| /linux/arch/csky/ |
| H A D | Kconfig | 14 select ARCH_INLINE_READ_LOCK if !PREEMPTION
15 select ARCH_INLINE_READ_LOCK_BH if !PREEMPTION
16 select ARCH_INLINE_READ_LOCK_IRQ if !PREEMPTION
17 select ARCH_INLINE_READ_LOCK_IRQSAVE if !PREEMPTION
18 select ARCH_INLINE_READ_UNLOCK if !PREEMPTION
19 select ARCH_INLINE_READ_UNLOCK_BH if !PREEMPTION
20 select ARCH_INLINE_READ_UNLOCK_IRQ if !PREEMPTION
21 select ARCH_INLINE_READ_UNLOCK_IRQRESTORE if !PREEMPTION
22 select ARCH_INLINE_WRITE_LOCK if !PREEMPTION
23 select ARCH_INLINE_WRITE_LOCK_BH if !PREEMPTION
[all …]
|
| /linux/Documentation/locking/ |
| H A D | preempt-locking.rst | 35 protect these situations by disabling preemption around them.
37 You can also use put_cpu() and get_cpu(), which will disable preemption.
44 Under preemption, the state of the CPU must be protected. This is arch-
47 section that must occur while preemption is disabled. Think what would happen
50 upon preemption, the FPU registers will be sold to the lowest bidder. Thus,
51 preemption must be disabled around such regions.
54 kernel_fpu_begin and kernel_fpu_end will disable and enable preemption.
72 Data protection under preemption is achieved by disabling preemption for the
84 n-times in a code path, and preemption will not be reenabled until the n-th
86 preemption is not enabled.
[all …]
|
| /linux/kernel/ |
| H A D | Kconfig.preempt | 11 select PREEMPTION
18 prompt "Preemption Model"
22 bool "No Forced Preemption (Server)"
26 This is the traditional Linux preemption model, geared towards
37 bool "Voluntary Kernel Preemption (Desktop)"
43 "explicit preemption points" to the kernel code. These new
44 preemption points have been selected to reduce the maximum
66 otherwise not be about to reach a natural preemption point.
76 bool "Scheduler controlled preemption model"
81 This option provides a scheduler driven preemption model that
[all …]
|
| H A D | Kconfig.locks | 104 # - DEBUG_SPINLOCK=n and PREEMPTION=n
142 depends on !PREEMPTION || ARCH_INLINE_SPIN_UNLOCK_IRQ
171 depends on !PREEMPTION || ARCH_INLINE_READ_UNLOCK
179 depends on !PREEMPTION || ARCH_INLINE_READ_UNLOCK_IRQ
208 depends on !PREEMPTION || ARCH_INLINE_WRITE_UNLOCK
216 depends on !PREEMPTION || ARCH_INLINE_WRITE_UNLOCK_IRQ
|
| /linux/drivers/gpu/drm/msm/adreno/ |
| H A D | a5xx_gpu.h | 58 * In order to do lockless preemption we use a simple state machine to progress
61 * PREEMPT_NONE - no preemption in progress. Next state START.
62 * PREEMPT_START - The trigger is evaulating if preemption is possible. Next
66 * PREEMPT_TRIGGERED: A preemption has been executed on the hardware. Next
68 * PREEMPT_FAULTED: A preemption timed out (never completed). This will trigger
70 * PREEMPT_PENDING: Preemption complete interrupt fired - the callback is
85 * CPU to store the state for preemption. The record itself is much larger
88 * There is a preemption record assigned per ringbuffer. When the CPU triggers a
89 * preemption, it fills out the record with the useful information (wptr, ring
91 * the preemption. When a ring is switched out, the CP will save the ringbuffer
[all …]
|
| H A D | a5xx_preempt.c | 9 * Try to transition the preemption state from old to new. Return
22 * Force the preemption state to the specified state. This is used in cases
30 * preemption or in the interrupt handler so barriers are needed in set_preempt_state()
90 DRM_DEV_ERROR(dev->dev, "%s: preemption timed out\n", gpu->name); in a5xx_preempt_timer()
94 /* Try to trigger a preemption switch */
106 * Serialize preemption start to ensure that we always make in a5xx_preempt_trigger()
113 * Try to start preemption by moving from NONE to START. If in a5xx_preempt_trigger()
114 * unsuccessful, a preemption is already in flight in a5xx_preempt_trigger()
128 * Its possible that while a preemption request is in progress in a5xx_preempt_trigger()
152 /* Set the address of the incoming preemption record */ in a5xx_preempt_trigger()
[all …]
|
| /linux/arch/loongarch/ |
| H A D | Kconfig | 34 select ARCH_INLINE_READ_LOCK if !PREEMPTION
35 select ARCH_INLINE_READ_LOCK_BH if !PREEMPTION
36 select ARCH_INLINE_READ_LOCK_IRQ if !PREEMPTION
37 select ARCH_INLINE_READ_LOCK_IRQSAVE if !PREEMPTION
38 select ARCH_INLINE_READ_UNLOCK if !PREEMPTION
39 select ARCH_INLINE_READ_UNLOCK_BH if !PREEMPTION
40 select ARCH_INLINE_READ_UNLOCK_IRQ if !PREEMPTION
41 select ARCH_INLINE_READ_UNLOCK_IRQRESTORE if !PREEMPTION
42 select ARCH_INLINE_WRITE_LOCK if !PREEMPTION
43 select ARCH_INLINE_WRITE_LOCK_BH if !PREEMPTION
[all …]
|
| /linux/Documentation/trace/rv/ |
| H A D | monitor_sched.rst | 88 The schedule called with preemption disabled (scpd) monitor ensures schedule is
89 called with preemption disabled::
110 does not enable preemption::
180 The need resched preempts (nrp) monitor ensures preemption requires
181 ``need_resched``. Only kernel preemption is considered, since preemption
184 A kernel preemption is whenever ``__schedule`` is called with the preemption
186 type of preemption occurs after the need for ``rescheduling`` has been set.
188 userspace preemption.
190 case, a task goes through the scheduler from a preemption context but it is
195 In theory, a preemption can only occur after the ``need_resched`` flag is set. In
[all …]
|
| H A D | monitor_wip.rst | 13 preemption disabled::
30 The wakeup event always takes place with preemption disabled because
|
| /linux/include/linux/ |
| H A D | preempt.h | 7 * preempt_count (used for kernel preemption, interrupt count, etc.)
15 * We put the hardirq and softirq counter into the preemption
18 * - bits 0-7 are the preemption count (max preemption depth: 256)
60 * Disable preemption until the scheduler is running -- use an unconditional
160 /* Locks on RT do not disable preemption */
169 * Which need to disable both preemption (CONFIG_PREEMPT_COUNT) and
281 * Even if we don't have any preemption, we need preempt disable/enable
301 * Modules have no business playing preemption tricks.
345 * preempt_notifier - key for installing preemption notifiers
429 * preempt_disable_nested - Disable preemption inside a normally preempt disabled section
[all …]
|
| /linux/Documentation/core-api/ |
| H A D | local_ops.rst | 42 making sure that we modify it from within a preemption safe context. It is
70 * Preemption (or interrupts) must be disabled when using local ops in
76 preemption already disabled. I suggest, however, to explicitly
77 disable preemption anyway to make sure it will still work correctly on
104 local atomic operations: it makes sure that preemption is disabled around write
110 If you are already in a preemption-safe context, you can use
161 * preemptible context (it disables preemption) :
|
| /linux/Documentation/RCU/ |
| H A D | NMI-RCU.rst | 45 The do_nmi() function processes each NMI. It first disables preemption
50 preemption is restored.
95 CPUs complete any preemption-disabled segments of code that they were
97 Since NMI handlers disable preemption, synchronize_rcu() is guaranteed
|
| /linux/drivers/gpu/drm/xe/ |
| H A D | xe_exec_queue_types.h | 117 /** @sched_props.preempt_timeout_us: preemption timeout in micro-seconds */
127 /** @lr.pfence: preemption fence */
129 /** @lr.context: preemption fence context */
131 /** @lr.seqno: preemption fence seqno */
197 /** @set_preempt_timeout: Set preemption timeout for exec queue */
|
| H A D | xe_pcode.c | 169 * @timeout_base_ms: timeout for polling with preemption enabled
174 * applying @reply_mask. Polling is first attempted with preemption enabled
176 * preemption disabled.
200 * the poll with preemption disabled to maximize the number of in xe_pcode_request()
207 "PCODE timeout, retrying with preemption disabled\n"); in xe_pcode_request()
|
| H A D | xe_hw_engine_types.h | 94 /** @sched_props.preempt_timeout_us: preemption timeout in micro-seconds */
96 /** @sched_props.preempt_timeout_min: min preemption timeout in micro-seconds */
98 /** @sched_props.preempt_timeout_max: max preemption timeout in micro-seconds */
|
| /linux/fs/ |
| H A D | stack.c | 18 * preemption (see include/linux/fs.h): we need nothing extra for in fsstack_copy_inode_size()
26 * i_blocks in sync despite SMP or PREEMPTION - though stat's in fsstack_copy_inode_size()
48 * i_blocks in sync despite SMP or PREEMPTION: use i_lock for that case in fsstack_copy_inode_size()
|
| /linux/arch/arm64/ |
| H A D | Kconfig | 63 select ARCH_INLINE_READ_LOCK if !PREEMPTION
64 select ARCH_INLINE_READ_LOCK_BH if !PREEMPTION
65 select ARCH_INLINE_READ_LOCK_IRQ if !PREEMPTION
66 select ARCH_INLINE_READ_LOCK_IRQSAVE if !PREEMPTION
67 select ARCH_INLINE_READ_UNLOCK if !PREEMPTION
68 select ARCH_INLINE_READ_UNLOCK_BH if !PREEMPTION
69 select ARCH_INLINE_READ_UNLOCK_IRQ if !PREEMPTION
70 select ARCH_INLINE_READ_UNLOCK_IRQRESTORE if !PREEMPTION
71 select ARCH_INLINE_WRITE_LOCK if !PREEMPTION
72 select ARCH_INLINE_WRITE_LOCK_BH if !PREEMPTION
[all …]
|
| /linux/kernel/trace/rv/monitors/opid/ |
| H A D | Kconfig | 13 interrupts and preemption disabled or during IRQs, where preemption
|
| /linux/arch/arm64/include/asm/ |
| H A D | percpu.h | 26 * Non-VHE hyp code runs with preemption disabled. No need to hazard in __hyp_my_cpu_offset()
140 * re-enabling preemption for preemptible kernels, but doing that in a way
144 * Not to mention it'll break the actual preemption model for missing a
145 * preemption point when TIF_NEED_RESCHED gets set while preemption is
|
| /linux/Documentation/mm/ |
| H A D | highmem.rst | 66 CPU while the mapping is active. Although preemption is never disabled by
73 As said, pagefaults and preemption are never disabled. There is no need to
74 disable preemption because, when context switches to a different task, the
110 effects of atomic mappings, i.e. disabling page faults or preemption, or both.
141 restrictions on preemption or migration. It comes with an overhead as mapping
|
| /linux/kernel/trace/ |
| H A D | trace_syscalls.c | 309 * Syscall probe called with preemption enabled, but the ring in ftrace_syscall_enter()
310 * buffer and per-cpu data require preemption to be disabled. in ftrace_syscall_enter()
354 * Syscall probe called with preemption enabled, but the ring in ftrace_syscall_exit()
355 * buffer and per-cpu data require preemption to be disabled. in ftrace_syscall_exit()
606 * Syscall probe called with preemption enabled, but the ring in perf_syscall_enter()
607 * buffer and per-cpu data require preemption to be disabled. in perf_syscall_enter()
715 * Syscall probe called with preemption enabled, but the ring in perf_syscall_exit()
716 * buffer and per-cpu data require preemption to be disabled. in perf_syscall_exit()
|
| /linux/Documentation/virt/kvm/devices/ |
| H A D | arm-vgic.rst | 99 maximum possible 128 preemption levels. The semantics of the register
100 indicate if any interrupts in a given preemption level are in the active
103 Thus, preemption level X has one or more active interrupts if and only if:
107 Bits for undefined preemption levels are RAZ/WI.
|
| /linux/arch/s390/include/asm/ |
| H A D | kmsan.h | 43 * KMSAN. Therefore, disable preemption here, and re-enable preemption in kmsan_virt_addr_valid()
|
| /linux/arch/riscv/include/asm/ |
| H A D | simd.h | 30 * RISCV_KERNEL_MODE_V is only set while preemption is disabled, in may_use_simd()
31 * and is clear whenever preemption is enabled. in may_use_simd()
|
| /linux/drivers/gpu/drm/amd/include/ivsrcid/vpe/ |
| H A D | irqsrcs_vpe_6_1.h | 30 #define VPE_6_1_SRCID__VPE_PREEMPT 4 // 0x4 Preemption
38 #define VPE_6_1_SRCID__VPE_IB_PREEMPT 12 // 0xC IB preemption
|