>
> Hello, Paul
>
> when running the rcutorture test with rcutorture.object_debug=1 in
> PREEMPT_RT kernel.
>
> BUG: sleeping function called from invalid context at
> kernel/locking/spinlock_rt.c:48
> [ 3.116503] in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid:
> 1, name: swapper/0
> [ 3.116503] preempt_count: 1, expected: 0
> [ 3.116503] RCU nest depth: 1, expected: 1
> [ 3.116503] 3 locks held by swapper/0/1:
> [ 3.116503] #0: ffffffff8f862d80 (fullstop_mutex){+.+.}-{4:4}, at:
> torture_init_begin+0x24/0xa0
> [ 3.116503] #1: ffffffff8f7b1400 (rcu_read_lock){....}-{1:3}, at:
> rcu_torture_init+0x20b9/0x2380
> [ 3.116503] #2: ffffffff8f89fa20 (vmap_area_lock){+.+.}-{3:3}, at:
> find_vmap_area+0x21/0x90
> [ 3.116503] irq event stamp: 571506
> [ 3.116503] hardirqs last enabled at (571505):
> [<ffffffff8d422ff5>] __call_rcu_common+0x3a5/0xdc0
> [ 3.116503] hardirqs last disabled at (571506):
> [<ffffffff900487f7>] rcu_torture_init+0x22b7/0x2380
> [ 3.116503] softirqs last enabled at (406916):
> [<ffffffff8d2e91f2>] __local_bh_enable_ip+0x132/0x170
> [ 3.116503] softirqs last disabled at (406910):
> [<ffffffff8e658e79>] inet_register_protosw+0x9/0x190
> [ 3.116503] Preemption disabled at:
> [ 3.116503] [<ffffffff9004865a>] rcu_torture_init+0x211a/0x2380
> [ 3.116503] CPU: 2 PID: 1 Comm: swapper/0 Tainted: G W
> 6.5.0-rc4-rt2-yocto-preempt-rt+ #7
> [ 3.116503] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009),
> BIOS rel-1.16.2-0-gea1b7a073390-prebuilt.qemu.org 04/01/2014
> [ 3.116503] Call Trace:
> [ 3.116503] <TASK>
> [ 3.116503] dump_stack_lvl+0x68/0xb0
> [ 3.116503] dump_stack+0x14/0x20
> [ 3.116503] __might_resched+0x221/0x2f0
> [ 3.116503] rt_spin_lock+0x60/0x130
> [ 3.116503] ? find_vmap_area+0x21/0x90
> [ 3.116503] find_vmap_area+0x21/0x90
> [ 3.116503] vmalloc_dump_obj+0x25/0xa0
> [ 3.116503] mem_dump_obj+0x22/0x90
> [ 3.116503] __call_rcu_common+0x920/0xdc0
> [ 3.116503] ? __pfx_rcu_torture_err_cb+0x10/0x10
> [ 3.116503] ? __pfx___call_rcu_common+0x10/0x10
> [ 3.116503] ? rcu_torture_init+0x22b7/0x2380
> [ 3.116503] ? debug_smp_processor_id+0x1b/0x30
> [ 3.116503] ? rcu_is_watching+0x3e/0xc0
> [ 3.116503] call_rcu_hurry+0x14/0x20
> [ 3.116503] rcu_torture_init+0x2186/0x2380
> [ 3.116503] ? __pfx_rcu_torture_init+0x10/0x10
> [ 3.116503] ? __pfx_rt_mutex_slowunlock+0x10/0x10
> [ 3.116503] ? __pfx_irq_debugfs_init+0x10/0x10
> [ 3.116503] ? __pfx_rcu_torture_leak_cb+0x10/0x10
> [ 3.116503] ? rt_spin_unlock+0x50/0xa0
> [ 3.116503] ? add_device_randomness+0xb1/0xe0
> [ 3.116503] ? __pfx_rcu_torture_leak_cb+0x10/0x10
> [ 3.116503] ? __pfx_rcu_torture_init+0x10/0x10
> [ 3.116503] do_one_initcall+0xd8/0x430
> [ 3.116503] ? __pfx_do_one_initcall+0x10/0x10
> [ 3.116503] ? __kasan_check_read+0x15/0x20
> [ 3.116503] kernel_init_freeable+0x3a0/0x670
> [ 3.116503] ? rt_spin_unlock+0x50/0xa0
> [ 3.116503] ? __pfx_kernel_init+0x10/0x10
> [ 3.116503] kernel_init+0x23/0x160
> [ 3.116503] ? __pfx_kernel_init+0x10/0x10
> [ 3.116503] ret_from_fork+0x40/0x60
> [ 3.116503] ? __pfx_kernel_init+0x10/0x10
> [ 3.116503] ret_from_fork_asm+0x1b/0x30
> [ 3.116503] </TASK>
> [ 3.116503] non-slab/vmalloc memory
>
>
> as described by Lei Zhen:
>
> v2 --> v3:
> 1. I made statistics about the source of 'rhp'. kmem_valid_obj() accounts for
> more than 97.5%, and vmalloc accounts for less than 1%. So change call
> mem_dump_obj() to call kmem_dump_obj() can meet debugging requirements and
> avoid the potential deadlock risk of vmalloc_dump_obj().
> - mem_dump_obj(rhp);
> + if (kmem_valid_obj(rhp))
> + kmem_dump_obj(rhp);
>
> The discussion about vmap_area_lock deadlock in v2:
> https://lkml.org/lkml/2022/11/11/493
>
> so whether mem_dump_obj() should also be replaced by kmem_dump_obj() in
> __call_rcu_common(), since the call_rcu() can be called in interrupt context,
> this may issue recursive spin vmap_area_lock lock calls.
>
> Any thoughts ?
or diff --git a/mm/util.c b/mm/util.c
index ddfbb22dc187..fbc007adc108 100644
--- a/mm/util.c
+++ b/mm/util.c
@@ -1066,10 +1066,9 @@ void mem_dump_obj(void *object)
if (kmem_dump_obj(object))
return;
- if (vmalloc_dump_obj(object))
- return;
-
- if (virt_addr_valid(object))
+ if (is_vmalloc_addr(object))
+ type = "vmalloc memory";
+ else if (virt_addr_valid(object))
type = "non-slab/vmalloc memory";
else if (object == NULL)
type = "NULL pointer";
Thanks
Zqiang
>
> Thanks
> Zqiang
