Message ID | 20220830081457.118960-1-aneesh.kumar@linux.ibm.com (mailing list archive) |
---|---|
State | New |
Headers | show |
Series | [mm-unstable] mm/demotion: Make toptier_distance inclusive upper bound of toptiers | expand |
On Tue, Aug 30, 2022 at 1:15 AM Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com> wrote: > > Based on suggestion from Wei Xu <weixugc@google.com>, having an > inclusive upper bound is found simpler. > > Fixes: mm/demotion: Update node_is_toptier to work with memory tiers > Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com> > --- > mm/memory-tiers.c | 5 +++-- > 1 file changed, 3 insertions(+), 2 deletions(-) > > diff --git a/mm/memory-tiers.c b/mm/memory-tiers.c > index c4bd6d052a33..c82eb0111383 100644 > --- a/mm/memory-tiers.c > +++ b/mm/memory-tiers.c > @@ -179,7 +179,7 @@ bool node_is_toptier(int node) > toptier = true; > goto out; > } > - if (memtier->adistance_start < top_tier_adistance) > + if (memtier->adistance_start <= top_tier_adistance) > toptier = true; > else > toptier = false; > @@ -361,7 +361,8 @@ static void establish_demotion_targets(void) > * abstract distance below the max value of this memtier below -> up to ? Looks good to me otherwise. Wei > * is considered toptier. > */ > - top_tier_adistance = memtier->adistance_start + MEMTIER_CHUNK_SIZE; > + top_tier_adistance = memtier->adistance_start + > + MEMTIER_CHUNK_SIZE - 1; > break; > } > } > -- > 2.37.2 >
diff --git a/mm/memory-tiers.c b/mm/memory-tiers.c index c4bd6d052a33..c82eb0111383 100644 --- a/mm/memory-tiers.c +++ b/mm/memory-tiers.c @@ -179,7 +179,7 @@ bool node_is_toptier(int node) toptier = true; goto out; } - if (memtier->adistance_start < top_tier_adistance) + if (memtier->adistance_start <= top_tier_adistance) toptier = true; else toptier = false; @@ -361,7 +361,8 @@ static void establish_demotion_targets(void) * abstract distance below the max value of this memtier * is considered toptier. */ - top_tier_adistance = memtier->adistance_start + MEMTIER_CHUNK_SIZE; + top_tier_adistance = memtier->adistance_start + + MEMTIER_CHUNK_SIZE - 1; break; } }
Based on suggestion from Wei Xu <weixugc@google.com>, having an inclusive upper bound is found simpler. Fixes: mm/demotion: Update node_is_toptier to work with memory tiers Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com> --- mm/memory-tiers.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)