[PATCH RFC v2 0/7] mm/vmscan: move vmscan tracepoints to a local header

[Ridong Chen <ridong.chen@linux.dev>]

From: Ridong Chen <chenridong@xiaomi.com>

This series started as an effort to add two tracepoints for MGLRU. In the
discussion, Steven suggested moving include/trace/events/vmscan.h to
mm/trace_vmscan.h [1], as several other subsystems already do. Once the
header lives under mm/, struct scan_control can be moved there too, letting
the trace events access its fields directly instead of having each callsite
copy the scalars out by hand.

The series builds in small steps:

  1) drop the now-unused vmscan trace include from memcontrol.c;
  2) move include/trace/events/vmscan.h to mm/trace_vmscan.h, using a
     relative TRACE_INCLUDE_PATH so no Makefile change is needed;
  3) move struct scan_control into a dedicated mm/vmscan.h, which
     trace_vmscan.h pulls in, guarded against the multi-read that
     define_trace.h performs;
  4-7) pass scan_control to the vmscan tracepoints and pick the fields
     out in TP_fast_assign, one group at a time: the reclaim-begin
     events, the LRU isolate/shrink events, mm_vmscan_reclaim_pages and
     mm_vmscan_balance_pgdat_end.

Besides the cleanup, this moves the code that computes the tracepoint
parameters into TP_fast_assign(), which is in a separate text section. It
removes code from the work flow, improving instruction cache.

There is no functional change and no ABI change: TP_STRUCT__entry and
TP_printk are untouched, so the exported event format is identical. This
was confirmed by diffing the tracefs format files before and after, and
by capturing the direct/memcg/node reclaim-begin events under load in a
QEMU guest (order, gfp_flags and memcg_id all match).

Where a value passed to a tracepoint is a local of the caller rather than
a scan_control field (highest_zoneidx for balance_pgdat_end, nr_reclaimed
for reclaim_pages), it stays a separate argument.

[1]: https://lore.kernel.org/linux-mm/20260916095122.50cbb620@robin/

---
Changes since v1:
  - Convert the remaining vmscan tracepoints too, not just reclaim-begin
    (4 -> 7 patches).
  - Put struct scan_control in its own mm/vmscan.h instead of directly in
    trace_vmscan.h suggested by Baoquan.

Ridong Chen (7):
  mm/memcontrol: drop unused vmscan tracepoint include
  mm/vmscan: move vmscan tracepoints to a local header
  mm/vmscan: move struct scan_control to a dedicated header
  mm/vmscan: pass scan_control to the reclaim-begin tracepoints
  mm/vmscan: pass scan_control to the LRU isolate/shrink tracepoints
  mm/vmscan: pass scan_control to mm_vmscan_reclaim_pages
  mm/vmscan: pass scan_control to mm_vmscan_balance_pgdat_end

 MAINTAINERS                                   |   4 +
 mm/memcontrol.c                               |   2 -
 mm/shrinker.c                                 |   2 +-
 .../events/vmscan.h => mm/trace_vmscan.h      |  80 ++++++-----
 mm/vmscan.c                                   | 136 ++----------------
 mm/vmscan.h                                   | 118 +++++++++++++++
 6 files changed, 178 insertions(+), 164 deletions(-)
 rename include/trace/events/vmscan.h => mm/trace_vmscan.h (90%)
 create mode 100644 mm/vmscan.h

-- 
2.34.1