maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/clang/test/Driver/rocm-device-libs.cl

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

181 lines
7.4 KiB
Common Lisp
Raw Normal View History

AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
// REQUIRES: clang-driver
// REQUIRES: amdgpu-registered-target
Disable test on windows
2020-04-11 06:47:48 +08:00
// REQUIRES: !system-windows
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
// Test flush-denormals-to-zero enabled uses oclc_daz_opt_on
// RUN: %clang -### -target amdgcn-amd-amdhsa \
// RUN: -x cl -mcpu=gfx900 \
recommit "[HIP] Add default header and include path" recommit 11d06b9511bd25aabbfad10dff548b0ce29135a5 with fix for lit tests.
2020-06-06 04:49:38 +08:00
// RUN: --rocm-path=%S/Inputs/rocm \
Remove dependency between test files. There seems to be no good reason for rocm-device-libs.cl to depend on opencl.cl. Removed this dependency to unbreak the tests in our setup.
2020-04-13 12:16:46 +08:00
// RUN: %s \
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
// RUN: 2>&1 | FileCheck -dump-input-on-failure --check-prefixes=COMMON,COMMON-DEFAULT,GFX900-DEFAULT,GFX900,WAVE64 %s
// Make sure the different denormal default is respected for gfx8
// RUN: %clang -### -target amdgcn-amd-amdhsa \
// RUN: -x cl -mcpu=gfx803 \
recommit "[HIP] Add default header and include path" recommit 11d06b9511bd25aabbfad10dff548b0ce29135a5 with fix for lit tests.
2020-06-06 04:49:38 +08:00
// RUN: --rocm-path=%S/Inputs/rocm \
Remove dependency between test files. There seems to be no good reason for rocm-device-libs.cl to depend on opencl.cl. Removed this dependency to unbreak the tests in our setup.
2020-04-13 12:16:46 +08:00
// RUN: %s \
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
// RUN: 2>&1 | FileCheck -dump-input-on-failure --check-prefixes=COMMON,COMMON-DEFAULT,GFX803-DEFAULT,GFX803,WAVE64 %s
// Make sure the non-canonical name works
// RUN: %clang -### -target amdgcn-amd-amdhsa \
// RUN: -x cl -mcpu=fiji \
recommit "[HIP] Add default header and include path" recommit 11d06b9511bd25aabbfad10dff548b0ce29135a5 with fix for lit tests.
2020-06-06 04:49:38 +08:00
// RUN: --rocm-path=%S/Inputs/rocm \
Remove dependency between test files. There seems to be no good reason for rocm-device-libs.cl to depend on opencl.cl. Removed this dependency to unbreak the tests in our setup.
2020-04-13 12:16:46 +08:00
// RUN: %s \
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
// RUN: 2>&1 | FileCheck -dump-input-on-failure --check-prefixes=COMMON,COMMON-DEFAULT,GFX803-DEFAULT,GFX803,WAVE64 %s
// RUN: %clang -### -target amdgcn-amd-amdhsa \
// RUN: -x cl -mcpu=gfx900 \
// RUN: -cl-denorms-are-zero \
recommit "[HIP] Add default header and include path" recommit 11d06b9511bd25aabbfad10dff548b0ce29135a5 with fix for lit tests.
2020-06-06 04:49:38 +08:00
// RUN: --rocm-path=%S/Inputs/rocm \
Remove dependency between test files. There seems to be no good reason for rocm-device-libs.cl to depend on opencl.cl. Removed this dependency to unbreak the tests in our setup.
2020-04-13 12:16:46 +08:00
// RUN: %s \
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
// RUN: 2>&1 | FileCheck -dump-input-on-failure --check-prefixes=COMMON,COMMON-DAZ,GFX900,WAVE64 %s
// RUN: %clang -### -target amdgcn-amd-amdhsa \
// RUN: -x cl -mcpu=gfx803 \
// RUN: -cl-denorms-are-zero \
recommit "[HIP] Add default header and include path" recommit 11d06b9511bd25aabbfad10dff548b0ce29135a5 with fix for lit tests.
2020-06-06 04:49:38 +08:00
// RUN: --rocm-path=%S/Inputs/rocm \
Remove dependency between test files. There seems to be no good reason for rocm-device-libs.cl to depend on opencl.cl. Removed this dependency to unbreak the tests in our setup.
2020-04-13 12:16:46 +08:00
// RUN: %s \
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
// RUN: 2>&1 | FileCheck -dump-input-on-failure --check-prefixes=COMMON,COMMON-DAZ,GFX803,WAVE64 %s
// RUN: %clang -### -target amdgcn-amd-amdhsa \
// RUN: -x cl -mcpu=gfx803 \
// RUN: -cl-finite-math-only \
recommit "[HIP] Add default header and include path" recommit 11d06b9511bd25aabbfad10dff548b0ce29135a5 with fix for lit tests.
2020-06-06 04:49:38 +08:00
// RUN: --rocm-path=%S/Inputs/rocm \
Remove dependency between test files. There seems to be no good reason for rocm-device-libs.cl to depend on opencl.cl. Removed this dependency to unbreak the tests in our setup.
2020-04-13 12:16:46 +08:00
// RUN: %s \
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
// RUN: 2>&1 | FileCheck -dump-input-on-failure --check-prefixes=COMMON,COMMON-FINITE-ONLY,GFX803,WAVE64 %s
// RUN: %clang -### -target amdgcn-amd-amdhsa \
// RUN: -x cl -mcpu=gfx803 \
// RUN: -cl-fp32-correctly-rounded-divide-sqrt \
recommit "[HIP] Add default header and include path" recommit 11d06b9511bd25aabbfad10dff548b0ce29135a5 with fix for lit tests.
2020-06-06 04:49:38 +08:00
// RUN: --rocm-path=%S/Inputs/rocm \
Remove dependency between test files. There seems to be no good reason for rocm-device-libs.cl to depend on opencl.cl. Removed this dependency to unbreak the tests in our setup.
2020-04-13 12:16:46 +08:00
// RUN: %s \
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
// RUN: 2>&1 | FileCheck -dump-input-on-failure --check-prefixes=COMMON,COMMON-CORRECT-SQRT,GFX803,WAVE64 %s
// RUN: %clang -### -target amdgcn-amd-amdhsa \
// RUN: -x cl -mcpu=gfx803 \
// RUN: -cl-fast-relaxed-math \
recommit "[HIP] Add default header and include path" recommit 11d06b9511bd25aabbfad10dff548b0ce29135a5 with fix for lit tests.
2020-06-06 04:49:38 +08:00
// RUN: --rocm-path=%S/Inputs/rocm \
Remove dependency between test files. There seems to be no good reason for rocm-device-libs.cl to depend on opencl.cl. Removed this dependency to unbreak the tests in our setup.
2020-04-13 12:16:46 +08:00
// RUN: %s \
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
// RUN: 2>&1 | FileCheck -dump-input-on-failure --check-prefixes=COMMON,COMMON-FAST-RELAXED,GFX803,WAVE64 %s
// RUN: %clang -### -target amdgcn-amd-amdhsa \
// RUN: -x cl -mcpu=gfx803 \
// RUN: -cl-unsafe-math-optimizations \
recommit "[HIP] Add default header and include path" recommit 11d06b9511bd25aabbfad10dff548b0ce29135a5 with fix for lit tests.
2020-06-06 04:49:38 +08:00
// RUN: --rocm-path=%S/Inputs/rocm \
Remove dependency between test files. There seems to be no good reason for rocm-device-libs.cl to depend on opencl.cl. Removed this dependency to unbreak the tests in our setup.
2020-04-13 12:16:46 +08:00
// RUN: %s \
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
// RUN: 2>&1 | FileCheck -dump-input-on-failure --check-prefixes=COMMON,COMMON-UNSAFE,GFX803,WAVE64 %s
// RUN: %clang -### -target amdgcn-amd-amdhsa \
// RUN: -x cl -mcpu=gfx1010 \
recommit "[HIP] Add default header and include path" recommit 11d06b9511bd25aabbfad10dff548b0ce29135a5 with fix for lit tests.
2020-06-06 04:49:38 +08:00
// RUN: --rocm-path=%S/Inputs/rocm \
Remove dependency between test files. There seems to be no good reason for rocm-device-libs.cl to depend on opencl.cl. Removed this dependency to unbreak the tests in our setup.
2020-04-13 12:16:46 +08:00
// RUN: %s \
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
// RUN: 2>&1 | FileCheck -dump-input-on-failure --check-prefixes=COMMMON,GFX1010,WAVE32 %s
// RUN: %clang -### -target amdgcn-amd-amdhsa \
// RUN: -x cl -mcpu=gfx1011 \
recommit "[HIP] Add default header and include path" recommit 11d06b9511bd25aabbfad10dff548b0ce29135a5 with fix for lit tests.
2020-06-06 04:49:38 +08:00
// RUN: --rocm-path=%S/Inputs/rocm \
Remove dependency between test files. There seems to be no good reason for rocm-device-libs.cl to depend on opencl.cl. Removed this dependency to unbreak the tests in our setup.
2020-04-13 12:16:46 +08:00
// RUN: %s \
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
// RUN: 2>&1 | FileCheck -dump-input-on-failure --check-prefixes=COMMMON,GFX1011,WAVE32 %s
// RUN: %clang -### -target amdgcn-amd-amdhsa \
// RUN: -x cl -mcpu=gfx1012 \
recommit "[HIP] Add default header and include path" recommit 11d06b9511bd25aabbfad10dff548b0ce29135a5 with fix for lit tests.
2020-06-06 04:49:38 +08:00
// RUN: --rocm-path=%S/Inputs/rocm \
Remove dependency between test files. There seems to be no good reason for rocm-device-libs.cl to depend on opencl.cl. Removed this dependency to unbreak the tests in our setup.
2020-04-13 12:16:46 +08:00
// RUN: %s \
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
// RUN: 2>&1 | FileCheck -dump-input-on-failure --check-prefixes=COMMMON,GFX1012,WAVE32 %s
// RUN: %clang -### -target amdgcn-amd-amdhsa \
// RUN: -x cl -mcpu=gfx1010 -mwavefrontsize64 \
recommit "[HIP] Add default header and include path" recommit 11d06b9511bd25aabbfad10dff548b0ce29135a5 with fix for lit tests.
2020-06-06 04:49:38 +08:00
// RUN: --rocm-path=%S/Inputs/rocm \
Remove dependency between test files. There seems to be no good reason for rocm-device-libs.cl to depend on opencl.cl. Removed this dependency to unbreak the tests in our setup.
2020-04-13 12:16:46 +08:00
// RUN: %s \
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
// RUN: 2>&1 | FileCheck -dump-input-on-failure --check-prefixes=COMMMON,GFX1010,WAVE64 %s
// RUN: %clang -### -target amdgcn-amd-amdhsa \
// RUN: -x cl -mcpu=gfx1010 -mwavefrontsize64 -mno-wavefrontsize64 \
recommit "[HIP] Add default header and include path" recommit 11d06b9511bd25aabbfad10dff548b0ce29135a5 with fix for lit tests.
2020-06-06 04:49:38 +08:00
// RUN: --rocm-path=%S/Inputs/rocm \
Remove dependency between test files. There seems to be no good reason for rocm-device-libs.cl to depend on opencl.cl. Removed this dependency to unbreak the tests in our setup.
2020-04-13 12:16:46 +08:00
// RUN: %s \
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
// RUN: 2>&1 | FileCheck -dump-input-on-failure --check-prefixes=COMMMON,GFX1010,WAVE32 %s
// Ignore -mno-wavefrontsize64 without wave32 support
// RUN: %clang -### -target amdgcn-amd-amdhsa \
// RUN: -x cl -mcpu=gfx803 -mno-wavefrontsize64 \
recommit "[HIP] Add default header and include path" recommit 11d06b9511bd25aabbfad10dff548b0ce29135a5 with fix for lit tests.
2020-06-06 04:49:38 +08:00
// RUN: --rocm-path=%S/Inputs/rocm \
Remove dependency between test files. There seems to be no good reason for rocm-device-libs.cl to depend on opencl.cl. Removed this dependency to unbreak the tests in our setup.
2020-04-13 12:16:46 +08:00
// RUN: %s \
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
// RUN: 2>&1 | FileCheck -dump-input-on-failure --check-prefixes=COMMMON,GFX803,WAVE64 %s
HIP: Merge builtin library handling Merge with the new --rocm-path handling used for OpenCL. This looks for a usable set of device libraries upfront, rather than giving a generic "no such file or directory error". If any of the required bitcode libraries are missing, this will now produce a "cannot find ROCm installation." error. This differs from the existing hip specific flags by pointing to a rocm root install instead of a single directory with bitcode files. This tries to maintain compatibility with the existing the --hip-device-lib and --hip-device-lib-path flags, as well as the HIP_DEVICE_LIB_PATH environment variable, or at least the range of uses with testcases. The existing range of uses and behavior doesn't entirely make sense to me, so some of the untested edge cases change behavior. Currently the two path forms seem to have the double purpose of a search path for an arbitrary --hip-device-lib, and for finding the stock set of libraries. Since the stock set of libraries This also changes the behavior when multiple paths are specified, and only takes the last one (and the environment variable only handles a single path). If --hip-device-lib is used, it now only treats --hip-device-lib-path as the search path for it, and does not attempt to find the rocm installation. If not, --hip-device-lib-path and the environment variable are used as the directory to search instead of the rocm root based path. This should also automatically fix handling of the options to use wave64.
2020-03-28 00:28:24 +08:00
// Test --hip-device-lib-path format
// RUN: %clang -### -target amdgcn-amd-amdhsa \
// RUN: -x cl -mcpu=gfx900 \
recommit "[HIP] Add default header and include path" recommit 11d06b9511bd25aabbfad10dff548b0ce29135a5 with fix for lit tests.
2020-06-06 04:49:38 +08:00
// RUN: --hip-device-lib-path=%S/Inputs/rocm/amdgcn/bitcode \
HIP: Merge builtin library handling Merge with the new --rocm-path handling used for OpenCL. This looks for a usable set of device libraries upfront, rather than giving a generic "no such file or directory error". If any of the required bitcode libraries are missing, this will now produce a "cannot find ROCm installation." error. This differs from the existing hip specific flags by pointing to a rocm root install instead of a single directory with bitcode files. This tries to maintain compatibility with the existing the --hip-device-lib and --hip-device-lib-path flags, as well as the HIP_DEVICE_LIB_PATH environment variable, or at least the range of uses with testcases. The existing range of uses and behavior doesn't entirely make sense to me, so some of the untested edge cases change behavior. Currently the two path forms seem to have the double purpose of a search path for an arbitrary --hip-device-lib, and for finding the stock set of libraries. Since the stock set of libraries This also changes the behavior when multiple paths are specified, and only takes the last one (and the environment variable only handles a single path). If --hip-device-lib is used, it now only treats --hip-device-lib-path as the search path for it, and does not attempt to find the rocm installation. If not, --hip-device-lib-path and the environment variable are used as the directory to search instead of the rocm root based path. This should also automatically fix handling of the options to use wave64.
2020-03-28 00:28:24 +08:00
// RUN: %S/opencl.cl \
// RUN: 2>&1 | FileCheck -dump-input-on-failure --check-prefixes=COMMON,COMMON-DEFAULT,GFX900-DEFAULT,GFX900,WAVE64 %s
// Test environment variable HIP_DEVICE_LIB_PATH
recommit "[HIP] Add default header and include path" recommit 11d06b9511bd25aabbfad10dff548b0ce29135a5 with fix for lit tests.
2020-06-06 04:49:38 +08:00
// RUN: env HIP_DEVICE_LIB_PATH=%S/Inputs/rocm/amdgcn/bitcode %clang -### -target amdgcn-amd-amdhsa \
HIP: Merge builtin library handling Merge with the new --rocm-path handling used for OpenCL. This looks for a usable set of device libraries upfront, rather than giving a generic "no such file or directory error". If any of the required bitcode libraries are missing, this will now produce a "cannot find ROCm installation." error. This differs from the existing hip specific flags by pointing to a rocm root install instead of a single directory with bitcode files. This tries to maintain compatibility with the existing the --hip-device-lib and --hip-device-lib-path flags, as well as the HIP_DEVICE_LIB_PATH environment variable, or at least the range of uses with testcases. The existing range of uses and behavior doesn't entirely make sense to me, so some of the untested edge cases change behavior. Currently the two path forms seem to have the double purpose of a search path for an arbitrary --hip-device-lib, and for finding the stock set of libraries. Since the stock set of libraries This also changes the behavior when multiple paths are specified, and only takes the last one (and the environment variable only handles a single path). If --hip-device-lib is used, it now only treats --hip-device-lib-path as the search path for it, and does not attempt to find the rocm installation. If not, --hip-device-lib-path and the environment variable are used as the directory to search instead of the rocm root based path. This should also automatically fix handling of the options to use wave64.
2020-03-28 00:28:24 +08:00
// RUN: -x cl -mcpu=gfx900 \
// RUN: %S/opencl.cl \
// RUN: 2>&1 | FileCheck -dump-input-on-failure --check-prefixes=COMMON,COMMON-DEFAULT,GFX900-DEFAULT,GFX900,WAVE64 %s
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
// COMMON: "-triple" "amdgcn-amd-amdhsa"
AMDGPU: Search for new ROCm bitcode library structure The current install situation is a mess, but I'm working on fixing it. Search for the target layout instead of one of the N options that exist today.
2020-04-11 00:56:20 +08:00
// COMMON-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/opencl.bc"
// COMMON-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/ocml.bc"
// COMMON-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/ockl.bc"
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
AMDGPU: Search for new ROCm bitcode library structure The current install situation is a mess, but I'm working on fixing it. Search for the target layout instead of one of the N options that exist today.
2020-04-11 00:56:20 +08:00
// GFX900-DEFAULT-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_daz_opt_off.bc"
// GFX803-DEFAULT-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_daz_opt_on.bc"
// GFX700-DEFAULT-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_daz_opt_on.bc"
// COMMON-DAZ-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_daz_opt_on.bc"
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
AMDGPU: Search for new ROCm bitcode library structure The current install situation is a mess, but I'm working on fixing it. Search for the target layout instead of one of the N options that exist today.
2020-04-11 00:56:20 +08:00
// COMMON-DEFAULT-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_unsafe_math_off.bc"
// COMMON-DEFAULT-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_finite_only_off.bc"
// COMMON-DEFAULT-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_correctly_rounded_sqrt_off.bc"
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
AMDGPU: Search for new ROCm bitcode library structure The current install situation is a mess, but I'm working on fixing it. Search for the target layout instead of one of the N options that exist today.
2020-04-11 00:56:20 +08:00
// COMMON-FINITE-ONLY-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_unsafe_math_off.bc"
// COMMON-FINITE-ONLY-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_finite_only_on.bc"
// COMMON-FINITE-ONLY-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_correctly_rounded_sqrt_off.bc"
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
AMDGPU: Search for new ROCm bitcode library structure The current install situation is a mess, but I'm working on fixing it. Search for the target layout instead of one of the N options that exist today.
2020-04-11 00:56:20 +08:00
// COMMON-CORRECT-SQRT-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_unsafe_math_off.bc"
// COMMON-CORRECT-SQRT-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_finite_only_off.bc"
// COMMON-CORRECT-SQRT-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_correctly_rounded_sqrt_on.bc"
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
AMDGPU: Search for new ROCm bitcode library structure The current install situation is a mess, but I'm working on fixing it. Search for the target layout instead of one of the N options that exist today.
2020-04-11 00:56:20 +08:00
// COMMON-FAST-RELAXED-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_unsafe_math_on.bc"
// COMMON-FAST-RELAXED-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_finite_only_on.bc"
// COMMON-FAST-RELAXED-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_correctly_rounded_sqrt_off.bc"
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
AMDGPU: Search for new ROCm bitcode library structure The current install situation is a mess, but I'm working on fixing it. Search for the target layout instead of one of the N options that exist today.
2020-04-11 00:56:20 +08:00
// COMMON-UNSAFE-MATH-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_unsafe_math_on.bc"
// COMMON-UNSAFE-MATH-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_finite_only_off.bc"
// COMMON-UNSAFE-MATH-SAME: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_correctly_rounded_sqrt_off.bc"
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
AMDGPU: Search for new ROCm bitcode library structure The current install situation is a mess, but I'm working on fixing it. Search for the target layout instead of one of the N options that exist today.
2020-04-11 00:56:20 +08:00
// WAVE64: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_wavefrontsize64_on.bc"
// WAVE32: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_wavefrontsize64_off.bc"
AMDGPU: Teach toolchain to link rocm device libs Currently the library is separately linked, but this isn't correct to implement fast math flags correctly. Each module should get the version of the library appropriate for its combination of fast math and related flags, with the attributes propagated into its functions and internalized. HIP already maintains the list of libraries, but this is not used for OpenCL. Unfortunately, HIP uses a separate --hip-device-lib argument, despite both languages using the same bitcode library. Eventually these two searches need to be merged. An additional problem is there are 3 different locations the libraries are installed, depending on which build is used. This also needs to be consolidated (or at least the search logic needs to deal with this unnecessary complexity).
2018-08-15 23:25:25 +08:00
AMDGPU: Search for new ROCm bitcode library structure The current install situation is a mess, but I'm working on fixing it. Search for the target layout instead of one of the N options that exist today.
2020-04-11 00:56:20 +08:00
// GFX900: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_isa_version_900.bc"
// GFX803: "-mlink-builtin-bitcode" "{{.*}}/amdgcn/bitcode/oclc_isa_version_803.bc"
Remove dependency between test files. There seems to be no good reason for rocm-device-libs.cl to depend on opencl.cl. Removed this dependency to unbreak the tests in our setup.
2020-04-13 12:16:46 +08:00
kernel void func(void);