On 1/30/20 8:49 AM, Michal Hocko wrote: > On Thu 30-01-20 13:24:38, Gavin Shan wrote: >> The inappropriate settings can be accepted by "enabled" and "defrag" >> interface, as the following example shows. This series reworks the >> related functions and fixes the issue by the way. > > Do we really care? This is a root only interface and we do expect root > to know how to use the interface. Also I think David already solved even this case with sysfs_streq() that Andrew suggested: https://lore.kernel.org/linux-mm/alpine.DEB.2.21.2001171411020.56385@xxxxxxxxxxxxxxxxxxxxxxxxx/ >> # cat /sys/kernel/mm/transparent_hugepage/enabled >> always [madvise] never >> # echo neveroff > /sys/kernel/mm/transparent_hugepage/enabled >> # cat /sys/kernel/mm/transparent_hugepage/enabled >> always madvise [never] >> >> Gavin Shan (3): >> mm: Rework {enabled,defrag}_store() >> mm: Rework {enabled,defrag}_show() >> mm: Rework setup_transparent_hugepage() >> >> mm/huge_memory.c | 216 +++++++++++++++++++++++++++-------------------- >> 1 file changed, 125 insertions(+), 91 deletions(-) >> >> -- >> 2.23.0 >> >
