Remove the local variable "array" and all the memcpy function calls
because this copy operation from different arrays to this variable is
unnecessary.
The vnt_control_out function already does a kmemdup copy of its const
char *buffer argument and this was made unnecessary by:
commit 12ecd24ef9
("staging: vt6656: use off stack for out buffer USB transfers.")
Author: Malcolm Priestley <tvboxspy@gmail.com>
Date: Sat Apr 22 11:14:57 2017 +0100
staging: vt6656: use off stack for out buffer USB transfers.
Since 4.9 mandated USB buffers be heap allocated this causes the driver
to fail.
Since there is a wide range of buffer sizes use kmemdup to create
allocated buffer.
So, the same result can be achieved using the arrays directly.
Signed-off-by: Oscar Carter <oscar.carter@gmx.com>
Link: https://lore.kernel.org/r/20200425151747.8199-2-oscar.carter@gmx.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Check the return value of vnt_control_out_* function calls. When
necessary modify the function prototype to be able to return the new
checked error code.
It's safe to modify all the function prototypes without fix the call
because the only change is the return value from void to int. If before
the call didn't check the return value, now neither.
Signed-off-by: Oscar Carter <oscar.carter@gmx.com>
Link: https://lore.kernel.org/r/20200425134257.4502-2-oscar.carter@gmx.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We need the staging fixes in here too, and this resolves a merge issue
with the vt6656 driver.
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
vnt_set_bss_mode needs to be called on all changes to BSS_CHANGED_BASIC_RATES,
BSS_CHANGED_ERP_PREAMBLE and BSS_CHANGED_ERP_SLOT
Remove all other calls and vnt_update_ifs which is called in vnt_set_bss_mode.
Fixes an issue that preamble mode is not being updated correctly.
Fixes: c12603576e ("staging: vt6656: Only call vnt_set_bss_mode on basic rates change.")
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Malcolm Priestley <tvboxspy@gmail.com>
Link: https://lore.kernel.org/r/44110801-6234-50d8-c583-9388f04b486c@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The problem is that the group key was saved as VNT_KEY_DEFAULTKEY
was over written by the VNT_KEY_GROUP_ADDRESS index.
mac80211 could not clear the mac_addr in the default key.
The VNT_KEY_DEFAULTKEY is not necesscary so remove it and set as
VNT_KEY_GROUP_ADDRESS.
mac80211 can clear any key using vnt_mac_disable_keyentry.
Fixes: f9ef05ce13 ("staging: vt6656: Fix pairwise key for non station modes")
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Malcolm Priestley <tvboxspy@gmail.com>
Link: https://lore.kernel.org/r/da2f7e7f-1658-1320-6eee-0f55770ca391@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Replace the for loops of the vnt_update_top_rates function by the fls
function.
The purpose of the two for loops is to find the most significant bit set
in a range of bits. So, they can be replace by the fls function (find
last set) with a previous mask to define the range.
This way avoid the iteration over unnecessary for loops.
The header "linux/bits.h" can be remove as it is included in the header
"linux/bitops.h".
Signed-off-by: Oscar Carter <oscar.carter@gmx.com>
Link: https://lore.kernel.org/r/20200420155246.4925-1-oscar.carter@gmx.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Replace the for loop by a ternary operator whose condition is an AND
bitmask against the priv->basic_rates variable.
The purpose of the for loop was to check if any of bits from RATE_54M to
RATE_6M was set, but it's not necessary to check every individual bit.
The same result can be achieved using only one single mask which
comprises all the commented bits.
This way avoid the iteration over an unnecessary for loop.
Also change the return type to bool because it's the type that this
function returns.
Signed-off-by: Oscar Carter <oscar.carter@gmx.com>
Link: https://lore.kernel.org/r/20200418134553.6415-1-oscar.carter@gmx.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Create a constant array with the values of the "phy->signal" for every
rate. Remove all "phy->signal" assignments inside the switch statement
and replace these with a single reading from the new vnt_phy_signal
array.
The constant array can be of one dimension because the OR mask with
BIT(3) or BIT(4) allow obtain a second value according to the rate,
the preamble_type and the pkt_type.
Signed-off-by: Oscar Carter <oscar.carter@gmx.com>
Link: https://lore.kernel.org/r/20200414153849.5785-5-oscar.carter@gmx.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Use the error code returned by the vnt_control_out function as the
returned value of the vnt_rf_write_embedded function instead of a
boolean value.
Then, fix all vnt_rf_write_embedded calls removing the "and" operations
and replace with a direct assignment to the ret variable and add a check
condition after every call.
Also replace the boolean values true or false in the vnt_rf_set_txpower
function to 0 or error code EINVAL to follow the coding style guide.
The vnt_rf_set_txpower function is called only in the vnt_rf_setpower
function that already returns error codes. The calls to this function
(vnt_rf_set_txpower) not use the returned values, so they not need to be
fixed.
Signed-off-by: Oscar Carter <oscar.carter@gmx.com>
Link: https://lore.kernel.org/r/20200414153849.5785-2-oscar.carter@gmx.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Four tables can be extracted from RSPINF_A_* based on BB_TYPE_11A or
else being GB rates.
Preamble short or long tables from fixed size len of 14 for RSPINF_B rates.
Remove function vnt_calculate_ofdm_rate and replace with the tables
calling RSPINF_A and RSPINF_B separately.
Signed-off-by: Malcolm Priestley <tvboxspy@gmail.com>
Link: https://lore.kernel.org/r/addca9e1-9ea2-59ee-dec9-3afb3c731c2e@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Create three arrays with the threshold data use in the switch statement
of the vnt_update_pre_ed_threshold function. These three arrays contains
elements of struct vnt_threshold new type.
Create a for loop in the vnt_update_pre_ed_threshold function to do
exactly the same that the if-elseif-else statements in the switch
statement.
Also, remove the if check against the !cr_201 && !cr_206 due to now it
is replace by the NULL check against the threshold pointer. When this
pointer is NULL means that the cr_201 and cr_206 variables have not been
assigned, that is the same that the old comparison against cr_201 and
cr_206 due to these variables were initialized with 0.
The statistics of the old baseband object file are:
section size addr
.text 3415 0
.data 576 0
.bss 0 0
.rodata 120 0
.comment 45 0
.note.GNU-stack 0 0
.note.gnu.property 28 0
Total 4184
The statistics of the new baseband object file are:
section size addr
.text 2209 0
.data 576 0
.bss 0 0
.rodata 344 0
.comment 45 0
.note.GNU-stack 0 0
.note.gnu.property 28 0
Total 3202
With this refactoring it increase a little the readonly data but it
decrease much more the .text section. This refactoring decrease the
footprint and makes the code more clear.
Signed-off-by: Oscar Carter <oscar.carter@gmx.com>
Reviewed-by: Quentin Deslandes <quentin.deslandes@itdev.co.uk>
Link: https://lore.kernel.org/r/20200328181706.14276-1-oscar.carter@gmx.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Replace the bit left shift operation with the BIT_ULL() macro and remove
the unnecessary mask off operation against the bit_nr variable.
This mask is only there for legacy reasons. Originally it was 31 (0x1f)
and the bit_nr variable spread across an mc_filter array with two u32
elements. Now, this mask is not needed because its value is 0x3f and the
mc_filter variable is an u64 type.
In this situation, the 26 bit right shift operation against the value
returned by the ether_crc function set the bit_nr variable to the
following value:
bit 31 to bit 6 -> All 0 (due to the bit shift operation happens over an
unsigned type).
bit 5 to bit 0 -> The 6 msb bits of the value returned by the ether_crc
function.
The purpose of the 0x3f mask against the bit_nr variable is to reset
(set to 0) the 26 msb bits (bit 31 to bit 6), but these bits are yet 0.
So, the mask off operation is now unnecessary.
Signed-off-by: Oscar Carter <oscar.carter@gmx.com>
Link: https://lore.kernel.org/r/20200314161441.4870-1-oscar.carter@gmx.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
