294bf50f68
Fix hipMemset stream resolution
2018-04-11 19:01:53 +05:30
412a35be20
hipMemset refactoring
2018-04-11 15:58:48 +05:30
03eca1c57e
hipMemcpyAsync returns success when trying to copy 0 bytes
...
Change-Id: I4c0ee7ccc7563e2df657b50356cdd7fec9a1ef15
2018-04-09 12:39:44 +05:30
1ba06f63c4
Apply .clangformat to all repo source files
...
Change-Id: I7e79c6058f0303f9a98911e3b7dd2e8596079344
2018-03-12 11:29:03 +05:30
dc7560ef22
Change directory name to match HIP lowercase style.
2018-02-22 13:15:10 +00:00
4b8ae78891
Merge pull request #321 from gargrahul/hipMemcpyArray_Functions
...
Added support for hipMemcpy Array functions-
2018-02-12 10:36:38 +05:30
24ab820a11
Fixed host allocated globals address lookup for host usage
...
Fixed texture driver APIs failure
2018-01-30 18:06:31 +05:30
487a430b5a
Added support for -
...
- hipMemcpyFromArray
- hipMemcpyAtoH
- hipMemcpyHtoA
2018-01-16 11:44:19 +05:30
115c7f2b79
Added support for
...
- 3D texture driver APIs
- hipMalloc3D
- hipMemcpy3D for destination other than array
2017-12-05 14:11:13 +05:30
9bba97fdcc
Fix some cppcheck style issues.
2017-12-01 20:45:34 +00:00
4313686d6e
Fix warning from default cppchek.
2017-12-01 20:45:33 +00:00
7c0b9a005b
Fix legacy mode detection of the address of an agent allocated variable. In this mode, there exist two executables per each code object, one created by HCC and one created by HIP. Since we dispatch through HCC in legacy mode, we should obtain the address for an agent allocated variable from the latter's executable. Also add two omitted validity checks, whose absence could lead to segfaults when the current process had no .kernel section and / or when an invalid or empty blob was extracted from the latter.
2017-11-30 03:29:04 +00:00
32e11e7dc6
Revert "Revert adoption of CUDA indexing in general - this can only work with later versions of the compiler, just like module based dispatch, and thus must be guarded against usage in earlier (e.g. 1.6) versions."
...
This reverts commit d2fd1f5
2017-11-29 21:49:10 +00:00
fbaf729f88
Revert "Revert adoption of CUDA indexing in general - this can only work with later versions of the compiler, just like module based dispatch, and thus must be guarded against usage in earlier (e.g. 1.6) versions."
...
This reverts commit d2fd1f5
2017-11-29 21:36:29 +00:00
6e4ca3fbb4
Change memset kernel to use memcpy instead of placement new. Simplify indexers.
2017-11-28 19:45:47 +00:00
dc67ca3feb
Merge remote-tracking branch 'origin/master' into feature_use_module_based_dispatch_instead_of_pfe
...
# Conflicts:
# src/hip_module.cpp
2017-11-28 17:29:11 +00:00
56862b1c35
Fixed review comments
2017-11-21 21:19:06 +05:30
9866fa250d
Changed function hipMemcpy_2D to hipMemcpyParam2D
2017-11-21 12:36:24 +05:30
9d088d2283
Refactor the __device__ versions of memset and memcpy to be less awkward i.e. not return nullptr as opposed to the destination pointer (it can only be assumed it was done for maximum confusion) and actually unroll as they claim to. Change all of the {to, from}Symbol functions to use hipModuleGetGlobal, as opposed to hc::accelerator::get_symbol_address which is no longer valid with module based dispatch.
2017-11-21 02:40:34 +00:00
ef09c4918d
Texture driver APIs support
2017-11-09 22:10:55 +05:30
d8e323d4b5
Clean up trailing whitespace so as to reduce noise in #246 .
2017-11-08 00:08:55 +00:00
bb1176001f
Merge remote-tracking branch 'origin/master' into feature_use_module_based_dispatch_instead_of_pfe
...
# Conflicts:
# src/hip_module.cpp
2017-11-03 10:53:39 +00:00
2cacda91bb
Correctly deal with functions from shared objects, wherein the program visible VA == so_base_va + st_value(function_symbol). Remove quaint usage of pfe for hipMemset (which is actually fill_n).
2017-11-01 22:33:13 +00:00
c2482d1255
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.
2017-11-01 15:09:59 +00:00
7d30f32332
Fix bug with peer-to-peer combined with context API
...
- Store context inside the tracker rather than using int deviceID that
was always mapped to primary context
- IsPeerWatcher now based on device IDs rather than specific peers.
2017-10-26 19:44:22 +00:00
3ed5f42dad
Merge pull request #179 from gargrahul/fix_hipmallocarray
...
Fixed hipMallocArray for 1D cases
2017-09-14 12:59:33 -05:00
cea80cd8b3
Add HIP_INIT_ALLOC to init allocated memory.
2017-09-13 23:31:48 +00:00
2a915b2790
Fixed hipMallocArray for 1D cases
2017-09-12 21:52:11 +05:30
38c9febe21
Null check on input pointer arguments
2017-08-25 08:46:34 +05:30
e40047f2bf
[texture] guard new HCC APIs under workweek
...
Change-Id: I4f60a64fb0b0496ca1eb01ffe6ddda121c25d976
2017-08-15 15:51:38 +05:30
4264a4ce56
[HIP Texture] The GPU virtual address for texture memory needs to be
...
aligned.
In hcc_am, a bigger buffer will be allocated for alignment purpose
and _unalignedDevicePointer is added in struct AmPointerInfo for
original allocated address.
2017-08-08 11:18:00 -04:00
0053e4db03
fix hipMemcpy2DAsync
2017-07-29 06:50:56 +05:30
526fcb1223
Merge pull request #116 from bensander/enable_coh_host
...
Make host memory allocations coherent by default.
2017-07-27 10:45:32 +05:30
8b8e97ff28
Make host memory allocations coherent by default.
...
Associated change is to optimize event recording so it uses
agent-scope releaes (since it was only using system-scope release
to support non-coherent host mem).
Flags and environment variables exist to obtain previous behavior
if desired. Options are documented in new performance guide.
2017-07-26 19:20:34 -05:00
e671cec367
HIP Texture Support
2017-07-17 15:16:12 -04:00
1df08626c8
Add support of HIP_HIDDEN_FREE_MEM, to deduct the returned available
...
memory from hipMemGetInfo API, measured in MB.
Change-Id: I7a8260c12e032e04e26611db4c38c893a29f2653
2017-06-26 15:29:38 -05:00
ba51d7f676
Validity check of input arguments in Ipc Mem APIs
...
Change-Id: Ia48e949d19f354f10c7e44cc2457fd4154bf6d76
2017-06-14 15:18:57 +05:30
99e9c7cca5
Use amHostCoherentFlag. Requires new HCC version.
2017-06-07 09:06:40 -05:00
75f691ec2f
Add hipHostMallocCoherent, hipHostMallocNonCoherent
...
Provide per-allocation control over coherent/non-coherent mem.
These overrid the default HIP_COHERENT_HOST_ALLOC setting.
2017-05-24 00:48:10 -05:00
d0ef9d8462
Remove HIP_MAX_QUEUES (replaced with HCC_MAX_QUEUES)
2017-05-23 23:48:01 -05:00
7cfe07cff4
Fix trace category for hipHostMalloc
2017-05-23 23:15:45 -05:00
46030bb2d2
Return precise address for hipHostGetDevicePointer.
2017-05-17 07:36:06 -05:00
27877f8854
Add initial HIP_SYNC_NULL_STREAM=0 mode.
...
This eliminates host-synchronization for null stream. Instead, the
null-stream uses GPU-side events to wait for other streams.
Default is OFF pending additional testing.
Add enhanced null-stream test.
Also refine HIP_TRACE_API.
2017-05-16 19:04:25 -05:00
0edab14139
Add HIP_TRACE_API=4. Only display memory allocation/free apis.
2017-05-16 19:04:25 -05:00
c3ccaa01e5
hipHostMalloc allocation are mapped to all devices by default.
...
Support hipHostMallocPortable flag.
Default flags are hipHostMallocPortable | hipHostMallocMapped.
Also:
-refactor tests to move addCount and addCountReverse into HipTest
namespace.
-test multi-GPU host memory.
2017-05-10 17:34:36 -05:00
687809104b
Fix some typos, add additional guidance for -BSymbolic
2017-05-05 17:29:04 -05:00
ec27c695c4
Added support for hipMemcpy2DAsync in HIP/HCC
...
Change-Id: Ia4a8306f2dc1e33a81a7195ec29aef652fcccc4b
2017-05-03 22:29:12 +05:30
59df70662a
Fix hipMalloc to return error code if allocation fails.
2017-04-24 22:30:54 -05:00
59ab3659ee
fixed build issues with hipPointerGetAttributes
...
Change-Id: I3f5fbc05bdaef720884ba949075928752a070377
2017-04-24 15:31:07 -05:00
72bcfa438b
changed arguments for hipPointerGetAttributes
...
Change-Id: Ia7a7c4722c1f7d0a23f0e5cc3dd6dea6c01c1fd8
2017-04-24 15:24:16 -05:00
Rahul Garg
Maneesh Gupta
Alex Voicu
Ben Sander
Weixing Zhang
Sun, Peng
Aditya Atluri