On Fri 25-07-25 10:16:54, Zhang Yi wrote:
> From: Zhang Yi <yi.zhang@xxxxxxxxxx>
>
> mb_set_largest_free_order() requires the parameter bb_largest_free_order
> and the list bb_largest_free_order_node to be initialized, and
> mb_update_avg_fragment_size() requires the parameter
> bb_avg_fragment_size_order and bb_avg_fragment_size_node to be
> initialized too. But the test_mb_mark_used kunit tests do not init these
> parameters, and trigger the following crash issue.
>
> Pid: 35, comm: kunit_try_catch Tainted: G W N 6.16.0-rc4-00031-gbbe11dd13a3f-dirty
> RIP: 0033:mb_set_largest_free_order+0x5c/0xc0
> RSP: 00000000a0883d98 EFLAGS: 00010206
> RAX: 0000000060aeaa28 RBX: 0000000060a2d400 RCX: 0000000000000008
> RDX: 0000000060aea9c0 RSI: 0000000000000000 RDI: 0000000060864000
> RBP: 0000000060aea9c0 R08: 0000000000000000 R09: 0000000060a2d400
> R10: 0000000000000400 R11: 0000000060a9cc00 R12: 0000000000000006
> R13: 0000000000000400 R14: 0000000000000305 R15: 0000000000000000
> Kernel panic - not syncing: Segfault with no mm
> CPU: 0 UID: 0 PID: 35 Comm: kunit_try_catch Tainted: G W N 6.16.0-rc4-00031-gbbe11dd13a3f-dirty #36 NONE
> Tainted: [W]=WARN, [N]=TEST
> Stack:
> 60210c60 00000200 60a9e400 00000400
> 40060300280 60864000 60a9cc00 60a2d400
> 00000400 60aea9c0 60a9cc00 60aea9c0
> Call Trace:
> [<60210c60>] ? ext4_mb_generate_buddy+0x1f0/0x230
> [<60215c3b>] ? test_mb_mark_used+0x28b/0x4e0
> [<601df5bc>] ? ext4_get_group_desc+0xbc/0x150
> [<600bf1c0>] ? ktime_get_ts64+0x0/0x190
> [<60086370>] ? to_kthread+0x0/0x40
> [<602b559b>] ? kunit_try_run_case+0x7b/0x100
> [<60086370>] ? to_kthread+0x0/0x40
> [<602b7850>] ? kunit_generic_run_threadfn_adapter+0x0/0x30
> [<602b7862>] ? kunit_generic_run_threadfn_adapter+0x12/0x30
> [<60086a51>] ? kthread+0xf1/0x250
> [<6004a541>] ? new_thread_handler+0x41/0x60
> [ERROR] Test: test_mb_mark_used: 0 tests run!
>
> Fixes: bbe11dd13a3f ("ext4: fix largest free orders lists corruption on mb_optimize_scan switch")
> Reported-by: Theodore Ts'o <tytso@xxxxxxx>
> Closes: https://lore.kernel.org/linux-ext4/20250724145437.GD80823@xxxxxxx/
> Signed-off-by: Zhang Yi <yi.zhang@xxxxxxxxxx>
> ---
> This patch applies to the kernel that has only merged bbe11dd13a3f
> ("ext4: fix largest free orders lists corruption on mb_optimize_scan
> switch"), but not merged 458bfb991155 ("ext4: convert free groups order
> lists to xarrays").
Hum, I think it would be best to just squash this into bbe11dd13a3f and
then just rebase & squash the other unittest fixup to the final commit when
we have to rebase anyway. Because otherwise backports to stable kernel will
quickly become rather messy.
Honza
> fs/ext4/mballoc-test.c | 8 ++++++++
> 1 file changed, 8 insertions(+)
>
> diff --git a/fs/ext4/mballoc-test.c b/fs/ext4/mballoc-test.c
> index d634c12f1984..ba939be0ec55 100644
> --- a/fs/ext4/mballoc-test.c
> +++ b/fs/ext4/mballoc-test.c
> @@ -802,6 +802,10 @@ static void test_mb_mark_used(struct kunit *test)
> KUNIT_ASSERT_EQ(test, ret, 0);
>
> grp->bb_free = EXT4_CLUSTERS_PER_GROUP(sb);
> + grp->bb_largest_free_order = -1;
> + grp->bb_avg_fragment_size_order = -1;
> + INIT_LIST_HEAD(&grp->bb_largest_free_order_node);
> + INIT_LIST_HEAD(&grp->bb_avg_fragment_size_node);
> mbt_generate_test_ranges(sb, ranges, TEST_RANGE_COUNT);
> for (i = 0; i < TEST_RANGE_COUNT; i++)
> test_mb_mark_used_range(test, &e4b, ranges[i].start,
> @@ -875,6 +879,10 @@ static void test_mb_free_blocks(struct kunit *test)
> ext4_unlock_group(sb, TEST_GOAL_GROUP);
>
> grp->bb_free = 0;
> + grp->bb_largest_free_order = -1;
> + grp->bb_avg_fragment_size_order = -1;
> + INIT_LIST_HEAD(&grp->bb_largest_free_order_node);
> + INIT_LIST_HEAD(&grp->bb_avg_fragment_size_node);
> memset(bitmap, 0xff, sb->s_blocksize);
>
> mbt_generate_test_ranges(sb, ranges, TEST_RANGE_COUNT);
> --
> 2.46.1
>
--
Jan Kara <jack@xxxxxxxx>
SUSE Labs, CR
