Linux kernel regression status

Latest five known activities:

• Re: [PATCH 0/2] execve scalability issues, part 1
  , by Mateusz Guzik
• Re: [PATCH] mm: Sync percpu mm RSS counters before querying
  , by Michal Hocko
• Re: [PATCH] mm: convert mm's rss stats into percpu_counter
  , by Shakeel Butt
• Re: [PATCH] mm: convert mm's rss stats into percpu_counter
  , by Dennis Zhou
• Re: [PATCH] mm: convert mm's rss stats into percpu_counter
  , by Jan Kara

Regzbot command history:

• note: https://lore.kernel.org/all/20230608111408.s2minsenlcjow7q3@quack3 [RFC PATCH 3/4] mm: Avoid percpu MM counters on single-threaded tasks [implicit due to link]
  , by Gabriel Krisman Bertazi
• note: https://lore.kernel.org/all/20230608111408.s2minsenlcjow7q3@quack3 [RFC PATCH 0/4] Optimize rss_stat initialization/teardown for single-threaded tasks [implicit due to link]
  , by Gabriel Krisman Bertazi
• note: https://lore.kernel.org/linux-mm/20230608111408.s2minsenlcjow7q3@quack3 [RFC PATCH 0/7] Reviving the slab destructor to tackle the percpu allocator scalability problem [implicit due to link]
  , by Harry Yoo
• note: https://lore.kernel.org/linux-mm/20230608111408.s2minsenlcjow7q3@quack3 [RFC PATCH 0/7] Reviving the slab destructor to tackle the percpu allocator scalability problem [implicit due to link]
  , by Harry Yoo
• note: https://lore.kernel.org/all/20230608111408.s2minsenlcjow7q3@quack3/ [RFC PATCH v2] mm: use per-numa-node atomics instead of percpu_counters [implicit due to link]
  , by Sweet Tea Dorminy
• note: https://lore.kernel.org/all/20230608111408.s2minsenlcjow7q3@quack3/ [PATCH] mm: use per-numa-node atomics instead of percpu_counters [implicit due to link]
  , by Sweet Tea Dorminy
• note: https://lore.kernel.org/all/20230608111408.s2minsenlcjow7q3@quack3/ [PATCH] mm: use per-numa-node atomics instead of percpu_counters [implicit due to link]
  , by Sweet Tea Dorminy
• linked: "Re: [PATCH 0/2] execve scalability issues, part 1" mentioned this regression
  , by Mateusz Guzik
• linked: "Re: [PATCH] mm: Sync percpu mm RSS counters before querying" mentioned this regression
  , by Michal Hocko
• backburner: not urgent according to reporter
  , by Linux regression tracking #adding (Thorsten Leemhuis)
• title: mm: performance regression on shell-heavy workloads
  , by Linux regression tracking #adding (Thorsten Leemhuis)
• introduced: f1a7941243c ^
  , by Linux regression tracking #adding (Thorsten Leemhuis)
• note: '18471cb160b2' in 'stable/linux-6.2.y' contains a 'Fixes:' tag for the culprit of this regression
  , by Mathieu Desnoyers
• note: 'b20b0368c614' in 'next' contains a 'Fixes:' tag for the culprit of this regression
  , by Mathieu Desnoyers

When fixing, add this to the commit message to make regzbot notice patch postings and commits to resolve the issue:

• Fixes: f1a7941243c ("mm: convert mm's rss stats into percpu_counter")
• Reported-by: Jan Kara <jack@suse.cz>
• Link: https://lore.kernel.org/r/20230608111408.s2minsenlcjow7q3@quack3/