The NI USB adapter sets the END bit in the status word when an error
occurs such as a read being interrupted by the setting of ATN. This
happens for example when a device clear is received from the
controller in charge during a read.
The common driver changes the error return to 0 whenever the END bit
is set in order to avoid errors such as timeout or interrupt to be
reported after the full message has actually been read. The behaviour
of the NI USB adapter in setting the END bit on errors was causing
actual errors (-EINTR, -ETIMEDOUT) not to be reported.
We avoid setting the END bit in the ni_usb_gpib driver when an error
is reported in error_code of the status from the adaptor.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://patch.msgid.link/20260422074807.3194-1-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The following errors were reported for a s390 randconfig build
of the fluke gpib driver:
>> drivers/gpib/eastwood/fluke_gpib.c:1002:23: error: call to undeclared function 'ioremap'; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration]
1002 | nec_priv->mmiobase = ioremap(e_priv->gpib_iomem_res->start,
| ^
>> drivers/gpib/eastwood/fluke_gpib.c:1002:21: error: incompatible integer to pointer conversion assigning to 'void *' from 'int' [-Wint-conversion]
1002 | nec_priv->mmiobase = ioremap(e_priv->gpib_iomem_res->start,
| ^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1003 | resource_size(e_priv->gpib_iomem_res));
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpib/eastwood/fluke_gpib.c:1036:33: error: incompatible integer to pointer conversion assigning to 'void *' from 'int' [-Wint-conversion]
1036 | e_priv->write_transfer_counter = ioremap(e_priv->write_transfer_counter_res->start,
| ^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1037 | resource_size(e_priv->write_transfer_counter_res));
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Add HAS_IOMEM dependency to Kconfig for fluke driver option
Suggested-by: Arnd Bergmann <arnd@arndb.de>
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202601221748.AFAqHieJ-lkp@intel.com/
Fixes: baf8855c91 ("staging: gpib: fix address space mixup")
Cc: stable <stable@kernel.org>
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://patch.msgid.link/20260202094755.4259-1-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Move the gpib drivers out of staging and into the "real" part of the
kernel. This entails:
- Remove the gpib Kconfig menu and Makefile build rule from staging.
- Remove gpib/uapi from the header file search path in subdir-ccflags
of the gpib Makefile
- move the gpib/uapi files to include/uapi/linux
- Move the gpib tree out of staging to drivers.
- Remove the word "Linux" from the gpib Kconfig file.
- Add the gpib Kconfig menu and Makefile build rule to drivers
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://patch.msgid.link/20251117144021.23569-5-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When the ATN (Attention) line is asserted during a read we get a
NIUSB_ATN_STATE_ERROR during a read. For the controller to send a
device clear it asserts ATN. Normally this is an error but in the case
of a device clear it should be regarded as an interrupt.
Return -EINTR when the Device Clear Active State (DCAS) is entered
else signal an error with dev_dbg with status instead of just dev_err.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Cc: stable <stable@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This driver was not sending device clear or trigger events when the
board entered the DCAS or DTAS state respectively in device mode.
DCAS is the Device Clear Active State which is entered on receiving a
selective device clear message (SDC) or universal device clear message
(DCL) from the controller in charge.
DTAS is the Device Trigger Active State which is entered on receiving
a group execute trigger (GET) message from the controller.
In order for an application, implementing a particular device, to
detect when one of these states is entered the driver needs to send
the appropriate event.
Send the appropriate gpib_event when DCAS or DTAS is set in the
reported status word. This sets the DCAS or DTAS bits in the board's
status word which can be monitored by the application.
Fixes: 4e127de14f ("staging: gpib: Add National Instruments USB GPIB driver")
Cc: stable <stable@kernel.org>
Tested-by: Dave Penkler <dpenkler@gmail.com>
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
EOI (End Or Identify) is a hardware line on the GPIB bus that can be
asserted with the last byte of a message to indicate the end of the
transfer to the receiving device.
In this driver, a write with send_eoi true is done in 3 parts:
Send first byte directly
Send remaining but 1 bytes using the fifo
Send the last byte directly with EOI asserted
The first byte in a write is always sent by writing to the tms9914
chip directly to setup for the subsequent fifo transfer. We were not
checking for a 1 byte write with send_eoi true resulting in EOI not
being asserted. Since the fifo transfer was not executed
(fifotransfersize == 0) the retval in the test after the fifo transfer
code was still 1 from the preceding direct write. This caused it to
return without executing the final direct write which would have sent
an unsollicited extra byte.
For a 2 byte message the first byte was sent directly. But since the
fifo transfer was not executed (fifotransfersize == 1) and the retval
in the test after the fifo transfer code was still 1 from the
preceding first byte write it returned before the final direct byte
write with send_eoi true. The second byte was then sent as a separate
1 byte write to complete the 2 byte write count again without EOI
being asserted as above.
Only send the first byte directly if more than 1 byte is to be
transferred with send_eoi true.
Also check for retval < 0 for the error return in case the fifo code
is not used (1 or 2 byte message with send_eoi true).
Fixes: 09a4655ee1 ("staging: gpib: Add HP/Agilent/Keysight 8235xx PCI GPIB driver")
Cc: stable <stable@kernel.org>
Tested-by: Dave Penkler <dpenkler@gmail.com>
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Ensure that all structs have a size that is an integral mutliple
of 32 bits.
Change trailing single bit field to __u32 in gpib_open_dev_ioctl
Add byte padding to gpib_serial_poll_ioctl and gpib_request_service2
Add bit padding to gpib_board_info_ioctl and gpib_ppoll_config_ioctl
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250603184320.30594-5-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The usbtmc488_ioctl_read_stb function relied on a positive return from
usbtmc_get_stb to reset the srq condition in the driver. The
USBTMC_IOCTL_GET_STB case tested for a positive return to return the stb
to the user.
Commit: <cac01bd178d6> ("usb: usbtmc: Fix erroneous get_stb ioctl
error returns") changed the return value of usbtmc_get_stb to 0 on
success instead of returning the value of usb_control_msg which is
positive in the normal case. This change caused the function
usbtmc488_ioctl_read_stb and the USBTMC_IOCTL_GET_STB ioctl to no
longer function correctly.
Change the test in usbtmc488_ioctl_read_stb to test for failure
first and return the failure code immediately.
Change the test for the USBTMC_IOCTL_GET_STB ioctl to test for 0
instead of a positive value.
Fixes: cac01bd178 ("usb: usbtmc: Fix erroneous get_stb ioctl error returns")
Cc: stable@vger.kernel.org
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250521121656.18174-3-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
GPIB secondary addresses have valid values between 0 and 31
inclusive. The Make Secondary Address function MSA, used to form
the protocol byte, was using the gpib_address_restrict function
erroneously restricting the address range to 0 through 30.
Remove the call to gpib_address_restrict and simply trim the
address to 5 bits.
Fixes: 2da03e7e31 ("staging: gpib: Add user api include files")
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250520155100.5808-1-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
wait_event_interruptible_timeout returns a long
The return was being assigned to an int causing an integer overflow when
the remaining jiffies > INT_MAX resulting in random error returns.
Use a long return value and convert to int ioctl return only on error.
When the return value of wait_event_interruptible_timeout was <= INT_MAX
the number of remaining jiffies was returned which has no meaning for the
user. Return 0 on success.
Reported-by: Michael Katzmann <vk2bea@gmail.com>
Fixes: dbf3e7f654 ("Implement an ioctl to support the USMTMC-USB488 READ_STATUS_BYTE operation.")
Cc: stable@vger.kernel.org
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250502070941.31819-2-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This addresses warnings produced by make W=1 with the configuration
parameter CONFIG_GPIB_PCMCIA=y
cb7210/cb7210.c:1251:28: warning: variable 'dev' set but not used [-Wunused-but-set-variable]
cb7210/cb7210.c:1250:31: warning: variable 'handle' set but not used [-Wunused-but-set-variable]
Remove the declarations and assignments of the unused variables.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250427093123.18565-1-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The IEEE-488 GPIB standard was designed to ensure that there is only ever
one controller-in-charge on the bus at any one time. If a board becomes
controller-in-charge on request_system_control there is no way to ensure
that there is not another board also acting as controller-in-charge.
This can lead to bus conflicts and hangs.
Remove the setting of controller-in-charge from request_system_control.
Fixes: fce79512a9 ("staging: gpib: Add LPVO DIY USB GPIB driver")
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250426150319.5580-1-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The yoga pin map is compatible with the others so that its led
can be used as the activity led. Using the LED subsytem required
adding a dtoverlay to the boot config as well as setting up the
activity led via sysfs. To simplify the setup we remove the
dependency on the LED subsystem and use the on board led as for
the other pin maps.
Tested-by: Dave Penkler <dpenkler@gmail.com>
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250418173434.2892-5-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The gpio board can only act as system controller.
By the IEEE488.1 standard a system controller becomes
controller-in-charge when it asserts the interface-clear control
line.
Remove the setting of controller-in-charge from bb_take_control
and move it to bb_interface_clear.
Fixes: 4cd654f847 ("staging: gpib: Add gpio bitbang GPIB driver")
Tested-by: Dave Penkler <dpenkler@gmail.com>
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250418173434.2892-3-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A number of drivers are unable to release control due to hardware or
software limitations. As request_system_control was defined as void,
no error could be signalled.
This patch changes the prototype of request_system_control to int and
adds the appropriate checking and returns. In the case that a board
cannot release control EINVAL is returned. If a driver does not
implement request_system_control EPERM is returned.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250418133537.22491-1-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
User code includes gpib_user.h. Since this include has diverged
from the original by
- removing unused functions and defines
- changing camel-case identifiers
- removing typedefs
we need to change the name of the kernel include. This include
will be included in the userland gpib_user.h ensuring backward
compatibility for application programmes.
Rename the file and change the references to it.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250418093111.8820-3-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If the agilent usb dongle is disconnected subsequent calls to the
driver cause a NULL dereference Oops as the bus_interface
is set to NULL on disconnect.
This problem was introduced by setting usb_dev from the bus_interface
for dev_xxx messages.
Previously bus_interface was checked for NULL only in the functions
directly calling usb_fill_bulk_urb or usb_control_msg.
Check for valid bus_interface on all interface entry points
and return -ENODEV if it is NULL.
Fixes: fbae7090f3 ("staging: gpib: Update messaging and usb_device refs in agilent_usb")
Cc: stable <stable@kernel.org>
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250222204515.5104-1-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If the usb dongle is disconnected subsequent calls to the
driver cause a NULL dereference Oops as the bus_interface
is set to NULL on disconnect.
This problem was introduced by setting usb_dev from the bus_interface
for dev_xxx messages.
Previously bus_interface was checked for NULL only in the the functions
directly calling usb_fill_bulk_urb or usb_control_msg.
Check for valid bus_interface on all interface entry points
and return -ENODEV if it is NULL.
Fixes: 4934b98bb2 ("staging: gpib: Update messaging and usb_device refs in ni_usb")
Cc: stable <stable@kernel.org>
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250222165817.12856-1-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Enable module name to be printed in pr_xxx and dev_xxx
Use DRV_NAME defined as KBUILD_MODNAME instead of hard coded
string "tnt4882_gpib" in pci_driver struct, request_region and
request_irq.
Remove unsupported chipset pr_err's
Remove messages on interrupted or timed out reads and writes.
Remove board not found messages and return -ENODEV
Remove "tnt4882:" prefix in messages as it will be printed by
pr_fmt and dev_fmt.
Change pr_err to dev_err where possible.
Remove irq and chipset pr_info's.
Replace error messages with appropriate error returns.
Remove call to mite_list_devices() and the function in mite.c
Remove PCMCIA debug comments, PCMCIA_DEBUG conditional compilation
symbol, the DEBUG macro definition and its uses.
Remove pr_info's in mite.c
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250220090920.32497-3-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Enable module name to be printed in pr_xxx and dev_xxx
Use DRV_NAME defined as KBUILD_MODNAME instead of hard coded
string "ines_gpib" in pci_request_regions, request_irq and
request_region.
Remove "ines:" and "ines_gpib:" string prefixes in messages
since module name printing is enabled.
Change pr_err to dev_err where possible.
Remove interrupt warnings.
Remove PCMCIA debug comments, PCMCIA_DEBUG conditional compilation
symbol, the DEBUG macro definition and its uses.
Change pr_debug to dev_dbg.
Remove pr_info
Remove commented printk.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250220090920.32497-2-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Enable module name to be printed in pr_xxx and dev_xxx
Use DRV_NAME defined as KBUILD_MODNAME instead of hard coded
string "cb7210" everywhere.
Remove "cb7210:" string prefix in messages since module name
printing is enabled.
Change pr_err to dev_err where possible.
Remove interrupt warnings.
Return consistent error codes with error messages:
-EBUSY when resources are busy
-ENODEV when device is not present
-EIO for others.
Return -ENOMEM for failed kmalloc (no message in driver)
Remove PCMCIA debug comments, PCMCIA_DEBUG conditional compilation
symbol, the DEBUG macro definition and its uses.
Change pr_warn to dev_warn and pr_err to dev_err where possible.
Remove commented printk.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250220090920.32497-1-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The drivers supporting the PCMCIA bus were using an intermediate
symbol GPIB_PCMCIA for compiling optional PCMCIA support.
This symbol is no longer needed for the in-tree drivers as the
Kconfig symbol is available.
Use the Kconfig symbol CONFIG_GPIB_PCMCIA directly for conditional
compilation of PCMCIA support.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250210151022.4358-2-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Enable module name to be printed in pr_xxx and dev_xxx
Remove pr_err on go_to_standby timeout.
Remove write wait and command wait interrupted messages.
Remove __func__ parameter on error messages
Change pr_err to dev_err where possible.
Remove commented printk
Uncomment dev_dbg in interrupt_have status
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250214114708.28947-17-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Enable module name to be printed in pr_xxx and dev_xxx
Use DRV_NAME defined as KBUILD_MODNAME instead of hard coded
string "ni_usb_gpib" in usb_driver struct.
Remove __func__ parameter from pr_err and dev_err.
Remove __func__ parameter from dev_dbg as this can be
enabled by dynamic debug.
Remove commented printk's and dev_err's.
Remove kmalloc failed messages.
Remove buffer over run bug dev_err message as this just checks
for a bug in the driver which does not exist.
Remove read/write length too long messages and return -EINVAL
Change dev_info to dev_dbg where possible.
Move attach message to the end of attach function.
Remove buffer overrun message. Use actual array indeces
instead of i and i++ to make code clear and check redundnant.
Remove module init and exit pr_info's.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250214114708.28947-15-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Enable module name to be printed by default in dev_dbg
Remove commented dev_dbg
Remove pr_err on timeout and gpib bus error.
Remove dev_dbg on wait interrupted, command timeout
also read / write timeout, gpib bus error and interrupt.
Remove commented printk and command variable initialization.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250214114708.28947-14-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Enable module name to be printed in pr_xxx and dev_xxx
Replace #define NAME "lpvo_usb_gpib" with KBUILD_MODNAME
Update comments about Diagnostics and Debug
Use dev_dbg instead of pr_alert in the DIA_LOG macro definition
Remove TTY_LOG macro and its uses. There are non-console applications
where this no longer makes sense.
Remove commented printk's
Remove now useless function printable().
Remove board parameter from SHOW_STATUS as it shadows
the variable in the function.
Remove the printing of the board pointer in SHOW_STATUS.
Change select DIA_LOG(0,... to dev_err where we need an error
message.
Remove loops for printing message buffer contents.
Remove dev_alerts for read errors.
Change dev_alert to dev_err.
Change some TTY_LOG to DIA_LOG(0,.. e.g. for attach and detach
messages.
Remove NAME parameter in dev_dbg as this is printed
with the module name.
Remove __func__ parameter in dev_dbg as this can be
enabled by dynamic debug.
Remove NOP message for unsuppoted functionality.
Remove "lpvo_usb_gpib:" prefix in register driver pr_err as it
is printed with the module name.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250214114708.28947-13-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Enable module name to be printed in pr_xxx and dev_xxx
Use DRV_NAME defined as KBUILD_MODNAME instead of hard coded
string "hp82341" in request_region and request_irq
Change pr_err to dev_err wherever possible
Remove pr_warn and pr_debug for timed out or interrupted
reads and writes.
Remove "hp_82341:" prefix in messages since this is
printed with the module name.
Remove __func__ parameter in pr_err.
Remove pr_info's.
Remove cpmmented printk's.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250214114708.28947-11-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Enable module name to be printed in pr_xxx and dev_xxx
Use DRV_NAME defined as KBUILD_MODNAME instead of hard coded
string "hp82335" in request_irq
Change pr_err to dev_err wherever possible
Remove pr_info's
Remove "hp83335:" prefix in messages since this is
printed with the module name.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250214114708.28947-10-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Enable module name to be printed in pr_xxx and dev_xxx
Change pr_info in the dbg_printk macro to dev_dbg.
In order for dbg_printk macro to have the board variable
defined the signatures and calls to bb_buffer_print and
set_atn were changed to include board as a parameter.
Remove the #ifdef CONFIG_GPIB_DEBUG code.
Remove commented dbk_printk's.
Change dbg_printk(0, to dev_err where an error message is needed.
Remove dbg_printk for "not implemented" functions.
Remove "gpib_bitbang:" prefix in pr_err as it will be printed
with the module name.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250214114708.28947-9-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Enable module name to be printed in pr_xxx and dev_xxx
Use DRV_NAME defined as KBUILD_MODNAME instead of hard coded
string "fmh_gpib" in platform_driver and pci_driver structs.
Remove commented printk's.
Change pr_err to dev_err wherever possible
Remove "fmh_gpib_gpib:" prefix in messages since this is
printed with the module name.
Remove write interrupted dev_dbg messages.
Remove read wait interrupted pr_warn.
Change dev_notice attach to dev_dbg
Change dev_info to dev_dbg.
Correct dev_err message erroneously containing cb7210 identifier.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250214114708.28947-8-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Enable module name to be printed in pr_xxx and dev_xxx
Use DRV_NAME defined as KBUILD_MODNAME instead of hard coded
string "fluke_gpib" in platform_driver struct.
Remove commented printk's.
Change pr_err to dev_err wherever possible
Remove "fluke_gpib:" prefix in messages since this is
printed with the module name.
Correct dev_err message erroneously containing cb7210 identifier.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250214114708.28947-7-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Enable module name to be printed in pr_xxx and dev_xxx
Change pr_err to dev_err wherever possible.
Remove dev_dbg messages on entry to some functions.
Remove error messages where userland can figure out
what went wrong through errno.
Remove __func__ and pid parameters in dev_dbg messages as
these can be enabled by dynamic debug.
Remove minor and "gpib" from dev_err and dev_dbg messages
since this information is printed by the dev name.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250214114708.28947-6-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Enable module name to be printed in pr_xxx and dev_xxx
Use DRV_NAME defined as KBUILD_MODNAME instead of hard coded
string "cec_gpib" in pci_driver struct.
Remove "cec_gpib:" string prefix in messages since module name
printing is enabled.
Change pr_err to dev_err where possible.
Return consistent error codes with error messages:
-EBUSY when resources are busy
-ENODEV when device is not present
-EIO for others.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250214114708.28947-5-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Enable module name to be printed in pr_xxx and dev_xxx
Use DRV_NAME defined as KBUILD_MODNAME instead of hard coded
string in usb_driver struct.
Remove __func__ parameter in dev_dbg messages as this can be
enabled with dynamic debug.
Remove __func__ parameter in dev_err messages as they are not
needed. The module name is sufficient.
Change pr_info to dev_dbg where needed and remove the rest
where possible.
Remove test and error messages for buffer over run in write and
read_registers as these are just trying to catch bugs in the driver.
Remove agilent_82357a string prefix in error messages.
Return -EIO for too many reads in read_registers instead of
continuing after printing an error message.
Change pr_warn to dev_warn.
Remove warning on calls for unsupported functionality.
Remove test and message for buffer overflow in agilent_82357_init
and agilent_82357a_go_idle which are just checking for a driver bug.
Use actual indeces in the array instead of i and then incrementing
i so that the code is clear and there is no need to check for
overflow or print a message.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250214114708.28947-3-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove dev_err messages on interrupted or timed-out
reads and writes. User land code can figure out what
went wrong with the errno return.
Return -ENODEV instead of -1 on attach failure when
no board is found.
Delete commented out console messages.
Use module name in pr_xxx and dev_xxx messages.
Remove const char * definition of driver_name and
extern definition in .h file.
Use DRV_NAME defined as KBUILD_MODNAME instead.
Remove driver_name parameter and agilent_82350b string prefix
in dev_xxx messages.
Use DRV_NAME instead of driver_name in pci_request_regions.
Use DRV_NAME instead of hard coded string in the pci_driver struct.
Change select dev_info's to dev_dbg.
Change pr_err to dev_err where possible.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250214114708.28947-2-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The dependecies on !X86_PAE were introduced to fix an i386
build issue due to drivers casting resource_type_t to void *
commit 48e8a8160d ("staging: gpib: Fix i386 build issue")
Subsequently
commit baf8855c91 ("staging: gpib: fix address space mixup")
properly resolved these issues by fixing the code in the
drivers.
Delete the lines "depends on !X86_PAE"
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250204174254.16576-1-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch fixes the following compile warning:
drivers/staging/gpib/hp_82341/hp_82341.c: In function 'hp_82341_attach':
./include/linux/kern_levels.h:5:25: warning: format '%lx' expects argument of type 'long unsigned int', but argument 2 has type 'u32' {aka 'unsigned int'} [-Wformat=]
It was introduced in
commit baf8855c91 ("staging: gpib: fix address space mixup")
but was not detected as the build of the driver depended on BROKEN.
Fixes: baf8855c91 ("staging: gpib: fix address space mixup")
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250124105900.27592-2-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A number of drivers were unnecessarily not declaring their entry
points as static.
Declare them as static.
In order to reduce the number of forward declarations the gpib
interface structure initializations have been moved to after the
code of the entry points that they use. This also makes the
structure of the drivers to be more consistent between them.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20250122103859.25499-3-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove useless #ifdef RESET_USB_CONFIG code.
Change kalloc / memset to kzalloc
The attach function was not freeing the private data on error
returns. Separate the releasing of urbs and private data and
add a common error exit for attach failure.
Set the board private data pointer to NULL after freeing
the private data.
Reduce console spam by emitting only one attach message.
Change last pr_err in attach to dev_err
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250118145046.12181-3-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When the detach function is called after a failed attach
the usb_dev initialization can cause a NULL pointer
dereference. This happens when the usb device is not found
in the attach procedure.
Remove the usb_dev variable and initialization and change the dev
in the dev_info message from the usb_dev to the gpib_dev.
Fixes: fbae7090f3 ("staging: gpib: Update messaging and usb_device refs in agilent_usb")
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250118145046.12181-2-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
During the initial checkpatch cleanup, when removing blanks
after open parentheses, a minus sign in the argument of a shift
operation was deleted by mistake. This transformed a pre-decrement
operation into a negation.
The result of a negative shift is undefined and a warning was
signalled by sparse.
Restore the pre-decrement operation.
Fixes: 6c52d5e3cd ("staging: gpib: Add common include files for GPIB drivers")
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250116110014.15577-1-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Some drivers were still using the old syntax for initializing
structs:
field : value;
This caused sparse to emit the following warning, for example:
common/gpib_os.c:2026:1: warning: obsolete struct initializer, use C99 syntax
Use C99 syntax:
.field = value;
Some local structs and arrays were not declared static causing
sparse to emit the following warning, for example:
warning: symbol 'ib_fops' was not declared. Should it be static?
Declare the local structs and arrays as static.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250114165403.16410-5-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A number of drivers were comparing request_region() with 0,
others were passing 0 instead of NULL as a pointer argument.
This led to the following sparse warning, for example:
cb7210/cb7210.c:1043:72: warning: Using plain integer as NULL pointer
Use !request_region() to test for NULL return and use NULL instead
of 0 as pointer parameter.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250114165403.16410-4-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The user buffers in copy_from_user, copy_to_user and some file ops
did not use the __user address space attribute.
This led to the following sparse warning, for example:
common/gpib_os.c:838:40: warning: incorrect type in argument 2 (different address spaces)
common/gpib_os.c:838:40: expected void const [noderef] __user *from
common/gpib_os.c:838:40: got void *
Add the __user address space attribute where needed.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250114165403.16410-3-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In a number of drivers the PCI memory pointers were declared simply as
void *.
This caused sparse to emit the following warning, for example:
agilent_82350b/agilent_82350b.c:44:58: warning: incorrect type in argument 2 (different address spaces)
agilent_82350b/agilent_82350b.c:44:58: expected void volatile [noderef] __iomem *addr
Declare the PCI memory pointers as void __iomem *addr.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250114165403.16410-2-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
All interface drivers were using the old style initialization of
this struct
field : value;
This generated the followng sparse warning, for example:
agilent_82357a/agilent_82357a.c:1492:1: warning: obsolete struct initializer, use C99 syntax
Change the initialization to use the C99 syntax
.field = value;
This also resolves the checkpatch constraint of no indentation
These structs were also not declared as static, unnecessarily polluting
the symbol namespace and generating the following sparse warnings,
for example:
agilent_82357a/agilent_82357a.c:1465:18: warning: symbol 'agilent_82357a_gpib_interface' was not declared. Should it be static?
Declare them as static and remove any conflicting extern declarations
in the corresponding include files.
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20250111160514.26954-1-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The USB_GPIB_SET_LINES command string used to be: "\nIBDC \n" but when
we were merging this code into the upstream kernel we deleted the space
character before the newline to make checkpatch happy. That turned
out to be a mistake.
The "\nIBDC" part of the string is a command that we pass to the
firmware and the next character is a variable u8 value.
It gets set in set_control_line().
msg[leng - 2] = value ? (retval & ~line) : retval | line;
where leng is the length of the command string.
Imagine the parameter was supposed to be "8".
With the pre-merge code the command string would be "\nIBDC8\n"
With the post-merge code the command string became "\nIBD8\n"
The firmware doesn't recognize "IBD8" as a valid command and rejects it.
Putting a "." where the parameter is supposed to go fixes the driver
and makes checkpatch happy. Same thing with the other define and
the in-line assignment.
Reported-by: Marcello Carla' <marcello.carla@gmx.com>
Fixes: fce79512a9 ("staging: gpib: Add LPVO DIY USB GPIB driver")
Co-developed-by: Marcello Carla' <marcello.carla@gmx.com>
Signed-off-by: Marcello Carla' <marcello.carla@gmx.com>
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20241205093442.5796-1-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
GPIB secondary addresses can be between 0 and 31 inclusive
unlike primary addresses where address 31 is not a valid device
address. When 31 is used as a primary talk address it
forms the UNT (Untalk) command and when used as a listener address it
forms the UNL (Unlisten) commmand.
The library was incorrectly not allowing a secondary address
with a value of 31 to be used.
Fixes: 9dde4559e9 ("staging: gpib: Add GPIB common core driver")
Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20241104175014.12317-13-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Dave Penkler