f0
[    4.965023] Call Trace:
[    4.965413]  <TASK>
[    4.966021]  ? __kasan_unpoison_range+0x26/0x50
[    4.966759]  alloc_slab_obj_exts_early.constprop.0+0x136/0x240
[    4.967021]  allocate_slab+0x107/0x4b0
[    4.968021]  ___slab_alloc+0x8f6/0xec0
[    4.969021]  ? kstrdup_const+0x2c/0x40
[    4.969615]  ? __xa_alloc+0x227/0x320
[    4.970021]  __slab_alloc.isra.0+0x35/0x90
[    4.970663]  __kmalloc_node_track_caller_noprof+0x4e2/0x7a0
[    4.971021]  ? kstrdup_const+0x2c/0x40
[    4.972021]  kstrdup+0x48/0xf0
[    4.972505]  ? kstrdup+0x48/0xf0
[    4.973021]  kstrdup_const+0x2c/0x40
[    4.973589]  alloc_vfsmnt+0xd5/0x680
[    4.974021]  vfs_create_mount.part.0+0x42/0x3e0
[    4.975021]  vfs_kern_mount.part.0+0x10c/0x150
[    4.975722]  vfs_kern_mount+0x13/0x40
[    4.976021]  devtmpfs_init+0xa8/0x430
[    4.977021]  ? __percpu_counter_init_many+0x199/0x360
[    4.977812]  ? __pfx_devtmpfs_init+0x10/0x10
[    4.978021]  ? page_offline_thaw+0x5/0x20
[    4.979021]  ? __kasan_check_write+0x14/0x30
[    4.979694]  driver_init+0x1a/0x60
[    4.980021]  kernel_init_freeable+0x7de/0xeb0
[    4.981021]  ? __pfx_kernel_init+0x10/0x10
[    4.981667]  kernel_init+0x1f/0x220
[    4.982021]  ? __pfx_kernel_init+0x10/0x10
[    4.983021]  ret_from_fork+0x2b8/0x3b0
[    4.983618]  ? __pfx_kernel_init+0x10/0x10
[    4.984021]  ret_from_fork_asm+0x1a/0x30
[    4.984639] RIP: 2e66:0x0
[    4.985021] Code: Unable to access opcode bytes at 0xffffffffffffffd6.
[    4.986021] RSP: 0084:0000000000000000 EFLAGS: 841f0f2e660000 ORIG_RAX: 2e66000000000084
[    4.987021] RAX: 0000000000000000 RBX: 2e66000000000084 RCX: 0000000000841f0f
[    4.988021] RDX: 000000841f0f2e66 RSI: 00841f0f2e660000 RDI: 1f0f2e6600000000
[    4.989021] RBP: 1f0f2e6600000000 R08: 1f0f2e6600000000 R09: 00841f0f2e660000
[    4.990021] R10: 000000841f0f2e66 R11: 0000000000841f0f R12: 00841f0f2e660000
[    4.991021] R13: 000000841f0f2e66 R14: 0000000000841f0f R15: 2e66000000000084
[    4.992022]  </TASK>
[    4.992372] ---[ end trace 0000000000000000 ]---

This warning is from kasan_unpoison():
	if (WARN_ON((unsigned long)addr & KASAN_GRANULE_MASK))
                return;

on x86_64, the address passed to kasan_{poison,unpoison}() should be at
least aligned with 8 bytes.

After manual investigation it turns out when the SLAB_STORE_USER flag is
specified, any metadata after the original kmalloc request size is
misaligned.

Questions:
- Could it cause any issues other than the one described above?
- Does KASAN even support architectures that have issues with unaligned
  accesses?
- How come we haven't seen any issues regarding this so far? :/
-- 
Cheers,
Harry / Hyeonggon[PATCH] mm/slab: ensure all metadata in slab object are word-alignedHarry Yoo <harry.yoo@oracle.com> undefinedVlastimil Babka <vbabka@suse.cz> undefined undefined undefined undefined undefined undefined undefined undefined undefined undefined undefined undefined undefined undefined undefined…kƒÂo