Last time I submitted a very long patch series: https://lore.kernel.org/git/pull.1016.git.1628842990.gitgitgadget@xxxxxxxxx/ My mentor Christian suggested to split the performance optimization part out, so this patch series used to optimize performance optimization in ref-filter. Changes in this patch series: 1. Skip parse_object_buffer(), which can reduce object content parsing. 2. Use linked list to record parsing results, which can reduce second format string parsing. 3. Reuse final buffer, which can reduce memcpy(); 4. Add a need_get_object_info flag instead of compare two object_info, which can reduce memcmp(); 5. Use ALLOC_ARRAY() instead of CALLOC_ARRAY(), which can reduce memset(); Since the relevant part of git cat-file --batch has been deleted, the execution time of git for-each-ref is very short, I haven’t added performance tests yet for git for-each-ref. ZheNing Hu (5): [GSOC] ref-filter: skip parse_object_buffer in some cases [GSOC] ref-filter: remove second parsing in format_ref_array_item [GSOC] ref-filter: reuse final buffer [GSOC] ref-filter: reduce unnecessary object_info comparisons [GSOC]: ref-filter: instead CALLOC_ARRAY to ALLOC_ARRAY builtin/branch.c | 2 + builtin/for-each-ref.c | 3 +- builtin/tag.c | 2 + builtin/verify-tag.c | 2 + ref-filter.c | 168 ++++++++++++++++++++++++++++++++--------- ref-filter.h | 17 ++++- 6 files changed, 154 insertions(+), 40 deletions(-) base-commit: f000ecbed922c727382651490e75014f003c89ca Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-1020%2Fadlternative%2Fref-filter-opt-perf-v1 Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-1020/adlternative/ref-filter-opt-perf-v1 Pull-Request: https://github.com/gitgitgadget/git/pull/1020 -- gitgitgadget
