- 21 Jun, 2021 40 commits
-
-
Sai Prakash Ranjan authored
During suspend/resume usecases and tests, it is common to see issues such as lockups either in suspend path or resume path because of the bugs in the corresponding device driver pm handling code. In such cases, it is important that watchdog is active to make sure that we either receive a watchdog pretimeout notification or a bite causing reset instead of a hang causing us to hard reset the machine. There are good reasons as to why we need this because: * We can have a watchdog pretimeout governor set to panic in which case we can have a backtrace which would help identify the issue with the particular driver and cause a normal reboot. * Even in case where there is no pretimeout support, a watchdog bite is still useful because some firmware has debug support to dump CPU core context on watchdog bite for post-mortem analysis. * One more usecase which comes to mind is of warm reboot. In case we hard reset the target, a cold reboot could be induced resulting in lose of ddr contents thereby losing all the debug info. Currently, the watchdog pm callback just invokes the usual suspend and resume callback which do not have any special ordering in the sense that a watchdog can be suspended before the buggy device driver suspend callback and watchdog resume can happen after the buggy device driver resume callback. This would mean that the watchdog will not be active when the buggy driver cause the lockups thereby hanging the system. So to make sure this doesn't happen, move the watchdog pm to use late/early system pm callbacks which will ensure that the watchdog is suspended late and resumed early so that it can catch such issues. Signed-off-by:
Sai Prakash Ranjan <saiprakash.ranjan@codeaurora.org> Reviewed-by:
Guenter Roeck <linux@roeck-us.net> Reviewed-by:
Douglas Anderson <dianders@chromium.org> Link: https://lore.kernel.org/r/20210310202004.1436-1-saiprakash.ranjan@codeaurora.orgSigned-off-by:
Wim Van Sebroeck <wim@linux-watchdog.org>
-
Bhaskar Chowdhury authored
s/freqency/frequency/ Signed-off-by:
Bhaskar Chowdhury <unixbhaskar@gmail.com> Acked-by:
Randy Dunlap <rdunlap@infradead.org> Reviewed-by:
-
Wei Yongjun authored
In case of error, the function device_node_to_regmap() returns ERR_PTR() and never returns NULL. The NULL test in the return value check should be replaced with IS_ERR(). Fixes: 6d532143 ("watchdog: jz4740: Use regmap provided by TCU driver") Reported-by:
Hulk Robot <hulkci@huawei.com> Signed-off-by:
Wei Yongjun <weiyongjun1@huawei.com> Acked-by:
Paul Cercueil <paul@crapouillou.net> Link: https://lore.kernel.org/r/20210304045909.945799-1-weiyongjun1@huawei.comSigned-off-by:
-
Christophe Leroy authored
Commit 92c8c16f ("powerpc/embedded6xx: Remove C2K board support") removed the last selector of CONFIG_MV64X60. Therefore CONFIG_MV64X60_WDT cannot be selected anymore and can be removed. Signed-off-by:
Christophe Leroy <christophe.leroy@csgroup.eu> Reviewed-by:
-
Wang Qing authored
Add description of pre-timeout in mtk-wdt. Signed-off-by:
Wang Qing <wangqing@vivo.com> Acked-by:
Rob Herring <robh@kernel.org> Link: https://lore.kernel.org/r/1619148020-2236-3-git-send-email-wangqing@vivo.comSigned-off-by:
-
Wang Qing authored
Use the bark interrupt as the pretimeout notifier if available. When the watchdog timer expires in dual mode, an interrupt will be triggered first, then the timing restarts. The reset signal will be initiated when the timer expires again. The pretimeout notification shall occur at timeout-sec/2. V2: - panic() by default if WATCHDOG_PRETIMEOUT_GOV is not enabled. V3: - Modify the pretimeout behavior, manually reset after the pretimeout - is processed and wait until timeout. V4: - Remove pretimeout related processing. - Add dual mode control separately. V5: - Fix some formatting and printing problems. V6: - Realize pretimeout processing through dualmode. V7: - Add set_pretimeout(). V8/V9: - Fix some formatting problems. Signed-off-by:
-
Lukas Bulwahn authored
The command 'find drivers/watchdog | xargs ./scripts/kernel-doc -none' reports a number of kernel-doc warnings in the watchdog subsystem. Address the kernel-doc warnings that were purely syntactic issues with kernel-doc comments. The remaining kernel-doc warnings are of type "Excess function parameter" and "Function parameter or member not described". These warnings would need to be addressed in a second pass with a bit more insight into the APIs and purpose of the functions in the watchdog subsystem. Signed-off-by:
Lukas Bulwahn <lukas.bulwahn@gmail.com> Reviewed-by:
-
Flavio Suligoi authored
Fix the following typo: "recommeded" --> "recommended" "firmare" --> "firmware" Signed-off-by:
Flavio Suligoi <f.suligoi@asem.it> Reviewed-by:
Mika Westerberg <mika.westerberg@linux.intel.com> Reviewed-by:
-
Álvaro Fernández Rojas authored
bcm7038_wdt can be used on bmips big endian (bcm63xx) devices too. Signed-off-by:
Álvaro Fernández Rojas <noltari@gmail.com> Reviewed-by:
-
Bhaskar Chowdhury authored
s/parmeter/parameter/ Signed-off-by:
-
Srinivas Neeli authored
"%p" is not printing the pointer value. In driver, printing pointer value is not useful so avoiding print. iSigned-off-by:
Srinivas Neeli <srinivas.neeli@xilinx.com> Reviewed-by:
-
Srinivas Neeli authored
clk is an optional property, if clock not defined, calling clk_prepare_enable() and devm_add_action_or_reset() are not useful. so calling these two apis only when clock is present. Addresses-Coverity:"FORWARD_NULL" Signed-off-by:
-
Srinivas Goud authored
This patch removes pr_info in stop function and adds dev_dbg() in start/stop function to display device specific debug info. Signed-off-by:
Srinivas Goud <srinivas.goud@xilinx.com> Signed-off-by:
Michal Simek <michal.simek@xilinx.com> Signed-off-by:
-
Srinivas Goud authored
Used BIT macro instead of mask value. Signed-off-by:
-
Srinivas Goud authored
Based on checkpatch every spinlock should be documented. The patch is fixing this issue: ./scripts/checkpatch.pl --strict -f drivers/watchdog/of_xilinx_wdt.c CHECK: spinlock_t definition without comment + spinlock_t spinlock; Signed-off-by:
-
Tian Tao authored
Use the device_get_match_data() helper instead of open coding. Signed-off-by:
Tian Tao <tiantao6@hisilicon.com> Reviewed-by:
-
Wong Vee Khee authored
s/thershold/threshold Cc: Vijayakannan Ayyathurai <vijayakannan.ayyathurai@intel.com> Signed-off-by:
Wong Vee Khee <vee.khee.wong@linux.intel.com> Reviewed-by:
-
Jiapeng Chong authored
Fix the following clang warning: drivers/watchdog/it87_wdt.c:155:20: warning: unused function 'superio_outw' [-Wunused-function]. Reported-by:
Abaci Robot <abaci@linux.alibaba.com> Signed-off-by:
Jiapeng Chong <jiapeng.chong@linux.alibaba.com> Reviewed-by:
-
Tao Ren authored
Fix hardware timeout calculation in aspeed_wdt_set_timeout function to ensure the reload value does not exceed the hardware limit. Fixes: efa859f7 ("watchdog: Add Aspeed watchdog driver") Reported-by:
Amithash Prasad <amithash@fb.com> Signed-off-by:
Tao Ren <rentao.bupt@gmail.com> Reviewed-by:
-
Randy Dunlap authored
Change a non-working ftp: URL to https:. Wrap long lines earlier. Spell "IP" with capital letters. Change "it`s" to "it's". The backtick (grave accent) is not an apostrophe. Signed-off-by:
-
Juerg Haefliger authored
Convert sprintf() in sysfs "show" functions to sysfs_emit() and sysfs_emit_at() in order to check for buffer overruns in sysfs outputs. Signed-off-by:
Juerg Haefliger <juergh@canonical.com> Reviewed-by:
-
Zou Wei authored
This driver's remove path calls del_timer(). However, that function does not wait until the timer handler finishes. This means that the timer handler may still be running after the driver's remove function has finished, which would result in a use-after-free. Fix by calling del_timer_sync(), which makes sure the timer handler has finished, and unable to re-schedule itself. Reported-by:
Zou Wei <zou_wei@huawei.com> Reviewed-by:
Vladimir Zapolskiy <vz@mleia.com> Link: https://lore.kernel.org/r/1620802676-19701-1-git-send-email-zou_wei@huawei.comSigned-off-by:
-
Zou Wei authored
This module's remove path calls del_timer(). However, that function does not wait until the timer handler finishes. This means that the timer handler may still be running after the driver's remove function has finished, which would result in a use-after-free. Fix by calling del_timer_sync(), which makes sure the timer handler has finished, and unable to re-schedule itself. Reported-by:
-
Juerg Haefliger authored
The valid range for the 'timeout' value is useful information so expose the min and max timeout values via sysfs. Signed-off-by:
-
Andy Shevchenko authored
Kernel doc validation script is not happy CHECK .../sp805_wdt.c .../sp805_wdt.c:73: warning: Function parameter or member 'rate' not described in 'sp805_wdt' Fix this by describing rate parameter. While at it, mark clk one optional. Fixes: dc0e4a3b ("watchdog: sp805: Add clock-frequency property") Cc: Srinath Mannam <srinath.mannam@broadcom.com> Signed-off-by:
Andy Shevchenko <andriy.shevchenko@linux.intel.com> Reviewed-by:
-
Andy Shevchenko authored
Replace open coded variants of devm_clk_get_optional(). While at it, drop unneeded OF and ACPI dependency as the APIs in use are provider agnostic. Cc: Srinath Mannam <srinath.mannam@broadcom.com> Signed-off-by:
Srinath Mannam <srinath.mannam@broadcom.com> Link: https://lore.kernel.org/r/20210517174456.22050-1-andriy.shevchenko@linux.intel.comSigned-off-by:
-
Jerry Hoemann authored
New hardware with new PCI ID info. Signed-off-by:
Jerry Hoemann <jerry.hoemann@hpe.com> Reviewed-by:
-
Jiapeng Chong authored
Clean up the following includecheck warning: ./drivers/watchdog/dw_wdt.c: linux/kernel.h is included more than once. Reported-by:
-
Zou Wei authored
This module's remove path calls del_timer(). However, that function does not wait until the timer handler finishes. This means that the timer handler may still be running after the driver's remove function has finished, which would result in a use-after-free. Fix by calling del_timer_sync(), which makes sure the timer handler has finished, and unable to re-schedule itself. Reported-by:
-
Robin Gong authored
Since watchdog_ping_work is not freezable so that it maybe scheduled before imx2_wdt_resume where watchdog clock enabled, hence, kernel will hang in imx2_wdt_ping without clock, and then watchdog reset happen. Add clk_is_on to prevent the above case by ignoring ping until watchdog driver resume back indeed. Signed-off-by:
Robin Gong <yibin.gong@nxp.com> Reviewed-by:
-
Randy Dunlap authored
The mtx-1_wdt driver does not need the au1000.h header file. Instead, the header file causes build errors, so drop it. This change fixes multiple build errors, all in au1000.h. E.g.: In file included from ../drivers/watchdog/mtx-1_wdt.c:44: ../arch/mips/include/asm/mach-au1x00/au1000.h: In function 'alchemy_rdsys': ../arch/mips/include/asm/mach-au1x00/au1000.h:603:36: error: implicit declaration of function 'KSEG1ADDR'; did you mean 'CKSEG1ADDR'? [-Werror=implicit-function-declaration] 603 | void __iomem *b = (void __iomem *)KSEG1ADDR(AU1000_SYS_PHYS_ADDR); ../arch/mips/include/asm/mach-au1x00/au1000.h:603:20: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast] 603 | void __iomem *b = (void __iomem *)KSEG1ADDR(AU1000_SYS_PHYS_ADDR); Fixes: da2a68b3 ("watchdog: Enable COMPILE_TEST where possible") Signed-off-by:
Florian Fainelli <f.fainelli@gmail.com> Link: https://lore.kernel.org/r/20210516211703.25349-1-rdunlap@infradead.orgSigned-off-by:
-
Shaokun Zhang authored
For Armv8.6, The frequency of CNTFRQ_EL0 is standardized to a frequency of 1GHz, so Arm Base System Architecture 1.0[1] has introduced watchdog revision 1 that increases the length the watchdog offset register to 48 bit, while other operation of the watchdog remains the same. The driver can determine which version of the watchdog is implemented through the watchdog interface identification register (W_IID). If the version is 0x1, the watchdog offset register will be 48 bit, otherwise it will be 32 bit. [1] https://developer.arm.com/documentation/den0094/latest Cc: Wim Van Sebroeck <wim@linux-watchdog.org> Cc: Guenter Roeck <linux@roeck-us.net> Cc: Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com> Cc: Al Stone <al.stone@linaro.org> Cc: Jianchao Hu <hujianchao@hisilicon.com> Cc: Huiqiang Wang <wanghuiqiang@huawei.com> Signed-off-by:
Shaokun Zhang <zhangshaokun@hisilicon.com> Reviewed-by:
-
Shruthi Sanil authored
Corrected typos, aligned the tabs and added new lines wherever required for better readability Reviewed-by:
Kris Pan <kris.pan@intel.com> Signed-off-by:
Shruthi Sanil <shruthi.sanil@intel.com> Link: https://lore.kernel.org/r/20210517174953.19404-10-shruthi.sanil@intel.comSigned-off-by:
-
Shruthi Sanil authored
Updated the WDT SMC handler MACRO name to make it clear that its a ARM SMC handler that helps in clearing the WDT interrupt bit. Reviewed-by:
-
Shruthi Sanil authored
In the TO ISR removed updating the Timeout value because its not serving any purpose as the timer would have already expired and the system would be rebooting. Fixes: fa0f8d51 ("watchdog: Add watchdog driver for Intel Keembay Soc") Reviewed-by:
-
Shruthi Sanil authored
Introduced MACRO's for WDT enable and disable values for better readability Reviewed-by:
-
Shruthi Sanil authored
Removed set timeout from the start WDT function. There is a function defined to set the timeout. Hence no need to set the timeout again in start function as the timeout would have been already updated before calling the start/enable. Fixes: fa0f8d51 ("watchdog: Add watchdog driver for Intel Keembay Soc") Reviewed-by:
-
Shruthi Sanil authored
During the interrupt service routine of the TimeOut interrupt and the ThresHold interrupt, the respective interrupt clear bit have to be cleared and not both. Fixes: fa0f8d51 ("watchdog: Add watchdog driver for Intel Keembay Soc") Reviewed-by:
-
Shruthi Sanil authored
The pretimeout has to be updated to zero during the ISR of the ThresHold interrupt. Else the TH interrupt would be triggerred for every tick until the timeout. Fixes: fa0f8d51 ("watchdog: Add watchdog driver for Intel Keembay Soc") Reviewed-by:
-
Shruthi Sanil authored
The pre-timeout value to be programmed to the register has to be calculated and updated for every change in the timeout value. Else the threshold time wouldn't be calculated to its corresponding timeout. Fixes: fa0f8d51 ("watchdog: Add watchdog driver for Intel Keembay Soc") Reviewed-by:
-
