Update software stack to NEO v25.27.34303 #513
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
michel2323
force-pushed
the
ms/update_stack
branch
from
71440b3 to
9f16d9e
Compare
michel2323
force-pushed
the
ms/update_stack
branch
from
9f16d9e to
15971c4
Compare
|
Your PR requires formatting changes to meet the project's style guidelines. Click here to view the suggested changes.diff --git a/deps/generate_helpers.jl b/deps/generate_helpers.jl
index 01f162d..5e715db 100644
--- a/deps/generate_helpers.jl
+++ b/deps/generate_helpers.jl
@@ -1,6 +1,6 @@
non_parametric_routines = ["init_matrix_handle", "release_matrix_handle", "set_matrix_property",
"init_matmat_descr", "release_matmat_descr", "set_matmat_data", "get_matmat_data", "matmat",
-"omatcopy", "sort_matrix", "optimize_gemv", "optimize_gemm", "optimize_trmv", "optimize_trsv", "optimize_trsm",
+ "omatcopy", "sort_matrix", "optimize_gemv", "optimize_gemm", "optimize_trmv", "optimize_trsv", "optimize_trsm",
"init_omatconvert_descr", "release_omatconvert_descr", "init_omatadd_descr", "release_omatadd_descr",
"omatconvert_buffer_size", "omatconvert_analyze", "omatconvert_get_nnz", "omatconvert",
"omatadd_buffer_size", "omatadd_analyze", "omatadd_get_nnz"]
diff --git a/deps/generate_interfaces.jl b/deps/generate_interfaces.jl
index a36fd0a..470c451 100644
--- a/deps/generate_interfaces.jl
+++ b/deps/generate_interfaces.jl
@@ -241,7 +241,7 @@ function generate_headers(library::String, filename::Vector{String}, output::Str
# The routine "optimize_trsm" has two versions.
suffix = ""
(name_routine == "optimize_trsm") && occursin("columns", header) && (suffix = "_advanced")
- (name_routine == "optimize_gemm") && occursin("columns", header) && (suffix = "_advanced")
+ (name_routine == "optimize_gemm") && occursin("columns", header) && (suffix = "_advanced")
name_routine ∈ ("set_csr_data", "set_coo_data") && occursin("int64_t", header) && (suffix = "_64")
occursin("batch", name_routine) && !occursin("**", header) && (suffix = "_strided")
diff --git a/lib/level-zero/libze.jl b/lib/level-zero/libze.jl
index d8ad0b8..167a511 100644
--- a/lib/level-zero/libze.jl
+++ b/lib/level-zero/libze.jl
@@ -11,7 +11,8 @@ end
function check(f)
res = retry_reclaim(err -> err == RESULT_ERROR_OUT_OF_HOST_MEMORY ||
- err == RESULT_ERROR_OUT_OF_DEVICE_MEMORY) do
+ err == RESULT_ERROR_OUT_OF_DEVICE_MEMORY
+ ) do
return f()
end
@@ -2164,9 +2165,11 @@ end
const ze_init_driver_type_flag_t = _ze_init_driver_type_flag_t
@checked function zeInitDrivers(pCount, phDrivers, desc)
- @ccall libze_loader.zeInitDrivers(pCount::Ptr{UInt32},
- phDrivers::Ptr{ze_driver_handle_t},
- desc::Ptr{ze_init_driver_type_desc_t})::ze_result_t
+ @ccall libze_loader.zeInitDrivers(
+ pCount::Ptr{UInt32},
+ phDrivers::Ptr{ze_driver_handle_t},
+ desc::Ptr{ze_init_driver_type_desc_t}
+ )::ze_result_t
end
@cenum _ze_api_version_t::UInt32 begin
@@ -3444,9 +3447,11 @@ end
const ze_kernel_get_binary_exp_version_t = _ze_kernel_get_binary_exp_version_t
@checked function zeKernelGetBinaryExp(hKernel, pSize, pKernelBinary)
- @ccall libze_loader.zeKernelGetBinaryExp(hKernel::ze_kernel_handle_t,
- pSize::Ptr{Csize_t},
- pKernelBinary::Ptr{UInt8})::ze_result_t
+ @ccall libze_loader.zeKernelGetBinaryExp(
+ hKernel::ze_kernel_handle_t,
+ pSize::Ptr{Csize_t},
+ pKernelBinary::Ptr{UInt8}
+ )::ze_result_t
end
@cenum _ze_driver_ddi_handles_ext_version_t::UInt32 begin
@@ -3492,39 +3497,49 @@ end
const ze_external_semaphore_ext_flag_t = _ze_external_semaphore_ext_flag_t
@checked function zeDeviceImportExternalSemaphoreExt(hDevice, desc, phSemaphore)
- @ccall libze_loader.zeDeviceImportExternalSemaphoreExt(hDevice::ze_device_handle_t,
- desc::Ptr{ze_external_semaphore_ext_desc_t},
- phSemaphore::Ptr{ze_external_semaphore_ext_handle_t})::ze_result_t
+ @ccall libze_loader.zeDeviceImportExternalSemaphoreExt(
+ hDevice::ze_device_handle_t,
+ desc::Ptr{ze_external_semaphore_ext_desc_t},
+ phSemaphore::Ptr{ze_external_semaphore_ext_handle_t}
+ )::ze_result_t
end
@checked function zeDeviceReleaseExternalSemaphoreExt(hSemaphore)
@ccall libze_loader.zeDeviceReleaseExternalSemaphoreExt(hSemaphore::ze_external_semaphore_ext_handle_t)::ze_result_t
end
-@checked function zeCommandListAppendSignalExternalSemaphoreExt(hCommandList, numSemaphores,
- phSemaphores, signalParams,
- hSignalEvent, numWaitEvents,
- phWaitEvents)
- @ccall libze_loader.zeCommandListAppendSignalExternalSemaphoreExt(hCommandList::ze_command_list_handle_t,
- numSemaphores::UInt32,
- phSemaphores::Ptr{ze_external_semaphore_ext_handle_t},
- signalParams::Ptr{ze_external_semaphore_signal_params_ext_t},
- hSignalEvent::ze_event_handle_t,
- numWaitEvents::UInt32,
- phWaitEvents::Ptr{ze_event_handle_t})::ze_result_t
-end
-
-@checked function zeCommandListAppendWaitExternalSemaphoreExt(hCommandList, numSemaphores,
- phSemaphores, waitParams,
- hSignalEvent, numWaitEvents,
- phWaitEvents)
- @ccall libze_loader.zeCommandListAppendWaitExternalSemaphoreExt(hCommandList::ze_command_list_handle_t,
- numSemaphores::UInt32,
- phSemaphores::Ptr{ze_external_semaphore_ext_handle_t},
- waitParams::Ptr{ze_external_semaphore_wait_params_ext_t},
- hSignalEvent::ze_event_handle_t,
- numWaitEvents::UInt32,
- phWaitEvents::Ptr{ze_event_handle_t})::ze_result_t
+@checked function zeCommandListAppendSignalExternalSemaphoreExt(
+ hCommandList, numSemaphores,
+ phSemaphores, signalParams,
+ hSignalEvent, numWaitEvents,
+ phWaitEvents
+ )
+ @ccall libze_loader.zeCommandListAppendSignalExternalSemaphoreExt(
+ hCommandList::ze_command_list_handle_t,
+ numSemaphores::UInt32,
+ phSemaphores::Ptr{ze_external_semaphore_ext_handle_t},
+ signalParams::Ptr{ze_external_semaphore_signal_params_ext_t},
+ hSignalEvent::ze_event_handle_t,
+ numWaitEvents::UInt32,
+ phWaitEvents::Ptr{ze_event_handle_t}
+ )::ze_result_t
+end
+
+@checked function zeCommandListAppendWaitExternalSemaphoreExt(
+ hCommandList, numSemaphores,
+ phSemaphores, waitParams,
+ hSignalEvent, numWaitEvents,
+ phWaitEvents
+ )
+ @ccall libze_loader.zeCommandListAppendWaitExternalSemaphoreExt(
+ hCommandList::ze_command_list_handle_t,
+ numSemaphores::UInt32,
+ phSemaphores::Ptr{ze_external_semaphore_ext_handle_t},
+ waitParams::Ptr{ze_external_semaphore_wait_params_ext_t},
+ hSignalEvent::ze_event_handle_t,
+ numWaitEvents::UInt32,
+ phWaitEvents::Ptr{ze_event_handle_t}
+ )::ze_result_t
end
@cenum _ze_device_cache_line_size_ext_version_t::UInt32 begin
@@ -3617,50 +3632,66 @@ mutable struct _ze_rtas_parallel_operation_ext_handle_t end
const ze_rtas_parallel_operation_ext_handle_t = Ptr{_ze_rtas_parallel_operation_ext_handle_t}
@checked function zeRTASBuilderCreateExt(hDriver, pDescriptor, phBuilder)
- @ccall libze_loader.zeRTASBuilderCreateExt(hDriver::ze_driver_handle_t,
- pDescriptor::Ptr{ze_rtas_builder_ext_desc_t},
- phBuilder::Ptr{ze_rtas_builder_ext_handle_t})::ze_result_t
+ @ccall libze_loader.zeRTASBuilderCreateExt(
+ hDriver::ze_driver_handle_t,
+ pDescriptor::Ptr{ze_rtas_builder_ext_desc_t},
+ phBuilder::Ptr{ze_rtas_builder_ext_handle_t}
+ )::ze_result_t
end
-@checked function zeRTASBuilderGetBuildPropertiesExt(hBuilder, pBuildOpDescriptor,
- pProperties)
- @ccall libze_loader.zeRTASBuilderGetBuildPropertiesExt(hBuilder::ze_rtas_builder_ext_handle_t,
- pBuildOpDescriptor::Ptr{ze_rtas_builder_build_op_ext_desc_t},
- pProperties::Ptr{ze_rtas_builder_ext_properties_t})::ze_result_t
+@checked function zeRTASBuilderGetBuildPropertiesExt(
+ hBuilder, pBuildOpDescriptor,
+ pProperties
+ )
+ @ccall libze_loader.zeRTASBuilderGetBuildPropertiesExt(
+ hBuilder::ze_rtas_builder_ext_handle_t,
+ pBuildOpDescriptor::Ptr{ze_rtas_builder_build_op_ext_desc_t},
+ pProperties::Ptr{ze_rtas_builder_ext_properties_t}
+ )::ze_result_t
end
@checked function zeDriverRTASFormatCompatibilityCheckExt(hDriver, rtasFormatA, rtasFormatB)
- @ccall libze_loader.zeDriverRTASFormatCompatibilityCheckExt(hDriver::ze_driver_handle_t,
- rtasFormatA::ze_rtas_format_ext_t,
- rtasFormatB::ze_rtas_format_ext_t)::ze_result_t
-end
-
-@checked function zeRTASBuilderBuildExt(hBuilder, pBuildOpDescriptor, pScratchBuffer,
- scratchBufferSizeBytes, pRtasBuffer,
- rtasBufferSizeBytes, hParallelOperation,
- pBuildUserPtr, pBounds, pRtasBufferSizeBytes)
- @ccall libze_loader.zeRTASBuilderBuildExt(hBuilder::ze_rtas_builder_ext_handle_t,
- pBuildOpDescriptor::Ptr{ze_rtas_builder_build_op_ext_desc_t},
- pScratchBuffer::Ptr{Cvoid},
- scratchBufferSizeBytes::Csize_t,
- pRtasBuffer::Ptr{Cvoid},
- rtasBufferSizeBytes::Csize_t,
- hParallelOperation::ze_rtas_parallel_operation_ext_handle_t,
- pBuildUserPtr::Ptr{Cvoid},
- pBounds::Ptr{ze_rtas_aabb_ext_t},
- pRtasBufferSizeBytes::Ptr{Csize_t})::ze_result_t
-end
-
-@checked function zeRTASBuilderCommandListAppendCopyExt(hCommandList, dstptr, srcptr, size,
- hSignalEvent, numWaitEvents,
- phWaitEvents)
- @ccall libze_loader.zeRTASBuilderCommandListAppendCopyExt(hCommandList::ze_command_list_handle_t,
- dstptr::Ptr{Cvoid},
- srcptr::Ptr{Cvoid},
- size::Csize_t,
- hSignalEvent::ze_event_handle_t,
- numWaitEvents::UInt32,
- phWaitEvents::Ptr{ze_event_handle_t})::ze_result_t
+ @ccall libze_loader.zeDriverRTASFormatCompatibilityCheckExt(
+ hDriver::ze_driver_handle_t,
+ rtasFormatA::ze_rtas_format_ext_t,
+ rtasFormatB::ze_rtas_format_ext_t
+ )::ze_result_t
+end
+
+@checked function zeRTASBuilderBuildExt(
+ hBuilder, pBuildOpDescriptor, pScratchBuffer,
+ scratchBufferSizeBytes, pRtasBuffer,
+ rtasBufferSizeBytes, hParallelOperation,
+ pBuildUserPtr, pBounds, pRtasBufferSizeBytes
+ )
+ @ccall libze_loader.zeRTASBuilderBuildExt(
+ hBuilder::ze_rtas_builder_ext_handle_t,
+ pBuildOpDescriptor::Ptr{ze_rtas_builder_build_op_ext_desc_t},
+ pScratchBuffer::Ptr{Cvoid},
+ scratchBufferSizeBytes::Csize_t,
+ pRtasBuffer::Ptr{Cvoid},
+ rtasBufferSizeBytes::Csize_t,
+ hParallelOperation::ze_rtas_parallel_operation_ext_handle_t,
+ pBuildUserPtr::Ptr{Cvoid},
+ pBounds::Ptr{ze_rtas_aabb_ext_t},
+ pRtasBufferSizeBytes::Ptr{Csize_t}
+ )::ze_result_t
+end
+
+@checked function zeRTASBuilderCommandListAppendCopyExt(
+ hCommandList, dstptr, srcptr, size,
+ hSignalEvent, numWaitEvents,
+ phWaitEvents
+ )
+ @ccall libze_loader.zeRTASBuilderCommandListAppendCopyExt(
+ hCommandList::ze_command_list_handle_t,
+ dstptr::Ptr{Cvoid},
+ srcptr::Ptr{Cvoid},
+ size::Csize_t,
+ hSignalEvent::ze_event_handle_t,
+ numWaitEvents::UInt32,
+ phWaitEvents::Ptr{ze_event_handle_t}
+ )::ze_result_t
end
@checked function zeRTASBuilderDestroyExt(hBuilder)
@@ -3668,13 +3699,17 @@ end
end
@checked function zeRTASParallelOperationCreateExt(hDriver, phParallelOperation)
- @ccall libze_loader.zeRTASParallelOperationCreateExt(hDriver::ze_driver_handle_t,
- phParallelOperation::Ptr{ze_rtas_parallel_operation_ext_handle_t})::ze_result_t
+ @ccall libze_loader.zeRTASParallelOperationCreateExt(
+ hDriver::ze_driver_handle_t,
+ phParallelOperation::Ptr{ze_rtas_parallel_operation_ext_handle_t}
+ )::ze_result_t
end
@checked function zeRTASParallelOperationGetPropertiesExt(hParallelOperation, pProperties)
- @ccall libze_loader.zeRTASParallelOperationGetPropertiesExt(hParallelOperation::ze_rtas_parallel_operation_ext_handle_t,
- pProperties::Ptr{ze_rtas_parallel_operation_ext_properties_t})::ze_result_t
+ @ccall libze_loader.zeRTASParallelOperationGetPropertiesExt(
+ hParallelOperation::ze_rtas_parallel_operation_ext_handle_t,
+ pProperties::Ptr{ze_rtas_parallel_operation_ext_properties_t}
+ )::ze_result_t
end
@checked function zeRTASParallelOperationJoinExt(hParallelOperation)
@@ -3693,11 +3728,15 @@ end
const ze_device_vector_sizes_ext_version_t = _ze_device_vector_sizes_ext_version_t
-@checked function zeDeviceGetVectorWidthPropertiesExt(hDevice, pCount,
- pVectorWidthProperties)
- @ccall libze_loader.zeDeviceGetVectorWidthPropertiesExt(hDevice::ze_device_handle_t,
- pCount::Ptr{UInt32},
- pVectorWidthProperties::Ptr{ze_device_vector_width_properties_ext_t})::ze_result_t
+@checked function zeDeviceGetVectorWidthPropertiesExt(
+ hDevice, pCount,
+ pVectorWidthProperties
+ )
+ @ccall libze_loader.zeDeviceGetVectorWidthPropertiesExt(
+ hDevice::ze_device_handle_t,
+ pCount::Ptr{UInt32},
+ pVectorWidthProperties::Ptr{ze_device_vector_width_properties_ext_t}
+ )::ze_result_t
end
@cenum _ze_cache_reservation_ext_version_t::UInt32 begin
@@ -4376,14 +4415,18 @@ const ze_mutable_command_list_exp_flag_t = _ze_mutable_command_list_exp_flag_t
pCommandId::Ptr{UInt64})::ze_result_t
end
-@checked function zeCommandListGetNextCommandIdWithKernelsExp(hCommandList, desc,
- numKernels, phKernels,
- pCommandId)
- @ccall libze_loader.zeCommandListGetNextCommandIdWithKernelsExp(hCommandList::ze_command_list_handle_t,
- desc::Ptr{ze_mutable_command_id_exp_desc_t},
- numKernels::UInt32,
- phKernels::Ptr{ze_kernel_handle_t},
- pCommandId::Ptr{UInt64})::ze_result_t
+@checked function zeCommandListGetNextCommandIdWithKernelsExp(
+ hCommandList, desc,
+ numKernels, phKernels,
+ pCommandId
+ )
+ @ccall libze_loader.zeCommandListGetNextCommandIdWithKernelsExp(
+ hCommandList::ze_command_list_handle_t,
+ desc::Ptr{ze_mutable_command_id_exp_desc_t},
+ numKernels::UInt32,
+ phKernels::Ptr{ze_kernel_handle_t},
+ pCommandId::Ptr{UInt64}
+ )::ze_result_t
end
@checked function zeCommandListUpdateMutableCommandsExp(hCommandList, desc)
@@ -4407,12 +4450,16 @@ end
phWaitEvents::Ptr{ze_event_handle_t})::ze_result_t
end
-@checked function zeCommandListUpdateMutableCommandKernelsExp(hCommandList, numKernels,
- pCommandId, phKernels)
- @ccall libze_loader.zeCommandListUpdateMutableCommandKernelsExp(hCommandList::ze_command_list_handle_t,
- numKernels::UInt32,
- pCommandId::Ptr{UInt64},
- phKernels::Ptr{ze_kernel_handle_t})::ze_result_t
+@checked function zeCommandListUpdateMutableCommandKernelsExp(
+ hCommandList, numKernels,
+ pCommandId, phKernels
+ )
+ @ccall libze_loader.zeCommandListUpdateMutableCommandKernelsExp(
+ hCommandList::ze_command_list_handle_t,
+ numKernels::UInt32,
+ pCommandId::Ptr{UInt64},
+ phKernels::Ptr{ze_kernel_handle_t}
+ )::ze_result_t
end
struct _ze_init_params_t
diff --git a/lib/support/liboneapi_support.jl b/lib/support/liboneapi_support.jl
index 9b5858f..29bc6f1 100644
--- a/lib/support/liboneapi_support.jl
+++ b/lib/support/liboneapi_support.jl
@@ -6613,23 +6613,31 @@ function onemklXsparse_release_matmat_descr(p_desc)
end
function onemklXsparse_init_omatconvert_descr(device_queue, p_descr)
- @ccall liboneapi_support.onemklXsparse_init_omatconvert_descr(device_queue::syclQueue_t,
- p_descr::Ptr{omatconvert_descr_t})::Cint
+ return @ccall liboneapi_support.onemklXsparse_init_omatconvert_descr(
+ device_queue::syclQueue_t,
+ p_descr::Ptr{omatconvert_descr_t}
+ )::Cint
end
function onemklXsparse_release_omatconvert_descr(device_queue, descr)
- @ccall liboneapi_support.onemklXsparse_release_omatconvert_descr(device_queue::syclQueue_t,
- descr::omatconvert_descr_t)::Cint
+ return @ccall liboneapi_support.onemklXsparse_release_omatconvert_descr(
+ device_queue::syclQueue_t,
+ descr::omatconvert_descr_t
+ )::Cint
end
function onemklXsparse_init_omatadd_descr(device_queue, p_omatadd_desc)
- @ccall liboneapi_support.onemklXsparse_init_omatadd_descr(device_queue::syclQueue_t,
- p_omatadd_desc::Ptr{omatadd_descr_t})::Cint
+ return @ccall liboneapi_support.onemklXsparse_init_omatadd_descr(
+ device_queue::syclQueue_t,
+ p_omatadd_desc::Ptr{omatadd_descr_t}
+ )::Cint
end
function onemklXsparse_release_omatadd_descr(device_queue, omatadd_desc)
- @ccall liboneapi_support.onemklXsparse_release_omatadd_descr(device_queue::syclQueue_t,
- omatadd_desc::omatadd_descr_t)::Cint
+ return @ccall liboneapi_support.onemklXsparse_release_omatadd_descr(
+ device_queue::syclQueue_t,
+ omatadd_desc::omatadd_descr_t
+ )::Cint
end
function onemklXsparse_omatcopy(device_queue, transpose_val, spMat_in, spMat_out)
@@ -6695,19 +6703,25 @@ function onemklXsparse_optimize_trsv(device_queue, uplo_val, opA, diag_val, A)
end
function onemklXsparse_optimize_gemm(device_queue, opA, A)
- @ccall liboneapi_support.onemklXsparse_optimize_gemm(device_queue::syclQueue_t,
- opA::onemklTranspose,
- A::matrix_handle_t)::Cint
-end
-
-function onemklXsparse_optimize_gemm_advanced(device_queue, layout_val, opA, opB, A,
- columns)
- @ccall liboneapi_support.onemklXsparse_optimize_gemm_advanced(device_queue::syclQueue_t,
- layout_val::onemklLayout,
- opA::onemklTranspose,
- opB::onemklTranspose,
- A::matrix_handle_t,
- columns::Int64)::Cint
+ return @ccall liboneapi_support.onemklXsparse_optimize_gemm(
+ device_queue::syclQueue_t,
+ opA::onemklTranspose,
+ A::matrix_handle_t
+ )::Cint
+end
+
+function onemklXsparse_optimize_gemm_advanced(
+ device_queue, layout_val, opA, opB, A,
+ columns
+ )
+ return @ccall liboneapi_support.onemklXsparse_optimize_gemm_advanced(
+ device_queue::syclQueue_t,
+ layout_val::onemklLayout,
+ opA::onemklTranspose,
+ opB::onemklTranspose,
+ A::matrix_handle_t,
+ columns::Int64
+ )::Cint
end
function onemklXsparse_optimize_trsm(device_queue, uplo_val, opA, diag_val, A)
diff --git a/test/onemkl.jl b/test/onemkl.jl
index e5b6541..a4fd56c 100644
--- a/test/onemkl.jl
+++ b/test/onemkl.jl
@@ -1132,7 +1132,7 @@ end
alpha = rand(T)
beta = rand(T)
- oneMKL.sparse_optimize_gemm!(transa, dA)
+ oneMKL.sparse_optimize_gemm!(transa, dA)
oneMKL.sparse_gemm!(transa, transb, alpha, dA, dB, beta, dC)
@test alpha * opa(A) * opb(B) + beta * C ≈ collect(dC)
end |
|
Rebased on @amontoison's #511 because it is related. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Edit: I reported local issues which just a wrong setting.