FYI, we noticed a -14.9% regression of aim7.jobs-per-min due to commit:
commit 012bc68ee96faf7b8d245edc270723544574bb11 ("fs: always set I_DIRTY_TIME to fsync correctly on lazytime")
in testcase: aim7
on test machine: 40 threads Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz with 384G memory
with following parameters:
test-description: AIM7 is a traditional UNIX system level benchmark suite which is used to test and measure the performance of multiuser system.
Results have been estimated based on internal Intel analysis and are provided
for informational purposes only. Any difference in system hardware or software
design or configuration may affect actual performance.
I've started regularly building a for-next topic branch here:
These are NFS/RDMA patches for the next kernel version. I'd like
to request that this topic branch is added to regular zero-day
and kbuild testing.