes, so synchronization errors more easily
exposed.


[ROCm/hip commit: 2e1fec47ab]
2017-05-16 19:04:25 -05:00
Maneesh Gupta 2799863a54 Fix compilation of some broken tests on nvcc path
Change-Id: I92406af00c2de09d728e9b7f661c1f3873470560


[ROCm/hip commit: f1c3dd0dff]
2017-02-27 13:17:51 +05:30
Maneesh Gupta 724a024bb6 directed tests: Updated to use new hip include path
Change-Id: Iab1aee638b7158eb9674a19625dedf6471947e51


[ROCm/hip commit: 7e04884d0f]
2016-10-04 22:20:50 +05:30
Maneesh Gupta a826689145 directed tests: Enable remaining tests in HIT infrastructure
Change-Id: I800ee00d6f7d339bb5d2ed898a16362a8541ed3a


[ROCm/hip commit: 4807279988]
2016-09-27 17:24:33 +05:30
Ben Sander 89a14a6d73 Refactor directed test infrastructue.
- Add hierarchy.  Tests now live in directories, each with its own
  CMakeFiles.txt.  Reduces merge conflicts.
- Change make_hip_executable -> build_hip_executable.
- Refresh docs.
- Enable some tests that were previously built but not run.

Change-Id: I8c5de3c954400bf233904282b8b42861a2b7c536


[ROCm/hip commit: 1b69b2e9f0]
2016-06-17 15:18:57 -05:00