On Sun, 19 Apr 2026 12:57:40 -0600 Nico Pache <[email protected]> wrote:
> The following cleanup reworks all the max_ptes_* handling into helper > functions. This increases the code readability and will later be used to > implement the mTHP handling of these variables. > > With these changes we abstract all the madvise_collapse() special casing > (dont respect the sysctls) away from the functions that utilize them. And > will later in this series to cleanly restrict mTHP collapses behaviors. > > Suggested-by: David Hildenbrand <[email protected]> > Signed-off-by: Nico Pache <[email protected]> > --- > mm/khugepaged.c | 114 +++++++++++++++++++++++++++++++++--------------- > 1 file changed, 78 insertions(+), 36 deletions(-) > The old code re-read khugepaged_max_ptes_* on every loop iteration; the new code snapshots them once per scan call. If userspace writes the sysctl mid-scan, old behavior reacted within the scan, new behavior uses the value sampled at entry. This is completely ok IMO, but might be good to call out. Also might be good to write no functional change intended apart from above in the commit message? Acked-by: Usama Arif <[email protected]>
