vkcube crashed with BASE_JD_EVENT_DATA_INVALID_FAULT

Hi,

When I run vkcube(https://github.com/KhronosGroup/Vulkan-Tools/blob/main/cube/cube.c ) based on malir48p0 with vulkan wsi layer(https://gitlab.freedesktop.org/mesa/vulkan-wsi-layer), I met below crash after several successful vkQueuePresentKHR().

command_queue = 0x1c48000
Got base event code: 0x58 , sw_type: 0x0, atom number 2
[ 2758.834483] (mali-cmar-backe) GPU addr: 0xb2b84300, jobtype=4
Exception type+stat: 0x00000001
Task idx: 0x00000000
Fault addr: 0x0000000000000000
Flags and indicies: 0x37ec000100010009
Next Job: 0xb2b87a00
7:0x00000000
8:0x00000000
9:0x20000000
10:0x08000000
11:0x00000000
12:0x00000000
13:0x00000000
14:0x00000000
15:0x00000000
16:0x00000002
17:0x00000000
18:0x00000000
19:0x00000000
20:0x00000000
21:0x00000000
22:0xb2b842b0
23:0x00000000
24:0x00000000
25:0x00000000
26:0x00000000
27:0x00000000
28:0xb2b843c0
29:0x00000000
30:0xb6193000
31:0x00000000
32:0x00000000
33:0x00000000
34:0x00000000
35:0x00000000
36:0x00000000
37:0x00000000
38:0x00000000
39:0x00000000
40:0x00000000
41:0x00000000
42:0x00000000
43:0x00000000
44:0xb2b842c0
45:0x00000000
46:0x00000000
47:0x00000000

[ 2758.834755] (mali-cmar-backe) GPU addr: 0xb2b87a00, jobtype=4
Exception type+stat: 0x00000001
Task idx: 0x00000000
Fault addr: 0x0000000000000000
Flags and indicies: 0x37ed000137ed0009
Next Job: 0xb2b84b00
7:0x00000000
8:0x00000000
9:0x20000000
10:0x08000000
11:0x00000000
12:0x00000000
13:0x00000000
14:0x00000000
15:0x00000000
16:0x00000002
17:0x00000000
18:0x00000000
19:0x00000000
20:0x00000000
21:0x00000000
22:0xb2b87970
23:0x00000000
24:0x00000000
25:0x00000000
26:0x00000000
27:0x00000000
28:0xb2b87b00
29:0x00000000
30:0xb2b87ac0
31:0x00000000
32:0x00000000
33:0x00000000
34:0x00000000
35:0x00000000
36:0x00000000
37:0x00000000
38:0x00000000
39:0x00000000
40:0x00000000
41:0x00000000
42:0x00000000
43:0x00000000
44:0xb2b83c00
45:0x00000000
46:0x00000000
47:0x00000000

[ 2758.835017] (mali-cmar-backe) GPU addr: 0xb2b84b00, jobtype=1
Exception type+stat: 0x00000001
Task idx: 0x00000000
Fault addr: 0x0000000000000000
Flags and indicies: 0x000037ed00030003
Next Job: 0xb2b87f40
7:0x00000000

[ 2758.835292] (mali-cmar-backe) GPU addr: 0xb2b87f40, jobtype=10
Exception type+stat: 0x00000000
Task idx: 0x00000000
Fault addr: 0x0000000000000000
Flags and indicies: 0x0008000100080015
Next Job: 0x00000000
7:0x00000000
8:0x00000000
9:0x20000000
10:0x18078008
11:0x00000000
12:0x00000000
13:0x00000000
14:0x00000000
15:0x00000000
16:0x3f800000
17:0x00000000
18:0xb2b83e80
19:0x00000000
20:0x00000000
21:0x00000000
22:0x00000000
23:0x00000000
24:0x00000000
25:0x00000000
26:0x00000000
27:0x00000000
28:0x00000000
29:0x00000000
30:0x00000000
31:0x00000000
32:0x9f9f00a3
33:0x00000000
34:0x00000000
35:0x00000000
36:0x00bd9340
37:0x00000001
38:0xb2b87ea0
39:0x00000000
40:0xb2b56100
41:0x00000000
42:0xb2b560c0
43:0x00000000
44:0x00000000
45:0x00000000
46:0xb2b570c0
47:0x00000000
48:0x00000000
49:0x00000000
50:0x00000000
51:0x00000000
52:0xb2b87e80
53:0x00000000
54:0xb2b57098
55:0x00000000
56:0xb2b87d20
57:0x00000000
58:0x00000000
59:0x00000000
60:0xb2b83c00
61:0x00000000
62:0x00000000
63:0x00000000
64:0x009f0003
65:0x00000000
66:0x00000000
67:0x00000000
68:0x00000000
69:0x00000000
70:0xb2b87ea0
71:0x00000000
72:0xb2b56100
73:0x00000000
74:0xb2b560c0
75:0x00000000
76:0xb2b880d0
77:0x00000000
78:0xb2b57040
79:0x00000000
80:0x00000000
81:0x00000000
82:0x00000000
83:0x00000000
84:0xb2b87e80
85:0x00000000
86:0xb2b57080
87:0x00000000
88:0x00000000
89:0x00000000
90:0x00000000
91:0x00000000
92:0xb2b83c00
93:0x00000000
94:0x00000000
95:0x00000000

BASE_JD_SW_EVENT_JOB event, udata->command != NULL, sending WORK_COMPLETED signal
Released atom id = 2
CMARP_SIGNAL_WORK_COMPLETED calling terminator
event_job c = 0x1da9000 , event->event_code = 0x58, atom_number 2

The same libs can work on r23p0, my question is, how this error happen? If there is data fault in command buffer, why it happen after some successful vkQueuePresentKHR(I can see the cube, just frozen).