| View previous topic :: View next topic |
| Author |
Message |
pa4wdh
l33t
Joined: 16 Dec 2005
Posts: 910
|
 Posted: Sat Jun 03, 2023 7:17 am Post subject: Big harddrive doesn't work (reported 1TB instead of 4TB) |
 |
|
Hi All,
I'm using an external 4TB harddrive for my backups. To make a backup i simply attach the drive and run a script to make the backup on the disk, this drive is used with 7 different systems, all running Gentoo.
On my most recent system i'm having problems using it, this is what dmesg says about it:
| Code: |
[ 3268.803578] usb 2-1: new SuperSpeed USB device number 4 using xhci_hcd
[ 3268.835364] usb 2-1: New USB device found, idVendor=152d, idProduct=2509, bcdDevice= 1.00
[ 3268.835384] usb 2-1: New USB device strings: Mfr=1, Product=11, SerialNumber=3
[ 3268.835394] usb 2-1: Product: SHARKOON COMBO
[ 3268.835401] usb 2-1: Manufacturer: JMicron
[ 3268.835408] usb 2-1: SerialNumber: 00A1234567D0
[ 3268.837624] usb-storage 2-1:1.0: USB Mass Storage device detected
[ 3268.838194] scsi host2: usb-storage 2-1:1.0
[ 3269.884006] scsi 2:0:0:0: Direct-Access SHARKOON COMBO 0000 PQ: 0 ANSI: 5
[ 3269.884504] sd 2:0:0:0: Attached scsi generic sg1 type 0
[ 3269.885438] sd 2:0:0:0: [sdb] 2147484931 512-byte logical blocks: (1.10 TB/1.00 TiB)
[ 3269.886233] sd 2:0:0:0: [sdb] Write Protect is off
[ 3269.886248] sd 2:0:0:0: [sdb] Mode Sense: 28 00 00 00
[ 3269.886773] sd 2:0:0:0: [sdb] No Caching mode page found
[ 3269.886787] sd 2:0:0:0: [sdb] Assuming drive cache: write through
[ 3269.888234] sd 2:0:0:0: [sdb] 2147483649 512-byte logical blocks: (1.10 TB/1.00 TiB)
[ 3269.889132] sdb: detected capacity change from 2147484931 to 2147483649
|
It may not look too bad, but this is a 4TB drive, not 1TB. All my other systems report this:
| Code: |
[5272797.012207] usb 3-3: new SuperSpeed USB device number 5 using xhci_hcd
[5272797.051682] usb-storage 3-3:1.0: USB Mass Storage device detected
[5272797.052044] scsi host2: usb-storage 3-3:1.0
[5272798.092186] scsi 2:0:0:0: Direct-Access SHARKOON COMBO 0000 PQ: 0 ANSI: 5
[5272798.092967] sd 2:0:0:0: [sdc] Very big device. Trying to use READ CAPACITY(16).
[5272798.093215] sd 2:0:0:0: [sdc] 7814037168 512-byte logical blocks: (4.00 TB/3.64 TiB)
[5272798.094700] sd 2:0:0:0: [sdc] Write Protect is off
[5272798.094709] sd 2:0:0:0: [sdc] Mode Sense: 28 00 00 00
[5272798.096227] sd 2:0:0:0: [sdc] No Caching mode page found
[5272798.096233] sd 2:0:0:0: [sdc] Assuming drive cache: write through
[5272798.154941] sdc: sdc1
[5272798.155270] sd 2:0:0:0: [sdc] Attached SCSI disk
|
I've had similar problems before with a raspberry pi where i tried to use the same disk, but i don't remember the solution  .
Does anyone know how to solve this?
With some internet searches i can only find references some kernel configuration option regarding to big block devices which doesn't exist anymore.
Thanks in advance.
_________________
The gentoo way of bringing peace to the world:
USE="-war" emerge --newuse @world
My shared code repository: https://code.pa4wdh.nl.eu.org
Music, Free as in Freedom: https://www.jamendo.com |
|
| Back to top |
|
 |
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54798
Location: 56N 3W
|
| Posted: Sat Jun 03, 2023 7:44 am Post subject: |
|
|
pa4wdh,
Either something does not understand 48bit LBA or you have GPT support missing in the kernel, or both.
I suspect the former as it goes wrong on hardware detection.
| Code: | [5272797.012207] usb 3-3: new SuperSpeed USB device number 5 using xhci_hcd
[5272797.051682] usb-storage 3-3:1.0: USB Mass Storage device detected
[5272797.052044] scsi host2: usb-storage 3-3:1.0
[5272798.092186] scsi 2:0:0:0: Direct-Access SHARKOON COMBO 0000 PQ: 0 ANSI: 5
[5272798.092967] sd 2:0:0:0: [sdc] Very big device. Trying to use READ CAPACITY(16).
[5272798.093215] sd 2:0:0:0: [sdc] 7814037168 512-byte logical blocks: (4.00 TB/3.64 TiB) |
When it goes wrong the | Code: | | [5272798.092967] sd 2:0:0:0: [sdc] Very big device. Trying to use READ CAPACITY(16). |
is missing and it not god as far as reading partition tables.
Also the error case says | Code: | | [ 3269.884504] sd 2:0:0:0: Attached scsi generic sg1 type 0 |
That's the wrong sort of device too.
Are you moving the drive between PCs including the USB to SATA bridge?
That's important as some USB to SATA bridge adaptors have odd limits because they cannot pass all SCSI commands.
Does the odd box have CONFIG_BLK_DEV_SD=y in the kernel?
Its not needed for nvme ... but USB attached storage requires it. That may be why the Attached scsi generic appears in dmesg.
While you are checking the kernel is a very good thing for USB3, so you might want to check for that.
Its not the cause of the problem but it allows DMA, NCQ and all those other go faster things over USB3, as long as its USB3 all the way.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
pa4wdh
l33t
Joined: 16 Dec 2005
Posts: 910
|
| Posted: Sat Jun 03, 2023 10:40 am Post subject: |
|
|
| NeddySeagoon wrote: |
Either something does not understand 48bit LBA or you have GPT support missing in the kernel, or both.
|
I came to a similar conclusion.
GPT is enabled (CONFIG_EFI_PARTITION=y), internal SSD is also partitioned with GPT because it boots with UEFI (i would have preferred legacy, but the system's BIOS can't do that anymore).
So far i've diff'ed the CONFIG_BLK, CONFIG_SCSI and CONFIG_USB options without success.
| Quote: |
When it goes wrong the | Code: | | [5272798.092967] sd 2:0:0:0: [sdc] Very big device. Trying to use READ CAPACITY(16). |
is missing and it not god as far as reading partition tables.
|
That's what i've been focusing on so far, even checking the code of drivers/scsi/sd.c to see the conditions that trigger it. There is no related CONFIG_ variable which made be think compile time configuration is not the problem.
| Quote: | Also the error case says | Code: | | [ 3269.884504] sd 2:0:0:0: Attached scsi generic sg1 type 0 |
That's the wrong sort of device too. |
Yep. I think that's because it can't properly read the partition table because of the incorrect detected size. Smaller devices (USB sticks) work ok, i don't have smaller sata devices available to attach to the bridge.
| Quote: | Are you moving the drive between PCs including the USB to SATA bridge?
That's important as some USB to SATA bridge adaptors have odd limits because they cannot pass all SCSI commands. |
Yes, i only move the USB connector from one PC to another, also tried USB2.0 and USB3.0 ports with no difference.
| Quote: | Does the odd box have CONFIG_BLK_DEV_SD=y in the kernel?
Its not needed for nvme ... but USB attached storage requires it. That may be why the Attached scsi generic appears in dmesg. |
Yes, it is also used by the internal sata disk (1TB samsung 870 EVO).
| Quote: | While you are checking the kernel is a very good thing for USB3, so you might want to check for that.
Its not the cause of the problem but it allows DMA, NCQ and all those other go faster things over USB3, as long as its USB3 all the way. |
I can give it a try, but the usb-sata bridge predates UAS and on other systems i've worked with UAS gave me more problems than joy  . It's also disabled on the other machines where the disk works properly.
_________________
The gentoo way of bringing peace to the world:
USE="-war" emerge --newuse @world
My shared code repository: https://code.pa4wdh.nl.eu.org
Music, Free as in Freedom: https://www.jamendo.com |
|
| Back to top |
|
|
pa4wdh
l33t
Joined: 16 Dec 2005
Posts: 910
|
| Posted: Sat Jun 03, 2023 11:40 am Post subject: |
|
|
I just recompiled my kernel with UAS support, there is no difference.
_________________
The gentoo way of bringing peace to the world:
USE="-war" emerge --newuse @world
My shared code repository: https://code.pa4wdh.nl.eu.org
Music, Free as in Freedom: https://www.jamendo.com |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54798
Location: 56N 3W
|
| Posted: Sat Jun 03, 2023 12:22 pm Post subject: |
|
|
pa4wdh,
Lets check the partition table in both good and bad conditions.
I don't think it that but it will rule another thing out.
Do | Code: | | fdisk -l -t dos /dev/... |
and | Code: | | fdisk -l -t gpt /dev/... |
on both systems.
On the 'bad' system I'm expecting fdisk to warn that there is at least one partition off the end of the device and complain that it cannot read the backup partition table.
On no account let it write anything.
-t dos should show
| Code: | Device Boot Start End Sectors Size Id Type
/dev/sda1 1 4294967295 4294967295 2T ee GPT |
as that's the protective MSDOS table that shows type 0xee to tell that its a fake and GPT in is use.
Its 2T as that is the most that the MSDOS partition table can describe. That's from an 8T drive.
I expect the GPT to read correctly ... other than the warnings in the 'bad' system.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
Anon-E-moose
Watchman
Joined: 23 May 2008
Posts: 6209
Location: Dallas area
|
| Posted: Sat Jun 03, 2023 1:55 pm Post subject: |
|
|
I've never liked the jmicron controllers, I usually try and avoid them (too much trouble in the past compared to asmedia)
I suppose it would be easy to grab the config files from both machines, 1 working, 1 not and strip out comments and look for differences in these areas
SCSI, BLK, USB
_________________
UM780, 6.12 zen kernel, gcc 13, openrc, wayland |
|
| Back to top |
|
|
pa4wdh
l33t
Joined: 16 Dec 2005
Posts: 910
|
| Posted: Sat Jun 03, 2023 5:23 pm Post subject: |
|
|
Thanks for you help Neddy.
| NeddySeagoon wrote: |
Do | Code: | | fdisk -l -t dos /dev/... |
and | Code: | | fdisk -l -t gpt /dev/... |
on both systems.
|
Good:
| Code: | Working:
pc12 ~ # fdisk -l -t dos /dev/sdc
Disk /dev/sdc: 3.64 TiB, 4000787030016 bytes, 7814037168 sectors
Disk model: COMBO
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: dos
Disk identifier: 0x00000000
Device Boot Start End Sectors Size Id Type
/dev/sdc1 1 4294967295 4294967295 2T ee GPT
pc12 ~ # fdisk -l -t gpt /dev/sdc
Disk /dev/sdc: 3.64 TiB, 4000787030016 bytes, 7814037168 sectors
Disk model: COMBO
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: gpt
Disk identifier: 77441E8F-17B4-43B6-9A37-8883F483ECE5
Device Start End Sectors Size Type
/dev/sdc1 2048 7814037134 7814035087 3.6T Linux filesystem |
Bad: | Code: | pc18 ~ # fdisk -l -t dos /dev/sdb
Disk /dev/sdb: 1 TiB, 1099511628288 bytes, 2147483649 sectors
Disk model: COMBO
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: dos
Disk identifier: 0x00000000
Device Boot Start End Sectors Size Id Type
/dev/sdb1 1 4294967295 4294967295 2T ee GPT
pc18 ~ # fdisk -l -t gpt /dev/sdb
Disk /dev/sdb: 1 TiB, 1099511628288 bytes, 2147483649 sectors
Disk model: COMBO
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes |
| Quote: | On the 'bad' system I'm expecting fdisk to warn that there is at least one partition off the end of the device and complain that it cannot read the backup partition table.
On no account let it write anything. |
Other than the output above there are no warnings whatsoever.
| Quote: | | I've never liked the jmicron controllers, I usually try and avoid them (too much trouble in the past compared to asmedia) |
So far it has been serving me well for many years.
| Quote: | I suppose it would be easy to grab the config files from both machines, 1 working, 1 not and strip out comments and look for differences in these areas
SCSI, BLK, USB |
Which was my initial idea too, but that didn't give any useful results.
_________________
The gentoo way of bringing peace to the world:
USE="-war" emerge --newuse @world
My shared code repository: https://code.pa4wdh.nl.eu.org
Music, Free as in Freedom: https://www.jamendo.com |
|
| Back to top |
|
|
Anon-E-moose
Watchman
Joined: 23 May 2008
Posts: 6209
Location: Dallas area
|
| Posted: Sat Jun 03, 2023 6:53 pm Post subject: |
|
|
I would try another usb port (even a usb 2 port just to see if it changes the recognition of the drive), if you haven't tried that already.
What does lspci say about USB on the machine that doesn't work, as well as one machine that does.
I had a problem in the past where the kernel didn't like the usb controller on the MB, it added some funny quirks, and disabled some of the use of it.
_________________
UM780, 6.12 zen kernel, gcc 13, openrc, wayland |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54798
Location: 56N 3W
|
| Posted: Sat Jun 03, 2023 7:24 pm Post subject: |
|
|
pa4wdh,
| Code: | | idVendor=152d, idProduct=2509 |
That leads to a JMicron USB to SATA2 bridge chip.
It can do USB2 or USB3 but its bulk mode only. It does not understand UAS.
The protective MSDOS partition table is correct. All 2TB of it.
The GPT is certainly there, just not displayed.
I'm with Anon-E-moose on the avoidance of JMicron.
While all of this is interesting, its not progressing the solution to your issue.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
Anon-E-moose
Watchman
Joined: 23 May 2008
Posts: 6209
Location: Dallas area
|
| Posted: Sat Jun 03, 2023 8:54 pm Post subject: |
|
|
What type machine is the one that doesn't work with the drive. Model, mb make, etc, the key may be there.
_________________
UM780, 6.12 zen kernel, gcc 13, openrc, wayland |
|
| Back to top |
|
|
pa4wdh
l33t
Joined: 16 Dec 2005
Posts: 910
|
| Posted: Sun Jun 04, 2023 7:28 am Post subject: |
|
|
The motherboard is an ASRock J5040-ITX: https://www.asrock.com.tw/mb/Intel/J5040-ITX/index.nl.asp
The actual SoC on there is an Intel J5040: https://www.intel.com/content/www/us/en/products/sku/197304/intel-pentium-silver-j5040-processor-4m-cache-up-to-3-20-ghz/specifications.html
The USB controller according to lspci: 00:15.0 USB controller: Intel Corporation Celeron/Pentium Silver Processor USB 3.0 xHCI Controller (rev 06)
USB Topology when the drive is connected:
| Code: | /: Bus 02.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/7p, 5000M
|__ Port 1: Dev 2, If 0, Class=Mass Storage, Driver=usb-storage, 5000M
/: Bus 01.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/9p, 480M
|__ Port 5: Dev 2, If 0, Class=Hub, Driver=hub/4p, 480M
|__ Port 6: Dev 3, If 0, Class=Hub, Driver=hub/2p, 480M |
USB device list:
| Code: | Bus 001 Device 003: ID 0409:005a NEC Corp. HighSpeed Hub
Bus 001 Device 002: ID 05e3:0610 Genesys Logic, Inc. USB2.0 Hub
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation xHCI Host Controller
Bus 002 Device 002: ID 152d:2509 JMicron Technology Corp. / JMicron USA Technology Corp. JMS539, JMS551 SATA 3Gb/s bridge
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation xHCI Host Controller |
lsusb -v for the USB-SATA bridge (https://en.sharkoon.com/product/13760)
| Code: | Bus 002 Device 002: ID 152d:2509 JMicron Technology Corp. / JMicron USA Technology Corp. JMS539, JMS551 SATA 3Gb/s bridge
Device Descriptor:
bLength 18
bDescriptorType 1
bcdUSB 3.00
bDeviceClass 0
bDeviceSubClass 0
bDeviceProtocol 0
bMaxPacketSize0 9
idVendor 0x152d JMicron Technology Corp. / JMicron USA Technology Corp.
idProduct 0x2509 JMS539, JMS551 SATA 3Gb/s bridge
bcdDevice 1.00
iManufacturer 1 JMicron
iProduct 11 SHARKOON COMBO
iSerial 3 00A1234567D0
bNumConfigurations 1
Configuration Descriptor:
bLength 9
bDescriptorType 2
wTotalLength 0x002c
bNumInterfaces 1
bConfigurationValue 1
iConfiguration 0
bmAttributes 0xc0
Self Powered
MaxPower 0mA
Interface Descriptor:
bLength 9
bDescriptorType 4
bInterfaceNumber 0
bAlternateSetting 0
bNumEndpoints 2
bInterfaceClass 8 Mass Storage
bInterfaceSubClass 6 SCSI
bInterfaceProtocol 80 Bulk-Only
iInterface 0
Endpoint Descriptor:
bLength 7
bDescriptorType 5
bEndpointAddress 0x81 EP 1 IN
bmAttributes 2
Transfer Type Bulk
Synch Type None
Usage Type Data
wMaxPacketSize 0x0400 1x 1024 bytes
bInterval 0
bMaxBurst 15
Endpoint Descriptor:
bLength 7
bDescriptorType 5
bEndpointAddress 0x02 EP 2 OUT
bmAttributes 2
Transfer Type Bulk
Synch Type None
Usage Type Data
wMaxPacketSize 0x0400 1x 1024 bytes
bInterval 0
bMaxBurst 15
Binary Object Store Descriptor:
bLength 5
bDescriptorType 15
wTotalLength 0x002a
bNumDeviceCaps 3
USB 2.0 Extension Device Capability:
bLength 7
bDescriptorType 16
bDevCapabilityType 2
bmAttributes 0x00000002
HIRD Link Power Management (LPM) Supported
SuperSpeed USB Device Capability:
bLength 10
bDescriptorType 16
bDevCapabilityType 3
bmAttributes 0x00
wSpeedsSupported 0x000e
Device can operate at Full Speed (12Mbps)
Device can operate at High Speed (480Mbps)
Device can operate at SuperSpeed (5Gbps)
bFunctionalitySupport 1
Lowest fully-functional device speed is Full Speed (12Mbps)
bU1DevExitLat 10 micro seconds
bU2DevExitLat 32 micro seconds
Container ID Device Capability:
bLength 20
bDescriptorType 16
bDevCapabilityType 4
bReserved 0
ContainerID {03020100-0504-0706-0800-000000000000}
Device Status: 0x0001
Self Powered |
_________________
The gentoo way of bringing peace to the world:
USE="-war" emerge --newuse @world
My shared code repository: https://code.pa4wdh.nl.eu.org
Music, Free as in Freedom: https://www.jamendo.com |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54798
Location: 56N 3W
|
| Posted: Sun Jun 04, 2023 8:02 am Post subject: |
|
|
pa4wdh,
I suspect its kernel rather than hardware related.
Please pastebin the kernel configs from both systems.
Ideally from /proc/config.gz. If do don have that proc entry, please check that the config you are posting belongs to the running kernel.
The entire dmesg from the dud may be interesting too.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
Anon-E-moose
Watchman
Joined: 23 May 2008
Posts: 6209
Location: Dallas area
|
| Posted: Sun Jun 04, 2023 9:56 am Post subject: |
|
|
I'm with Neddy, I'm beginning to think something is not set right on the machine that doesn't work.
It does use the scsi interface, and usb storage does need BLK_DEV_SD set,
the help for USB_STORAGE mentions this but doesn't enforce it.
From my config
| Code: | CONFIG_BLK_DEV=y
CONFIG_BLK_DEV_BSG=y
CONFIG_BLK_DEV_BSG_COMMON=y
CONFIG_BLK_DEV_FD=m
CONFIG_BLK_DEV_LOOP=y
CONFIG_BLK_DEV_LOOP_MIN_COUNT=8
CONFIG_BLK_DEV_NVME=y
CONFIG_BLK_DEV_SD=y
CONFIG_BLK_DEV_SR=y
...
CONFIG_SCSI=y
CONFIG_SCSI_COMMON=y
CONFIG_SCSI_DMA=y
CONFIG_SCSI_MOD=y
CONFIG_SCSI_SCAN_ASYNC=y
CONFIG_SCSI_SPI_ATTRS=m |
_________________
UM780, 6.12 zen kernel, gcc 13, openrc, wayland |
|
| Back to top |
|
|
pa4wdh
l33t
Joined: 16 Dec 2005
Posts: 910
|
|
| Back to top |
|
|
pa4wdh
l33t
Joined: 16 Dec 2005
Posts: 910
|
| Posted: Sun Jun 04, 2023 12:50 pm Post subject: |
|
|
I've been monitoring my USB bus with wireshark to see what's happening, and it seems my USB-Sata bridge is responding different.
On a good system (a different system than the config.gz is from):
| Code: | Frame 88: 72 bytes on wire (576 bits), 72 bytes captured (576 bits) on interface usbmon1, id 0
USB URB
USB Mass Storage
SCSI Payload (Read Capacity(10) Response Data)
[LUN: 0x0000]
[Command Set:Direct Access Device (0x00) ]
[SBC Opcode: Read Capacity(10) (0x25)]
[Request in: 85]
[Response in: 90]
LBA: 4294967295 (2047 GB)
Block size in bytes: 512 |
On the bad system:
| Code: | Frame 147: 72 bytes on wire (576 bits), 72 bytes captured (576 bits) on interface usbmon0, id 0
USB URB
USB Mass Storage
SCSI Payload (Read Capacity(10) Response Data)
[LUN: 0x0000]
[Command Set:Direct Access Device (0x00) ]
[SBC Opcode: Read Capacity(10) (0x25)]
[Request in: 144]
[Response in: 149]
LBA: 2147484930 (1024 GB)
Block size in bytes: 512 |
Note the difference in reported capacity. Keep in mind this is the READ_CAPACITY(10) response, so it's expected to report the wrong capacity. However, on the good system 4294967295 sectors is 0xffffffff, which is the value checked in drivers/scsi/sd.c line 2443 and triggers the use of READ_CAPACITY(16) as reported in the working dmesg. Of course this doesn't tell why it's responding different, but it explains the difference in response from the kernel when the drive is detected.
_________________
The gentoo way of bringing peace to the world:
USE="-war" emerge --newuse @world
My shared code repository: https://code.pa4wdh.nl.eu.org
Music, Free as in Freedom: https://www.jamendo.com |
|
| Back to top |
|
|
pa4wdh
l33t
Joined: 16 Dec 2005
Posts: 910
|
| Posted: Sun Jun 04, 2023 2:57 pm Post subject: |
|
|
As an experiment i've addad my USB-Sata bridge to the unusual devs file (drivers/usb/storage/unusual_devs.h) with this entry:
| Code: |
UNUSUAL_DEV( 0x152d, 0x2509, 0x0100, 0x0100,
"JMicron",
"SHARKOON COMBO",
USB_SC_DEVICE, USB_PR_DEVICE, NULL,
US_FL_NEEDS_CAP16),
|
After a recompile/reboot dmesg tells the good news:
| Code: | [ 77.646154] usb 2-1: new SuperSpeed USB device number 2 using xhci_hcd
[ 77.677868] usb 2-1: New USB device found, idVendor=152d, idProduct=2509, bcdDevice= 1.00
[ 77.677888] usb 2-1: New USB device strings: Mfr=1, Product=11, SerialNumber=3
[ 77.677898] usb 2-1: Product: SHARKOON COMBO
[ 77.677905] usb 2-1: Manufacturer: JMicron
[ 77.677911] usb 2-1: SerialNumber: 00A1234567D0
[ 77.680203] usb-storage 2-1:1.0: USB Mass Storage device detected
[ 77.680758] usb-storage 2-1:1.0: Quirks match for vid 152d pid 2509: 400000
[ 77.680936] scsi host2: usb-storage 2-1:1.0
[ 78.746775] scsi 2:0:0:0: Direct-Access SHARKOON COMBO 0000 PQ: 0 ANSI: 5
[ 78.747349] sd 2:0:0:0: Attached scsi generic sg1 type 0
[ 78.747967] sd 2:0:0:0: [sdb] 7814037168 512-byte logical blocks: (4.00 TB/3.64 TiB)
[ 78.748484] sd 2:0:0:0: [sdb] Write Protect is off
[ 78.748499] sd 2:0:0:0: [sdb] Mode Sense: 28 00 00 00
[ 78.749199] sd 2:0:0:0: [sdb] No Caching mode page found
[ 78.749213] sd 2:0:0:0: [sdb] Assuming drive cache: write through
[ 79.072687] sdb: sdb1
[ 79.073189] sd 2:0:0:0: [sdb] Attached SCSI disk |
The drive works as expected and can be mounted/used.
Now i'm a bit unsure it this is the right solution, but at least it's a workaround for now.
_________________
The gentoo way of bringing peace to the world:
USE="-war" emerge --newuse @world
My shared code repository: https://code.pa4wdh.nl.eu.org
Music, Free as in Freedom: https://www.jamendo.com |
|
| Back to top |
|
|
Anon-E-moose
Watchman
Joined: 23 May 2008
Posts: 6209
Location: Dallas area
|
| Posted: Sun Jun 04, 2023 3:32 pm Post subject: |
|
|
IMO, that is one of the right ways to fix the problem, I've had to add to quirks files before.
Files like that are there for a reason.
_________________
UM780, 6.12 zen kernel, gcc 13, openrc, wayland |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54798
Location: 56N 3W
|
| Posted: Sun Jun 04, 2023 4:03 pm Post subject: |
|
|
pa4wdh,
Is that already in the kernel on the working system?
That's one of the better ways to fix it. Its device specific that way.
Is it worth a Gentoo bug.?
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
pa4wdh
l33t
Joined: 16 Dec 2005
Posts: 910
|
| Posted: Sun Jun 04, 2023 5:11 pm Post subject: |
|
|
| Quote: | | Is that already in the kernel on the working system? |
I've made the change on the not-working system, all other systems are left as they were.
| Quote: | | Is it worth a Gentoo bug.? |
Would be nice if it could be part of the genpatches or upstreamed via the gentoo kernel project. There's not really a nice way to submit changes like this as an end user.
_________________
The gentoo way of bringing peace to the world:
USE="-war" emerge --newuse @world
My shared code repository: https://code.pa4wdh.nl.eu.org
Music, Free as in Freedom: https://www.jamendo.com |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54798
Location: 56N 3W
|
| Posted: Sun Jun 04, 2023 5:36 pm Post subject: |
|
|
pa4wdh,
Start with a gentoo bug then.
Let the kernel devs decide if it should bu pushed upstream or not.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
pietinger
Moderator
Joined: 17 Oct 2006
Posts: 5353
Location: Bavaria
|
| Posted: Sun Jun 04, 2023 6:25 pm Post subject: |
|
|
pa4wdh,
I have compared both kernel config-files and there are two main differences which could have an impact. On your bad machine you have:
| Code: | # CONFIG_CC_OPTIMIZE_FOR_PERFORMANCE is not set
CONFIG_CC_OPTIMIZE_FOR_SIZE=y
# CONFIG_INTEL_IOMMU is not set |
(On your good machine you have OPTIMIZE_FOR_PERFORMANCE and INTEL_IOMMU is enabled; without this you will miss also other options like e.g. PCI_PRI ...)
Maybe try to change only one of them and see if it helps (to be honest: OPTIMIZE_FOR_SIZE is really bad and you should never use it for standard desktop/server machines) |
|
| Back to top |
|
|
pa4wdh
l33t
Joined: 16 Dec 2005
Posts: 910
|
| Posted: Sun Jun 04, 2023 6:48 pm Post subject: |
|
|
@pietinger:
Thanks for pointing those differences out.
I wasn't aware of OPTIMIZE_FOR_SIZE (i started off with the install iso, maybe it's worth checking the config there), so I'll change it to OPTIMIZE_FOR_PERFORMANCE. INTEL_IOMMU seems to make sense too (although i don't expect it'll solve this issue).
Edit: I've changed both options, it doesn't make a difference for this problem. I'll keep them this way anyway because i think it makes sense
@NeddySeagoon:
I'll file the bug after i checked pietinger's suggestions.
In the mean time i made a user patch to easily apply the same change to future versions if needed:
| Code: | --- linux-6.1.19-gentoo/drivers/usb/storage/unusual_devs.h 2022-12-11 23:15:18.000000000 +0100
+++ linux-6.1.19-gentoo-mod/drivers/usb/storage/unusual_devs.h 2023-06-04 20:24:09.374711468 +0200
@@ -2425,6 +2425,12 @@
"Digital MP3 Audio Player",
USB_SC_DEVICE, USB_PR_DEVICE, NULL, US_FL_NOT_LOCKABLE ),
+UNUSUAL_DEV( 0x152d, 0x2509, 0x0100, 0x0100,
+ "JMicron",
+ "SHARKOON COMBO",
+ USB_SC_DEVICE, USB_PR_DEVICE, NULL,
+ US_FL_NEEDS_CAP16),
+
/* Unusual uas devices */
#if IS_ENABLED(CONFIG_USB_UAS)
#include "unusual_uas.h" |
_________________
The gentoo way of bringing peace to the world:
USE="-war" emerge --newuse @world
My shared code repository: https://code.pa4wdh.nl.eu.org
Music, Free as in Freedom: https://www.jamendo.com |
|
| Back to top |
|
|
pa4wdh
l33t
Joined: 16 Dec 2005
Posts: 910
|
| Posted: Mon Jun 05, 2023 5:13 pm Post subject: |
|
|
I'm still in doubt if this is the right way to go.
The thing is: The same hardware with the same software, even with my limited number of systems, works in 87.5% of the situations, the patch is only to make the other 12.5% work. It feels bad to blame the USB-Sata bridge, there must be something else misbehaving, although I don't know what.
_________________
The gentoo way of bringing peace to the world:
USE="-war" emerge --newuse @world
My shared code repository: https://code.pa4wdh.nl.eu.org
Music, Free as in Freedom: https://www.jamendo.com |
|
| Back to top |
|
|
NeddySeagoon
Administrator
Joined: 05 Jul 2003
Posts: 54798
Location: 56N 3W
|
| Posted: Mon Jun 05, 2023 5:17 pm Post subject: |
|
|
pa4wdh,
Does same == identical?
Especially kernel versions and configuration?
I don't think its hardware either.
_________________
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
| Back to top |
|
|
pa4wdh
l33t
Joined: 16 Dec 2005
Posts: 910
|
| Posted: Mon Jun 05, 2023 5:46 pm Post subject: |
|
|
| NeddySeagoon wrote: | pa4wdh,
Does same == identical?
Especially kernel versions and configuration?
I don't think its hardware either. |
Of course configurations differ because different hardware needs a different configuration, but there are no relevant differences are far as i can see. Kernels are 6.1.12 (working) and 6.1.19 (not working), but there have been no changes in drivers/scsi/sd.c.
To be 100% i can upgrade the 6.1.12 kernel to 6.1.19.
_________________
The gentoo way of bringing peace to the world:
USE="-war" emerge --newuse @world
My shared code repository: https://code.pa4wdh.nl.eu.org
Music, Free as in Freedom: https://www.jamendo.com |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
Kernel & Hardware
