e305d3fc94
Add missing file.
2017-11-29 02:16:44 +00:00
37c1811b2a
Fix oversight in selection mechanism which led to erroneous code to be compiled for the grid_launch_GGL component.
2017-11-29 01:37:52 +00:00
c6ab6f292b
Choose whether or not to use functional grid_launch based on the version of HCC used to compile.
2017-11-29 00:17:44 +00:00
45a9f4f7b1
Remove leftover agent allocated globals.
2017-11-28 19:56:04 +00:00
d5c8de3f41
Change memset kernel to use memcpy instead of placement new. Simplify indexers.
2017-11-28 19:45:47 +00:00
6a0efb7ed2
Re-sync with upstream and re-factor platform global management for texture references.
2017-11-28 19:15:29 +00:00
d37a5a6008
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
cc418da654
Fix float2int rounding functions
...
Change-Id: I67943859a6344c5eec0eaa23418c9b802ef72468
2017-11-28 17:23:43 +00:00
bb35299560
This actually (tries) to do the right thing all the way, by using memcpy for bitcasting, and not rely on undefined behaviour of a different flavour as a substitute for the original undefined behaviour. Note that the compiler will (should) optimise down to the same emitted code, since this is a pattern it understands.
2017-11-28 17:23:06 +00:00
4b7cb0624e
This fixes some outright quaint choices made when implementing HIP's bitwise conversion functions, by using simple reinterpret_casts, as is idiomatic. These functions are supposed to be re-entrant, correct and efficient. Sadly, they were neither: they hid a massive race condition against a value stored in global memory, which means that they were also unreasonably slow if they ever managed to be correct, and relied on union based type punning which is in a grey area of the standard. It is difficult to ascertain what may have been the reason for coming up with this quirky solution.
2017-11-28 17:23:06 +00:00
e93a24bdbe
Merge pull request #256 from gargrahul/texture_driver_api_support
...
Texture driver APIs support
2017-11-27 13:52:39 -06:00
822743845c
Fix float2int rounding functions
...
Change-Id: I67943859a6344c5eec0eaa23418c9b802ef72468
2017-11-23 09:57:24 +05:30
38029f2849
Fixed review comments
2017-11-21 21:19:06 +05:30
fb1021cc0a
This corrects how addresses are formed for symbols which reside in shared objects. For this case, the .value component of an ELF symbol holds the offset from the base VA where the shared object was loaded. Thus, to correctly obtain the VA of the object refered by the symbol, we must add the offset to the VA where the shared object is loaded. We were already doing this correctly for symbols denoting functions, but we were incorrect for those denoting objects.
2017-11-21 13:15:13 +00:00
24307fe5c4
Changed function hipMemcpy_2D to hipMemcpyParam2D
2017-11-21 12:36:24 +05:30
f8c1c1b38e
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
24f8a93ff7
Clean-up some remaining noise in program_state.cpp.
2017-11-20 22:41:46 +00:00
27bc487635
Correct ill-formed merge in earlier commit and adjust for differences with the new CUDA natural indexing mechanism.
2017-11-20 16:33:52 +00:00
30d90dab38
Re-sync with upstream.
2017-11-20 15:34:50 +00:00
a1e0c96689
Merge pull request #264 from pzins/missing_end_marker
...
Fix missing MARKER_END
2017-11-20 06:08:01 -06:00
eff09fcd86
-Moved coGlobals in hipModule class (takes care of multi module case)
...
-Used mutex scope for updating coGlobals
2017-11-20 16:23:18 +05:30
76f174b536
Update hipModuleGetTexRef API
2017-11-19 22:10:46 +05:30
0cc921f103
This implements the trivial change needed to move back from the hip{Something}_{x, y, z} macros to the natural CUDA syntax of Something.{x, y, z}. This is contained in lines 384-404 in hip_runtime.h. All of the other changes have to do with changing unit tests to use this syntax. The macros are retained for backwards compatibility.
2017-11-19 01:54:12 +00:00
e2e499461a
Removed redundant desc variable
2017-11-15 18:28:27 +05:30
edbfeeb141
-Fixed texture driver API sample
...
-Added hipTexRefSetAddress and hipTexRefSetAddress2D APIs
2017-11-15 18:23:28 +05:30
e536eb038d
Texture code reorganized
2017-11-14 11:09:35 +05:30
e60a95d7dd
Fix missing MARKER_END
...
Logging status of hipCtxSynchronize was missing
Test if hip profiling is active for MARKER_END in ihipPostLaunchKernel
Add MARKER_END after the completion of a kernel launched through
the "grid launch"
2017-11-13 16:13:19 -05:00
1518611976
Merge remote-tracking branch 'origin/master' into feature_use_module_based_dispatch_instead_of_pfe
2017-11-09 23:43:07 +00:00
1851c153f6
Texture driver APIs support
2017-11-09 22:10:55 +05:30
925f97fac7
Merge pull request #250 from AlexVlx/feature_add_agent_global_support
...
Support for agent globals
2017-11-09 07:52:09 +05:30
f7919a8321
Merge remote-tracking branch 'origin/master' into feature_use_module_based_dispatch_instead_of_pfe
...
# Conflicts:
# tests/src/runtimeApi/stream/hipStreamSync2.cpp
2017-11-08 10:26:30 +00:00
0ad1308efd
Clean up trailing whitespace so as to reduce noise in #246 .
2017-11-08 00:08:55 +00:00
8854dec897
Merge remote-tracking branch 'origin/master' into feature_use_module_based_dispatch_instead_of_pfe
2017-11-07 00:01:22 +00:00
e71456cc6c
Check for null event in hipEventElapsedTime
2017-11-06 23:49:31 +00:00
b5911acce2
hipStreamWaitEvent returns success if event created but not recorded
2017-11-06 23:49:31 +00:00
955cfbfdc7
Make hipEvent_t thread safe.
...
Support re-recording of same event by different threads.
- Add criticalData structure to hipEvent_t, similar to mechanism used
for streams, contexts, device. Events are always locked
after streams to avoid deadlock.
- ihipEvent_t::locked_copyCrit can be used to copy critical state
including marker. The critical state in the event can then
be re-recorded.
- refactor hipEventElapsedTime. Remmove stale debug code, native signal
refs.
2017-11-06 23:49:25 +00:00
dbb6b2bab8
Merge pull request #251 from ROCm-Developer-Tools/fix_event_state
...
Set event state AFTER it is recorded.
2017-11-06 07:28:11 +05:30
958ca62c63
Merge pull request #249 from bensander/warn_event
...
Add HIP_DB=warn + message if sync on dangerous event.
2017-11-06 07:25:40 +05:30
1c65bb1684
Set event state AFTER it is recorded.
2017-11-05 10:33:18 -06:00
17753cbd92
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
d12cf0da7d
This introduces correct support for agent global variables, and implements hipModuleGetGlobal as an actual equivalent for cuModuleGetGlobal.
2017-11-03 01:44:48 +00:00
4693c5e56c
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
fe32685fbc
Merge pull request #237 from bensander/use_ctxptr_for_p2p
...
Use ctxptr for p2p
2017-11-01 18:55:25 +01:00
172b9f96e6
Add HIP_DB=warn + message if sync on dangerous event.
2017-11-01 10:44:34 -07:00
5a80c4108e
Merge pull request #245 from scchan/centos_fixes
...
various fixes for centos/rhel
2017-11-01 18:10:29 +01:00
28f87f7d2e
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
c1a4ff7b20
Centos/RHEL - remove usage of constexpr since libc++ doesn't enable ctor for constexpr pair in C++11
2017-10-31 18:16:12 +00:00
dc7d993a02
Add ns-level timer for HIP API routines
...
Refactor some miuses of ihipLogStatus, these should only be in top-level
HIP APIs and should be paired with HIP_API_INIT calls.
2017-10-30 20:20:51 +00:00
7e8b39fc96
Merge pull request #222 from bensander/fix_device_prop
...
Fix device prop
2017-10-30 17:58:48 +01:00
4c7b2be1c2
Check for null copyEngine before looking at peers.
2017-10-30 16:58:03 +00:00
Alex Voicu
Maneesh Gupta
Ben Sander
Rahul Garg
Pierre
Siu Chi Chan