- 30 Jul, 2021 37 commits
-
-
Fabio Aiuto authored
fix the following checkpatch issues: CHECK: Unnecessary parentheses around 'pnetwork->network.ie_length > target->ie_length' 33: FILE: drivers/staging/rtl8723bs/core/rtw_mlme.c:604: + if ((pnetwork->network.ie_length > target->ie_length) && (target->reserved[0] == 1)) CHECK: Unnecessary parentheses around 'target->reserved[0] == 1' 33: FILE: drivers/staging/rtl8723bs/core/rtw_mlme.c:604: + if ((pnetwork->network.ie_length > target->ie_length) && (target->reserved[0] == 1)) CHECK: Unnecessary parentheses around 'target->reserved[0] != 2' 39: FILE: drivers/staging/rtl8723bs/core/rtw_mlme.c:608: + if ((target->reserved[0] != 2) && + (target->reserved[0] >= pnetwork->network.reserved[0]) ) { CHECK: Unnecessary parentheses around 'target->reserved[0] >= pnetwork->network.reserved[0]' 39: FILE: drivers/staging/rtl8723bs/core/rtw_mlme.c:608: + if ((target->reserved[0] != 2) && + (target->reserved[0] >= pnetwork->network.reserved[0]) ) { Signed-off-by:Fabio Aiuto <fabioaiuto83@gmail.com> Link: https://lore.kernel.org/r/fca83a857ebb158cf395ea31f1834c04402c39e4.1627656773.git.fabioaiuto83@gmail.comSigned-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Fabio Aiuto authored
fix camel case issue in field Reserved in struct wlan_bssid_ex Signed-off-by:
-
Greg Kroah-Hartman authored
The driver seems to want to include a specific directory for all include files on the build path, but that breaks when trying to build only the module directory, or when building with "O=" option. Fix this up by making all includes for the driver be relative locations. Reported-by:
Mark Brown <broonie@kernel.org> Link: https://lore.kernel.org/r/20210730144227.1770212-1-gregkh@linuxfoundation.orgSigned-off-by:
-
Greg Kroah-Hartman authored
It is a wrapper around usb_control_msg() that does nothing, so remove the macro and just call the correct USB function instead in the one place it is used. Link: https://lore.kernel.org/r/20210730092417.1014392-12-gregkh@linuxfoundation.orgSigned-off-by:
-
Greg Kroah-Hartman authored
No one is using this wrapper macro, so just remove it as it is pointless. Link: https://lore.kernel.org/r/20210730092417.1014392-11-gregkh@linuxfoundation.orgSigned-off-by:
-
Greg Kroah-Hartman authored
No one was including this file, so just remove it. Link: https://lore.kernel.org/r/20210730092417.1014392-10-gregkh@linuxfoundation.orgSigned-off-by:
-
Greg Kroah-Hartman authored
None of these enumerated values were used anywhere in the driver, so just remove them all. Link: https://lore.kernel.org/r/20210730092417.1014392-9-gregkh@linuxfoundation.orgSigned-off-by:
-
Greg Kroah-Hartman authored
It is just a wrapper around kfree(), so just replace all calls with kfree() instead. The pointer is also set to NULL and the length set to 0 to emulate the original logic, just to be "safe" as I do not have a device to test with, but odds are that can be removed later on. Link: https://lore.kernel.org/r/20210730092417.1014392-8-gregkh@linuxfoundation.orgSigned-off-by:
-
Greg Kroah-Hartman authored
It is just a wrapper around kfree(), so remove it and just call kfree() instead. Link: https://lore.kernel.org/r/20210730092417.1014392-7-gregkh@linuxfoundation.orgSigned-off-by:
-
Greg Kroah-Hartman authored
No one is using the defines in this file, so remove it as it is not needed anywhere. Link: https://lore.kernel.org/r/20210730092417.1014392-6-gregkh@linuxfoundation.orgSigned-off-by:
-
Greg Kroah-Hartman authored
No one uses this file, so just remove it. Also remove the pointless Makefile rule that was attempting to create it. Link: https://lore.kernel.org/r/20210730092417.1014392-5-gregkh@linuxfoundation.orgSigned-off-by:
-
Greg Kroah-Hartman authored
No one uses this file, so remove it from the tree. Link: https://lore.kernel.org/r/20210730092417.1014392-4-gregkh@linuxfoundation.orgSigned-off-by:
-
Greg Kroah-Hartman authored
ODM_GetRightChnlPlaceforIQK() is only called in one place, so move the call to the single location and make the file static. This lets us remove the hal/HalPhyRf.c and include/HalPhyRf.h files as they are now empty. Link: https://lore.kernel.org/r/20210730092417.1014392-3-gregkh@linuxfoundation.orgSigned-off-by:
-
Greg Kroah-Hartman authored
The ODM_TARGET_CHNL_NUM_2G_5G define is only used in one place, so move it to the .c file it is used in instead of being in a .h file. Link: https://lore.kernel.org/r/20210730092417.1014392-2-gregkh@linuxfoundation.orgSigned-off-by:
-
Greg Kroah-Hartman authored
ODM_ResetIQKResult() is empty and does nothing, so remove it. Link: https://lore.kernel.org/r/20210730092417.1014392-1-gregkh@linuxfoundation.orgSigned-off-by:
-
Fabio Aiuto authored
BIT(x) macro used all over the driver is defined in include/vsdo/bit.h as - #define BIT(nr) (UL(1) << (nr)) which is safer than the local BIT macros declared. Local macros shift a signed integer which brings unespected results. For example: (unsigned long)(1 << 31) => 0xffffffff80000000 shift.c: int main() { printf("%lx\n", (unsigned long)(1 << 31)); printf("%lx\n", (unsigned long)(1U << 31)); return 0; } ... $ ./shift ffffffff80000000 80000000 ... So just remove redundant, less safe macro declarations. Suggested-by:David Sterba <dsterba@suse.cz> Signed-off-by:
-
Michael Straube authored
Clean up comparsions to NULL in the os_dep directory reported by checkpatch. x == NULL -> !x x != NULL -> x Signed-off-by:
Michael Straube <straube.linux@gmail.com> Link: https://lore.kernel.org/r/20210730130204.18229-1-straube.linux@gmail.comSigned-off-by:
-
Fabio M. De Francesco authored
Remove the functions and variables from rtw_security.c that are no more necessary since the patch that replaces getcrc32() with crc32_le(). Signed-off-by:
Fabio M. De Francesco <fmdefrancesco@gmail.com> Link: https://lore.kernel.org/r/20210730103716.27210-3-fmdefrancesco@gmail.comSigned-off-by:
-
Fabio M. De Francesco authored
Use crc32_le() in place of the custom getcrc32(). Signed-off-by:
-
Michael Straube authored
Use clamp() to simplify odm_evm_db_to_percentage(). Signed-off-by:
-
Michael Straube authored
Rename parameter of odm_evm_db_to_percentage() to avoid camel case. Value -> value Signed-off-by:
-
Michael Straube authored
Rename odm_EVMdbToPercentage() to avoid camel case. odm_EVMdbToPercentage() -> odm_evm_db_to_percentage() Signed-off-by:
-
Larry Finger authored
The driver can call rtw_addbareq_cmd() from an interrupt routine. resulting in the following splat: [21760.582207] BUG: sleeping function called from invalid context at include/linux/sched/mm.h:201 [21760.582219] in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 0, name: swapper/0 [21760.582224] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G C O 5.14.0-rc2-00135-g774ba5f0db67 #3 [21760.582228] Hardware name: TOSHIBA TECRA A50-A/TECRA A50-A, BIOS Version 4.50 09/29/2014 [21760.582230] Call Trace: [21760.582232] <IRQ> [21760.582233] dump_stack_lvl+0x34/0x44 [21760.582245] ? rtw_addbareq_cmd+0x2a/0xba [r8188eu] [21760.582306] ___might_sleep.cold+0x88/0x95 [21760.582311] kmem_cache_alloc_trace+0x25a/0x2d0 [21760.582315] rtw_addbareq_cmd+0x2a/0xba [r8188eu] [21760.582368] rtw_issue_addbareq_cmd+0xda/0x17f [r8188eu] [21760.582404] rtw_dump_xframe+0xa6/0x266 [r8188eu] [21760.582433] xmitframe_direct+0x40/0x57 [r8188eu] [21760.582459] pre_xmitframe+0x72/0x110 [r8188eu] [21760.582485] rtl8188eu_hal_xmit+0xa/0xb [r8188eu] [21760.582508] rtw_hal_xmit+0x1b/0x1c [r8188eu] [21760.582539] rtw_xmit+0xd6/0x1cb [r8188eu] [21760.582564] rtw_xmit_entry+0xe4/0x1d2 [r8188eu] [21760.582584] xmit_one.constprop.0+0x98/0x170 [21760.582588] dev_hard_start_xmit+0x40/0x90 [21760.582591] sch_direct_xmit+0xfa/0x230 [21760.582596] __dev_xmit_skb+0x27c/0x530 [21760.582598] __dev_queue_xmit+0x327/0x4e0 [21760.582602] ip_finish_output2+0x279/0x5c0 [21760.582605] __ip_queue_xmit+0x167/0x3f0 [21760.582607] __tcp_transmit_skb+0x4ff/0x9a0 [21760.582610] tcp_rcv_established+0x5c1/0x6f0 [21760.582613] tcp_v4_do_rcv+0x13e/0x240 [21760.582616] tcp_v4_rcv+0xbad/0xc70 [21760.582618] ? ip_rcv_finish_core.constprop.0+0x13c/0x2f0 [21760.582621] ip_protocol_deliver_rcu+0x30/0x1f0 [21760.582623] ip_local_deliver_finish+0x4b/0x60 [21760.582625] __netif_receive_skb_one_core+0x63/0x90 [21760.582628] process_backlog+0x8f/0x140 [21760.582631] __napi_poll+0x2e/0x180 [21760.582634] net_rx_action+0x11b/0x260 [21760.582637] __do_softirq+0xcb/0x2f5 [21760.582640] irq_exit_rcu+0x9e/0xc0 [21760.582643] common_interrupt+0x83/0xa0 [21760.582647] </IRQ> [21760.582647] asm_common_interrupt+0x1e/0x40 [21760.582650] RIP: 0010:cpuidle_enter_state+0xdf/0x5c0 [21760.582655] Code: 49 89 c6 0f 1f 44 00 00 31 ff e8 4c a1 98 ff 45 84 ff 74 12 9c 58 f6 c4 02 0f 85 10 03 00 00 31 ff e8 d5 73 9f ff fb 45 85 ed <0f> 88 59 01 00 00 4d 63 e5 49 83 fc 09 0f 87 05 04 00 00 4b 8d 04 Fix by changing kmalloc() flags argument from GFP_KERNEL to GFP_ATOMIC. Signed-off-by:
Larry Finger <Larry.Finger@lwfinger.net> Link: https://lore.kernel.org/r/20210730003822.12471-1-Larry.Finger@lwfinger.netSigned-off-by:
-
Agam Kohli authored
Fixed a coding style issue Signed-off-by:
Agam Kohli <agamkohli9@gmail.com> Link: https://lore.kernel.org/r/YQLveNI2UrnpeeAt@Agam.localdomainSigned-off-by:
-
Larry Finger authored
Compiler gcc11, and possibly others, emit a warning when a fall-through case is found in a switch statement. Add a "fallthrough" statement to eliminate this warning. Signed-off-by:
-
Larry Finger authored
This file contains only the definition of the driver version. This statement is moved into a header called by all the affected source files, and the now empty header is deleted. Signed-off-by:
-
Larry Finger authored
This header only includes one prototype and one define statement. The new definition is used once, thus it can be removed. The prototype is moved to a header that is already called by the supplier and the user of that routine, thus the small header is removed. Signed-off-by:
-
Larry Finger authored
This header file merely defines a single struct. Move that definition to the header file that uses it, and delete rtw_qos.h. Signed-off-by:
-
Larry Finger authored
In the GitHub repository from which this driver was derived, old kernels had to be supported. Now that the driver is included in the kernel, this code can be removed as well as all mention of version.h. Signed-off-by:
-
Larry Finger authored
File include/drv_types_linux.h is empty. Remove it. Signed-off-by:
-
Larry Finger authored
Before this driver can be incorporated in the drivers/net/wireless tree, the copyright info in all files must be converted to SPDX notation. This patch converts the C source files in hal/. Signed-off-by:
-
Larry Finger authored
Before this driver can be incorporated in the drivers/net/wireless tree, the copyright info in all files must be converted to SPDX notation. This patch converts the 9 C source files in os_dep/. Signed-off-by:
-
Larry Finger authored
Before this driver can be incorporated in the drivers/net/wireless tree, the copyright info in all files must be converted to SPDX notation. This patch converts the 23 C source files in core/. Signed-off-by:
-
Larry Finger authored
Before this driver can be incorporated in the drivers/net/wireless tree, the copyright info in all files must be converted to SPDX notation. This patch converts the next 49 files. Signed-off-by:
-
Larry Finger authored
Before this driver can be incorporated in the drivers/net/wireless tree, the copyright info in all files must be converted to SPDX notation. This patch converts the next 23 files. Signed-off-by:
-
Larry Finger authored
Before this driver can be incorporated in the drivers/net/wireless tree, the copyright info in all files must be converted to SPDX notation. This patch converts the first 20 files. Signed-off-by:
-
Greg Kroah-Hartman authored
The vt6655 and vt6656 drivers have an unused CONFIG_PATH define floating around in the code, but it is never used. Remove it as drivers should never be reading from config files anyway, even if these were valid files. Reported-by:
Joe Perches <joe@perches.com> Reported-by:
Randy Dunlap <rdunlap@infradead.org> Acked-by:
-
- 29 Jul, 2021 2 commits
-
-
Len Baker authored
strcpy() performs no bounds checking on the destination buffer. This could result in linear overflows beyond the end of the buffer, leading to all kinds of misbehaviors. The safe replacement is strscpy(). Signed-off-by:
Len Baker <len.baker@gmx.com> Link: https://lore.kernel.org/r/20210718131217.3806-1-len.baker@gmx.comSigned-off-by:
-
Fabio Aiuto authored
remove some unused static variables. One of them is used to toggle on the BT coexistence mechanism, but it is always enabled and it's not conditioned to the value of the related parameter. Remove unused field of the registry_priv struct as well, they were intended for hosting the parameters this patch rids. Signed-off-by:
-
- 28 Jul, 2021 1 commit
-
-
Benjamin Philip authored
The struct lynxfb_crtc has a member named vScreen. This name is CamelCase and is frowned upon. This commit renames it to v_screen and makes the necessary changes for the module to build. This change also fixes the following checkpatch CHECKs: CHECK: Avoid CamelCase: <vScreen> 454: FILE: sm750.c:454: + memset_io(crtc->vScreen, 0x0, crtc->vidmem_size); CHECK: Avoid CamelCase: <vScreen> 136: FILE: sm750.h:136: + unsigned char __iomem *vScreen; /* virtual address of on_screen */ Signed-off-by:
Benjamin Philip <benjamin.philip495@gmail.com> Link: https://lore.kernel.org/r/52db14bfc2f39e246aed7fdf866845a72e503b36.1627413010.git.benjamin.philip495@gmail.comSigned-off-by:
-
