On 9/11/2026 6:29 PM, Ridong Chen wrote:
From: Ridong Chen <[email protected]> Background ========== MGLRU currently has no tracepoints of its own. The scan and evict paths reuse the classic-LRU tracepoints (trace_mm_vmscan_lru_isolate() and trace_mm_vmscan_lru_shrink_inactive()), which predate MGLRU and carry no generation, sequence, memcg or swappiness context. A trace of a running system therefore cannot tell which memcg a given scan belongs to, how far reclaim has progressed through the generations, or when a new generation is created - so how MGLRU actually operates is effectively invisible. Implementation ============== Patch 1 is a cleanup that factors the per-generation page-count summation into a helper, lru_gen_seq_nr_pages(), reused by patch 3. Patch 2 adds mm_mglru_scan_folios on the scan path, and patch 3 adds mm_mglru_inc_max_seq on the aging path. Both carry the memcg id and the generation window (min_seq/max_seq), live at MGLRU-specific layers with no classic-LRU counterpart, and are guarded so the hot paths stay zero-cost when disabled. Effect ====== Paired, the two tracepoints make the full aging-to-eviction window observable per memcg: aging advances max_seq (the leading edge), scanning consumes the oldest generations, and the min_seq/max_seq pair on each event shows how the generation window moves over time. A sample trace, with the classic-LRU tracepoints left enabled to show how they interleave: mm_vmscan_lru_isolate: classzone=4 order=0 nr_requested=36 nr_scanned=31 nr_skipped=0 nr_taken=29 lru=inactive_file mm_mglru_scan_folios: memcg_id=73 classzone=4 order=0 nr_requested=36 nr_scanned=31 nr_sorted=2 nr_skipped=0 nr_taken=29 lru=inactive_file max_seq=3 tier=3 min_seq=0 mm_vmscan_lru_shrink_inactive: nid=0 nr_scanned=31 nr_reclaimed=29 nr_dirty=0 nr_writeback=0 nr_congested=0 nr_immediate=0 nr_activate_anon=0 nr_activate_file=0 nr_ref_keep=0 nr_unmap_fail=0 priority=5 flags=RECLAIM_WB_FILE|RECLAIM_WB_ASYNC mm_mglru_inc_max_seq: memcg_id=73 max_seq=4 anon_min_seq=1 file_min_seq=1 nr_anon={0x0,0x0,0x1,0x0} nr_file={0x0,0x427,0x6,0x4a} mm_vmscan_lru_isolate: classzone=4 order=0 nr_requested=5 nr_scanned=5 nr_skipped=0 nr_taken=1 lru=inactive_file mm_mglru_scan_folios: memcg_id=73 classzone=4 order=0 nr_requested=5 nr_scanned=5 nr_sorted=4 nr_skipped=0 nr_taken=1 lru=inactive_file max_seq=4 tier=3 min_seq=1 mm_vmscan_lru_shrink_inactive: nid=0 nr_scanned=5 nr_reclaimed=1 nr_dirty=0 nr_writeback=0 nr_congested=0 nr_immediate=0 nr_activate_anon=0 nr_activate_file=0 nr_ref_keep=0 nr_unmap_fail=0 priority=5 flags=RECLAIM_WB_FILE|RECLAIM_WB_ASYNC The mm_mglru_scan_folios lines carry the memcg id and the generation window (max_seq/min_seq) that the bare mm_vmscan_lru_isolate lines above them cannot - those cannot even say which memcg they came from. The mm_mglru_inc_max_seq line then shows a new generation being created (max_seq 3 -> 4) with the per-generation page counts for both types. --- v2: address Sashiko AI review [1] - pass struct lruvec * into the tracepoints and do the memcg lookup inside TP_fast_assign. [1] https://sashiko.dev/#/patchset/[email protected]?part=2 Ridong Chen (3): mm/mglru: factor out lru_gen_seq_nr_pages() mm/mglru: add tracepoint for scan_folios() mm/mglru: add tracepoint for inc_max_seq() include/trace/events/vmscan.h | 101 ++++++++++++++++++++++++++++++++++ mm/vmscan.c | 53 +++++++++++++++--- 2 files changed, 146 insertions(+), 8 deletions(-)
Hi Steven, Baolin and Baoquan, Thanks again for reviewing v2. Following Steven's suggestion in this thread, I first moved the vmscan trace header under mm/ and exposed struct scan_control to it in a separate RFC series. Patches [2/3] and [3/3] of v2 have now been reworked on top of that and posted together with it as one series: [PATCH v4 0/9] mm/vmscan: move tracepoints to mm/ and add MGLRU tracepoints https://lore.kernel.org/all/[email protected]/ The two MGLRU tracepoints are [8/9] and [9/9] there. Compared with v2: - Both events now take only lruvec/sc plus the local counters, and derive everything else (memcg id, order, reclaim_idx, lru, max_seq, min_seq and the per-generation page counts) in TP_fast_assign() (Steven). - Add nid, derived from lruvec_pgdat(lruvec) (Baoquan). - Group the int fields at the end of the entry to avoid padding on 64-bit (Steven). - Drop v2 [1/3] "mm/mglru: factor out lru_gen_seq_nr_pages()": the per-generation page count is now summed inside TP_fast_assign(), so the helper has no user left. Since [8/9] and [9/9] changed substantially, I did not carry over your Reviewed-by tags. I would really appreciate it if you could take another look at v4. -- Best regards Ridong
