Public bug reported:
Hi,
I am using Ubuntu 14.04, ubuntu-14 3.13.0-74-generic #118-Ubuntu SMP Thu Dec 17
22:52:10 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux
also tried Linux dmy-ubuntu-14 4.3.3-040303-generic #201512150130 SMP Tue Dec
15 06:32:30 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux
same behaviour in both the kernels
For the past couple of days I am having problems related to libusb bulk
transfers ( async ). When the debug prints are enabled from libusb, I
could see more packets ( example 5 packets each 163840 bytes ) were
received by libusb. However, the callback was called only once ( example
only 163840 bytes ) were received by the application using libsub.
We are seeing this issue only past fews days and computers that has
latest kernel updates. For example computers that did not update or
kernel dated July 2015 or Jan 2105 are still working.
Today I upgraded Linux kernel to 3.19 to test but no difference. With
3.19 kernel I goet following message in dmesg,
[14720.137271] xhci_hcd 0000:08:00.0: ERROR Transfer event TRB DMA ptr not part
of current TD ep_index 4 comp_code 1 [14720.137282] xhci_hcd 0000:08:00.0:
Looking for event-dma 00000000b25bd400 trb-start 00000000b25bd2d0 trb-end
00000000b25bd360 seg-start 00000000b25bd000 seg-end 00000000b25bdff0
[14720.138698] xhci_hcd 0000:08:00.0: ERROR Transfer event TRB DMA ptr not part
of current TD ep_index 4 comp_code 1 [14720.138706] xhci_hcd 0000:08:00.0:
Looking for event-dma 00000000b25bd4a0 trb-start 00000000b25bd2d0 trb-end
00000000b25bd360 seg-start 00000000b25bd000 seg-end 00000000b25bdff0
[14720.140146] xhci_hcd 0000:08:00.0: ERROR Transfer event TRB DMA ptr not part
of current TD ep_index 4 comp_code 1 [14720.140155] xhci_hcd 0000:08:00.0:
Looking for event-dma 00000000b25bd540 trb-start 00000000b25bd2d0 trb-end
00000000b25bd360 seg-start 00000000b25bd000 seg-end 00000000b25bdff0
[14720.140188] xhci_hcd 0000:08:00.0: ERROR Transfer event TRB DMA ptr not part
of current TD ep_index 4 comp_code 13 [14720.140192] xhci_hcd 0000:08:00.0:
Looking for event-dma 00000000b25bd550 trb-start 00000000b25bd2d0 trb-end
00000000b25bd360 seg-start 00000000b25bd000 seg-end 00000000b25bdff0
Do you see any possible conflicts, with latest Kernels and libusb) (
esply for USB 3.0 ) Have anyone seen similar issues recently ?
This is easily reproducible.
Scenario that fails:
--------------------------
1. USB 3.0 device that can stream 16KB chunks.
2. Use libusb and submit more than 16K bulk read requests e.g. 160KB bulkread
request to libusb.
3. start the data transfer from USB 3.0 device and the transfer failes at the
end of getting data, i.e. when a short packet is transferred.
However, the same logic worked for kernel versions below 3.13.0-71 ( including
this version )
Scenario that works with new kernel ( but fails after sometime that might be
our device and stack through put issue )
--------------------------------------------------
1. in the above setup if I just reduce the bulk request size to 16KB instead of
160KB, then it works.
TIA.
/Dinesh
** Affects: linux (Ubuntu)
Importance: Undecided
Status: Incomplete
** Tags: trusty
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1542246
Title:
ERROR Transfer event TRB DMA ptr not part of current TD ep_index 4
comp_code 1
Status in linux package in Ubuntu:
Incomplete
Bug description:
Hi,
I am using Ubuntu 14.04, ubuntu-14 3.13.0-74-generic #118-Ubuntu SMP Thu Dec
17 22:52:10 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux
also tried Linux dmy-ubuntu-14 4.3.3-040303-generic #201512150130 SMP Tue Dec
15 06:32:30 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux
same behaviour in both the kernels
For the past couple of days I am having problems related to libusb
bulk transfers ( async ). When the debug prints are enabled from
libusb, I could see more packets ( example 5 packets each 163840 bytes
) were received by libusb. However, the callback was called only once
( example only 163840 bytes ) were received by the application using
libsub.
We are seeing this issue only past fews days and computers that has
latest kernel updates. For example computers that did not update or
kernel dated July 2015 or Jan 2105 are still working.
Today I upgraded Linux kernel to 3.19 to test but no difference. With
3.19 kernel I goet following message in dmesg,
[14720.137271] xhci_hcd 0000:08:00.0: ERROR Transfer event TRB DMA ptr not
part of current TD ep_index 4 comp_code 1 [14720.137282] xhci_hcd 0000:08:00.0:
Looking for event-dma 00000000b25bd400 trb-start 00000000b25bd2d0 trb-end
00000000b25bd360 seg-start 00000000b25bd000 seg-end 00000000b25bdff0
[14720.138698] xhci_hcd 0000:08:00.0: ERROR Transfer event TRB DMA ptr not
part of current TD ep_index 4 comp_code 1 [14720.138706] xhci_hcd 0000:08:00.0:
Looking for event-dma 00000000b25bd4a0 trb-start 00000000b25bd2d0 trb-end
00000000b25bd360 seg-start 00000000b25bd000 seg-end 00000000b25bdff0
[14720.140146] xhci_hcd 0000:08:00.0: ERROR Transfer event TRB DMA ptr not
part of current TD ep_index 4 comp_code 1 [14720.140155] xhci_hcd 0000:08:00.0:
Looking for event-dma 00000000b25bd540 trb-start 00000000b25bd2d0 trb-end
00000000b25bd360 seg-start 00000000b25bd000 seg-end 00000000b25bdff0
[14720.140188] xhci_hcd 0000:08:00.0: ERROR Transfer event TRB DMA ptr not
part of current TD ep_index 4 comp_code 13 [14720.140192] xhci_hcd
0000:08:00.0: Looking for event-dma 00000000b25bd550 trb-start 00000000b25bd2d0
trb-end 00000000b25bd360 seg-start 00000000b25bd000 seg-end 00000000b25bdff0
Do you see any possible conflicts, with latest Kernels and libusb) (
esply for USB 3.0 ) Have anyone seen similar issues recently ?
This is easily reproducible.
Scenario that fails:
--------------------------
1. USB 3.0 device that can stream 16KB chunks.
2. Use libusb and submit more than 16K bulk read requests e.g. 160KB bulkread
request to libusb.
3. start the data transfer from USB 3.0 device and the transfer failes at the
end of getting data, i.e. when a short packet is transferred.
However, the same logic worked for kernel versions below 3.13.0-71 (
including this version )
Scenario that works with new kernel ( but fails after sometime that might be
our device and stack through put issue )
--------------------------------------------------
1. in the above setup if I just reduce the bulk request size to 16KB instead
of 160KB, then it works.
TIA.
/Dinesh
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1542246/+subscriptions
--
Mailing list: https://launchpad.net/~kernel-packages
Post to : kernel-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kernel-packages
More help : https://help.launchpad.net/ListHelp
