与 Ext4 (Linux) 相比，为什么 lstat 在 APFS (OSX) 上的性能较差
Why is performance of lstat poor on APFS (OSX) compared to Ext4 (Linux)
在大型存储库上分析 Git 时，我发现 git 状态在 mac 上比 linux 慢得多（10 倍）。 git status 在存储库中的每个文件上运行 lstat，这是缓慢的来源。
与 linux 相比，此系统调用在 mac 上慢得多是否有特殊原因？
首先检查您的 Git 版本，因为最近的 Git 版本（如 , and even the upcoming 2.27 with submodules）有所改进。
甚至git status improved with Git 2.24
The feature.manyFiles setting is suitable for repos with many files in the working directory.
By setting index.version=4 and core.untrackedCache=true, commands such as 'git status' should improve.
在“Global Kernel Locks in APFS" by Gregory Szorc:
中对 APFS 中 lstats 与 Ext4 的差异进行了一项分析
It is apparent that macOS 10.14 Mojave has received performance work relative to macOS 10.13!
Despite those improvements, APFS is still spending a lot of CPU time in the kernel. And the kernel CPU time is still comparatively very high compared to Linux/EXT4, even for single process operation.
While the source code for APFS is not available for me to confirm, the profiling results showing excessive time spent in lck_mtx_lock_grab_mutex() combined with the fact that execution time decreases when the parallel process count decreases leads me to the conclusion that APFS obtains a global kernel lock during read-only operations such as readdir().
In other words, APFS slows down when attempting to perform parallel read-only I/O.
在大型存储库上分析 Git 时，我发现 git 状态在 mac 上比 linux 慢得多（10 倍）。 git status 在存储库中的每个文件上运行 lstat，这是缓慢的来源。
与 linux 相比，此系统调用在 mac 上慢得多是否有特殊原因？
首先检查您的 Git 版本，因为最近的 Git 版本（如
甚至git status improved with Git 2.24
The
feature.manyFilessetting is suitable for repos with many files in the working directory.
By settingindex.version=4andcore.untrackedCache=true, commands such as 'git status' should improve.
在“Global Kernel Locks in APFS" by Gregory Szorc:
中对 APFS 中 lstats 与 Ext4 的差异进行了一项分析It is apparent that macOS 10.14 Mojave has received performance work relative to macOS 10.13!
Despite those improvements, APFS is still spending a lot of CPU time in the kernel. And the kernel CPU time is still comparatively very high compared to Linux/EXT4, even for single process operation.While the source code for APFS is not available for me to confirm, the profiling results showing excessive time spent in
lck_mtx_lock_grab_mutex()combined with the fact that execution time decreases when the parallel process count decreases leads me to the conclusion that APFS obtains a global kernel lock during read-only operations such as readdir().In other words, APFS slows down when attempting to perform parallel read-only I/O.