- Apr 02, 2021
-
-
Qiheng Lin authored
In case of error, the function pci_iomap() returns NULL pointer not ERR_PTR(). The IS_ERR() test in the return value check should be replaced with NULL test. Reported-by:
Hulk Robot <hulkci@huawei.com> Signed-off-by:
Qiheng Lin <linqiheng@huawei.com> Link: https://lore.kernel.org/r/20210330013659.916-1-linqiheng@huawei.com Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
- Mar 28, 2021
-
-
Xu Yilun authored
This patch adds description for UIO support for dfl devices on DFL bus. Reviewed-by:
Tom Rix <trix@redhat.com> Reviewed-by:
Wu Hao <hao.wu@intel.com> Signed-off-by:
Xu Yilun <yilun.xu@intel.com> Link: https://lore.kernel.org/r/1615168776-8553-3-git-send-email-yilun.xu@intel.com Signed-off-by:
-
Xu Yilun authored
This patch supports the DFL drivers be written in userspace. This is realized by exposing the userspace I/O device interfaces. The driver now only binds the ether group feature, which has no irq. So the irq support is not implemented yet. Reviewed-by:
-
Mihai Carabas authored
Add support for pvpanic PCI device added in qemu [1]. At probe time, obtain the address where to read/write pvpanic events and pass it to the generic handling code. Will follow the same logic as pvpanic MMIO device driver. At remove time, unmap base address and disable PCI device. [1] https://github.com/qemu/qemu/commit/9df52f58e76e904fb141b10318362d718f470db2 Signed-off-by:
Mihai Carabas <mihai.carabas@oracle.com> Link: https://lore.kernel.org/r/1616597356-20696-4-git-send-email-mihai.carabas@oracle.com Signed-off-by:
-
Mihai Carabas authored
Create the mecahism that allows multiple pvpanic instances to call pvpanic_probe and receive panic events. A global list will retain all the mapped addresses where to write panic events. Signed-off-by:
-
Mihai Carabas authored
Split-up generic and platform dependent code in order to be able to re-use generic event handling code in pvpanic PCI device driver in the next patches. The code from pvpanic.c was split in two new files: - pvpanic.c: generic code that handles pvpanic events - pvpanic-mmio.c: platform/bus dependent code Signed-off-by:
-
Xu Jia authored
When cmd > 6 or copy_to_user() fail, The variable 'ret' would not be returned back. Fix the 'ret' set but not used. Reviewed-by:
Arnd Bergmann <arnd@arndb.de> Signed-off-by:
Xu Jia <xujia39@huawei.com> Link: https://lore.kernel.org/r/20210324072031.941791-1-xujia39@huawei.com Signed-off-by:
-
Bhaskar Chowdhury authored
s/Initilize/Initialize/ Signed-off-by:
Bhaskar Chowdhury <unixbhaskar@gmail.com> Link: https://lore.kernel.org/r/20210326091443.26525-1-unixbhaskar@gmail.com Signed-off-by:
-
Kai Ye authored
delete unneeded variable initialization. Signed-off-by:
Kai Ye <yekai13@huawei.com> Link: https://lore.kernel.org/r/1616749747-3882-1-git-send-email-yekai13@huawei.com Signed-off-by:
-
Daniel Thompson authored
Currently kgdbts can get stuck waiting for do_sys_open() to be called in some of the current tests. This is because C compilers often automatically inline this function, which is a very thin wrapper around do_sys_openat2(), into some of its callers. gcc-10 does this on (at least) both x86 and arm64. We can fix the test suite by placing the breakpoints on do_sys_openat2() instead since that isn't (currently) inlined. However do_sys_openat2() is a static function so we cannot simply use an addressof. Since we are testing debug machinery it is acceptable to use kallsyms to lookup a suitable address because this is more or less what kdb does in the same circumstances. Re-implement lookup_addr() to be based on kallsyms rather than function pointers. Signed-off-by:
Daniel Thompson <daniel.thompson@linaro.org> Link: https://lore.kernel.org/r/20210325094807.3546702-1-daniel.thompson@linaro.org Signed-off-by:
-
Bhaskar Chowdhury authored
s/requsted/requested/ s/equests/requests/ s/occured/occurred/ s/conditon/condition/ s/emtpy/empty/ Acked-by:
Randy Dunlap <rdunlap@infradead.org> Signed-off-by:
-
Zou Wei authored
The sparse tool complains as follows: drivers/misc/sgi-xp/xp_main.c:24:22: warning: symbol 'xp_dbg_name' was not declared. Should it be static? drivers/misc/sgi-xp/xp_main.c:28:15: warning: symbol 'xp_dbg_subname' was not declared. Should it be static? These symbols are not used outside of xp_main.c, so this commit marks them static. Reported-by:
Zou Wei <zou_wei@huawei.com> Link: https://lore.kernel.org/r/20210324084823.7393-1-zou_wei@huawei.com Signed-off-by:
-
- Mar 24, 2021
-
-
Qiang Ma authored
These two lines of code don't meet the linux kernel style, and use a space after these keywords: if, switch, case, for, do, while. Do not add spaces around (inside) parenthesized expressions. So remove the redundant space. Signed-off-by:
Qiang Ma <maqianga@uniontech.com> Link: https://lore.kernel.org/r/20210309072059.22107-1-maqianga@uniontech.com Signed-off-by:
-
Andy Shevchenko authored
Rearrange optional stuff in pps_gpio_setup() so it will go after mandatory one and with reduced indentation. This will increase readability of the sources. Acked-by:
Rodolfo Giometti <giometti@enneenne.com> Signed-off-by:
Andy Shevchenko <andriy.shevchenko@linux.intel.com> Link: https://lore.kernel.org/r/20210318130321.24227-7-andriy.shevchenko@linux.intel.com Signed-off-by:
-
Andy Shevchenko authored
In most parts of the code the platform device is not used. Use struct device pointer directly to reduce code size and increase readability. Acked-by:
-
Andy Shevchenko authored
Device property API allows to gather device resources from different sources, such as ACPI. Convert the drivers to unleash the power of device property API. Acked-by:
-
Andy Shevchenko authored
Platform data is a legacy interface to supply device properties to the driver. In this case we even don't have in-kernel users for it. Just remove it for good. Acked-by:
-
Andy Shevchenko authored
The timer along with GPIO API are NULL-aware, there is no need to test against existing GPIO echo line. Acked-by:
-
Andy Shevchenko authored
When GPIO APIs return -EPROBE_DEFER there is no need to print the message, especially taking into consideration that it may repeat several times. Use dev_err_probe() to avoid log noise in such cases. Acked-by:
-
Andy Shevchenko authored
When requesting optional GPIO echo line, bail out on error, so user will know that something wrong with the existing property. Acked-by:
-
Marco Ballesio authored
User space needs to know if binder transactions occurred to frozen processes. Introduce a new BINDER_GET_FROZEN ioctl and keep track of transactions occurring to frozen proceses. Signed-off-by:
Marco Ballesio <balejs@google.com> Signed-off-by:
Li Li <dualli@google.com> Acked-by:
Todd Kjos <tkjos@google.com> Link: https://lore.kernel.org/r/20210316011630.1121213-4-dualli@chromium.org Signed-off-by:
-
Marco Ballesio authored
when interrupted by a signal, binder_wait_for_work currently returns -ERESTARTSYS. This error code isn't propagated to user space, but a way to handle interruption due to signals must be provided to code using this API. Replace this instance of -ERESTARTSYS with -EINTR, which is propagated to user space. binder_wait_for_work Signed-off-by:
-
Marco Ballesio authored
Frozen tasks can't process binder transactions, so a way is required to inform transmitting ends of communication failures due to the frozen state of their receiving counterparts. Additionally, races are possible between transitions to frozen state and binder transactions enqueued to a specific process. Implement BINDER_FREEZE ioctl for user space to inform the binder driver about the intention to freeze or unfreeze a process. When the ioctl is called, block the caller until any pending binder transactions toward the target process are flushed. Return an error to transactions to processes marked as frozen. Co-developed-by:
-
Gleb Fotengauer-Malinovskiy authored
__stringify macro function expands its arguments, but in this messages we expect to see ioctl request name instead of it's _IOC macro expansion. $ cat stringify.c #include <stdio.h> #include <asm/ioctl.h> #define __stringify_1(x) #x #define __stringify(x) __stringify_1(x) #define VMCI_DO_IOCTL(ioctl_name) \ char *name = __stringify(IOCTL_VMCI_ ## ioctl_name); int main() { VMCI_DO_IOCTL(INIT_CONTEXT) printf("%s\n", name); } $ cc stringify.c $ ./a.out (((0U) << (((0+8)+8)+14)) | (((7)) << (0+8)) | (((0xa0)) << 0) | ((0) << ((0+8)+8))) Signed-off-by:Gleb Fotengauer-Malinovskiy <glebfm@altlinux.org> Link: https://lore.kernel.org/r/20210302153346.300416-1-glebfm@altlinux.org Signed-off-by:
-
Yang Li authored
This fixes the following sparse warnings: drivers/misc/sgi-xp/xpc_main.c:210:23: warning: Using plain integer as NULL pointer Reported-by:
Abaci Robot <abaci@linux.alibaba.com> Reviewed-by:
Steve Wahl <steve.wahl@hpe.com> Signed-off-by:
Yang Li <yang.lee@linux.alibaba.com> Link: https://lore.kernel.org/r/1615885041-68750-1-git-send-email-yang.lee@linux.alibaba.com Signed-off-by:
-
Bhaskar Chowdhury authored
s/mesured/measured/ .......twice Acked-by:
-
dongjian authored
Use kobj_to_dev() instead of open-coding it Signed-off-by:
dongjian <dongjian@yulong.com> Link: https://lore.kernel.org/r/1615877987-32163-1-git-send-email-dj0227@163.com Signed-off-by:
-
Dan Carpenter authored
It's slightly cleaner to use the clamp() macro instead of open coding this. Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Link: https://lore.kernel.org/r/YEedHNwqEH8fvjkD@mwanda Signed-off-by:
-
Yang Li authored
This function always return '0' and no callers use the return value. So make it a void function. This eliminates the following coccicheck warning: ./drivers/char/mwave/tp3780i.c:182:5-11: Unneeded variable: "retval". Return "0" on line 187 Reported-by:
-
Shuo Liu authored
This fixes below build warning with extra build checks. $ make W=1 ../drivers/virt/acrn/vm.c:105: warning: expecting prototype for acrn_inject_msi(). Prototype was for acrn_msi_inject() instead Fixes: c7cf8d27 ("virt: acrn: Introduce interrupt injection interfaces") Signed-off-by:
Shuo Liu <shuo.a.liu@intel.com> Link: https://lore.kernel.org/r/20210311015206.19715-1-shuo.a.liu@intel.com Signed-off-by:
-
Laurent Dufour authored
It is better to rely on the API provided by the MM layer instead of directly manipulating the mm_users field. Acked-by:
Frederic Barrat <fbarrat@linux.ibm.com> Acked-by:
Andrew Donnellan <ajd@linux.ibm.com> Signed-off-by:
Laurent Dufour <ldufour@linux.ibm.com> Link: https://lore.kernel.org/r/20210310174405.51044-1-ldufour@linux.ibm.com Signed-off-by:
-
Bhaskar Chowdhury authored
s/filesytem/filesystem/ s/symantics/semantics/ Acked-by:
-
Hans de Goede authored
Logging an error when kmalloc fails is not necessary (and in general should be avoided) because the malloc failure will already complain loudly itself. Signed-off-by:
Hans de Goede <hdegoede@redhat.com> Link: https://lore.kernel.org/r/20210217102501.31758-3-hdegoede@redhat.com Signed-off-by:
-
Hans de Goede authored
Modern HP laptops do not necessarily actually contain a lis3lv02d sensor, yet they still define a HPQ6007 device in there ACPI tables. This leads to the following messages being logged in dmesg: [ 17.376342] hp_accel: laptop model unknown, using default axes configuration [ 17.399766] lis3lv02d: unknown sensor type 0x0 [ 17.399804] hp_accel: probe of HPQ6007:00 failed with error -22 The third message is unnecessary and does not provide any useful info, change the return value for unknown sensors to -ENODEV. This is the proper return value to indicate that the driver will not be handling the device and it silences the pr_warn printing the third message. Signed-off-by:
-
Hans de Goede authored
Before this commit lis3lv02d_get_pwron_wait() had a WARN_ONCE() to catch a potential divide by 0. WARN macros should only be used to catch internal kernel bugs and that is not the case here. We have been receiving a lot of bug reports about kernel backtraces caused by this WARN. The div value being checked comes from the lis3->odrs[] array. Which is sized to be a power-of-2 matching the number of bits in lis3->odr_mask. The only lis3 model where this array is not entirely filled with non zero values. IOW the only model where we can hit the div == 0 check is the 3dc ("8 bits 3DC sensor") model: int lis3_3dc_rates[16] = {0, 1, 10, 25, 50, 100, 200, 400, 1600, 5000}; Note the 0 value at index 0, according to the datasheet an odr index of 0 means "Power-down mode". HP typically uses a lis3 accelerometer for HDD fall protection. What I believe is happening here is that on newer HP devices, which only contain a SDD, the BIOS is leaving the lis3 device powered-down since it is not used for HDD fall protection. Note that the lis3_3dc_rates array initializer only specifies 10 values, which matches the datasheet. So it also contains 6 zero values at the end. Replace the WARN with a normal check, which treats an odr index of 0 as power-down and uses a normal dev_err() to report the error in case odr index point past the initialized part of the array. Fixes: 1510dd59 ("lis3lv02d: avoid divide by zero due to unchecked") Cc: stable@vger.kernel.org Signed-off-by: -
Geert Uytterhoeven authored
Remove the license boilerplate (containing an obsolete address), because we now have the SPDX header. Signed-off-by:
Geert Uytterhoeven <geert@linux-m68k.org> Link: https://lore.kernel.org/r/20210322141748.1062733-1-geert@linux-m68k.org Signed-off-by:
-
Arnd Bergmann authored
gcc-11 starts warning about misleading indentation inside of macros: drivers/misc/kgdbts.c: In function ‘kgdbts_break_test’: drivers/misc/kgdbts.c:103:9: error: this ‘if’ clause does not guard... [-Werror=misleading-indentation] 103 | if (verbose > 1) \ | ^~ drivers/misc/kgdbts.c:200:9: note: in expansion of macro ‘v2printk’ 200 | v2printk("kgdbts: breakpoint complete\n"); | ^~~~~~~~ drivers/misc/kgdbts.c:105:17: note: ...this statement, but the latter is misleadingly indented as if it were guarded by the ‘if’ 105 | touch_nmi_watchdog(); \ | ^~~~~~~~~~~~~~~~~~ The code looks correct to me, so just reindent it for readability. Fixes: e8d31c20 ("kgdb: add kgdb internal test suite") Acked-by: -
Nava kishore Manne authored
This patch adds support for Xilinx Dynamic Function eXchange(DFX) AXI shutdown manager IP. It can be used to safely handling the AXI traffic on a Reconfigurable Partition when it is undergoing dynamic reconfiguration and there by preventing system deadlock that may occur if AXI transactions are interrupted during reconfiguration. PR-Decoupler and AXI shutdown manager are completely different IPs. But both the IP registers are compatible and also both belong to the same sub-system (fpga-bridge).So using same driver for both IP's. Signed-off-by:
Nava kishore Manne <nava.manne@xilinx.com> Reviewed-by:
Moritz Fischer <mdf@kernel.org> Signed-off-by:
-
Nava kishore Manne authored
This patch Adds compatible value for Xilinx Dynamic Function eXchnage(DFX) AXI Shutdown manager IP. Signed-off-by:
Rob Herring <robh@kernel.org> Signed-off-by:
-
Michal Simek authored
Use already prepared dev_err_probe() introduced by commit a787e540 ("driver core: add device probe log helper"). It simplifies EPROBE_DEFER handling. Signed-off-by:
Michal Simek <michal.simek@xilinx.com> Signed-off-by:
-
