seduction
 Language:
Welcome, Guest. Please login or register.
Did you miss your activation email?
2020/05/31, 00:59:24


Help

Author [EN] [PL] [ES] [PT] [IT] [DE] [FR] [NL] [TR] [SR] [AR] [RU] Topic: (solved) Terrible slow usb 3.0 ssd  (Read 11722 times)

0 Members and 1 Guest are viewing this topic.

Offline orinoco

  • User
  • Posts: 172
(solved) Terrible slow usb 3.0 ssd
« on: 2015/03/11, 23:55:31 »
Hello,

I bought an external ssd with usb 3.0.

Code: [Select]
Mär 11 23:29:00 jake kernel: usb 3-2: new SuperSpeed USB device number 5 using xhci_hcd
Mär 11 23:29:00 jake kernel: usb 3-2: New USB device found, idVendor=174c, idProduct=07d1
Mär 11 23:29:00 jake kernel: usb 3-2: New USB device strings: Mfr=2, Product=3, SerialNumber=1
Mär 11 23:29:00 jake kernel: usb 3-2: Product: TS512GESD400K
Mär 11 23:29:00 jake kernel: usb 3-2: Manufacturer: StoreJet
Mär 11 23:29:00 jake kernel: usb 3-2: SerialNumber: xxxxxxxxxxxxxxxxxxxx
Mär 11 23:29:00 jake kernel: scsi host12: uas
Mär 11 23:29:00 jake kernel: blk-mq: reduced tag depth to 10240
Mär 11 23:29:00 jake kernel: ^[[1;39mscsi 12:0:0:0: Direct-Access     StoreJet TS512GESD400K    0    PQ: 0 ANSI: 6
Mär 11 23:29:00 jake kernel: ^[[1;39msd 12:0:0:0: Attached scsi generic sg8 type 0
Mär 11 23:29:00 jake mtp-probe[8869]: checking bus 3, device 5: "/sys/devices/pci0000:00/0000:00:05.0/0000:02:00.0/usb3/3-2"
Mär 11 23:29:00 jake mtp-probe[8869]: bus: 3, device: 5 was not an MTP device
Mär 11 23:29:05 jake kernel: ^[[1;39msd 12:0:0:0: [sdg] 1000215216 512-byte logical blocks: (512 GB/476 GiB)
Mär 11 23:29:05 jake kernel: ^[[1;39msd 12:0:0:0: [sdg] Write Protect is off
Mär 11 23:29:05 jake kernel: sd 12:0:0:0: [sdg] Mode Sense: 43 00 00 00
Mär 11 23:29:05 jake kernel: ^[[1;39msd 12:0:0:0: [sdg] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
Mär 11 23:29:05 jake kernel:  sdg: sdg1
Mär 11 23:29:05 jake kernel: ^[[1;39msd 12:0:0:0: [sdg] Attached SCSI disk

Mounting works fine

Code: [Select]
Mär 11 23:34:44 jake kernel: EXT4-fs (sdg1): mounted filesystem with ordered data mode. Opts: (null)
Mär 11 23:34:44 jake udisksd[32266]: Mounted /dev/sdg1 at /media/FLASH_SSD on behalf of uid 1000

My computer has usb 3.0 onboard.

Code: [Select]
02:00.0 USB controller: NEC Corporation uPD720200 USB 3.0 Host Controller (rev 03) (prog-if 30 [XHCI])
        Subsystem: Gigabyte Technology Co., Ltd Device 5007
        Flags: bus master, fast devsel, latency 0, IRQ 17
        Memory at fdbfe000 (64-bit, non-prefetchable) [size=8K]
        Capabilities: [50] Power Management version 3
        Capabilities: [70] MSI: Enable- Count=1/8 Maskable- 64bit+
        Capabilities: [90] MSI-X: Enable+ Count=8 Masked-
        Capabilities: [a0] Express Endpoint, MSI 00
        Capabilities: [100] Advanced Error Reporting
        Capabilities: [140] Device Serial Number ff-ff-ff-ff-ff-ff-ff-ff
        Capabilities: [150] Latency Tolerance Reporting
        Kernel driver in use: xhci_hcd

If I copy some large files transferspeed seems to be okay (>160MB/s). After some seconds the copy process stops.

Code: [Select]
Mär 11 23:40:39 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 1 inflight: CMD IN
Mär 11 23:40:39 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:40:39 jake kernel: ATA command pass through(12)/Blank: a1 08 2e 00 01 00 00 00 00 ec 00 00
Mär 11 23:41:35 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 24 inflight: CMD OUT
Mär 11 23:41:35 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:41:35 jake kernel: Write(10): 2a 00 1d c4 0a 00 00 00 48 00
Mär 11 23:41:35 jake kernel: scsi host12: uas_eh_bus_reset_handler start
Mär 11 23:41:35 jake kernel: sd 12:0:0:0: [sdg] uas_zap_pending 0 uas-tag 23 inflight: CMD
Mär 11 23:41:35 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:41:35 jake kernel: Write(10): 2a 00 03 62 e4 00 00 00 a0 00
Mär 11 23:41:35 jake kernel: usb 3-2: reset SuperSpeed USB device number 5 using xhci_hcd
Mär 11 23:41:35 jake kernel: xhci_hcd 0000:02:00.0: xHCI xhci_drop_endpoint called with disabled ep ffff88041aa37800
Mär 11 23:41:35 jake kernel: xhci_hcd 0000:02:00.0: xHCI xhci_drop_endpoint called with disabled ep ffff88041aa37848
Mär 11 23:41:35 jake kernel: xhci_hcd 0000:02:00.0: xHCI xhci_drop_endpoint called with disabled ep ffff88041aa37890
Mär 11 23:41:35 jake kernel: xhci_hcd 0000:02:00.0: xHCI xhci_drop_endpoint called with disabled ep ffff88041aa378d8
Mär 11 23:41:35 jake kernel: scsi host12: uas_eh_bus_reset_handler success

After two minutes the copy process gets active again.

Code: [Select]
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 14 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f 78 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 13 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f 74 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 15 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f 7c 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 16 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f 80 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 17 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f 84 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 2 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f 88 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 18 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f 8c 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 19 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f 90 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 20 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f 94 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 21 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f 98 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 22 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f 9c 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 23 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f a0 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 24 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f a4 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 25 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f a8 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 26 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f ac 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 27 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f b0 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 28 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f b4 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 29 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f b8 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 3 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f bc 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 4 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f c0 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 5 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f c4 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 6 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f c8 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 7 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f cc 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 1 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f d0 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 8 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f d4 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 9 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f d8 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 10 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f dc 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 11 inflight: CMD OUT
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Write(10): 2a 00 03 7f e0 00 00 04 00 00
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 12 inflight: CMD IN
Mär 11 23:43:20 jake kernel: sd 12:0:0:0: [sdg] CDB:
Mär 11 23:43:20 jake kernel: Read(10): 28 00 03 80 08 00 00 00 08 00
Mär 11 23:43:20 jake kernel: scsi host12: uas_eh_bus_reset_handler start
Mär 11 23:43:20 jake kernel: usb 3-2: reset SuperSpeed USB device number 5 using xhci_hcd
Mär 11 23:43:20 jake kernel: xhci_hcd 0000:02:00.0: xHCI xhci_drop_endpoint called with disabled ep ffff88041aa37800
Mär 11 23:43:20 jake kernel: xhci_hcd 0000:02:00.0: xHCI xhci_drop_endpoint called with disabled ep ffff88041aa37848
Mär 11 23:43:20 jake kernel: xhci_hcd 0000:02:00.0: xHCI xhci_drop_endpoint called with disabled ep ffff88041aa37890
Mär 11 23:43:20 jake kernel: xhci_hcd 0000:02:00.0: xHCI xhci_drop_endpoint called with disabled ep ffff88041aa378d8
Mär 11 23:43:20 jake kernel: scsi host12: uas_eh_bus_reset_handler success

Again after some seconds the copy process stops again and the procedure repeats again.

Code: [Select]
Linux jake 3.19.1-towo.1-siduction-amd64 #1 SMP PREEMPT siduction 3.19-4 (2015-03-07) x86_64 GNU/Linux
Whats going on here?
« Last Edit: 2015/03/24, 11:36:51 by piper »

Offline orinoco

  • User
  • Posts: 172
Re: Terrible slow usb 3.0 ssd
« Reply #1 on: 2015/03/17, 20:51:04 »
More from journald


Code: [Select]
Mär 17 20:45:00 jake kernel: sd 10:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 6 inflight: CMD OUT
Mär 17 20:45:00 jake kernel: sd 10:0:0:0: [sdg] CDB:
Mär 17 20:45:00 jake kernel: Write(10): 2a 00 1d c4 15 50 00 00 08 00
Mär 17 20:46:31 jake kernel: sd 10:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 1 inflight: CMD IN
Mär 17 20:46:31 jake kernel: sd 10:0:0:0: [sdg] CDB:
Mär 17 20:46:31 jake kernel: Read(10): 28 00 06 c0 09 08 00 00 08 00
Mär 17 20:46:31 jake kernel: scsi host10: uas_eh_bus_reset_handler start
Mär 17 20:46:31 jake kernel: sd 10:0:0:0: [sdg] uas_zap_pending 0 uas-tag 2 inflight: CMD
Mär 17 20:46:31 jake kernel: sd 10:0:0:0: [sdg] CDB:
Mär 17 20:46:31 jake kernel: Read(10): 28 00 06 c0 09 00 00 00 08 00
Mär 17 20:46:31 jake kernel: usb 3-1.1: reset SuperSpeed USB device number 3 using xhci_hcd
Mär 17 20:46:31 jake kernel: xhci_hcd 0000:02:00.0: xHCI xhci_drop_endpoint called with disabled ep ffff880387403200
Mär 17 20:46:31 jake kernel: xhci_hcd 0000:02:00.0: xHCI xhci_drop_endpoint called with disabled ep ffff880387403248
Mär 17 20:46:31 jake kernel: xhci_hcd 0000:02:00.0: xHCI xhci_drop_endpoint called with disabled ep ffff880387403290
Mär 17 20:46:31 jake kernel: xhci_hcd 0000:02:00.0: xHCI xhci_drop_endpoint called with disabled ep ffff8803874032d8
Mär 17 20:46:31 jake kernel: scsi host10: uas_eh_bus_reset_handler success
Mär 17 20:46:31 jake kernel: sd 10:0:0:0: [sdg] FAILED Result: hostbyte=DID_TIME_OUT driverbyte=DRIVER_OK
Mär 17 20:46:31 jake kernel: sd 10:0:0:0: [sdg] CDB:
Mär 17 20:46:31 jake kernel: Read(10): 28 00 06 c0 09 08 00 00 08 00
Mär 17 20:46:31 jake kernel: blk_update_request: I/O error, dev sdg, sector 113248520
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 18 inflight: CMD
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] CDB:
Mär 17 20:48:13 jake kernel: Synchronize Cache(10): 35 00 00 00 00 00 00 00 00 00
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 9 inflight: CMD OUT
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] CDB:
Mär 17 20:48:13 jake kernel: Write(10): 2a 00 21 3f d9 80 00 04 00 00
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 10 inflight: CMD OUT
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] CDB:
Mär 17 20:48:13 jake kernel: Write(10): 2a 00 21 3f dd 80 00 04 00 00
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 1 inflight: CMD OUT
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] CDB:
Mär 17 20:48:13 jake kernel: Write(10): 2a 00 21 3f e1 80 00 04 00 00
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 2 inflight: CMD OUT
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] CDB:
Mär 17 20:48:13 jake kernel: Write(10): 2a 00 21 3f e5 80 00 04 00 00
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 11 inflight: CMD OUT
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] CDB:
Mär 17 20:48:13 jake kernel: Write(10): 2a 00 21 3f e9 80 00 04 00 00
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 12 inflight: CMD OUT
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] CDB:
Mär 17 20:48:13 jake kernel: Write(10): 2a 00 21 3f ed 80 00 04 00 00
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 13 inflight: CMD OUT
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] CDB:
Mär 17 20:48:13 jake kernel: Write(10): 2a 00 21 3f f1 80 00 04 00 00
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 14 inflight: CMD OUT
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] CDB:
Mär 17 20:48:13 jake kernel: Write(10): 2a 00 21 3f f5 80 00 04 00 00
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 15 inflight: CMD OUT
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] CDB:
Mär 17 20:48:13 jake kernel: Write(10): 2a 00 21 3f f9 80 00 04 00 00
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 16 inflight: CMD OUT
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] CDB:
Mär 17 20:48:13 jake kernel: Write(10): 2a 00 21 3f fd 80 00 04 00 00
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 17 inflight: CMD OUT
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] CDB:
Mär 17 20:48:13 jake kernel: Write(10): 2a 00 21 40 01 80 00 04 00 00
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 19 inflight: CMD OUT
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] CDB:
Mär 17 20:48:13 jake kernel: Write(10): 2a 00 21 40 05 80 00 02 80 00
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] uas_eh_abort_handler 0 uas-tag 3 inflight: CMD IN
Mär 17 20:48:13 jake kernel: sd 10:0:0:0: [sdg] CDB:
Mär 17 20:48:13 jake kernel: Read(10): 28 00 21 40 08 08 00 00 08 00
Mär 17 20:48:13 jake kernel: scsi host10: uas_eh_bus_reset_handler start
Mär 17 20:48:13 jake kernel: usb 3-1.1: reset SuperSpeed USB device number 3 using xhci_hcd
Mär 17 20:48:13 jake kernel: xhci_hcd 0000:02:00.0: xHCI xhci_drop_endpoint called with disabled ep ffff880387403200
Mär 17 20:48:13 jake kernel: xhci_hcd 0000:02:00.0: xHCI xhci_drop_endpoint called with disabled ep ffff880387403248
Mär 17 20:48:13 jake kernel: xhci_hcd 0000:02:00.0: xHCI xhci_drop_endpoint called with disabled ep ffff880387403290
Mär 17 20:48:13 jake kernel: xhci_hcd 0000:02:00.0: xHCI xhci_drop_endpoint called with disabled ep ffff8803874032d8
Mär 17 20:48:13 jake kernel: scsi host10: uas_eh_bus_reset_handler success

Code: [Select]
Linux jake 3.19.1-towo.3-siduction-amd64 #1 SMP PREEMPT siduction 3.19-6 (2015-03-15) x86_64 GNU/Linux

Offline towo

  • Global Moderator
  • User
  • *****
  • Posts: 2.352
Re: Terrible slow usb 3.0 ssd
« Reply #2 on: 2015/03/17, 21:27:57 »
usb3 is terrible crap on linux, it may depend on the controller, but even on my system usb3 is unstable like hell.
The actual kernel has some xhci fixes in, maybe try it, i don't know, why you don't use the recent kernel.
I have given up to use mass-storage on usb 3, i have now only eSata in use for that.
Ich gehe nicht zum Karneval, ich verleihe nur manchmal mein Gesicht.

Offline braveheartleo

  • User
  • Posts: 23
Re: Terrible slow usb 3.0 ssd
« Reply #3 on: 2015/03/18, 04:12:59 »
<snipped>

Again after some seconds the copy process stops again and the procedure repeats again.

Code: [Select]
Linux jake 3.19.1-towo.1-siduction-amd64 #1 SMP PREEMPT siduction 3.19-4 (2015-03-07) x86_64 GNU/Linux
Whats going on here?

Hi orinoco,

Your system is using USB Attached SCSI for your usb3 host and slave devices, and that is where you might be experiencing your issues (as is evident from your logs).

You can try this workaround and see if it helps, create a new file like quirks.conf in /etc/modprobe.d/ with the following content:

Code: [Select]
options usb-storage quirks=0x174c:0x07d1:u
The format is quirks=<VID>:<PID>:u where <VID> is the VendorId and <PID> is the ProductId of your quirky usb device, which may be found via lsusb. The 'u' flag disables uas for the device.

It is advisable that you do this before you attach the device to the system.

While USB Attached SCSI (UAS) is preferrable over USB Mass Storage Bulk-only-transfers (BOT) for faster transfer speeds, USB 3.0 over at Linux still has room for improvements. Disabling UAS for the device will switch it to using BOT.

Lemme know if this helps. Goodluck. ;-)
« Last Edit: 2015/03/18, 05:17:26 by braveheartleo »

Offline braveheartleo

  • User
  • Posts: 23
Re: Terrible slow usb 3.0 ssd
« Reply #4 on: 2015/03/18, 04:32:59 »
usb3 is terrible crap on linux, it may depend on the controller, but even on my system usb3 is unstable like hell.

In my case I have a usb 3.0 dock (which uses a Jmicron JMS539 usb3-to-sata2 bridge) that would sometimes spontaneously disconnect over usb 3.0, but not 2.0. I still haven't found the cause for this erratic behavior, but I learned to live with it.

USBs are still very much my preferred external I/O device compared to other standards like eSATA / Firewire. They are ubiquitous and self-powered compared to eSATAs (except eSATAp, which may not even be that common). I could only hope that the usb3 state and uasp support in Linux improves.

Offline orinoco

  • User
  • Posts: 172
Re: Terrible slow usb 3.0 ssd
« Reply #5 on: 2015/03/23, 23:48:33 »
[
Code: [Select]
options usb-storage quirks=0x174c:0x07d1:u
Braveheartleo, that was the trick. Thank you very much. I made extensive tests during the weekend with large vdr recordings und your workaround helped a lot.

Offline braveheartleo

  • User
  • Posts: 23
Re: Terrible slow usb 3.0 ssd
« Reply #6 on: 2015/03/24, 04:27:49 »
Glad I could help. Good to know it worked.  8)

Offline ayla

  • User
  • Posts: 1.706
Re: Terrible slow usb 3.0 ssd
« Reply #7 on: 2015/03/26, 23:06:35 »
usb3 is terrible crap on linux, it may depend on the controller, but even on my system usb3 is unstable like hell.
The actual kernel has some xhci fixes in, maybe try it, i don't know, why you don't use the recent kernel.
I have given up to use mass-storage on usb 3, i have now only eSata in use for that.
Code: [Select]
Kernel: 3.19.2-towo.2-siduction-amd64 x86_64
Bus 002 Device 003: ID 174c:5136 ASMedia Technology Inc.
Bus 002 Device 004: ID 174c:07d1 ASMedia Technology Inc.

hmm, nothing to complain here, also no problems with the last two Kernel versions (at least):
USB 3.0 capable hub in use.
Code: [Select]
/dev/sde:
 Timing O_DIRECT cached reads:   730 MB in  2.00 seconds = 364.55 MB/sec
 Timing O_DIRECT disk reads: 1256 MB in  3.00 seconds = 418.16 MB/sec
/dev/sdf:
 Timing O_DIRECT cached reads:   400 MB in  2.00 seconds = 199.81 MB/sec
 Timing O_DIRECT disk reads: 640 MB in  3.01 seconds = 212.78 MB/sec

modprobe -V uas
kmod version 20

just for the record...

greets
ayla
Kernel: 4.18.11-towo.1-siduction-amd64 x86_64
Desktop: KDE Plasma 5.13.5
(mostly) kept current siduction "18.3.0 Patience"
HP EliteBook 8570p; MoBo: HP model: 17A7 v: KBC Version 42.35
Dual core  i5-3320M (-MT-MCP-)
Intel 3. Gen Graphics (Ivybridge) Driver:i915/ intel
 Wifi Intel Centrino N6205

holgerw

  • Guest
Re: (solved) Terrible slow usb 3.0 ssd
« Reply #8 on: 2015/04/14, 07:22:55 »
Hi @braveheartleo,

thanks a lot for your hint, would you like it to describe this trick with usb3 in siduction wiki?

I thougt, that something is wrong with my usb3 slots on my mainboard, during copy actions I got interrupts, sometimes the usb3 device has lost its connection and so on.

Your trick works, it is not siduction specific, it works also with openSUSE Thumbleweed.

It would be nice, if such an entry in /etc/modprobe.d/usb3.conf is generated automatically while connecting an usb3 device.

Perhaps somebody write a bash script, I would do this, but I've too little experinence with scripting.

Kind regards,
  Holger