From ec1fea4516ac2f0c08d31d6308929298d1b281d0 Mon Sep 17 00:00:00 2001 From: Serapheim Dimitropoulos Date: Fri, 26 Jun 2020 18:06:50 -0700 Subject: Use percpu_counter for obj_alloc counter of Linux-backed caches A previous commit enabled the tracking of object allocations in Linux-backed caches from the SPL layer for debuggability. The commit is: 9a170fc6fe54f1e852b6c39630fe5ef2bbd97c16 Unfortunately, it also introduced minor performance regressions that were highlighted by the ZFS perf test-suite. Within Delphix we found that the regression would be from -1%, all the way up to -8% for some workloads. This commit brings performance back up to par by creating a separate counter for those caches and making it a percpu in order to avoid lock-contention. The initial performance testing was done by myself, and the final round was conducted by @tonynguien who was also the one that discovered the regression and highlighted the culprit. Reviewed-by: Matt Ahrens Reviewed-by: Brian Behlendorf Signed-off-by: Serapheim Dimitropoulos Closes #10397 --- config/kernel.m4 | 2 ++ 1 file changed, 2 insertions(+) (limited to 'config/kernel.m4') diff --git a/config/kernel.m4 b/config/kernel.m4 index 78b0ce4d3..ec52f014a 100644 --- a/config/kernel.m4 +++ b/config/kernel.m4 @@ -121,6 +121,7 @@ AC_DEFUN([ZFS_AC_KERNEL_TEST_SRC], [ ZFS_AC_KERNEL_SRC_TOTALRAM_PAGES_FUNC ZFS_AC_KERNEL_SRC_TOTALHIGH_PAGES ZFS_AC_KERNEL_SRC_KSTRTOUL + ZFS_AC_KERNEL_SRC_PERCPU AC_MSG_CHECKING([for available kernel interfaces]) ZFS_LINUX_TEST_COMPILE_ALL([kabi]) @@ -216,6 +217,7 @@ AC_DEFUN([ZFS_AC_KERNEL_TEST_RESULT], [ ZFS_AC_KERNEL_TOTALRAM_PAGES_FUNC ZFS_AC_KERNEL_TOTALHIGH_PAGES ZFS_AC_KERNEL_KSTRTOUL + ZFS_AC_KERNEL_PERCPU ]) dnl # -- cgit v1.2.3