On 2024/4/2 22:20, Oscar Salvador wrote:
On Fri, Mar 29, 2024 at 12:54:30PM +0800, Kefeng Wang wrote:I still see the following memory leak, could you check it? /mnt/arm64 # cat /sys//kernel/debug/kmemleak unreferenced object 0xffff000200d91000 (size 16): comm "kworker/4:0", pid 42, jiffies 4294892753 hex dump (first 16 bytes): 60 37 57 c1 00 00 ff ff 00 00 00 00 00 00 00 00 `7W............. backtrace (crc 4458f477): [<(____ptrval____)>] kmemleak_alloc+0x3c/0x50 [<(____ptrval____)>] kmalloc_trace+0x20c/0x2e0 [<(____ptrval____)>] __set_page_owner+0x1d0/0x2a0 [<(____ptrval____)>] prep_new_page+0x108/0x138 [<(____ptrval____)>] get_page_from_freelist+0x79c/0x16b8 [<(____ptrval____)>] __alloc_pages+0x1bc/0x440 [<(____ptrval____)>] new_slab+0x104/0x3c8 [<(____ptrval____)>] ___slab_alloc+0x368/0xb20 [<(____ptrval____)>] __slab_alloc.isra.0+0x3c/0x88 [<(____ptrval____)>] kmalloc_trace+0x280/0x2e0 [<(____ptrval____)>] __set_page_owner+0x1d0/0x2a0 [<(____ptrval____)>] prep_new_page+0x108/0x138 [<(____ptrval____)>] get_page_from_freelist+0x79c/0x16b8 [<(____ptrval____)>] __alloc_pages+0x1bc/0x440 [<(____ptrval____)>] new_slab+0x104/0x3c8 [<(____ptrval____)>] ___slab_alloc+0x368/0xb20Hi Kefeng Wang You seem to be missing [1] Could you try with that patch applied? [1] https://lore.kernel.org/linux-mm/20240315222610.6870-1-osalvador@xxxxxxx/
Hi, re-run with clean v6.9-rc2 (already include 7844c0147211 mm,page_owner: fix recursion), after booting and wait some time, the kmemleak is still occurred.
Thanks
