| Age | Commit message (Collapse) | Author |
|---|
|
git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging
Pull staging driver updates from Greg KH:
"Here is the big set of staging driver changes for 6.13-rc1.
Lots of changes this merge cycle, drivers removed and drivers added.
Highlights include:
- removals of the following staging drivers due to no forward
progress and no one having either the hardware or the time/energy
to deal with them anymore:
- fieldbus
- gdm724x
- olpc_dcon
- rtl8712
- rts5208
- vt6655
- vt6656
If anyone has this hardware and wants to work on the drivers, it
can be an easy revert to get them back.
- addition of the gpib driver subsystem. Lots of drivers for really
old and semi-old interfaces to lab equipments. We expect lots of
churn in these drivers as they get cleaned up to "working" order.
These were added at the request of a user and the maintainer/author
of them is helping out with the effort
- loads and loads of tiny coding style cleanups for almost all
staging drivers. Too many to list, see the shortlog for details.
All of these have been in linux-next for a very long time with no
reported issues"
* tag 'staging-6.13-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging: (216 commits)
Staging: gpib: gpib_os.c - Remove unnecessary OOM message
staging: gpib: avoid unintended sign extension
staging: vchiq_debugfs: Use forward declarations
staging: vchiq_core: Rectify header include for vchiq_dump_state()
staging: vc04_services: Cleanup TODO entry
staging: most: Remove TODO contact information
staging: rtl8723bs: Remove TODO contact information
staging: sm750fb: Remove TODO contact information
staging: iio: Remove TODO file
staging: greybus: uart: Fix atomicity violation in get_serial_info()
staging: rtl8723bs: Remove unused function Efuse_GetCurrentSize
staging: rtl8723bs: Remove unused function efuse_WordEnableDataRead
staging: rtl8723bs: Remove function hal_EfusePgPacketWrite1ByteHeader
staging: rtl8723bs: Remove function hal_EfusePgPacketWrite2ByteHeader
staging: rtl8723bs: Remove unused function hal_EfusePgCheckAvailableAddr
staging: rtl8723bs: Remove unused function hal_EfuseConstructPGPkt
staging: rtl8723bs: Remove unused function hal_EfusePartialWriteCheck
staging: rtl8723bs: Remove unused function hal_EfusePgPacketWriteHeader
staging: rtl8723bs: Remove unused function hal_EfusePgPacketWriteData
staging: rtl8723bs: Remove unused function Hal_EfusePgPacketWrite_BT
...
|
|
Use forward declarations for struct vchiq_state and vchiq_instance.
We can then drop the vchiq_core.h header from vchiq_debugfs.h.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20241107194806.90408-4-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The header vchiq_core.h does not need <linux/debugfs.h>. It needs
the <linux/seq_file.h> for vchiq_dump_state() to dump the vchiq
state through vchiq_debugfs.[ch].
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20241107194806.90408-3-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The TODO entry "Fix behvaiour of message handling" no longer
applies due to killable completions [1]. Drop the entry from TODO list.
[1] https://lore.kernel.org/all/20240918163100.870596-1-umang.jain@ideasonboard.com/
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20241107194806.90408-2-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
A bulk callback transfer can be initiated from two places -
inside kernel interface or from user interface. However,
the callback data pointer 'cb_data' is used for tracking both
sets of data pointer. This commit tracks the callback
data pointer from user interface (named as 'cb_userdata') separately,
in the bulk transfer service callback.
This is esentially done by adding a 'void __user *cb_userdata' for
tracking __user pointers in vchiq_bulk and vchiq_completion_data
structs. Furthermore, the 'cb_userdata' data pointer is appended to
the vchiq_service's callback signature.
Separating the two callback data pointers ('cb_data' and 'cb_userdata')
fixes the sparse warnings around mixing userspace and kernel space
pointers.
As there are no additional sparse warnings left for vc04_services,
drop the relevant entry from the TODO.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20241023110406.885199-7-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Pass struct vchiq_bulk pointer to make_service_callback() instead of
just passing the bulk->cb_data. This is a preparatory change when we
need to pass the callback data user pointer (__user) in a subsequent
commit.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20241023110406.885199-6-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
In a previous commit, struct vchiq_bulk 'userdata' got renamed to
'cb_data' since it is the data pointer passed in
VCHIQ_BULK_CALLBACK_MODE's callback. Since struct vchiq_completion_data*
structs also has 'bulk_userdata' for completion records, rename
'bulk_userdata' member to 'cb_data' for these structs as well.
This brings consistency and clarity for the struct members. No
functional change in this patch.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20241023110406.885199-5-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Rename the struct vchiq_bulk 'userdata' member to 'cb_data' to clarify
its purpose. 'cb_data' is meant to be passed to service callback
function in VCHIQ_BULK_MODE_CALLBACK mode.
No functional changes in this patch.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20241023110406.885199-4-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Currently, struct bulk_waiter is allocated for VCHIQ_BULK_MODE_BLOCKING
bulk transfer and its pointer is assigned to vchiq_bulk->userdata. Avoid
this kind of piggybacking and introduce a dedicate 'waiter' member in
struct vchiq_bulk.
The 'userdata' is meant for VCHIQ_BULK_MODE_CALLBACK mode, to pass user
specified parameter to the actual callback function.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20241023110406.885199-3-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Rename the struct vchiq_bulk's 'data' member to 'dma_addr' for
better readability. No functional changes intended in this patch.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20241023110406.885199-2-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
This commit cleans up the formatting in by removing extraneous
blank lines, improving code readability without changing functionality.
Signed-off-by: Rohit Chavan <roheetchavan@gmail.com>
Link: https://lore.kernel.org/r/20241029063901.1857067-1-roheetchavan@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The struct drv_mgmt 'mgmt' is currently allocated dynamically using
kzalloc(). Unfortunately, it is subjected to memory leaks in the error
handling paths of the probe() function.
To address this issue, use device resource management
helper devm_kzalloc(), to ensure cleanup after the allocation.
Cc: stable@vger.kernel.org
Fixes: 1c9e16b73166 ("staging: vc04_services: vchiq_arm: Split driver static and runtime data")
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20241016130225.61024-3-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The struct vchiq_arm_state 'platform_state' is currently allocated
dynamically using kzalloc(). Unfortunately, it is never freed and is
subjected to memory leaks in the error handling paths of the probe()
function.
To address the issue, use device resource management helper
devm_kzalloc(), to ensure cleanup after its allocation.
Fixes: 71bad7f08641 ("staging: add bcm2708 vchiq driver")
Cc: stable@vger.kernel.org
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20241016130225.61024-2-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The 'userdata' local pointer can be dropped which is set to bulk_waiter.
We can directly pass the waiter->bulk_waiter pointer to
vchiq_bulk_xfer_waiting().
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20241017133629.216672-7-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
vchiq_bulk_xfer_queue_msg_killable() is a common function between
various bulk transfer code paths (blocking, callback and no-callback).
These code paths were simplified earlier by passing a populated
struct vchiq_bulk pointer in order to avoid open-coding the parameters
required to initiate a bulk transfer.
Now simplify the vchiq_bulk_xfer_queue_msg_killable() in a similar way
i.e. avoid open-coding the function parameters and pass the struct
vchiq_bulk pointer directly, coming from the various bulk transfer code
paths.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20241017133629.216672-6-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The (no)callback mode bulk transfer tends to open-code every function
parameter needed to initiate the bulk transfer. Instead of doing that,
simply pass a populated struct vchiq_bulk down the function chain.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20241017133629.216672-5-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Blocking bulk transfer functions tend to open-code every function
parameter needed to initiate the bulk transfer. Instead of doing
that, simply pass a populated struct vchiq_bulk down the function
chain.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20241017133629.216672-4-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Two functions create_pagelist() and vchiq_prepare_bulk_data()
open code bulk data arguments ('size' and 'dir') in their function
signatures which can easily be obtained by struct vchiq_bulk pointer.
Retrieve the arguments from vchiq_bulk pointer instead and reduce
the number of arguments passed in create_pagelist() and
vchiq_bulk_prepare_data().
No functional changes intended in this patch.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20241017133629.216672-3-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Subsume offset and uoffset inside struct vchiq_bulk instead of
open-coding them in vchiq_prepare_bulk_data() function. It helps in
reducing function parameters and can be easily retrieved from the
struct vchiq_bulk pointer for creating pagelist.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20241017133629.216672-2-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
An error path was introduced without including the required call to
of_node_put() to decrement the node's refcount and avoid leaking memory.
If the call to kzalloc() for 'mgmt' fails, the probe returns without
decrementing the refcount.
Use the automatic cleanup facility to fix the bug and protect the code
against new error paths where the call to of_node_put() might be missing
again.
Cc: stable@vger.kernel.org
Fixes: 1c9e16b73166 ("staging: vc04_services: vchiq_arm: Split driver static and runtime data")
Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20241014-vchiq_arm-of_node_put-v2-2-cafe0a4c2666@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The 'failed_platform_init' and 'error_exit' labels do not simplify the
code, there is a single jump to them in the code, and the actions taken
from then on can be easily carried out where the goto occurs.
Signed-off-by: Javier Carrasco <javier.carrasco.cruz@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Reviewed-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20241014-vchiq_arm-of_node_put-v2-1-cafe0a4c2666@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Reduce indentation of the conditional nesting in
vchiq_close_service_internal() switch case by checking the error paths
first and break early. This helps to reduce conditional branching and
reduce indentation levels.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20241013112128.397249-7-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
If the service is not in VCHIQ_SRVSTATE_LISTENING state, it is
implied that the message is dealt with and parse_open() should return.
If this is the case, simply jump the code flow to return site using
'goto done;' statement.
This helps to lower the indentation of
if (service->srvstate == VCHIQ_SRVSTATE_LISTENING)
conditional branch.
No functional changes intended in this patch.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Stefan Wahren <wahrenst@gmx.net>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20241013112128.397249-6-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Move the statistics and bulk completion events handling to a separate
function. This helps to improve readability for notify_bulks().
No functional changes intended in this patch.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Kieran Bingham <kieran.bingham@ideasonboard.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20241013112128.397249-5-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Fix the copy_message_data() indentation in queue_message_sync().
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Stefan Wahren <wahrenst@gmx.net>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20241013112128.397249-4-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Do not log a dev_dbg() with a separate scope. Drop the {..}
scope and align the dev_dbg() to make it more readable.
No functional changes intended in this patch.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Stefan Wahren <wahrenst@gmx.net>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20241013112128.397249-3-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Locally cache 'cache_line_size' information in a variable instead of
repeatedly accessing it from drv_mgmt->info. This helps to reflow lines
under 80 columns.
No functional change intended in this patch.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Stefan Wahren <wahrenst@gmx.net>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20241013112128.397249-2-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Removes multiple blank lines to adhere to kernel coding
style. No functional changes intended in this patch.
Signed-off-by: Danilo Pereira <dpereira@lkcamp.dev>
Link: https://lore.kernel.org/r/20241012233931.30720-1-dpereira@lkcamp.dev
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Recent tests on Raspberry Pi 3 B Plus have shown that one
iteration is not enough to discover issues reliable. So
switch back to the defaults (1000 iterations).
Link: https://lore.kernel.org/linux-staging/c7e302b6-fc62-4754-ab1d-7c2771cccf60@gmx.net/
Signed-off-by: Stefan Wahren <wahrenst@gmx.net>
Reviewed-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20241011100119.111399-1-wahrenst@gmx.net
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The two static asserts are under 80 columns width, hence indent them
on the same line.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Stefan Wahren <wahrenst@gmx.net>
Link: https://lore.kernel.org/r/20241010102250.236545-3-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Replace spaces with tabs to adhere to kernel coding style. No
functional changes intended in this patch.
Signed-off-by: Hridesh MG <hridesh699@gmail.com>
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Stefan Wahren <wahrenst@gmx.net>
Link: https://lore.kernel.org/r/20241010102250.236545-2-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
vchiq_pagelist.h only defines one struct and a couple of macros.
It can be merged with vchiq_core since all the pagelist related
function helpers are now in vchiq_core for bulk transfers.
Move the struct and related macros to vchiq_core header and drop
vchiq_pagelist.h.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20240919142130.1331495-4-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Bulk transfers core logic lives in vchiq_core.c, hence move all
the preparatory bulk data allocation helpers to vchiq_core.c (from
vchiq_arm).
The discrepancy was noticed when vchiq_prepare_bulk_data() and
vchiq_complete_bulk() are being used vchiq_core.c but are defined
in vchiq_arm. Now that they are now confined to vchiq_core.c,
they can be made static and their signatures from vchiq_core header
can be dropped.
vchiq_prepare_bulk_data() and vchiq_complete_bulk() depends on
struct vchiq_pagelist_info, cleanup_pagelist(), free_pagelist() and
create_pagelist() hence they are pulled in from vchiq_arm as well,
as part of this commit.
No functional changes intended in this patch.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20240919142130.1331495-3-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The function remote_event_signal() is declared in vchiq_core.h while
defined in vchiq_arm.c and used only in vchiq_core.c. Move the
definition to vchiq_core.c as it is only used in this file.
Also convert it to static and drop the function signature from
vchiq_core.h header. BELL2 doorbell macro is also moved from vchiq_arm
to vchiq_core as part of this change.
No functional changes intended in this patch.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20240919142130.1331495-2-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
-EINTR is returned by vchiq_queue_message() on receiving a fatal
signal to the process. Since the process is deemed to be terminated
anyway, do not retry queuing with vchiq_queue_message() on -EINTR.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20240918163100.870596-7-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
-EINTR is returned by various vchiq bulk transfer code paths
on receiving a fatal signal to the process. Since the process is
deemed to be terminated anyway, do not retry the bulk transfer
on -EINTR.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20240918163100.870596-6-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Bulk transfers for various VCHIQ modes use mutex_lock_killable() and
wait_for_completion_killable() variations. Currently, -EAGAIN is
returned if these are interrupted by a fatal signal.
-EAGAIN may mislead the caller into thinking the operation can be
retried, while in reality, the process has received a fatal signal
and is terminating. Therefore, we should update the return value to
align with what these killable functions would return, specifically
-EINTR (Interrupted system call).
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20240918163100.870596-5-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
queue_message() uses mutex_lock_killable() and
wait_for_completion_killable() variations of locking and wait event
completions respectively. These functions return either 0 (on success)
or -EINTR, if interrupted by a fatal signal (as documented in the
kernel).
However, queue_message() is currently returning -EAGAIN if these
killable functions are interrupted by fatal signals. Bubbling up
-EAGAIN might give a sense to the caller, that the code path can
be re-tried however, in actual sense, a fatal signal has been
received by the process and the process is going away.
Hence, we should align the return value with what these killable
versions will return i.e. -EINTR (Interrupted system call).
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20240918163100.870596-4-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
In vchiq_connect_internal(), a MAKE_CONNECT message is queued
if the connection is disconnected, but only -EAGAIN error is
checked on the error path and returned.
However, queue_message() can fail with other errors as well hence,
vchiq_connect_internal() should return in those cases as well.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20240918163100.870596-3-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
commit f27e47bc6b8b ("staging: vchiq: use completions instead of
semaphores") introduced completions for events in vchiq interface.
It introduced _interruptible() version of completions for waiting
on events. However, it missed a subtle down_interruptible() macro
override in vchiq_killable.h, which used to mask most of the signals
and only interrupt on fatal ones.
The above issue was fixed in commit a772f116702e ("staging: vchiq: switch
to wait_for_completion_killable"). Given the override logic of
down_interruptible() that existed in vchiq_killable.h, that commit
fixed the completions with the correct variation i.e. killable() family
of functions.
However, commit a772f116702e ("staging: vchiq: switch to
wait_for_completion_killable") later got reverted [1] due to high CPU
load noticed by various downstream and upstream distributions [2].
Reverting the commit solved this problem but the root cause was never
diagonsed and the entire commit was reverted.
This patch brings back killable version of wait events but only for
bulk transfers and queue_message() transfer code paths.
The idea is to bring back killable versions for various event
completions in a phased manner so that we do not re-regress again as
noticed in [2]. Hence, no other wait events are converted from
interruptible -> killable in this patch.
Since the bulk transfers are no longer interruptible (but killable),
drop the "_interruptible" suffix from all vchiq_bulk_xfer_* functions.
[1]: commit 086efbabdc04 ("staging: vchiq: revert "switch to wait_for_completion_killable"")
[2]: https://patchwork.kernel.org/project/linux-arm-kernel/cover/20190509143137.31254-1-nsaenzjulienne@suse.de/
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20240918163100.870596-2-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
After commit 0edb555a65d1 ("platform: Make platform_driver::remove()
return void") .remove() is (again) the right callback to implement for
platform drivers.
Convert all staging drivers to use .remove(), with the eventual goal to
drop struct platform_driver::remove_new(). As .remove() and .remove_new()
have the same prototypes, conversion is done by just changing the structure
member name in the driver initializer.
Signed-off-by: Sergio Paracuellos <sergio.paracuellos@gmail.com>
Link: https://lore.kernel.org/r/20241001085751.282113-1-sergio.paracuellos@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Pass proper enumerated flag which exists, instead of an integer while
calling queue_message(). It helps with readability of the code.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Stefan Wahren <wahrenst@gmx.net>
Tested-by: Stefan Wahren <wahrenst@gmx.net>
Link: https://lore.kernel.org/r/20240910051007.297227-8-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The argument 'is_blocking' in queue_message_sync() is not
used in the function. Drop it.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Stefan Wahren <wahrenst@gmx.net>
Tested-by: Stefan Wahren <wahrenst@gmx.net>
Link: https://lore.kernel.org/r/20240910051007.297227-7-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Drop vchiq_bulk_transfer() as every VCHIQ_BULK_MODE_* mode
now have their own dedicated functions to execute bulk transfers.
Also, drop the temporary label we introduced earlier in vchiq-dev.c
to jump over the vchiq_bulk_transfer() call when each separate mode
helper was being developed.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Tested-by: Stefan Wahren <wahrenst@gmx.net>
Link: https://lore.kernel.org/r/20240910051007.297227-6-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Factor out bulk transfer for VCHIQ_BULK_MODE_NOCALLBACK and
VCHIQ_BULK_MODE_CALLBACK mode into a separate dedicated function
bulk_xfer_callback_interruptible(). It is suffixed by "_interruptible"
to denote that it can be interrupted and -EAGAIN can be returned. It
would be up to the users of the function to retry the call in those cases.
bulk_xfer_callback_interruptible() also takes in 'mode' parameter to
differentiate between VCHIQ_BULK_MODE_NOCALLBACK and
VCHIQ_BULK_MODE_CALLBACK, which then is directly passed to
vchiq_bulk_xfer_queue_msg_interruptible() inside the function.
Adjust the calls to vchiq-dev.c ioctl interface and vchiq_arm.c
for the respective bulk transfers.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Tested-by: Stefan Wahren <wahrenst@gmx.net>
Link: https://lore.kernel.org/r/20240910051007.297227-5-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Factor out bulk transfer for blocking mode into a separate dedicated
function bulk_xfer_blocking_interruptible(). It is suffixed by
"_interruptible" to denote that it can be interrupted and -EAGAIN
can be returned. It would be up to the users of the function to retry
the call in those cases.
Adjust the calls to vchiq-dev.c ioctl interface and vchiq_arm.c
for blocking bulk transfers.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Tested-by: Stefan Wahren <wahrenst@gmx.net>
Link: https://lore.kernel.org/r/20240910051007.297227-4-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
Factor out core logic for preparing bulk data transfer(mutex locking,
waits on vchiq_bulk_queue wait-queue, initialising the bulk transfer)
out of the vchiq_bulk_transfer(). This simplifies the existing
vchiq_bulk_transfer() and makes it more readable since all the core
logic is handled in vchiq_bulk_xfer_queue_msg_interruptible(). It
will also help us to refactor vchiq_bulk_transfer() easily for different
vchiq bulk transfer modes.
No functional changes intended in this patch.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Tested-by: Stefan Wahren <wahrenst@gmx.net>
Link: https://lore.kernel.org/r/20240910051007.297227-3-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
The bulk transfer is VCHIQ_BULK_MODE_WAITING is used by VCHIQ ioctl
interface. It is factored out to a separate function from
vchiq_bulk_transfer() to bulk_xfer_waiting_interruptible().
This is a part of vchiq_bulk_transfer refactoring. Each bulk mode
will have their dedicated functions to execute bulk transfers.
Each mode will be handled separately in subsequent patches.
bulk_xfer_waiting_interruptible() is suffixed with "_interruptible"
to denote that it can be interrupted when a signal is received.
-EAGAIN maybe returned in those cases, similar to what
vchiq_bulk_transfer() does.
Adjust the vchiq_irq_queue_bulk_tx_rx() in the vchiq-dev.c to call
bulk_xfer_waiting_interruptible() for waiting mode. A temporary
goto label has been introduced to jump the call execution over
vchiq_bulk_transfer() for waiting mode only. When all dedicated bulk
transfer calls are introduced, this label shall be dropped.
No function changes intended in this patch.
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Tested-by: Stefan Wahren <wahrenst@gmx.net>
Link: https://lore.kernel.org/r/20240910051007.297227-2-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
Pull driver core updates from Greg KH:
"Here is the big set of driver core changes for 6.11-rc1.
Lots of stuff in here, with not a huge diffstat, but apis are evolving
which required lots of files to be touched. Highlights of the changes
in here are:
- platform remove callback api final fixups (Uwe took many releases
to get here, finally!)
- Rust bindings for basic firmware apis and initial driver-core
interactions.
It's not all that useful for a "write a whole driver in rust" type
of thing, but the firmware bindings do help out the phy rust
drivers, and the driver core bindings give a solid base on which
others can start their work.
There is still a long way to go here before we have a multitude of
rust drivers being added, but it's a great first step.
- driver core const api changes.
This reached across all bus types, and there are some fix-ups for
some not-common bus types that linux-next and 0-day testing shook
out.
This work is being done to help make the rust bindings more safe,
as well as the C code, moving toward the end-goal of allowing us to
put driver structures into read-only memory. We aren't there yet,
but are getting closer.
- minor devres cleanups and fixes found by code inspection
- arch_topology minor changes
- other minor driver core cleanups
All of these have been in linux-next for a very long time with no
reported problems"
* tag 'driver-core-6.11-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core: (55 commits)
ARM: sa1100: make match function take a const pointer
sysfs/cpu: Make crash_hotplug attribute world-readable
dio: Have dio_bus_match() callback take a const *
zorro: make match function take a const pointer
driver core: module: make module_[add|remove]_driver take a const *
driver core: make driver_find_device() take a const *
driver core: make driver_[create|remove]_file take a const *
firmware_loader: fix soundness issue in `request_internal`
firmware_loader: annotate doctests as `no_run`
devres: Correct code style for functions that return a pointer type
devres: Initialize an uninitialized struct member
devres: Fix memory leakage caused by driver API devm_free_percpu()
devres: Fix devm_krealloc() wasting memory
driver core: platform: Switch to use kmemdup_array()
driver core: have match() callback in struct bus_type take a const *
MAINTAINERS: add Rust device abstractions to DRIVER CORE
device: rust: improve safety comments
MAINTAINERS: add Danilo as FIRMWARE LOADER maintainer
MAINTAINERS: add Rust FW abstractions to FIRMWARE LOADER
firmware: rust: improve safety comments
...
|
|
The various kthreads thread functions (slot_handler_func, sync_func,
recycle_func) in vchiq_core and vchiq_keepalive_thread_func in
vchiq_arm should be stopped when the module is unloaded.
Previous attempt were made to address this but later reverted [1]
due to VC04 firmware corruption. The issue around
wait_event_interruptible() handling on stopping a kthread has been
handled in the previous commit. Hence, it is now safe to stop kthreads
on module unload, without any firmware corruption.
This also completes the "Fix kernel module support" TODO item, hence
drop it from the list.
[1] commit ebee9ca2f59e ("Revert "staging: vc04_services: vchiq_core: Stop kthreads on shutdown"")
Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Tested-by: Stefan Wahren <wahrenst@gmx.net>
Link: https://lore.kernel.org/r/20240703131052.597443-3-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
