summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* kconfig: split menu.c out of parser.yMasahiro Yamada2021-04-144-5/+14
| | | | | | Compile menu.c as an independent compilation unit. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: nconf: refactor in print_in_middle()Masahiro Yamada2021-04-143-34/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This helper is the same as the sample code in the NCURSES HOWTO [1], but it is over-engineering to be used for nconf. I do not see any good reason to use the 'float' type just for the division by 2. All the call-sites pass a non-NULL pointer to the first argument, so 'if (win == NULL) win = stdscr;' is dead code. 'if (startx != 0) x = startx;' is dead code because 'x' will be overridden some lines below, by 'x = startx + (int)temp;'. All the call-sites pass a non-zero value to the second argument, so 'if (starty != 0)' is always true. getyx(win, y, x) is also dead-code because both 'y' and 'x' are overridden. All the call-sites pass 0 to the third parameter, so 'startx' can be removed. All the call-sites pass a non-zero value to the fourth parameter, so 'if (width == 0) width = 80;' is dead code. The window will be refreshed later, so there is no need to call refresh() in this function. Change the type of the last parameter from 'chtype' to 'int' to be aligned with the prototype, 'int wattrset(WINDOW *win, int attrs);' I also slightly cleaned up the indentation style. [1]: https://tldp.org/HOWTO/NCURSES-Programming-HOWTO/color.html Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: nconf: remove meaningless wattrset() call from show_menu()Masahiro Yamada2021-04-141-1/+0
| | | | | | | | | This attribute is not used because it will be overridden some lines below: wattrset(main_window, attr_main_menu_box); Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: nconf: change set_config_filename() to void functionMasahiro Yamada2021-04-141-2/+1
| | | | | | No one uses the return value of this function. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: nconf: refactor attributes setup codeMasahiro Yamada2021-04-143-175/+144
| | | | | | | | | | | | | | | | | | | | The current attributes setup code is strange; the array attribute[] is set to values outside the range of the attribute_t enum. At least, attributes_t attributes[ATTR_MAX+1] = {0}; ... should be int attribute[ATTR_MAX+1] = {0}; Also, there is no need to hard-code the color-pair numbers in attributes_t. The current code is messy. Rewrite it. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: nconf: remove unneeded default for menu promptMasahiro Yamada2021-04-141-4/+2
| | | | | | | | | | | | The rootmenu always has a prompt even if the 'mainmenu' statement is missing in the top Kconfig file. conf_parse() calls menu_add_prompt(P_MENU, "Main menu", NULL) in this case. So, every 'menu' has a prompt. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: nconf: get rid of (void) casts from wattrset() callsMasahiro Yamada2021-04-142-17/+17
| | | | | | | | | | | | | | | | | | | | | | | | | This reverts commit 10175ba65fde ("nconfig: Silence unused return values from wattrset"). With this patch applied, recent GCC versions can cleanly build nconf without "value computed is not used" warnings. The wattrset() used to be implemented as a macro, like this: #define wattrset(win,at) \ (NCURSES_OK_ADDR(win) \ ? ((win)->_attrs = NCURSES_CAST(attr_t, at), \ OK) \ : ERR) The GCC bugzilla [1] reported a false-positive -Wunused-value warning in a similar test case. It was fixed by GCC 4.4.1. Let's revert that commit, and see if somebody will claim the issue. [1]: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=39889 Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: nconf: fix NORMAL attributesMasahiro Yamada2021-04-141-2/+2
| | | | | | | The lower 8-bit of attributes should be 0, but this code wrongly sets it to NORMAL (=1). The correct one is A_NORMAL. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: mconf,nconf: remove unneeded '\0' termination after snprintf()Masahiro Yamada2021-04-142-17/+6
| | | | | | | | | | snprintf() always terminates the destination buffer with '\0' even if the buffer is not long enough. (In this case, the last element of the buffer becomes '\0'.) The explicit termination is unneeded. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: use /boot/config-* etc. as DEFCONFIG_LIST only for native buildMasahiro Yamada2021-04-142-4/+9
| | | | | | | | | | | | | | | | When the .config file is missing, 'make config', 'make menuconfig', etc. uses a file listed in DEFCONFIG_LIST, if found, as base configuration. Ususally, /boot/config-$(uname -r) exists, and is used as default. However, when you are cross-compiling the kernel, it does not make sense to use /boot/config-* on the build host. It should default to arch/$(SRCARCH)/configs/$(KBUILD_DEFCONFIG). UML previously did not use DEFCONFIG_LIST at all, but it should be able to use arch/um/configs/$(KBUILD_DEFCONFIG) as a base config file. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: change sym_change_count to a boolean flagMasahiro Yamada2021-04-147-24/+18
| | | | | | | | | | | | sym_change_count has no good reason to be 'int' type. sym_set_change_count() compares the old and new values after casting both of them to (bool). I do not see any practical diffrence between sym_set_change_count(1) and sym_add_change_count(1). Use the boolean flag, conf_changed. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: nconf: fix core dump when searching in empty menuMasahiro Yamada2021-04-141-3/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The following code in get_mext_match(): index = (index + items_num) % items_num; ... makes the program crash when items_num is zero (that is, the menu is empty). A menu can be empty when all the options in it are hidden by unmet 'depends on'. For example, menu "This menu will be empty" config FOO bool "foo" depends on BROKEN endmenu If you visit this menu and press a '/' key and then another key, nconf crashes with: Floating point exception (core dumped) When the number of items is zero, it does not make sense to search in the menu. In this case, current_item() returns NULL, and item_index() ERR, but get_mext_match() does not check it. Let's make get_mext_match() just return if the menu is empty. While I am here, change items_num from 'int' to 'unsigned int' because it should never become negative. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: lxdialog: A spello fix and a punctuation addedBhaskar Chowdhury2021-04-141-2/+2
| | | | | | | | s/propperly/properly/ s/thats/that\'s/ Signed-off-by: Bhaskar Chowdhury <unixbhaskar@gmail.com> Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: streamline_config.pl: Couple of typo fixesBhaskar Chowdhury2021-04-141-2/+2
| | | | | | | | s/configuraton/configuration/ s/orignal/original/ Signed-off-by: Bhaskar Chowdhury <unixbhaskar@gmail.com> Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: use true and false for bool variableYang Li2021-04-141-1/+1
| | | | | | | | | | fixed the following coccicheck: ./scripts/kconfig/confdata.c:36:9-10: WARNING: return of 0/1 in function 'is_dir' with return type bool Reported-by: Abaci Robot <abaci@linux.alibaba.com> Signed-off-by: Yang Li <yang.lee@linux.alibaba.com> Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: change "modules" from sub-option to first-level attributeMasahiro Yamada2021-04-149-27/+13
| | | | | | | | Now "modules" is the only member of the "option" property. Remove "option", and move "modules" to the top level property. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: remove allnoconfig_y optionMasahiro Yamada2021-04-147-24/+1
| | | | | | | Now that the only user, CONFIG_EMBEDDED has stopped using this option, remove it entirely. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: do not use allnoconfig_y optionMasahiro Yamada2021-04-143-2/+3
| | | | | | | | | | | | | | | | | | | allnoconfig_y is an ugly hack that sets a symbol to 'y' by allnoconfig. allnoconfig does not mean a minimal set of CONFIG options because a bunch of prompts are hidden by 'if EMBEDDED' or 'if EXPERT', but I do not like to hack Kconfig this way. Use the pre-existing feature, KCONFIG_ALLCONFIG, to provide a one liner config fragment. CONFIG_EMBEDDED=y is still forced when allnoconfig is invoked as a part of tinyconfig. No change in the .config file produced by 'make tinyconfig'. The output of 'make allnoconfig' will be changed; we will get CONFIG_EMBEDDED=n because allnoconfig literally sets all symbols to n. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: move default KBUILD_DEFCONFIG back to scripts/kconfig/MakefileMasahiro Yamada2021-04-142-3/+4
| | | | | | | | | | This is a partial revert of commit 2a86f6612164 ("kbuild: use KBUILD_DEFCONFIG as the fallback for DEFCONFIG_LIST"). Now that the reference to $(DEFCONFIG_LIST) was removed from init/Kconfig, the default KBUILD_DEFCONFIG can go back home. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: change defconfig_list option to environment variableMasahiro Yamada2021-04-1412-43/+50
| | | | | | | | | | | | | | | | "defconfig_list" is a weird option that defines a static symbol that declares the list of base config files in case the .config does not exist yet. This is quite different from other normal symbols; we just abused the "string" type and the "default" properties to list out the input files. They must be fixed values since these are searched for and loaded in the parse stage. It is an ugly hack, and should not exist in the first place. Providing this feature as an environment variable is a saner approach. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: move JUMP_NB to mconf.cMasahiro Yamada2021-04-142-2/+2
| | | | | | This macro is only used in mconf.c. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: move conf_set_all_new_symbols() to conf.cMasahiro Yamada2021-04-143-187/+193
| | | | | | | | | This function is only used in conf.c. Move it there together with the randomize_choice_values() helper. Define 'enum conf_def_mode' locally in conf.c as well. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: move conf_rewrite_mod_or_yes() to conf.cMasahiro Yamada2021-03-253-16/+15
| | | | | | This function is only used in conf.c. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: remove assignment for Kconfig fileMasahiro Yamada2021-03-251-2/+1
| | | | | | Pass av[optind] to conf_parse() directly. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: add help messages for --help (-h) and --silent (-s)Masahiro Yamada2021-03-251-3/+7
| | | | | | Add missing options and make the help message more readable. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: add long options --help and --silentMasahiro Yamada2021-03-251-0/+2
| | | | | | They are long options for -h and -s, respectively. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: refactor option parse codeMasahiro Yamada2021-03-251-48/+43
| | | | | | | | | | | | | | | | | | | | | | The current option parse code is clumsy. The 's' option is separately handled in an if-conditional due to the following code: input_mode = (enum input_mode)opt; If 's' is moved to the switch statement, the invalid value 's' would be assigned to the input_mode. Another potential problem is that we are mixing 'enum input_mode' and ASCII characters. They could overwrap if we add more input modes. To separate them out, set the flag field of long options to a pointer of input_mode_opt. For mode select options, getopt_long() returns 0, which never causes overwrap with ASCII characters that represent short options. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* kconfig: split randconfig setup code into set_randconfig_seed()Masahiro Yamada2021-03-251-23/+31
| | | | | | | | | This code is too big to be placed in the switch statement. Move the code into a new helper function. I slightly refactor the code without changing the behavior. Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
* Linux 5.12-rc3v5.12-rc3Linus Torvalds2021-03-141-1/+1
|
* prctl: fix PR_SET_MM_AUXV kernel stack leakAlexey Dobriyan2021-03-141-1/+1
| | | | | | | | | | | | | | | | | | Doing a prctl(PR_SET_MM, PR_SET_MM_AUXV, addr, 1); will copy 1 byte from userspace to (quite big) on-stack array and then stash everything to mm->saved_auxv. AT_NULL terminator will be inserted at the very end. /proc/*/auxv handler will find that AT_NULL terminator and copy original stack contents to userspace. This devious scheme requires CAP_SYS_RESOURCE. Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* Merge tag 'irq-urgent-2021-03-14' of ↵Linus Torvalds2021-03-147-18/+8
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip Pull irq fixes from Thomas Gleixner: "A set of irqchip updates: - Make the GENERIC_IRQ_MULTI_HANDLER configuration correct - Add a missing DT compatible string for the Ingenic driver - Remove the pointless debugfs_file pointer from struct irqdomain" * tag 'irq-urgent-2021-03-14' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: irqchip/ingenic: Add support for the JZ4760 dt-bindings/irq: Add compatible string for the JZ4760B irqchip: Do not blindly select CONFIG_GENERIC_IRQ_MULTI_HANDLER ARM: ep93xx: Select GENERIC_IRQ_MULTI_HANDLER directly irqdomain: Remove debugfs_file from struct irq_domain
| * Merge tag 'irqchip-fixes-5.12-1' of ↵Thomas Gleixner2021-03-1411444-286202/+488784
| |\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/maz/arm-platforms into irq/urgent Pull irqchip fixes from Marc Zyngier: - More compatible strings for the Ingenic irqchip (introducing the JZ4760B SoC) - Select GENERIC_IRQ_MULTI_HANDLER on the ARM ep93xx platform - Drop all GENERIC_IRQ_MULTI_HANDLER selections from the irqchip Kconfig, now relying on the architecture to get it right - Drop the debugfs_file field from struct irq_domain, now that debugfs can track things on its own
| | * irqchip/ingenic: Add support for the JZ4760Paul Cercueil2021-03-092-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | Add support for the interrupt controller found in the JZ4760 SoC, which works exactly like the one in the JZ4770. Signed-off-by: Paul Cercueil <paul@crapouillou.net> Signed-off-by: Marc Zyngier <maz@kernel.org> Link: https://lore.kernel.org/r/20210307172014.73481-2-paul@crapouillou.net
| | * dt-bindings/irq: Add compatible string for the JZ4760BPaul Cercueil2021-03-091-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add the ingenic,jz4760b-intc compatible string with a fallback to the ingenic,jz4760-intc compatible string. Signed-off-by: Paul Cercueil <paul@crapouillou.net> Acked-by: Rob Herring <robh@kernel.org> Signed-off-by: Marc Zyngier <maz@kernel.org> Link: https://lore.kernel.org/r/20210307172014.73481-1-paul@crapouillou.net
| | * irqchip: Do not blindly select CONFIG_GENERIC_IRQ_MULTI_HANDLERMarc Zyngier2021-03-081-9/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Implementing CONFIG_GENERIC_IRQ_MULTI_HANDLER is a decision that is made at the architecture level, and shouldn't involve the irqchip at all (we even provide a fallback helper when the option isn't selected). Drop all instances of such selection from non-arch code. Signed-off-by: Marc Zyngier <maz@kernel.org> Link: https://lore.kernel.org/r/20210217142800.2547737-1-maz@kernel.org Signed-off-by: Mark Rutland <mark.rutland@arm.com> Tested-by: Hector Martin <marcan@marcan.st> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: James Morse <james.morse@arm.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Will Deacon <will@kernel.org> Signed-off-by: Marc Zyngier <maz@kernel.org>
| | * ARM: ep93xx: Select GENERIC_IRQ_MULTI_HANDLER directlyMarc Zyngier2021-03-081-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ep93xx currently relies of CONFIG_ARM_VIC to select GENERIC_IRQ_MULTI_HANDLER. Given that this is logically a platform architecture property, add the selection of GENERIC_IRQ_MULTI_HANDLER at the platform level. Further patches will remove the selection from the irqchip side. Reported-by: Marc Rutland <mark.rutland@arm.com> Signed-off-by: Marc Zyngier <maz@kernel.org> Signed-off-by: Mark Rutland <mark.rutland@arm.com> Tested-by: Hector Martin <marcan@marcan.st> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: James Morse <james.morse@arm.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Will Deacon <will@kernel.org> Signed-off-by: Marc Zyngier <maz@kernel.org>
| | * irqdomain: Remove debugfs_file from struct irq_domainGreg Kroah-Hartman2021-03-082-9/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There's no need to keep around a dentry pointer to a simple file that debugfs itself can look up when we need to remove it from the system. So simplify the code by deleting the variable and cleaning up the logic around the debugfs file. Cc: Marc Zyngier <maz@kernel.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-kernel@vger.kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Marc Zyngier <maz@kernel.org> Link: https://lore.kernel.org/r/YCvYV53ZdzQSWY6w@kroah.com
* | | Merge tag 'timers-urgent-2021-03-14' of ↵Linus Torvalds2021-03-141-21/+39
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip Pull timer fix from Thomas Gleixner: "A single fix in for hrtimers to prevent an interrupt storm caused by the lack of reevaluation of the timers which expire in softirq context under certain circumstances, e.g. when the clock was set" * tag 'timers-urgent-2021-03-14' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: hrtimer: Update softirq_expires_next correctly after __hrtimer_get_next_event()
| * | | hrtimer: Update softirq_expires_next correctly after __hrtimer_get_next_event()Anna-Maria Behnsen2021-03-081-21/+39
| | |/ | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | hrtimer_force_reprogram() and hrtimer_interrupt() invokes __hrtimer_get_next_event() to find the earliest expiry time of hrtimer bases. __hrtimer_get_next_event() does not update cpu_base::[softirq_]_expires_next to preserve reprogramming logic. That needs to be done at the callsites. hrtimer_force_reprogram() updates cpu_base::softirq_expires_next only when the first expiring timer is a softirq timer and the soft interrupt is not activated. That's wrong because cpu_base::softirq_expires_next is left stale when the first expiring timer of all bases is a timer which expires in hard interrupt context. hrtimer_interrupt() does never update cpu_base::softirq_expires_next which is wrong too. That becomes a problem when clock_settime() sets CLOCK_REALTIME forward and the first soft expiring timer is in the CLOCK_REALTIME_SOFT base. Setting CLOCK_REALTIME forward moves the clock MONOTONIC based expiry time of that timer before the stale cpu_base::softirq_expires_next. cpu_base::softirq_expires_next is cached to make the check for raising the soft interrupt fast. In the above case the soft interrupt won't be raised until clock monotonic reaches the stale cpu_base::softirq_expires_next value. That's incorrect, but what's worse it that if the softirq timer becomes the first expiring timer of all clock bases after the hard expiry timer has been handled the reprogramming of the clockevent from hrtimer_interrupt() will result in an interrupt storm. That happens because the reprogramming does not use cpu_base::softirq_expires_next, it uses __hrtimer_get_next_event() which returns the actual expiry time. Once clock MONOTONIC reaches cpu_base::softirq_expires_next the soft interrupt is raised and the storm subsides. Change the logic in hrtimer_force_reprogram() to evaluate the soft and hard bases seperately, update softirq_expires_next and handle the case when a soft expiring timer is the first of all bases by comparing the expiry times and updating the required cpu base fields. Split this functionality into a separate function to be able to use it in hrtimer_interrupt() as well without copy paste. Fixes: 5da70160462e ("hrtimer: Implement support for softirq based hrtimers") Reported-by: Mikael Beckius <mikael.beckius@windriver.com> Suggested-by: Thomas Gleixner <tglx@linutronix.de> Tested-by: Mikael Beckius <mikael.beckius@windriver.com> Signed-off-by: Anna-Maria Behnsen <anna-maria@linutronix.de> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Ingo Molnar <mingo@kernel.org> Link: https://lore.kernel.org/r/20210223160240.27518-1-anna-maria@linutronix.de
* | | Merge tag 'sched-urgent-2021-03-14' of ↵Linus Torvalds2021-03-142-67/+63
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip Pull scheduler fixes from Thomas Gleixner: "A set of scheduler updates: - Prevent a NULL pointer dereference in the migration_stop_cpu() mechanims - Prevent self concurrency of affine_move_task() - Small fixes and cleanups related to task migration/affinity setting - Ensure that sync_runqueues_membarrier_state() is invoked on the current CPU when it is in the cpu mask" * tag 'sched-urgent-2021-03-14' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: sched/membarrier: fix missing local execution of ipi_sync_rq_state() sched: Simplify set_affinity_pending refcounts sched: Fix affine_move_task() self-concurrency sched: Optimize migration_cpu_stop() sched: Collate affine_move_task() stoppers sched: Simplify migration_cpu_stop() sched: Fix migration_cpu_stop() requeueing
| * | | sched/membarrier: fix missing local execution of ipi_sync_rq_state()Mathieu Desnoyers2021-03-061-3/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The function sync_runqueues_membarrier_state() should copy the membarrier state from the @mm received as parameter to each runqueue currently running tasks using that mm. However, the use of smp_call_function_many() skips the current runqueue, which is unintended. Replace by a call to on_each_cpu_mask(). Fixes: 227a4aadc75b ("sched/membarrier: Fix p->mm->membarrier_state racy load") Reported-by: Nadav Amit <nadav.amit@gmail.com> Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Cc: stable@vger.kernel.org # 5.4.x+ Link: https://lore.kernel.org/r/74F1E842-4A84-47BF-B6C2-5407DFDD4A4A@gmail.com
| * | | sched: Simplify set_affinity_pending refcountsPeter Zijlstra2021-03-061-12/+20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Now that we have set_affinity_pending::stop_pending to indicate if a stopper is in progress, and we have the guarantee that if that stopper exists, it will (eventually) complete our @pending we can simplify the refcount scheme by no longer counting the stopper thread. Fixes: 6d337eab041d ("sched: Fix migrate_disable() vs set_cpus_allowed_ptr()") Cc: stable@kernel.org Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Reviewed-by: Valentin Schneider <valentin.schneider@arm.com> Link: https://lkml.kernel.org/r/20210224131355.724130207@infradead.org
| * | | sched: Fix affine_move_task() self-concurrencyPeter Zijlstra2021-03-061-3/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Consider: sched_setaffinity(p, X); sched_setaffinity(p, Y); Then the first will install p->migration_pending = &my_pending; and issue stop_one_cpu_nowait(pending); and the second one will read p->migration_pending and _also_ issue: stop_one_cpu_nowait(pending), the _SAME_ @pending. This causes stopper list corruption. Add set_affinity_pending::stop_pending, to indicate if a stopper is in progress. Fixes: 6d337eab041d ("sched: Fix migrate_disable() vs set_cpus_allowed_ptr()") Cc: stable@kernel.org Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Reviewed-by: Valentin Schneider <valentin.schneider@arm.com> Link: https://lkml.kernel.org/r/20210224131355.649146419@infradead.org
| * | | sched: Optimize migration_cpu_stop()Peter Zijlstra2021-03-061-1/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When the purpose of migration_cpu_stop() is to migrate the task to 'any' valid CPU, don't migrate the task when it's already running on a valid CPU. Fixes: 6d337eab041d ("sched: Fix migrate_disable() vs set_cpus_allowed_ptr()") Cc: stable@kernel.org Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Reviewed-by: Valentin Schneider <valentin.schneider@arm.com> Link: https://lkml.kernel.org/r/20210224131355.569238629@infradead.org
| * | | sched: Collate affine_move_task() stoppersPeter Zijlstra2021-03-061-15/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The SCA_MIGRATE_ENABLE and task_running() cases are almost identical, collapse them to avoid further duplication. Fixes: 6d337eab041d ("sched: Fix migrate_disable() vs set_cpus_allowed_ptr()") Cc: stable@kernel.org Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Reviewed-by: Valentin Schneider <valentin.schneider@arm.com> Link: https://lkml.kernel.org/r/20210224131355.500108964@infradead.org
| * | | sched: Simplify migration_cpu_stop()Peter Zijlstra2021-03-061-48/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When affine_move_task() issues a migration_cpu_stop(), the purpose of that function is to complete that @pending, not any random other p->migration_pending that might have gotten installed since. This realization much simplifies migration_cpu_stop() and allows further necessary steps to fix all this as it provides the guarantee that @pending's stopper will complete @pending (and not some random other @pending). Fixes: 6d337eab041d ("sched: Fix migrate_disable() vs set_cpus_allowed_ptr()") Cc: stable@kernel.org Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Reviewed-by: Valentin Schneider <valentin.schneider@arm.com> Link: https://lkml.kernel.org/r/20210224131355.430014682@infradead.org
| * | | sched: Fix migration_cpu_stop() requeueingPeter Zijlstra2021-03-061-11/+28
| |/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When affine_move_task(p) is called on a running task @p, which is not otherwise already changing affinity, we'll first set p->migration_pending and then do: stop_one_cpu(cpu_of_rq(rq), migration_cpu_stop, &arg); This then gets us to migration_cpu_stop() running on the CPU that was previously running our victim task @p. If we find that our task is no longer on that runqueue (this can happen because of a concurrent migration due to load-balance etc.), then we'll end up at the: } else if (dest_cpu < 1 || pending) { branch. Which we'll take because we set pending earlier. Here we first check if the task @p has already satisfied the affinity constraints, if so we bail early [A]. Otherwise we'll reissue migration_cpu_stop() onto the CPU that is now hosting our task @p: stop_one_cpu_nowait(cpu_of(rq), migration_cpu_stop, &pending->arg, &pending->stop_work); Except, we've never initialized pending->arg, which will be all 0s. This then results in running migration_cpu_stop() on the next CPU with arg->p == NULL, which gives the by now obvious result of fireworks. The cure is to change affine_move_task() to always use pending->arg, furthermore we can use the exact same pattern as the SCA_MIGRATE_ENABLE case, since we'll block on the pending->done completion anyway, no point in adding yet another completion in stop_one_cpu(). This then gives a clear distinction between the two migration_cpu_stop() use cases: - sched_exec() / migrate_task_to() : arg->pending == NULL - affine_move_task() : arg->pending != NULL; And we can have it ignore p->migration_pending when !arg->pending. Any stop work from sched_exec() / migrate_task_to() is in addition to stop works from affine_move_task(), which will be sufficient to issue the completion. Fixes: 6d337eab041d ("sched: Fix migrate_disable() vs set_cpus_allowed_ptr()") Cc: stable@kernel.org Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Reviewed-by: Valentin Schneider <valentin.schneider@arm.com> Link: https://lkml.kernel.org/r/20210224131355.357743989@infradead.org
* | | Merge tag 'objtool-urgent-2021-03-14' of ↵Linus Torvalds2021-03-142-6/+7
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip Pull objtool fix from Thomas Gleixner: "A single objtool fix to handle the PUSHF/POPF validation correctly for the paravirt changes which modified arch_local_irq_restore not to use popf" * tag 'objtool-urgent-2021-03-14' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: objtool,x86: Fix uaccess PUSHF/POPF validation
| * | | objtool,x86: Fix uaccess PUSHF/POPF validationPeter Zijlstra2021-03-122-6/+7
| |/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Commit ab234a260b1f ("x86/pv: Rework arch_local_irq_restore() to not use popf") replaced "push %reg; popf" with something like: "test $0x200, %reg; jz 1f; sti; 1:", which breaks the pushf/popf symmetry that commit ea24213d8088 ("objtool: Add UACCESS validation") relies on. The result is: drivers/gpu/drm/amd/amdgpu/si.o: warning: objtool: si_common_hw_init()+0xf36: PUSHF stack exhausted Meanwhile, commit c9c324dc22aa ("objtool: Support stack layout changes in alternatives") makes that we can actually use stack-ops in alternatives, which means we can revert 1ff865e343c2 ("x86,smap: Fix smap_{save,restore}() alternatives"). That in turn means we can limit the PUSHF/POPF handling of ea24213d8088 to those instructions that are in alternatives. Fixes: ab234a260b1f ("x86/pv: Rework arch_local_irq_restore() to not use popf") Reported-by: Borislav Petkov <bp@alien8.de> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Acked-by: Josh Poimboeuf <jpoimboe@redhat.com> Link: https://lkml.kernel.org/r/YEY4rIbQYa5fnnEp@hirez.programming.kicks-ass.net
* | | Merge tag 'locking-urgent-2021-03-14' of ↵Linus Torvalds2021-03-143-10/+9
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip Pull locking fixes from Thomas Gleixner: "A couple of locking fixes: - A fix for the static_call mechanism so it handles unaligned addresses correctly. - Make u64_stats_init() a macro so every instance gets a seperate lockdep key. - Make seqcount_latch_init() a macro as well to preserve the static variable which is used for the lockdep key" * tag 'locking-urgent-2021-03-14' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: seqlock,lockdep: Fix seqcount_latch_init() u64_stats,lockdep: Fix u64_stats_init() vs lockdep static_call: Fix the module key fixup
generated by cgit v1.2.3 (git 2.39.1) at 2025-01-30 02:07:49 +0100