c3fdb8ab8e
Merge remote-tracking branch 'origin/master' into feature_use_module_based_dispatch_instead_of_pfe
...
[ROCm/hip commit: adaf6b8dff
2017-11-07 00:01:22 +00:00
33836c9521
Merge pull request #251 from ROCm-Developer-Tools/fix_event_state
...
Set event state AFTER it is recorded.
[ROCm/hip commit: 1131c9e41a
2017-11-06 07:28:11 +05:30
8845e38efd
Merge pull request #249 from bensander/warn_event
...
Add HIP_DB=warn + message if sync on dangerous event.
[ROCm/hip commit: a62d5aa875
2017-11-06 07:25:40 +05:30
3746211481
Merge pull request #239 from phani544/newTests
...
[newTests]Adding tests for device APIs
[ROCm/hip commit: 2f4e1983e2
2017-11-06 07:22:13 +05:30
6a4c50cf6c
Set event state AFTER it is recorded.
...
[ROCm/hip commit: 4c3b65a5cd
2017-11-05 10:33:18 -06:00
1c77e8d6b1
Merge remote-tracking branch 'origin/master' into feature_use_module_based_dispatch_instead_of_pfe
...
[ROCm/hip commit: 2932706d47
2017-11-03 10:57:12 +00:00
1f911cd23a
Merge remote-tracking branch 'origin/master' into feature_use_module_based_dispatch_instead_of_pfe
...
# Conflicts:
# src/hip_module.cpp
[ROCm/hip commit: bb1176001f
2017-11-03 10:53:39 +00:00
9d7c904529
[newTests]modified the test
...
[ROCm/hip commit: 0599d13b0d
2017-11-03 14:52:19 +05:30
0026b0a3d9
[newTests]Modified the test as per review comments
...
[ROCm/hip commit: a764a0980d
2017-11-03 10:39:47 +05:30
04d9ec1400
Merge pull request #248 from mangupta/SWDEV-134531
...
Remove libsupc++ dependency from HIP
[ROCm/hip commit: d90d431490
2017-11-03 08:01:59 +05:30
94a22fe015
Remove libsupc++ dependency from HIP
...
Change-Id: Ib843f99dfb3057b7c28029250c2449732ea7f17f
[ROCm/hip commit: f34f5c8684
2017-11-02 12:29:09 +05:30
b8a81d2b75
This corrects some interesting choices that were present in the HIP
...
unit tests such as e.g. de-allocating memory allocated with new[] using
free. All of these were identified via cppcheck.
[ROCm/hip commit: a619ac6f92
2017-11-01 22:51:22 +00:00
dab971370e
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).
...
[ROCm/hip commit: 2cacda91bb
2017-11-01 22:33:13 +00:00
2979a99888
Merge pull request #237 from bensander/use_ctxptr_for_p2p
...
Use ctxptr for p2p
[ROCm/hip commit: 09d866a639
2017-11-01 18:55:25 +01:00
6c60773e82
Add HIP_DB=warn + message if sync on dangerous event.
...
[ROCm/hip commit: 70c25bdf8e
2017-11-01 10:44:34 -07:00
6f3b28d27c
Merge pull request #245 from scchan/centos_fixes
...
various fixes for centos/rhel
[ROCm/hip commit: 86f62accfd
2017-11-01 18:10:29 +01:00
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
d938703f9b
Centos/RHEL - remove usage of constexpr since libc++ doesn't enable ctor for constexpr pair in C++11
...
[ROCm/hip commit: 99d32a195f
2017-10-31 18:16:12 +00:00
b6265074fa
use libc++ on centos 7 and rhel 7.4
...
[ROCm/hip commit: cd6996614f
2017-10-31 18:14:07 +00:00
e6e90e9cfa
Merge pull request #197 from bensander/update_coherency_tests
...
Update coherency tests
[ROCm/hip commit: f27c2c1715
2017-10-31 17:26:50 +05:30
07b52ec043
Merge pull request #241 from ROCm-Developer-Tools/multi_host
...
Inital codes to remove x86_64 dependency in HIP source build
[ROCm/hip commit: 4d85b6ab29
2017-10-31 16:35:12 +05:30
e88ef63bc8
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.
[ROCm/hip commit: 7e908bdec8
2017-10-30 20:20:51 +00:00
afb75b0b2c
Inital codes to remove x86_64 dependency in HIP source build
...
[ROCm/hip commit: 92fb244841
2017-10-30 15:19:23 -05:00
51ee7807db
Merge pull request #222 from bensander/fix_device_prop
...
Fix device prop
[ROCm/hip commit: 2e8ec71e40
2017-10-30 17:58:48 +01:00
ccb8b441a7
Check for null copyEngine before looking at peers.
...
[ROCm/hip commit: d610f16c47
2017-10-30 16:58:03 +00:00
8242c45eda
Merge pull request #226 from scchan/add_printf3
...
add printf to HIP device functions
[ROCm/hip commit: f8843ae415
2017-10-30 17:08:18 +01:00
a36c3c6d8b
[newTests]Adding tests for device APIs
...
[ROCm/hip commit: 511de63bcb
2017-10-30 14:34:24 +05:30
27c5e94c81
[HIPIFY] fix typo - missing )
...
[ROCm/hip commit: 44c74b6511
2017-10-27 23:31:43 +03:00
b9a52020b2
Merge pull request #238 from ChrisKitching/statistics
...
[HIPIFY] Decouple the statistics system from the code rewriter
[ROCm/hip commit: b28a69785b
2017-10-27 23:17:20 +03:00
193235bd08
Merge pull request #236 from ChrisKitching/friendlyCmake
...
[HIPIFY] Make the cmake build system more friendly
[ROCm/hip commit: c9b7c43e1c
2017-10-27 22:35:13 +03:00
54f786583b
Remove commented else-block
...
A warning statement for _string literals_ seems a bit unhelpful.
There's no value in this being here.
[ROCm/hip commit: 20871a3a07
2017-10-27 20:12:33 +01:00
05699e3779
Decouple the statistics system from the code translation
...
The original implementation had the statistics system woken very
tightly into things like PPCallbacks, with counters duplicated
in two places, and all the output code duplicated. This made it
very difficult to alter the structure of the program without
breaking the statistics system.
Since the planned approach for solving the remaining preprocessor
bugs needs the introduction of a custom FrontendAction, and such
a restructure was incompatible with the way the statistics system
was set up, this rewrite was required.
'tis rather simpler now, mind you :D
This commit also fixes an issue where some stats were counted
twice, and allows `-print-stats` to operate independently of
`-stat-output`, allowing you to print stats to a file without
printing them to a terminal (or vice-versa).
[ROCm/hip commit: b303ffe53e
2017-10-27 20:12:33 +01:00
1454bf9651
Copy-paste less in the statistics printing code
...
[ROCm/hip commit: 5699c18adc
2017-10-27 20:12:33 +01:00
e4569bc84e
Inline updateCountersExt
...
[ROCm/hip commit: 50448aec3b
2017-10-27 20:12:32 +01:00
a3c1d30745
Update counter maps sanely
...
operator[] default-constructs the map value if no value exists
for that key. Default-construction of int yields a zero. So all
the manual faffing around is just unnecessary.
[ROCm/hip commit: d8beee8918
2017-10-27 20:12:32 +01:00
51df5b20c9
Prefer references to pointers in updateCountersExt()
...
[ROCm/hip commit: 00bb447e55
2017-10-27 20:12:32 +01:00
25517e41fd
Move string utility functions into their own translation unit
...
[ROCm/hip commit: ee8e11a720
2017-10-27 20:12:32 +01:00
85fd2e6f51
Extract LLVM compatibility code into its own translation unit
...
[ROCm/hip commit: 1bd837b4b1
2017-10-27 20:12:32 +01:00
389fa2e68e
Remove unused field
...
[ROCm/hip commit: 0c09bdf523
2017-10-27 20:12:32 +01:00
48e7403762
Remove CUDA_EXCLUDES
...
An artefact from a now-defunct hack to avoid corrupting programs
[ROCm/hip commit: c6707ef33c
2017-10-27 20:12:32 +01:00
199d75adc0
Make unsupported actually be a bool...
...
[ROCm/hip commit: 2f376c9b25
2017-10-27 20:12:31 +01:00
2285482075
Describe the LLVM we found
...
[ROCm/hip commit: 69e67fe25a
2017-10-27 19:39:41 +01:00
7a0eedabbe
Update hipify-clang readme for simplified build process
...
[ROCm/hip commit: 82d05ee6f4
2017-10-27 19:39:41 +01:00
e3338fe5a4
hipify does not add the hipLaunchParm option any more
...
This was removed a while ago - seems like it uses a different
variant of the launch kernel function now, so this is redundant.
[ROCm/hip commit: b412802c66
2017-10-27 19:39:41 +01:00
fd656a0fbb
Use cmake's builtin mechanism for handling library locations
...
See [the documentation](https://cmake.org/cmake/help/v3.0/command/find_package.html )
for exactly how the search procedure works. If you want to use an
LLVM from a specific location, use CMAKE_PREFIX_PATH as normal.
No longer do we have a nonstandard HIPIFY_CLANG_LLVM_DIR variable
for people to learn about.
[ROCm/hip commit: 8fefc6a2b7
2017-10-27 19:39:40 +01:00
1461c69757
Move the "LLVM found" print adjacent to the find_package call
...
Very surprising that LLVM's finder module doesn't print this
itself like _literally every other finder module_. Blarg.
[ROCm/hip commit: 92c90a7068
2017-10-27 19:39:40 +01:00
cc1bc495a0
We no longer rely on HIPIFY_CLANG_LLVM_DIR to disable hipify-clang
...
Since there's now an option for toggling hipify-clang, omitting the
path is no longer something we need to check for. We'll still
abort if LLVM isn't found, due to `REQUIRED`.
[ROCm/hip commit: c60c8d417e
2017-10-27 19:39:40 +01:00
d92f3c8d76
Don't attempt to find test dependencies if tests are disabled
...
And while we're at it, introduce a handy program-finder macro
[ROCm/hip commit: 921ff4c8a3
2017-10-27 19:39:40 +01:00
b61c4a241f
Use add_dependencies to avoid duplication of pkg_hip_base
...
[ROCm/hip commit: 56b4222043
2017-10-27 19:39:40 +01:00
7947e406c7
Make BUILD_HIPIFY_CLANG a cmake option
...
Instead of deciding whether to build hipify-clang based on
the presence of an LLVM path on the command line, have an
explicit option.
Do we want this default-on or default-off? I've defaulted it to
on for now, but maybe we want the opposite?
[ROCm/hip commit: a4ecd4eb31
2017-10-27 19:39:39 +01:00
Alex Voicu
Maneesh Gupta
Ben Sander
Phaneendr-kumar Lanka
Siu Chi Chan
Wen-Heng (Jack) Chung
Evgeny Mankov
Evgeny Mankov
Chris Kitching