Author: Alan Carvalho de Assis <acassis@gmail.com>
Run nxstyle on .c and .h files and fix it
Author: Alin Jerpelea <alin.jerpelea@sony.com>
drivers: usbdev: minor fix
drivers: usbdev: usbmsc full speed not available
Change transfer size to be based on maxpacket size.
drivers: usbdev: Fix string ID calculation
For *_STRBASE defines, it already unnecessary because composite device setup
has been changed, it would be calculated by *_composite.c in board sources.
drivers: usbdev: Fix invalid/unsupported command processing
Mass Storage Class shall stall when invalid or unsupported commands
has been recieved.
drivers: usbdev: Remove unnecessary reset logic
drivers: usbdev: Flags comparison fix
drivers: usbdev: Descriptor type mismatch fix when dual speed is enabled
Author: Alan Carvalho de Assis <acassis@gmail.com>
Run nxstyle against .c .h files and fix it
Author: Adam Feuer <adam@starcat.io>
Summary
Adds CDC ECM Ethernet over USB High Speed for SAMA5D36-Xplained
(and maybe other boards) (most of the code was there already,
but didn't work out of the box for the SAMA5D36-Xplained)
Only SAMA5D36-Xplained has been tested so far
Impact
None if you don't use the CDC ECM Ethernet driver
On SAMA5D36-Xplained, this adds high-speed Internet connectivity
over USB 2.0 High Speed. via the USB CDC ECM Gadget driver.
It may work on other boards too.
This also fixed full-speed (low-speed) mode for the board too.
Limitations
Hasn't been tested on anything other than SAMA5D36-Xplained board.
TODO
Ideally this would include a composite RNDIS device so it would
also work seamlessly on Windows. That is for a future PR
Ideally this would include software to help configuration via
mDNS/DNS-SD for plug and play compatibility with Linux and macOS.
That is for a future PR.
Detail
Only a few lines of C driver code needed to be changed, since the
capability was there already. The rest is config and documentation.
Changes the SAMA5D3-Xplained board bringup to match the SAMA5D3-EK
board bringup
A helper script to configure Linux routing and iptables NAT is also
provided, along with documentation on how to use it.
Testing
Manual, on a Ubuntu Linux 19.10 system and MacOS 10.14.6 Mojave
MacBook Pro.
How To Verify
Follow the new CDC ECM Ethernet over USB instructions in the board
README.txt file
Commits:
remove non-UTF-8 chars in comment and reformat
removed unneeded comment markers
instructions for using the defconfigs
removed EMAC from config
- to prove this example only needs the CDC ECM Ethernet over USB to work
added CDC-ECM Ethernet over USB info to README
added U-Boot image
added netusb helper script
- this can configure the Linux network interface and routes
so you can ping or access the NuttX system via TCP/IP.
renamed defconfig dirs to be ethernet-over-usb
- was usb-over-ethernet which is not right
added USB DMA to defconfigs
updated readme with autoboot and debugging info
bringing ethernet-over-usb examples into parity
added cdc ecm ethernet over usb with telnetd config
added defconfig
only use phy interrupt if netdevices is ethernet
- because now netdevice could be CDC ECM ethernet over usb
which has no PHY interrupt
add bringup to Makefile
add bringup
app init cleanup
init cdc ecm driver and rndis driver; some cleanup
fixed some typos and odd characters
usb over ethernet working over usb 2.0 hs
* Simplify EINTR/ECANCEL error handling
1. Add semaphore uninterruptible wait function
2 .Replace semaphore wait loop with a single uninterruptible wait
3. Replace all sem_xxx to nxsem_xxx
* Unify the void cast usage
1. Remove void cast for function because many place ignore the returned value witout cast
2. Replace void cast for variable with UNUSED macro
Squashed commit of the following:
Author: Gregory Nutt <gnutt@nuttx.org>
Ran nxstyle against many of the affected files. But this job was too big for today. Many of the network drivers under arch are highly non-compiant and generate many, many faults from nxstyle. Those will have to be visited again another day.
Author: Xiang Xiao <xiaoxiang@xiaomi.com>
This effects all network drivers as well as timing related portions of net/: devif_poll_tcp_timer shouldn't be skipped in the multiple card case. devif_timer will be called multiple time in one period if the multiple card exist, the elapsed time calculated for the first callback is right, but the flowing callback in the same period is wrong(very short) because the global variable g_polltimer is used in the calculation. So let's pass the delay time to devif_timer and remove g_polltimer.
drivers: usbdev: usbmsc: fix usbmsc_exportluns declaration
* drivers: usbdev: usbmsc: fix usbmsc_exportluns declaration
fixes the following error
apps/system/usbmsc/usbmsc_main.c:567:
undefined reference to `usbmsc_exportluns'
Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
* configs: spresense: nsh: register the procfs
this is a configuration change in sync with the other ones
Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
* configs: spresense: enable READLINE_CMD_HISTORY
we are following the default spresense board configuration
regarding CMD HISTORY configuration
Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
* configs: spresense: enable by default SPI 4 and 5
we are following the default spresense board configuration
regarding SPI configuration
Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
* configs: spresense: add I2C configuration
in sync with the default spresense board configuration
Signed-off-by: Alin Jerpelea <alin.jerpelea@sony.com>
Approved-by: Gregory Nutt <gnutt@nuttx.org>
Initial OTG support for STM32H7 and some minor improvements
arch/arm/src/stm32h7/stm32h7x3xx_rcc.c: add support for HSI48 clock
configs/nucleo-f207zg: add support for USB and initialize CDCACM if configured
configs/nucleo-h743zi: add support for USB and initialize CDCACM if configured
drivers/usbdev/Kconfig: fix type for USBDEV_TRACE_INITIALIDSET
configs/teensy-3.x/usbnsh/defconfig: update config according to change in USBDEV_TRACE_INITIALIDSET
Approved-by: Gregory Nutt <gnutt@nuttx.org>
1. lost data when receiving buffer is full;
2. low-water mask implement issue;
3. re-flush cdc buffer when enabling
4. serial dma is conflict with cdc , modify the serial.h
RNDIS composite support
* NuttX usb/composite.h: Forward-declare composite_devdesc_s.
This avoids "error: conflicting types for 'composite_initialize'"
on some versions of GCC. Because of the cross-inclusion between
usbdev.h and composite.h, the full declaration is not always
available.
* NuttX: USB Composite driver: Fix strid comparison
The last string ID used by composite driver is 4, and
the number of IDs used is 5 (0..4). The comparison
strid <= COMPOSITE_NSTRIDS caused composite driver to
reply with -EINVAL for id 5, even though it should be
available for subdevices to use.
* NuttX: RNDIS USB driver: Add support for composite configuration.
Approved-by: GregoryN <gnutt@nuttx.org>
These Microsoft-only descriptors help in loading the correct driver on Windows.
They are especially helpful to give libusb access to a custom device without
having to manually configure/install WinUSB driver.
With this change DFU interface works automatically on
Windows 10 with dfu-util 0.9 and libusb 1.0.22. On Windows 7
it still appears to need driver installation.
drivers/usbdev: Fix buffer overrun check in rndis.c
The rndis driver has been working since 13 Nov 2017.
However, I finally found that it depends on network and buffer
configurations. If a receiving TCP packet is devided into smaller
ones based on USB max packet size, this condition check works
correctly.
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
Approved-by: GregoryN <gnutt@nuttx.org>
This makes the user interface a little hostile. People thing of an MTU of 1500 bytes, but the corresponding packet is really 1514 bytes (including the 14 byte Ethernet header). A more friendly solution would configure the MTU (as before), but then derive the packet buffer size by adding the MAC header length. Instead, we define the packet buffer size then derive the MTU.
The MTU is not common currency in networking. On the wire, the only real issue is the MSS which is derived from MTU by subtracting the IP header and TCP header sizes (for the case of TCP). Now it is derived for the PKTSIZE by subtracting the IP header, the TCP header, and the MAC header sizes. So we should be all good and without the recurring 14 byte error in MTU's and MSS's.
Squashed commit of the following:
Trivial update to fix some spacing issues.
net/: Rename several macros containing _MTU to _PKTSIZE.
net/: Rename CONFIG_NET_SLIP_MTU to CONFIG_NET_SLIP_PKTSIZE and similarly for CONFIG_NET_TUN_MTU. These are not the MTU which does not include the size of the link layer header. These are the full size of the packet buffer memory (minus any GUARD bytes).
net/: Rename CONFIG_NET_6LOWPAN_MTU to CONFIG_NET_6LOWPAN_PKTSIZE and similarly for CONFIG_NET_TUN_MTU. These are not the MTU which does not include the size of the link layer header. These are the full size of the packet buffer memory (minus any GUARD bytes).
net/: Rename CONFIG_NET_ETH_MTU to CONFIG_NET_ETH_PKTSIZE. This is not the MTU which does not include the size of the link layer header. This is the full size of the packet buffer memory (minus any GUARD bytes).
net/: Rename the file d_mtu in the network driver structure to d_pktsize. That value saved there is not the MTU. The packetsize is the memory large enough to hold the maximum packet PLUS the size of the link layer header. The MTU does not include the link layer header.
RNDIS support on STM32F4Discovery
* stm32f4discovery: Add stm32_netinit.c to avoid a compilation error
* stm32f4discovery: Add rndis initialization in stm32_bringup.c
NOTE: MAC address for the host side starts 0xaa. This assignment
scheme should be fixed later.
* stm32f4discovery: Add rndis configuration
NOTE: STM32F4Discovery + DM-STF4BB
* drivers/usbdev/rndis.c: Fix some issues in rndis.c
Introduce rndis_transmit() and change rndis_rxdispatch() to avoid
packet corruption. Introduce max packet size for dual speed which
is mainly used for high speed mode. Fix adjusting MTU warning
on Linux host. Fix data corruption if a packet size excceds MTU.
NOTE: Max packet size is not configured dynamically. This should
be fixed in the future version.
* stm32f4discovery: Modify #ifdef condition for rndis in stm32_bringup()
* drivers/usbdev/rndis.c: Change HPWORK to ETHWORK
* drivers/usbdev/rndis.c: Merge the commit 07b98ccbb5
Previous commit b09365784a was based on old rndis.c thus resulted
in reverting the commit 07b98ccbb5. This change merges the commit.
NOTE: In the commit 07b98ccbb5, max packet size of bulkout was
assumed to be 64. In this commit, priv->epbulkout->maxpacket is
used instead.
Approved-by: Gregory Nutt <gnutt@nuttx.org>
This commit backs out most of commit b4747286b1. That change was added because sem_wait() would sometimes cause cancellation points inappropriated. But with these recent changes, nxsem_wait() is used instead and it is not a cancellation point.
In the OS, all calls to sem_wait() changed to nxsem_wait(). nxsem_wait() does not return errors via errno so each place where nxsem_wait() is now called must not examine the errno variable.
In all OS functions (not libraries), change sem_wait() to nxsem_wait(). This will prevent the OS from creating bogus cancellation points and from modifying the per-task errno variable.
sched/semaphore: Add the function nxsem_wait(). This is a new internal OS interface. It is functionally equivalent to sem_wait() except that (1) it is not a cancellation point, and (2) it does not set the per-thread errno value on return.
sched/semaphore: Add nxsem_post() which is identical to sem_post() except that it never modifies the errno variable. Changed all references to sem_post in the OS to nxsem_post().
sched/semaphore: Add nxsem_destroy() which is identical to sem_destroy() except that it never modifies the errno variable. Changed all references to sem_destroy() in the OS to nxsem_destroy().
libc/semaphore and sched/semaphore: Add nxsem_getprotocol() and nxsem_setprotocola which are identical to sem_getprotocol() and set_setprotocol() except that they never modifies the errno variable. Changed all references to sem_setprotocol in the OS to nxsem_setprotocol(). sem_getprotocol() was not used in the OS
libc/semaphore: Add nxsem_getvalue() which is identical to sem_getvalue() except that it never modifies the errno variable. Changed all references to sem_getvalue in the OS to nxsem_getvalue().
sched/semaphore: Rename all internal private functions from sem_xyz to nxsem_xyz. The sem_ prefix is (will be) reserved only for the application semaphore interfaces.
libc/semaphore: Add nxsem_init() which is identical to sem_init() except that it never modifies the errno variable. Changed all references to sem_init in the OS to nxsem_init().
sched/semaphore: Rename sem_tickwait() to nxsem_tickwait() so that it is clear this is an internal OS function.
sched/semaphoate: Rename sem_reset() to nxsem_reset() so that it is clear this is an internal OS function.
drivers/usbdev/cdcacm.c: Change design for queuing RX packets that cannot be processed. Previous design had a logic problem that could cause data loss.
drivers/usbdev/cdcacm: Fixes one of two know design issues.
drivers/usbdev/cdcacm: First attempt to plug data leak in input flow control design. Still missing a few things.
commit 69fcf3e849
Author: Alan Carvalho de Assis <acassis@gmail.com>
Date: Sun Jul 16 08:39:33 2017 -0600
Fix spark/stm32_composite.c: board_composite_connect cannot be static
commit 28eb253401
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Sun Jul 16 08:36:01 2017 -0600
Composite: Final review for coding style before merge
commit e6af1b9994
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Sun Jul 16 07:41:38 2017 -0600
Composite: Simplify some intiialization of data structures.
commit 771c367411
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Sun Jul 16 07:15:08 2017 -0600
Cosmetic changes to alignment.
commit 5d67ddda4e
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Sun Jul 16 07:00:48 2017 -0600
USBMSC: Add missing logic to saved device description.
commit 0729151d29
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Sat Jul 15 17:11:41 2017 -0600
Trivial, costmetic
commit 74b916ff84
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Sat Jul 15 14:50:29 2017 -0600
Composite: Private functions need to be marked static. Move static functions out of 'Public Functions' to 'Private Functions' where they belong. Disable composite configuration 1 in all STM32 F1 configurations.
commit cfaa4ece13
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Sat Jul 15 13:20:34 2017 -0600
Add some comments.
commit 8143563be6
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Sat Jul 15 12:33:10 2017 -0600
Spark: Need to condition out MSC logic in composite setup if there is no MSC in the composite.
commit 69d3a91ef1
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Sat Jul 15 11:03:33 2017 -0600
Composite: Remove all dependencies on CONFIG_SYSTEM_COMPOSITE_* configuration settings. Nothing in the OS can depend on external application settings.
commit 55a4388bbd
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Sat Jul 15 09:59:31 2017 -0600
All composite configurations now also support a dual CDC/ACM configuration.
commit 428f2147af
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Sat Jul 15 09:24:32 2017 -0600
Composite: Move board_msc* interfaces from apps/system/composite to the board specific OS logic where they belong.
commit f1cc168a5c
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Sat Jul 15 08:56:33 2017 -0600
Refresh all composite configurations.
commit 246afcaa10
Merge: 919877191d02c6672868
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Sat Jul 15 08:22:26 2017 -0600
Merge remote-tracking branch 'origin/master' into composite
commit 919877191d
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Sat Jul 15 08:20:02 2017 -0600
Composite setup: Remove useless board_cdc* wrapper.
commit 82129cf8c6
Merge: f2cb8b252a6537e4ea20
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Fri Jul 14 16:23:57 2017 -0600
Merge remote-tracking branch 'origin/master' into composite
commit f2cb8b252a
Author: Alan Carvalho de Assis <acassis@gmail.com>
Date: Fri Jul 14 10:19:35 2017 -0600
Composite: Fix for another cloned typo.
commit 676cfd526a
Author: Alan Carvalho de Assis <acassis@gmail.com>
Date: Fri Jul 14 09:11:37 2017 -0600
Composite: Fix some typos
commit 1ea0368c18
Author: Alan Carvalho de Assis <acassis@gmail.com>
Date: Fri Jul 14 09:10:18 2017 -0600
Composite: ./stm3210e-eval/src/stm32_composite.c
commit e485caced9
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Fri Jul 14 09:08:17 2017 -0600
Composite: I don't think the original code should have forced minor=0.
commit 6443c29621
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Fri Jul 14 07:15:38 2017 -0600
Composite: Flesh out support for all of other configurations that support composite.
commit 23cbc28b05
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Fri Jul 14 06:59:45 2017 -0600
Detangle use of board_xyzclassobject() and board_xyzuninitialize()
commit 1674cb8c8e
Merge: 6bc881a1925033a6def7
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Thu Jul 13 13:57:40 2017 -0600
Merge remote-tracking branch 'origin/master' into composite
commit 6bc881a192
Merge: fe3af4941d85b8d16d8c
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Tue Jul 11 12:24:07 2017 -0600
Merge remote-tracking branch 'origin/master' into composite
commit fe3af4941d
Merge: 0f9ad16e181bc0eea143
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Mon Jul 10 11:07:36 2017 -0600
Merge remote-tracking branch 'origin/master' into composite
commit 0f9ad16e18
Merge: a4cd90d4efaa2e9c15a5
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Fri Jul 7 20:26:53 2017 -0600
Merge remote-tracking branch 'origin/master' into composite
commit a4cd90d4ef
Merge: 8a4be7175e31f832d8c5
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Wed Jul 5 11:12:52 2017 -0600
Merge remote-tracking branch 'origin/master' into composite
commit 8a4be7175e
Merge: 18a32ed2caae1771454a
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Fri Jun 30 16:14:04 2017 -0600
Merge remote-tracking branch 'origin/master' into composite
commit 18a32ed2ca
Merge: aaa81ce4976d8df90b79
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Thu Jun 29 10:18:16 2017 -0600
Merge remote-tracking branch 'origin/master' into composite
commit aaa81ce497
Merge: 4eb548226b8cb4636bb1
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Mon Jun 26 11:56:11 2017 -0600
Merge remote-tracking branch 'origin/master' into composite
commit 4eb548226b
Merge: 2327f5a1b4dc8eec0b61
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Mon Jun 19 17:27:00 2017 -0600
Merge remote-tracking branch 'origin/master' into composite
commit 2327f5a1b4
Merge: 49cd279fc699bf0b522b
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Fri Jun 16 17:30:03 2017 -0600
Merge remote-tracking branch 'origin/master' into composite
commit 49cd279fc6
Merge: bb6a13f30a46f86982ee
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Wed Jun 14 09:17:49 2017 -0600
Merge remote-tracking branch 'origin/master' into composite
commit bb6a13f30a
Merge: 918480047aac93d4bda9
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Mon Jun 5 17:40:06 2017 -0600
Merge remote-tracking branch 'origin/master' into composite
commit 918480047a
Merge: e4d262436c4526cd665e
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Sat Jun 3 08:52:31 2017 -0600
Merge remote-tracking branch 'origin/master' into composite
commit e4d262436c
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Fri Jun 2 07:19:27 2017 -0600
SAMV71-Xult Composite: Now can switch between two different composite configurations dynamically.
commit 815257743d
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Fri Jun 2 07:11:57 2017 -0600
usbdev composite and SAMV7-Xult: Move board-specific USB composite configuration out of boardctl.c and into board-specific logic where it belongs. Add a configuration option to the boardctl() calls to support multiple composite device configurations dynamically.
commit ac13619dc5
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Thu Jun 1 17:03:58 2017 -0600
Cosmetic
commit 9dd41bdd2f
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Thu Jun 1 17:02:43 2017 -0600
Composite: More compile-related fixes
commit fc1438c95d
Merge: 049ccbfcbeff2b54a5e0
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Thu Jun 1 16:35:41 2017 -0600
Merge remote-tracking branch 'origin/master' into composite
commit 049ccbfcbe
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Thu Jun 1 16:35:16 2017 -0600
Composite: Add some structure definitions missed in first application of the patch.
commit ef33329e3a
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Thu Jun 1 16:14:46 2017 -0600
Add a warning
commit 89f77cd91a
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Thu Jun 1 16:11:27 2017 -0600
Fix some incomplete name changes
commit 0bb7af549a
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Thu Jun 1 15:09:50 2017 -0600
It is unnecessary to pack a structure that consists only of uint8_t data fields.
commit bd9b548914
Author: Gregory Nutt <gnutt@nuttx.org>
Date: Thu Jun 1 15:05:41 2017 -0600
Remove COMPILE_TIME_ASSERTION
commit 7e6f481581
Author: Frank Benkert <Frank.Benkert@avat.de>
Date: Thu Jun 1 14:58:04 2017 -0600
Part II of the same big commit
commit dcc9b07715
Author: Frank Benkert <Frank.Benkert@avat.de>
Date: Thu Jun 1 14:08:22 2017 -0600
[[This is part 1 or several commits]]
We developed a huge Changeset over a year ago to make USB Composite configuration dynamical and be able to instanciate the CDCACM multiple times inside this device. We use this feature to switch between one in normal and up to three CDCACMs in maintenance boot.
The control path starts in “boardctl.c” where the configuration for the device is constructed. There are still a few issues which I’ll ask you to have a look at before this beast can be merged.
1. To be able to construct the data dynamically I have changed some USB-Structs to be packed. Maybe there are additional structs to change (just for completeness – not for current functionality).
2. I’ve added the Macro “COMPILE_TIME_ASSERTION” two times (in usbmsc_desc.c and in cdcacm_desc.c) to stay private. Maybe you’ll find a better place. It’s used to check the size of the structs against the assumptions.
3. I’ve changed the interface for some USB-Functions to receive also the dynamic configuration. Maybe this can be done more elegant.
4. The original NuttX (without the patch) seems to have problems with a Composite device holding a CDCACM and an MSC. The “USB SET CONFIGURATION” request does not to work at all. This makes the test fail under Windows and under Linux. Applying this patch doesn’t change anything – because it only changes the configuration – not the behavior. Maybe you’ll have a look at this problem before applying the patch.
The scsi thread was waiting for the wrong condition.
However, this was masked by the fact that the code creating the scsi thread
was also holding usbmsc_scsi_lock(priv) while initializing data, hence this
lock synchronized the scsi thread start with init completion.
bulk IN endpoint before the loop that frees the requests via cdcasm_freereq.
That function checks the parameters and skips the freeing if either is NULL.
Freeing the bulk IN enpoint will cause the first param to be NULL, thereby
bypassing the free operation. To fix, I moved the release of the bulk IN
endpoint until after to loop (much as was the case for the OUT and read
requests, which did not exhibit the problem).
