mm/page_alloc: effectively disable pcp with CONFIG_SMP=n

mm/page_alloc: effectively disable pcp with CONFIG_SMP=n

Patch series "mm/page_alloc: pcp locking cleanup".

This is a followup to the hotfix 038a102535eb ("mm/page_alloc: prevent pcp
corruption with SMP=n"), to simplify the code and deal with the original
issue properly.  The previous RFC attempt [1] argued for changing the UP
spinlock implementation, which was discouraged, but thanks to David's
off-list suggestion, we can achieve the goal without changing the spinlock
implementation.

The main change in Patch 1 relies on the fact that on UP we don't need the
pcp lists for scalability, so just make them always bypassed during
alloc/free by making the pcp trylock an unconditional failure.

The various drain paths that use pcp_spin_lock_maybe_irqsave() continue to
exist but will never do any work in practice.  In Patch 2 we can again
remove the irq saving from them that commit 038a102535eb added.

Besides simpler code with all the ugly UP_flags removed, we get less bloat
with CONFIG_SMP=n for mm/page_alloc.o as a result:

add/remove: 25/28 grow/shrink: 4/5 up/down: 2105/-6665 (-4560)
Function                                     old     new   delta
get_page_from_freelist                      5689    7248   +1559
free_unref_folios                           2006    2324    +318
make_alloc_exact                             270     286     +16
__zone_watermark_ok                          306     322     +16
drain_pages_zone.isra                        119     109     -10
decay_pcp_high                               181     149     -32
setup_pcp_cacheinfo                          193     147     -46
__free_frozen_pages                         1339    1089    -250
alloc_pages_bulk_noprof                     1054     419    -635
free_frozen_page_commit                      907       -    -907
try_to_claim_block                          1975       -   -1975
__rmqueue_pcplist                           2614       -   -2614
Total: Before=54624, After=50064, chg -8.35%

This patch (of 3):

The page allocator has been using a locking scheme for its percpu page
caches (pcp) based on spin_trylock() with no _irqsave() part.  The trick
is that if we interrupt the locked section, we fail the trylock and just
fallback to the slowpath taking the zone lock.  That's more expensive, but
rare, so we don't need to pay the irqsave/restore cost all the time in the
fastpaths.

It's similar to but not exactly local_trylock_t (which is also newer
anyway) because in some cases we do lock the pcp of a non-local cpu to
drain it, in a way that's cheaper than using IPI or queue_work_on().

The complication of this scheme has been UP non-debug spinlock
implementation which assumes spin_trylock() can't fail on UP and has no
state to track whether it's locked.  It just doesn't anticipate this usage
scenario.  So to work around that we disable IRQs only on UP, complicating
the implementation.  Also recently we found years old bug in where we
didn't disable IRQs in related paths - see 038a102535eb ("mm/page_alloc:
prevent pcp corruption with SMP=n").

We can avoid this UP complication by realizing that we do not need the pcp
caching for scalability on UP in the first place.  Removing it completely
with #ifdefs is not worth the trouble either.  Just make
pcp_spin_trylock() return NULL unconditionally on CONFIG_SMP=n.  This
makes the slowpaths unconditional, and we can remove the IRQ save/restore
handling in pcp_spin_trylock()/unlock() completely.

Link: https://lkml.kernel.org/r/20260227-b4-pcp-locking-cleanup-v1-0-f7e22e603447@kernel.org
Link: https://lkml.kernel.org/r/20260227-b4-pcp-locking-cleanup-v1-1-f7e22e603447@kernel.org
Link: https://lore.kernel.org/all/d762c46b-36f0-471a-b5b4-23c8cf5628ae@suse.cz/
Signed-off-by: Vlastimil Babka (SUSE) <vbabka@kernel.org>
Suggested-by: David Hildenbrand (Arm) <david@kernel.org>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Brendan Jackman <jackmanb@google.com>
Cc: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Zi Yan <ziy@nvidia.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>