From: Scott Mitchell <[email protected]> This series optimizes RTE_PTR_ADD and RTE_PTR_SUB by using char* pointer arithmetic instead of uintptr_t casts when operating on pointer types. This enables better compiler optimization, particularly for Clang which can now recognize simple pointer patterns and apply vectorization, loop unrolling, and improved assembly.
The implementation uses C11 _Generic to dispatch based on input type, maintaining full API compatibility while enabling significant performance improvements in checksum operations (40% for small buffers, up to 8x for larger buffers). The first patch contains the core optimization changes and unit tests. The second patch adds a .mailmap entry for consistent git attribution. Note on checkpatch warnings: The patches generate checkpatch warnings due to C11 _Generic syntax not being fully recognized by checkpatch.pl (designed for kernel C). The warnings are false positives - the code is valid C11 and follows the same pattern as existing DPDK code (e.g., __rte_constant macro, rte_bitops.h). The COMPLEX_MACRO warning is explicitly acknowledged by checkpatch itself as expected for statement expression macros. Changes in v4: - Fixed duplicate "v3 v3" in subject line from v3 Changes in v3: - Fixed CI build failures on GCC - Simplified const pointer handling: GCC type-checks all _Generic branches including unselected ones, causing errors with the nested _Generic/ternary approach used in v2. Now uses simple char* cast for all pointer types, which compiles on both GCC and Clang while still achieving the optimization goal (char* arithmetic enables compiler optimizations vs uintptr_t). - Removed Clang-specific pragma for -Wconditional-type-mismatch (no longer needed without the ternary operator) Changes in v2: - Fixed copyright in test file to Apple Inc. (was Intel Corporation) - Added checkpatch warning explanation to cover letter Scott Mitchell (2): eal: RTE_PTR_ADD/SUB char* for compiler optimizations mailmap: add Scott Mitchell .mailmap | 1 + app/test/meson.build | 1 + app/test/test_ptr_add_sub.c | 190 +++++++++++++++++++++++++++++++++++ lib/eal/include/rte_common.h | 60 +++++++++++ 4 files changed, 252 insertions(+) create mode 100644 app/test/test_ptr_add_sub.c -- 2.39.5 (Apple Git-154)
