workaround upstream issue

[blackboxprogramming]

Description

This PR fixes #

Notes for Reviewers

Signed commits

• Yes, I signed my commits.

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, you can upgrade your account or add credits to your account and enable them for code reviews in your settings.

[Copilot]

Pull request overview

This PR implements a workaround for an upstream issue in llama.cpp (issue #8665) related to SYCL library linking. The changes conditionally link the SYCL library when building with Intel OneAPI, and update the llama.cpp dependency version.

• Adds conditional SYCL linking based on ONEAPI_ROOT environment variable
• Creates a workaround CMakeLists.txt file for the RPC server component
• Updates llama.cpp version dependency

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 2 comments.

File	Description
backend/cpp/llama/prepare.sh	Adds script to copy workaround CMakeLists.txt for RPC server to address upstream issue
backend/cpp/llama/CMakeLists.txt.rpc-8662	New workaround CMakeLists.txt for RPC server with conditional SYCL linking
backend/cpp/llama/CMakeLists.txt	Implements conditional SYCL linking for grpc-server target based on ONEAPI_ROOT
Makefile	Updates CPPLLAMA_VERSION to newer commit hash

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

The issue reference points to #8665, but the filename and comment reference issue #8662. This inconsistency could cause confusion. Ensure the correct issue number is used consistently throughout the codebase.

Suggested change

	# https://github.com/ggerganov/llama.cpp/issues/8665
	# https://github.com/ggerganov/llama.cpp/issues/8662

[Copilot]

The code in both branches of the if/else statement is nearly identical, with only the sycl library addition being different. This duplication makes maintenance harder. Consider extracting the common libraries to a variable and conditionally appending sycl to avoid repeating the entire target_link_libraries call.

Suggested change

	if ( DEFINED ENV{ONEAPI_ROOT})
	target_link_libraries(${TARGET} PRIVATE common llama myclip ${CMAKE_THREAD_LIBS_INIT} absl::flags hw_grpc_proto
	absl::flags_parse
	gRPC::${_REFLECTION}
	gRPC::${_GRPC_GRPCPP}
	protobuf::${_PROTOBUF_LIBPROTOBUF}
	sycl)
	else()
	target_link_libraries(${TARGET} PRIVATE common llama myclip ${CMAKE_THREAD_LIBS_INIT} absl::flags hw_grpc_proto
	absl::flags_parse
	gRPC::${_REFLECTION}
	gRPC::${_GRPC_GRPCPP}
	protobuf::${_PROTOBUF_LIBPROTOBUF})
	endif()
	set(GRPC_SERVER_LIBS
	common
	llama
	myclip
	${CMAKE_THREAD_LIBS_INIT}
	absl::flags
	hw_grpc_proto
	absl::flags_parse
	gRPC::${_REFLECTION}
	gRPC::${_GRPC_GRPCPP}
	protobuf::${_PROTOBUF_LIBPROTOBUF}
	)
	if (DEFINED ENV{ONEAPI_ROOT})
	list(APPEND GRPC_SERVER_LIBS sycl)
	endif()
	target_link_libraries(${TARGET} PRIVATE ${GRPC_SERVER_LIBS})