Volker Halle wrote:
> Only the SDA> USB SHOW OHCI UCB command may show useful information
>

On the system where the USB is still working normally:

SDA> USB SHOW OHCI UCB
Open Host Controller UCB extensions
Direct DMA window base address 0000000000000000
I/O handle ffffffff826eb380
Pointer to USB structure 826e5590
Pointer to saved device registers 826e4c80
Pointer to HCCA 826f0400
HCCA PCI buss address c00e0400
Controler revision 00000110
Root hub address 00000001
Number of ports 00000002
Status change request queue
Flink 826fbe10
Blink 826fbe10
Connected end points queue
Flink 826dd1a0
Blink 826dd1a0
Free Endpoints queue
Flink 826f48c0
Blink 826f47c0
Count of free ED 0000005d
Free transfer descriptor queue

Show OHCI
--------------------
Flink 826f29c0
Blink 826f3940
Count of free TD 0000005c
Used transfer descriptor queue
Flink 826f200c
Blink 826f3a0c
Count of used TD 00000023
Stalled endpoint descriptor queue
Flink 826dd1cc
Blink 826dd1cc
Running endpoint descriptor queue
Flink 826dd1d4
Blink 826dd1d4
Paused endpoint descriptor queue
Flink 826dd1dc
Blink 826dd1dc
ED lists will not be displayed
Frame Hight part 01a28000

how OHCI
--------------------
Available Bandwidth 00000064
Used Bandwidth 00000000
SO count 00000000
SO stall Frame 00000000
SO Limit Frame 00000000
SO Limit Hit 00000000
SO Stall Hit 00000000
UCM message mailbox 00000000
Vendor ID 00001080
Device ID 0000c693
Map registers supported 00000001
BAP MIN 0000000000000000
BAP MAX 0000000000000000
BAP IN USE 0000000000000000
BAP PEAK USE 0000000000000000
DMA_MAPPED 00000000000959ea
DMA_FREED fffffffffff6a617

On 2022-03-03, Joukj <joukj@hrem.nano.tudelft.nl> wrote:
> Volker Halle wrote:
>> Only the SDA> USB SHOW OHCI UCB command may show useful information
>>
>
> On the system where the USB is still working normally:
>
>
> SDA> USB SHOW OHCI UCB
> Open Host Controller UCB extensions
> Direct DMA window base address 0000000000000000
> I/O handle ffffffff826eb380

[snip]

OHCI and EHCI are not the same thing.

OHCI controllers are for full-speed and low-speed devices while EHCI
controllers are for the high-speed devices.

Volker, does VMS use this OHCI command to show the EHCI data, or is
there also an EHCI show command as well ?

For this command to be of any use, it needs to show the detailed
TD entry data as well so you can see the status of each TD.

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Walking destinations on a map are further away than they appear.

Simon,

there are no SDA> USB SHOW EHCI commands.

The only command, which seems to provide any references to TDs is:

SDA> usb show ohci ucb
Open Host Controller UCB extensions
Direct DMA window base address 0000000000000000
I/O handle ffffffff84bdf900
Pointer to USB structure 84bdb710
Pointer to saved device registers 84be7480
Pointer to HCCA 84b74700
HCCA PCI buss address 45320700
Controler revision 00000010
Root hub address 00000001
Number of ports 00000003
Status change request queue
Flink 84bf6290
Blink 84bf6290
Connected end points queue
Flink 84bb4698
Blink 84bb4698
Free Endpoints queue
Flink 84bee8c0
Blink 84bee7c0
Count of free ED 0000005d
Free transfer descriptor queue
Flink 84becd40
Blink 84bede40
Count of free TD 0000005c
Used transfer descriptor queue
Flink 84bec00c
Blink 84becbcc
Count of used TD 00000023
Stalled endpoint descriptor queue <<< ?
Flink 84bb46c4
Blink 84bb46c4
Running endpoint descriptor queue
Flink 84bb46cc
Blink 84bb46cc
Paused endpoint descriptor queue
Flink 84bb46d4
Blink 84bb46d4
ED lists will not be displayed
Frame Hight part bb9d8000
Available Bandwidth 00000064
Used Bandwidth 00000000
SO count 00000000
SO stall Frame 00000000
SO Limit Frame 00000000
SO Limit Hit 00000000
SO Stall Hit 00000000
UCM message mailbox 00000000
Vendor ID 00001033
Device ID 00000035

Jouk,

looks like the USB disks are not serviced by a disk class driver, so there are no CDDBs (and CDRPs ?).

So the question remains: where is the current IRP stuck in case of the problem ?

Regards,
Volker.

Jouk,

would you be willing to make a forced crashdump file available, which shows this situation. So that I can make an attempt to find the hanging IRP ?

Volker.

Volker Halle wrote:
>
> would you be willing to make a forced crashdump file available, which shows this situation. So that I can make an attempt to find the hanging IRP ?
>
I contacted you "offline"

Jouk,

when looking for the 'current' (hanging) DNA2: IRP address in all of nonpaged pool, I found only 2 occurences of this address:

- UCB$L_IRP in DNA2: UCB
- KPB$PS_IRP of a KPB (Kernel Process Block), which is in state EXE$KP_STALL_GENERAL

Although the SDA> USB SHOW commands in V8.4-2L2 are much more detailled than in previous versions, I can't find anything, which could have caused SYS$DNDRIVER to stall this KPB. DNDRIVER sources are not available.

So the only path to a possible resolution would be a support call to VSI.

Sorry,

Volker.