Графік комітів

6528 Коміти

Автор SHA1 Повідомлення Дата
Vladislav Sytchenko 4ca75d8ee2 Take into an account the number of channels...
when querying the element size of an array.

Change-Id: Id57d3374b14d80a59230ec8286704f2fbabb0fae


[ROCm/clr commit: c9084d0ad2]
2020-04-03 15:43:18 -04:00
German Andryeyev d14399f60f SWDEV-184709 - support hipLaunchCooperativeKernel()
- Add validation checks for cooperative launch to pass Cuda test

Change-Id: Ie296f0c3f113909d9a357879db3b2a833ab314c5


[ROCm/clr commit: 5efb3f26c0]
2020-04-03 15:18:21 -04:00
Michael Hong Bin Liao a54b720eba Merge "Fix size type in __hipRegisterVar" into amd-master-next
[ROCm/clr commit: 16ac35c4d5]
2020-04-03 15:12:21 -04:00
Saleel Kudchadker 67c6b46cf0 Wake up commandQueue before returning
Change-Id: I87eb5a22c81a9cb807474a960b5987d5fb6c2b86


[ROCm/clr commit: e2def55164]
2020-04-03 10:23:36 -07:00
Aaron En Ye Shi 074efe1568 Merge "Fix path for hip-clang when using hipcc (#1961)" into amd-master-next
[ROCm/clr commit: 0bb217245c]
2020-04-03 12:28:50 -04:00
Vladislav Sytchenko a6d0416ab6 Merge branch 'amd-master-next' into amd-npi-next
Change-Id: I3b91e45da7cb38c2a2c6d8675e7651de754764ec


[ROCm/clr commit: b82ab8b329]
2020-04-03 12:11:22 -04:00
Michael LIAO efe48959e9 Fix size type in __hipRegisterVar
Change-Id: I6b667600ae8f133583b768ab963318882b84179f


[ROCm/clr commit: d904f30f9e]
2020-04-03 10:51:58 -04:00
Michael Hong Bin Liao 1daf292d78 Merge "[hip] Clean up unnecessary casting." into amd-master-next
[ROCm/clr commit: 66aca97d6f]
2020-04-03 10:50:40 -04:00
German Andryeyev 8a6d893130 SWDEV-184709 - support hipLaunchCooperativeKernel()
- Enable cooperative tests for single and multiple devices

Change-Id: I54b6713f578b6b5e670f117b17469c0091028c99


[ROCm/clr commit: 2e948e4034]
2020-04-02 12:55:05 -04:00
Michael LIAO 978a25d960 [hip] Clean up unnecessary casting.
Change-Id: I64b08aaef5c67ffb49330c9c605611f1fbd3f5a2


[ROCm/clr commit: e3795436b2]
2020-04-02 12:46:15 -04:00
Paul Fultz II c196ac897b Fix path for hip-clang when using hipcc (#1961)
* Fix path for hip-clang when using hipcc

* Fix typo

* Update regex

Change-Id: I31bbee2e70d58b89191f970f5c6ae7e1c8b40900


[ROCm/clr commit: 58f04fb774]
2020-04-02 12:09:31 -04:00
Paul Fultz II eb2336fd32 Add missing flags for hip::device target on hip-clang (#1230)
This adds the missing compilation flags to hip::device so it can compile with hip-clang compiler.

Change-Id: Ie2b30ea606bfca385a0e84ae03ee0a8d828ad16a


[ROCm/clr commit: 631f82ab82]
2020-04-02 12:09:03 -04:00
Saleel Kudchadker 22a021d12e Merge "OpenCL2.2 Header changes" into amd-master-next
[ROCm/clr commit: 5a357a795a]
2020-04-02 02:46:49 -04:00
Vladislav Sytchenko ee88afced4 Add entry points for hipTexObject*() API
Even though the runtime and driver texture object API is one to one, the structs used by these APIs are not. See hipResourceDesc vs HIP_RESOURCE_DESC differences.

These differences are not trivial and most likely won't be able to handled by hipify, so we need new API entry points.

Change-Id: Id4bcb1ad0ae15378dbdb5a2ed07e5ea30f320082


[ROCm/clr commit: a09fadecf2]
2020-04-01 14:51:51 -04:00
eshcherb 2fa4dc4b57 adding hipApiString (#1989)
[ROCm/clr commit: 1e5ea70588]
2020-04-01 12:10:33 +05:30
ansurya 79c1ac76a4 tex1Dfetch behaviour for different address mode and filter mode (#1772)
Fixes github issue: #1754

- When ResourceDesc::resType is hipResourceTypeLinear ignore address mode and filter mode.
- When textureDesc::normalizedCoords is set to zero, AddressModeWrap and AddressModeMirror won't be supported and will be switched to AddressModeClamp.

[ROCm/clr commit: 68f1b814b1]
2020-04-01 12:10:17 +05:30
Saleel Kudchadker 35e7b758e3 Merge "Cleanup stream from hip:Event class." into amd-master-next
[ROCm/clr commit: 99a024fc14]
2020-03-31 20:14:48 -04:00
Vladislav Sytchenko 870a37307f (SWDEV-229354)
This patch is a workaround to support user pitch for hipMemcpy{2D/3D}.

Historically OpenCL didn't support pitch with clEnqueueFillBuffer(), so neither did we in VDI. Adding it now will be slightly nontrivial, since the fill kernel and runtime in many places will need to be modified.

As a temporary workaround for cases when pitch > width, we can just enqueue a fill for each row separately. This implementation is slow, but it satisfies the correctness criteria.

Change-Id: Idfeca349288b51d6ff84a7cf001fb63c6a66818a


[ROCm/clr commit: 6e0722a5d0]
2020-03-31 18:12:56 -04:00
Reshabh Sharma 69bd482d26 Output file name should not change flags picked for compiler (#1938)
Fixes SWDEV-207362,

The output file name should not contribute to picking up the right flags for the compiler. This fix solves issues when the output has conflicting extensions which confuses hipcc to treat them as the source files and add the required flags for them.

PS: Output file refers to the file followed by -o

Change-Id: I1095966c11143ad73e81fabc35b4e9de5d3afada
Example: hipcc test.o -o test.hip will add the flags for .hip compilation ignoring the fact that it is an output file


[ROCm/clr commit: 36f24a40e5]
2020-03-31 16:13:43 -04:00
Saleel Kudchadker 6629500e70 Cleanup stream from hip:Event class.
Change-Id: I98de07d33bb7fea8f5e2d32b288c15f10ce58902


[ROCm/clr commit: 65722f8cca]
2020-03-31 11:22:00 -07:00
Michael Hong Bin Liao 04fe29d1e0 Merge "[hipcc] Remove the previous workaround." into amd-master-next
[ROCm/clr commit: 9420f56c66]
2020-03-31 10:05:56 -04:00
Michael LIAO dc0ddc29fe [vdi] Fix hipGetSymbol{Address|Size}
- Use symbol value as the qeury key. Compared to the symbol name, the
  symbol value is more robust as developers may use unqualified or
  qualified identifiers. It also removes the mangling and/or demangling
  requirement for the runtime API.

Change-Id: I9d4259f3842612c7cc98551269fc2092d8b5c19e


[ROCm/clr commit: a14695d4eb]
2020-03-31 00:26:53 -04:00
Christophe Paquot 1e254f8432 Do not retry to allocate when OOM. Shouldn't be needed since we idle on Free.
SWDEV-229214

Change-Id: I183006f409388e3c7981f2569649d01d6378be46


[ROCm/clr commit: 47718cbf16]
2020-03-30 12:49:48 -07:00
kjayapra-amd bce2b56ea0 SWDEV-216213 - Use different static & dynamic module maps for faster lookup.
Change-Id: Ia605e76a411ad5be04046b9d61f1ac111d49bb4a


[ROCm/clr commit: 7356a74d35]
2020-03-30 14:28:07 -04:00
Anusha Godavarthy Surya 8b0fec4333 Merge "Update Enable/Disable peers to match cuda behaviour" into amd-master-next
[ROCm/clr commit: 9d213070b0]
2020-03-30 13:46:42 -04:00
agodavar d329bd0684 Update Enable/Disable peers to match cuda behaviour
Change-Id: I67194ccf77a0019368579ff7d95b7790fcf228f3


[ROCm/clr commit: 96d119ac77]
2020-03-30 12:49:16 -04:00
Tao Sang 623168a082 Merge "enable HCC printf when using hip-clang" into amd-master-next
[ROCm/clr commit: 44a145b2a3]
2020-03-30 11:37:23 -04:00
Michael LIAO ee25f80e54 [hipcc] Remove the previous workaround.
Change-Id: Ie1c2aa21689bc343c10fd4b699546cf500ced5b2


[ROCm/clr commit: 64bbce8e73]
2020-03-30 10:18:47 -04:00
Michael LIAO 9c361faab3 [vdi] Fix calculation of MaxWaves
- Consider the case where `usedVGPRs` is zero.
- This fixes [SWDEV-228537](http://ontrack-internal.amd.com/browse/SWDEV-228537)

Change-Id: I8675311f5fe24fb59c5d45bada122afefb55b128


[ROCm/clr commit: 55d869df99]
2020-03-30 09:10:16 -04:00
Sarbojit2019 7e05ef29eb Fix for segfault seen in hipMemcpyDtoD (#1964)
* Fixes SWDEV-227444.

[ROCm/clr commit: b55775a49e]
2020-03-28 17:29:49 +05:30
Maneesh Gupta 3d1b086400 Remove address_space(1) typecast and use __ockl_atomic_add_noret_f32 (#1956)
* Remove address_space(1) typecast for ockl_global_atomic_add_f32
* use __ockl_atomic_add_noret_f32

[ROCm/clr commit: 062938a14c]
2020-03-28 17:28:33 +05:30
Sameer Sahasrabuddhe 207d9b3897 enable HCC printf when using hip-clang
This is cherry-picked from PR#1947 that was committed to the
github repo. It allows printf to work with hip-clang and HCC
runtime.

Change-Id: I754753250ea1e694cf3441722e2d4c9d25fa75bc


[ROCm/clr commit: c014c733c6]
2020-03-28 00:18:21 -04:00
Saleel Kudchadker eba4688437 Merge "Check event status before notify" into amd-master-next
[ROCm/clr commit: a5c287554e]
2020-03-27 20:19:40 -04:00
Vladislav Sytchenko d71bdef595 (SWDEV-228794)
Adjust the origin of the copy if the user passes a pointer that wasn't allocated by the runtime.

Change-Id: I0aeb20195ed730857a461a53f537626ec2573fd1


[ROCm/clr commit: 86c969fcea]
2020-03-27 16:33:16 -04:00
Vladislav Sytchenko 3f6db04b75 (SWDEV-228794)
Add hipMallocHost()

Change-Id: Ia3c7c5ca94b39fe30f3a51d1b60782d3472259ff


[ROCm/clr commit: 08729d0f43]
2020-03-27 15:57:48 -04:00
Vladislav Sytchenko 004dff62c4 (SWDEV-228782)
The only requirment from hipMallocPitch() is that the returned pitch is aligned to the HW image pitch alignment. There is no restriction on the size of the allocation, since the memory might not be used for images.

Change-Id: I97438e5fe4012ca4721b14b85f514dbac803c17c


[ROCm/clr commit: 111897dbea]
2020-03-27 15:52:17 -04:00
Tao Sang 0e3a7ed058 Merge "Support hipFuncGetAttributes with hip-clang+Hcc RT" into amd-master-next
[ROCm/clr commit: b2afe3c250]
2020-03-27 14:36:41 -04:00
Saleel Kudchadker 4abf9f5c38 Check event status before notify
Change-Id: I68f6bbbf236e49b859be2d5afbe0c8282fe15dd3


[ROCm/clr commit: fa7dea42b5]
2020-03-27 11:32:46 -07:00
satyanveshd fbc9771f63 [dtests] Added few Negative tests (#1735)
[ROCm/clr commit: 4554bbd9f9]
2020-03-27 14:10:12 +05:30
Siu Chi Chan 922219117b don't expose symbols from code_object_bundle (#1971)
Change-Id: I56479485aad42c3d517fe6d9055be1cd846eeb00

[ROCm/clr commit: 6ab1e864b6]
2020-03-27 14:09:07 +05:30
Sarbojit2019 4ef95f4e73 Fix few memory leaks in HIP (#1969)
[ROCm/clr commit: 4a68ab5a8c]
2020-03-27 14:08:30 +05:30
Jatin Chaudhary d919e2223d Packaging changes to bring them to sync with generated ones (#1918)
[ROCm/clr commit: 2449e0de8b]
2020-03-27 14:08:12 +05:30
Tao Sang 8218ede5e8 Support hipFuncGetAttributes with hip-clang+Hcc RT
Fix issues of missing kernel function symbols and missing argument list via
using __hipRegister* functions.
Then the following tests can pass,
directed_tests/runtimeApi/module/hipFuncGetAttributes
directed_tests/runtimeApi/module/hipExtLaunchMultiKernelMultiDevice
directed_tests/gcc/LaunchKernel

Change-Id: I52135b61e8283eb4f9f10f77895151e4e55418d9


[ROCm/clr commit: 4a113bb669]
2020-03-26 23:28:35 -04:00
Saleel Kudchadker 09140bb72e OpenCL2.2 Header changes
Change-Id: I3285a8ffc7e34333feb68e925ebd460831c8e3d4


[ROCm/clr commit: 388e656c4e]
2020-03-26 17:35:09 -07:00
Vladislav Sytchenko 96dad9615a Add support for formating hipExtent objects
Change-Id: Iea54a510e81a856c0c450305b3e5a7179ee48295


[ROCm/clr commit: 2b538b09e6]
2020-03-26 14:45:20 -04:00
Vladislav Sytchenko e9eababfdd Enable initial sRGB support
Instead of using the sampler field force_degamma to perform sRGB->linear conversion during pixel sampling, we use an appropriate image format instead. The overhead of this is having to create an image view when creating a texture object from an array.

Change-Id: I1ca368c312c1fd4b6f784a3a1b35b5eeb28070ff


[ROCm/clr commit: 23211f05d6]
2020-03-26 14:45:20 -04:00
Vladislav Sytchenko cfc508eef0 Add initial entry points for mipmapped array API
Change-Id: Icd59cc7323ddcb6773da6105260415a1e6f4cdcb


[ROCm/clr commit: 06bb183558]
2020-03-26 14:45:20 -04:00
Vladislav Sytchenko 21587cb0fa Handle offsets for dptr <-> image copies
Change-Id: I7a4a56ee07a26a741d2aac35502446d248f720ad


[ROCm/clr commit: af1c5a0015]
2020-03-26 14:45:20 -04:00
Vladislav Sytchenko cf49aaa7fc Replace hip::TextureObject with __hip_texture
This avoids the use of extra casts when obtaining a texture object handle.

Change-Id: I42df22bdad0ab9ac6c33cb8b282dee65fe7cfd6e


[ROCm/clr commit: 6cfbe19160]
2020-03-26 14:45:20 -04:00
Vladislav Sytchenko fe369a0e18 Correctly format hipResourceDesc objects
The struct consists of a union - only the active object should be read.

Change-Id: I1c40965b61518acd91a2dcbae92a015ac9be346a


[ROCm/clr commit: 1cacb5c9b7]
2020-03-26 14:45:20 -04:00