- 13 Jan, 2023 21 commits
-
-
Peter Zijlstra authored
Current arch_cpu_idle() is called with IRQs disabled, but will return with IRQs enabled. However, the very first thing the generic code does after calling arch_cpu_idle() is raw_local_irq_disable(). This means that architectures that can idle with IRQs disabled end up doing a pointless 'enable-disable' dance. Therefore, push this IRQ disabling into the idle function, meaning that those architectures can avoid the pointless IRQ state flipping. Signed-off-by:
Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by:
Ingo Molnar <mingo@kernel.org> Tested-by:
Tony Lindgren <tony@atomide.com> Tested-by:
Ulf Hansson <ulf.hansson@linaro.org> Reviewed-by:
Gautham R. Shenoy <gautham.shenoy@amd.com> Acked-by: Mark Rutland <mark.rutland@arm.com> [arm64] Acked-by:
Rafael J. Wysocki <rafael.j.wysocki@intel.com> Acked-by:
Guo Ren <guoren@kernel.org> Acked-by:
Frederic Weisbecker <frederic@kernel.org> Link: https://lore.kernel.org/r/20230112195540.618076436@infradead.org
-
Peter Zijlstra authored
objtool to the rescue: vmlinux.o: warning: objtool: intel_idle_ibrs+0x17: call to spec_ctrl_current() leaves .noinstr.text section vmlinux.o: warning: objtool: intel_idle_ibrs+0x27: call to wrmsrl.constprop.0() leaves .noinstr.text section Signed-off-by:
-
Peter Zijlstra authored
Fix instrumentation bugs objtool found: vmlinux.o: warning: objtool: intel_idle_s2idle+0xd5: call to fpu_idle_fpregs() leaves .noinstr.text section vmlinux.o: warning: objtool: intel_idle_xstate+0x11: call to fpu_idle_fpregs() leaves .noinstr.text section vmlinux.o: warning: objtool: fpu_idle_fpregs+0x9: call to xfeatures_in_use() leaves .noinstr.text section Signed-off-by:
-
Peter Zijlstra authored
So objtool found this bug: vmlinux.o: warning: objtool: intel_idle_irq+0x10c: call to trace_hardirqs_off() leaves .noinstr.text section As per commit 32d4fd57 ("cpuidle,intel_idle: Fix CPUIDLE_FLAG_IRQ_ENABLE"): "must not have tracing in idle functions" Clearly people can't read and tinker along until splat dissapears. This straight up reverts commit d295ad34 ("intel_idle: Fix false positive RCU splats due to incorrect hardirqs state"). It doesn't re-introduce the problem because preceding patches fixed it properly. Fixes: d295ad34 ("intel_idle: Fix false positive RCU splats due to incorrect hardirqs state") Signed-off-by:
-
Peter Zijlstra authored
Idle code is very like entry code in that RCU isn't available. As such, add a little validation. Signed-off-by:
Geert Uytterhoeven <geert@linux-m68k.org> Acked-by:
-
Peter Zijlstra authored
The __cpuidle functions will become a noinstr class, as such they need explicit annotations. Signed-off-by:
-
Peter Zijlstra authored
All the idle routines are called with RCU disabled, as such there must not be any tracing inside. While there; clean-up the io-port idle thing. Signed-off-by:
-
Peter Zijlstra authored
All callers should still have RCU enabled. Signed-off-by:
Mark Rutland <mark.rutland@arm.com> Acked-by:
-
Peter Zijlstra authored
The whole disable-RCU, enable-IRQS dance is very intricate since changing IRQ state is traced, which depends on RCU. Add two helpers for the cpuidle case that mirror the entry code: ct_cpuidle_enter() ct_cpuidle_exit() And fix all the cases where the enter/exit dance was buggy. Signed-off-by:
-
Peter Zijlstra authored
Doing RCU-idle outside the driver, only to then temporarily enable it again before going idle is suboptimal. Notably: this converts all dt_init_idle_driver() and __CPU_PM_CPU_IDLE_ENTER() users for they are inextrably intertwined. Signed-off-by:
-
Peter Zijlstra authored
Doing RCU-idle outside the driver, only to then temporarily enable it again, some *four* times, before going idle is suboptimal. Notably three times explicitly using RCU_NONIDLE() and once implicitly through cpu_pm_*(). Signed-off-by:
-
Peter Zijlstra authored
Doing RCU-idle outside the driver, only to then temporarily enable it again before going idle is suboptimal. Notably the cpu_pm_*() calls implicitly re-enable RCU for a bit. Signed-off-by:
-
Peter Zijlstra authored
Doing RCU-idle outside the driver, only to then teporarily enable it again before going idle is suboptimal. Notably the cpu_pm_*() calls implicitly re-enable RCU for a bit. Signed-off-by:
-
Peter Zijlstra authored
Doing RCU-idle outside the driver, only to then temporarily enable it again, at least twice, before going idle is suboptimal. Notably both cpu_pm_enter() and cpu_cluster_pm_enter() implicity re-enable RCU. Signed-off-by:
-
Peter Zijlstra authored
Doing RCU-idle outside the driver, only to then temporarily enable it again, at least twice, before going idle is suboptimal. Notably once implicitly through the cpu_pm_*() calls and once explicitly doing ct_irq_*_irqon(). Signed-off-by:
Kajetan Puchalski <kajetan.puchalski@arm.com> Tested-by:
-
Peter Zijlstra authored
Doing RCU-idle outside the driver, only to then temporarily enable it again, at least twice, before going idle is suboptimal. Notably once implicitly through the cpu_pm_*() calls and once explicitly doing RCU_NONIDLE(). Signed-off-by:
-
Peter Zijlstra authored
Doing RCU-idle outside the driver, only to then temporarily enable it again, at least twice, before going idle is suboptimal. That is, once implicitly through the cpu_pm_*() calls and once explicitly doing ct_irq_*_irqon(). Signed-off-by:
Anup Patel <anup@brainfault.org> Reviewed-by:
-
Peter Zijlstra authored
Make cpuidle_enter_state() consistent with the s2idle variant and verify ->enter() always returns with interrupts disabled. Signed-off-by:
-
Peter Zijlstra authored
Make cpuidle_state::enter() methods IRQ state invariant on exit. Additionally make sure to use raw_local_irq_*() methods since this cpuidle callback will be called with RCU already disabled. Signed-off-by:
-
Peter Zijlstra authored
Typical boot time setup; no need to suffer an indirect call for that. Signed-off-by:
-
Peter Zijlstra authored
The perf_lopwr_cb() function is called from the idle routines; there is no RCU there, we must not enter tracing. Use __always_inline, noidle annotations and existing no-trace methods. Signed-off-by:
-
- 12 Jan, 2023 2 commits
-
-
Mathieu Desnoyers authored
Two new auxiliary vector entries are introduced for rseq without matching increment of the AT_VECTOR_SIZE_BASE, which causes failures with CONFIG_HARDENED_USERCOPY=y. Fixes: 317c8194 ("rseq: Introduce feature size and alignment ELF auxiliary vector entries") Reported-by:
Nathan Chancellor <nathan@kernel.org> Signed-off-by:
Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Signed-off-by:
-
Mathieu Desnoyers authored
The mm_numa_cid related rseq patches from the series were not picked up into the tip tree, so enabling the mm_numa_cid test needs to be reverted. This reverts commit b344b8f2. Reported-by:
kernel test robot <oliver.sang@intel.com> Signed-off-by:
-
- 11 Jan, 2023 1 commit
-
-
Ingo Molnar authored
The following commit: c8997020 ("cputime: remove cputime_to_nsecs fallback") Removed an <asm/cputime.h> inclusion from <linux/sched/cputime.h>, but this broke the IA64 build: arch/ia64/kernel/time.c:110:6: warning: no previous prototype for 'arch_vtime_task_switch' [-Wmissing-prototypes] Add in the missing <asm/cputime.h> header to fix it. Fixes: c8997020 ("cputime: remove cputime_to_nsecs fallback") Reported-by:
kernel test robot <lkp@intel.com> Signed-off-by:
-
- 07 Jan, 2023 5 commits
-
-
Michal Clapinski authored
Keep track of previously issued registrations and compare the result with MEMBARRIER_CMD_GET_REGISTRATIONS return value. Signed-off-by:
Michal Clapinski <mclapinski@google.com> Signed-off-by:
Paul E. McKenney <paulmck@kernel.org> Link: https://lore.kernel.org/r/20221207164338.1535591-3-mclapinski@google.com
-
Michal Clapinski authored
Provide a method to query previously issued registrations. Signed-off-by:
-
Lukasz Luba authored
The max CPU capacity is the same for all CPUs sharing frequency domain. There is a way to avoid heavy operations in a loop for each CPU by leveraging this knowledge. Thus, simplify the looping code in the sugov_next_freq_shared() and drop heavy multiplications. Instead, use simple max() to get the highest utilization from these CPUs. This is useful for platforms with many (4 or 6) little CPUs. We avoid heavy 2*PD_CPU_NUM multiplications in that loop, which is called billions of times, since it's not limited by the schedutil time delta filter in sugov_should_update_freq(). When there was no need to change frequency the code bailed out, not updating the sg_policy::last_freq_update_time. Then every visit after delta_ns time longer than the sg_policy::freq_update_delay_ns goes through and triggers the next frequency calculation code. Although, if the next frequency, as outcome of that, would be the same as current frequency, we won't update the sg_policy::last_freq_update_time and the story will be repeated (in a very short period, sometimes a few microseconds). The max CPU capacity must be fetched every time we are called, due to difficulties during the policy setup, where we are not able to get the normalized CPU capacity at the right time. The fetched CPU capacity value is than used in sugov_iowait_apply() to calculate the right boost. This required a few changes in the local functions and arguments. The capacity value should hopefully be fetched once when needed and then passed over CPU registers to those functions. Signed-off-by:
Lukasz Luba <lukasz.luba@arm.com> Signed-off-by:
-
Chengming Zhou authored
ttwu_do_activate() is used for a complete wakeup, in which we will activate_task() and use ttwu_do_wakeup() to mark the task runnable and perform wakeup-preemption, also call class->task_woken() callback and update the rq->idle_stamp. Since ttwu_runnable() is not a complete wakeup, don't need all those done in ttwu_do_wakeup(), so we can move those to ttwu_do_activate() to simplify ttwu_do_wakeup(), making it only mark the task runnable to be reused in ttwu_runnable() and try_to_wake_up(). This patch should not have any functional changes. Suggested-by:
Chengming Zhou <zhouchengming@bytedance.com> Signed-off-by:
-
Chengming Zhou authored
ttwu_runnable() is used as a fast wakeup path when the wakee task is running on CPU or runnable on RQ, in both cases we can just set its state to TASK_RUNNING to prevent a sleep. If the wakee task is on_cpu running, we don't need to update_rq_clock() or check_preempt_curr(). But if the wakee task is on_rq && !on_cpu (e.g. an IRQ hit before the task got to schedule() and the task been preempted), we should check_preempt_curr() to see if it can preempt the current running. This also removes the class->task_woken() callback from ttwu_runnable(), which wasn't required per the RT/DL implementations: any required push operation would have been queued during class->set_next_task() when p got preempted. ttwu_runnable() also loses the update to rq->idle_stamp, as by definition the rq cannot be idle in this scenario. Suggested-by:
Valentin Schneider <vschneid@redhat.com> Suggested-by:
Steven Rostedt (Google) <rostedt@goodmis.org> Link: https://lore.kernel.org/r/20221223103257.4962-1-zhouchengming@bytedance.com
-
- 05 Jan, 2023 2 commits
-
-
Qais Yousef authored
Add a document explaining the util clamp feature: what it is and how to use it. The new document hopefully covers everything one needs to know about uclamp. Signed-off-by:
Qais Yousef <qais.yousef@arm.com> Signed-off-by:
Qais Yousef (Google) <qyousef@layalina.io> Signed-off-by:
-
Bing Huang authored
sched_init_domains() is only used in initialization Signed-off-by:
Bing Huang <huangbing@kylinos.cn> Signed-off-by:
-
- 02 Jan, 2023 1 commit
-
-
Mathieu Desnoyers authored
sched_mm_cid_after_execve() does not expect NULL t->mm, but it may happen if a usermodehelper kthread fails when attempting to execute a binary. sched_mm_cid_fork() can be issued from a usermodehelper kthread, which has t->flags PF_KTHREAD set. Fixes: af7f588d ("sched: Introduce per-memory-map concurrency ID") Reported-by:
kernel test robot <yujie.liu@intel.com> Reported-by:
Borislav Petkov <bp@alien8.de> Signed-off-by:
-
- 27 Dec, 2022 8 commits
-
-
Nicholas Piggin authored
The archs that use cputime_to_nsecs() internally provide their own definition and don't need the fallback. cputime_to_usecs() unused except in this fallback, and is not defined anywhere. This removes the final remnant of the cputime_t code from the kernel. Signed-off-by:
Nicholas Piggin <npiggin@gmail.com> Signed-off-by:
Alexander Gordeev <agordeev@linux.ibm.com> Link: https://lore.kernel.org/r/20221220070705.2958959-1-npiggin@gmail.com
-
Hao Jia authored
When select_idle_capacity() starts scanning for an idle CPU, it starts with target CPU that has already been checked in select_idle_sibling(). So we start checking from the next CPU and try the target CPU at the end. Similarly for task_numa_assign(), we have just checked numa_migrate_on of dst_cpu, so start from the next CPU. This also works for steal_cookie_task(), the first scan must fail and start directly from the next one. Signed-off-by:
Hao Jia <jiahao.os@bytedance.com> Signed-off-by:
Vincent Guittot <vincent.guittot@linaro.org> Acked-by:
Mel Gorman <mgorman@techsingularity.net> Link: https://lore.kernel.org/r/20221216062406.7812-3-jiahao.os@bytedance.com
-
Hao Jia authored
In update_numa_stats() we try to find an idle cpu on the NUMA node, preferably an idle core. we can stop looking for the next idle core or idle cpu after finding an idle core. But we can't stop the whole loop of scanning the CPU, because we need to calculate approximate NUMA stats at a point in time. For example, the src and dst nr_running is needed by task_numa_find_cpu(). Signed-off-by:
-
Matthew Wilcox (Oracle) authored
With a modified container_of() that preserves constness, the compiler finds some pointers which should have been marked as const. task_of() also needs to become const-preserving for the !FAIR_GROUP_SCHED case so that cfs_rq_of() can take a const argument. No change to generated code. Signed-off-by:
Matthew Wilcox (Oracle) <willy@infradead.org> Signed-off-by:
-
Mathieu Desnoyers authored
Add mm_numa_cid tests to the run_param_test.sh test script. Signed-off-by:
-
Mathieu Desnoyers authored
Add the mm_cid field to the rseq_update event, allowing tracers to follow which mm_cid is observed by user-space, and whether negative mm_cid values are visible in case of internal scheduler implementation issues. Signed-off-by:
-
Mathieu Desnoyers authored
Report and abort when a negative concurrency ID value is observed by the spinlock test. Signed-off-by:
-
Mathieu Desnoyers authored
Adapt to the rseq.h API changes introduced by commits "selftests/rseq: <arch>: Template memory ordering and percpu access mode". Build a new param_test_mm_cid, param_test_mm_cid_benchmark, and param_test_mm_cid_compare_twice executables to test the new "mm_cid" rseq field. Signed-off-by:
-
