Hello linux community expert: >> Added null pointer check to avoid system crash. >> >> Unable to handle kernel NULL pointer dereference at >> virtual address 00000000000001a8 >> pc : gs_read_complete+0x58/0x240 >> lr : usb_gadget_giveback_request+0x40/0x160 >> sp : ffffffc00f1539c0 >> x29: ffffffc00f1539c0 x28: ffffff8002a30000 x27: 0000000000000000 >> x26: ffffff8002a30000 x25: 0000000000000000 x24: ffffff8002a30000 >> x23: ffffff8002ff9a70 x22: ffffff898e7a7b00 x21: ffffff803c9af9d8 >> x20: ffffff898e7a7b00 x19: 00000000000001a8 x18: ffffffc0099fd098 >> x17: 0000000000001000 x16: 0000000080000000 x15: 0000000ac1200000 >> x14: 0000000000000003 x13: 000000000000d5e8 x12: 0000000355c314ac >> x11: 0000000000000015 x10: 0000000000000012 x9 : 0000000000000008 >> x8 : 0000000000000000 x7 : 0000000000000000 x6 : ffffff887cd12000 >> x5 : 0000000000000002 x4 : ffffffc00f9b07f0 x3 : ffffffc00f1538d0 >> x2 : 0000000000000001 x1 : 0000000000000000 x0 : 00000000000001a8 >> Call trace: >> gs_read_complete+0x58/0x240 >> usb_gadget_giveback_request+0x40/0x160 >> dwc3_remove_requests+0x170/0x484 >> dwc3_ep0_out_start+0xb0/0x1d4 >> __dwc3_gadget_start+0x25c/0x720 >> kretprobe_trampoline.cfi_jt+0x0/0x8 >> kretprobe_trampoline.cfi_jt+0x0/0x8 >> udc_bind_to_driver+0x1d8/0x300 >> usb_gadget_probe_driver+0xa8/0x1dc >> gadget_dev_desc_UDC_store+0x13c/0x188 >> configfs_write_iter+0x160/0x1f4 >> vfs_write+0x2d0/0x40c >> ksys_write+0x7c/0xf0 >> __arm64_sys_write+0x20/0x30 >> invoke_syscall+0x60/0x150 >> el0_svc_common+0x8c/0xf8 >> do_el0_svc+0x28/0xa0 >> el0_svc+0x24/0x84 >> el0t_64_sync_handler+0x88/0xec >> el0t_64_sync+0x1b4/0x1b8 >> Code: aa1f03e1 aa1303e0 52800022 2a0103e8 (88e87e62) >> ---[ end trace 938847327a739172 ]--- >> Kernel panic - not syncing: Oops: Fatal exception >> >> Signed-off-by: Lianqin Hu <hulianqin@xxxxxxxx> >> --- >> v1: > - Optimize code comments, delete log printing >Also, same comment as before, what: > - causes port to be NULL When unbind, the gser_unbind function call gserial_disconnect to set gser->ioport = NULL, and struct gs_port *port = gser->ioport, resulting in port being NULL. It seems that there is a potential competition between the gs_read_complete function and the gserial_disconnect function. > - prevents it from changing to NULL right after you check it I've seen other functions, such as gserial_suspend and gserial_resume, which all perform port empty detection operations. >Both need to be answered before we can take this change. Thanks
