On Sat, May 25, 2024 at 10:24 AM Alan Stern <stern@xxxxxxxxxxxxxxxxxxx> wrote: > If alauda_init_media() returns an error, leave info->initialized > unchanged. alauda_check_media() will return an error also, so the bad > division won't take place. > > Alan Stern Thanks! You also remind me that the return value from alauda_init_media() is never used! By this way, the workflow now seems to work correctly. It tries to initialize multiple times, and finally disconnects due to no response. Now if possible, I will post a [PATCH v4] for this bug soon. I want to know whether it is possible to add some tags like Suggested-by or Reviewed-by for Dear Alan Stern, gregkh and oneukum as thanks for your discussions. e.g. ==== kernel log [ 47.266129][ T4125] alauda_check_media: before init_media, status[0]: 0000000000000036 [ 47.266555][ T4125] alauda_check_media: enter init_media [ 47.467314][ T9] usb 1-1: USB disconnect, device number 2 root@syzkaller:~# [ 47.485304][ T4125] alauda_get_media_status: data=54, rc=4 [ 47.485640][ T4125] alauda_init_media: exit in 391 [ 47.486104][ T41] sd 2:0:0:0: [sdb] Read Capacity(10) failed: Result: hostbyte=DID_ERROR driverbyte=DRIVER_OK [ 47.486591][ T41] sd 2:0:0:0: [sdb] Sense not available. [ 47.486889][ T41] sd 2:0:0:0: [sdb] 0 512-byte logical blocks: (0 B/0 B) [ 47.487212][ T41] sd 2:0:0:0: [sdb] 0-byte physical blocks [ 47.487515][ T41] sd 2:0:0:0: [sdb] Write Protect is off [ 47.487813][ T41] sd 2:0:0:0: [sdb] Asking for cache data failed [ 47.488104][ T41] sd 2:0:0:0: [sdb] Assuming drive cache: write through [ 47.491396][ T41] sd 2:0:0:0: [sdb] Attached SCSI removable disk [ 48.105309][ T1198] not responding...
