On Wed, Jan 12, 2022 at 02:32:31AM +0300, Vitaliy Makkoveev wrote: > Does it make sense to document that kernel lock protects > `ps_wxcounter’? We have such [K] in other structure definitions. > > + u_int64_t ps_wxcounter; /* [K] number of W^X violations */
Yes, that would be in order. I'll make sure to follow up with a comment once/if progress is made -- until then I avoid changing the header so as to avoid rebuilds due to a mere comment. > The rest of diff looks good to me. So far, this diff runs fine on amd64, sparc64, arm64 doing releases, regress, ports builds and daily usage for me. Others have reported no regression on their octeon and i386 boxes with varying workloads, as well as additional builds on the architectures I already tested.
