// SPDX-License-Identifier: GPL-2.0-or-later /* * Derived from "arch/i386/kernel/process.c" * Copyright (C) 1995 Linus Torvalds * * Updated and modified by Cort Dougan (cort@cs.nmt.edu) and * Paul Mackerras (paulus@cs.anu.edu.au) * * PowerPC version * Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org)
*/
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM /* * Are we running in "Suspend disabled" mode? If so we have to block any * sigreturn that would get us into suspended state, and we also warn in some * other paths that we should never reach with suspend disabled.
*/ bool tm_suspend_disabled __ro_after_init = false;
staticvoid check_if_tm_restore_required(struct task_struct *tsk)
{ /* * If we are saving the current thread's registers, and the * thread is in a transactional state, set the TIF_RESTORE_TM * bit so that we know to restore the registers before * returning to userspace.
*/ if (tsk == current && tsk->thread.regs &&
MSR_TM_ACTIVE(tsk->thread.regs->msr) &&
!test_thread_flag(TIF_RESTORE_TM)) {
regs_set_return_msr(&tsk->thread.ckpt_regs,
tsk->thread.regs->msr);
set_thread_flag(TIF_RESTORE_TM);
}
}
/* * Make sure the floating-point register state in the * the thread_struct is up to date for task tsk.
*/ void flush_fp_to_thread(struct task_struct *tsk)
{ if (tsk->thread.regs) { /* * We need to disable preemption here because if we didn't, * another process could get scheduled after the regs->msr * test but before we have finished saving the FP registers * to the thread_struct. That process could take over the * FPU, and then when we get scheduled again we would store * bogus values for the remaining FP registers.
*/
preempt_disable(); if (tsk->thread.regs->msr & MSR_FP) { /* * This should only ever be called for current or * for a stopped child process. Since we save away * the FP register state on context switch, * there is something wrong if a stopped child appears * to still have its FP state in the CPU registers.
*/
BUG_ON(tsk != current);
giveup_fpu(tsk);
}
preempt_enable();
}
}
EXPORT_SYMBOL_GPL(flush_fp_to_thread);
if (current->thread.regs && (current->thread.regs->msr & MSR_FP)) {
check_if_tm_restore_required(current); /* * If a thread has already been reclaimed then the * checkpointed registers are on the CPU but have definitely * been saved by the reclaim code. Don't need to and *cannot* * giveup as this would save to the 'live' structure not the * checkpointed structure.
*/ if (!MSR_TM_ACTIVE(cpumsr) &&
MSR_TM_ACTIVE(current->thread.regs->msr)) return;
__giveup_fpu(current);
}
}
EXPORT_SYMBOL(enable_kernel_fp); #else staticinlinevoid __giveup_fpu(struct task_struct *tsk) { } #endif/* CONFIG_PPC_FPU */
if (current->thread.regs && (current->thread.regs->msr & MSR_VEC)) {
check_if_tm_restore_required(current); /* * If a thread has already been reclaimed then the * checkpointed registers are on the CPU but have definitely * been saved by the reclaim code. Don't need to and *cannot* * giveup as this would save to the 'live' structure not the * checkpointed structure.
*/ if (!MSR_TM_ACTIVE(cpumsr) &&
MSR_TM_ACTIVE(current->thread.regs->msr)) return;
__giveup_altivec(current);
}
}
EXPORT_SYMBOL(enable_kernel_altivec);
/* * Make sure the VMX/Altivec register state in the * the thread_struct is up to date for task tsk.
*/ void flush_altivec_to_thread(struct task_struct *tsk)
{ if (tsk->thread.regs) {
preempt_disable(); if (tsk->thread.regs->msr & MSR_VEC) {
BUG_ON(tsk != current);
giveup_altivec(tsk);
}
preempt_enable();
}
}
EXPORT_SYMBOL_GPL(flush_altivec_to_thread); #endif/* CONFIG_ALTIVEC */
if (current->thread.regs &&
(current->thread.regs->msr & (MSR_VSX|MSR_VEC|MSR_FP))) {
check_if_tm_restore_required(current); /* * If a thread has already been reclaimed then the * checkpointed registers are on the CPU but have definitely * been saved by the reclaim code. Don't need to and *cannot* * giveup as this would save to the 'live' structure not the * checkpointed structure.
*/ if (!MSR_TM_ACTIVE(cpumsr) &&
MSR_TM_ACTIVE(current->thread.regs->msr)) return;
__giveup_vsx(current);
}
}
EXPORT_SYMBOL(enable_kernel_vsx);
/* * The exception exit path calls restore_math() with interrupts hard disabled * but the soft irq state not "reconciled". ftrace code that calls * local_irq_save/restore causes warnings. * * Rather than complicate the exit path, just don't trace restore_math. This * could be done by having ftrace entry code check for this un-reconciled * condition where MSR[EE]=0 and PACA_IRQ_HARD_DIS is not set, and * temporarily fix it up for the duration of the ftrace call.
*/ void notrace restore_math(struct pt_regs *regs)
{ unsignedlong msr; unsignedlong new_msr = 0;
msr = regs->msr;
/* * new_msr tracks the facilities that are to be restored. Only reload * if the bit is not set in the user MSR (if it is set, the registers * are live for the user thread).
*/ if ((!(msr & MSR_FP)) && should_restore_fp())
new_msr |= MSR_FP;
if ((!(msr & MSR_VEC)) && should_restore_altivec())
new_msr |= MSR_VEC;
/* Deliver the signal to userspace */
force_sig_ptrace_errno_trap(breakpt, /* breakpoint or watchpoint id */
(void __user *)address);
} #else/* !CONFIG_PPC_ADV_DEBUG_REGS */
staticvoid do_break_handler(struct pt_regs *regs)
{ struct arch_hw_breakpoint null_brk = {0}; struct arch_hw_breakpoint *info;
ppc_inst_t instr = ppc_inst(0); int type = 0; int size = 0; unsignedlong ea; int i;
/* * If underneath hw supports only one watchpoint, we know it * caused exception. 8xx also falls into this category.
*/ if (nr_wp_slots() == 1) {
__set_breakpoint(0, &null_brk);
current->thread.hw_brk[0] = null_brk;
current->thread.hw_brk[0].flags |= HW_BRK_FLAG_DISABLED; return;
}
/* Otherwise find out which DAWR caused exception and disable it. */
wp_get_instr_detail(regs, &instr, &type, &size, &ea);
for (i = 0; i < nr_wp_slots(); i++) {
info = ¤t->thread.hw_brk[i]; if (!info->address) continue;
/* * We reach here only when watchpoint exception is generated by ptrace * event (or hw is buggy!). Now if CONFIG_HAVE_HW_BREAKPOINT is set, * watchpoint is already handled by hw_breakpoint_handler() so we don't * have to do anything. But when CONFIG_HAVE_HW_BREAKPOINT is not set, * we need to manually handle the watchpoint here.
*/ if (!IS_ENABLED(CONFIG_HAVE_HW_BREAKPOINT))
do_break_handler(regs);
/* Deliver the signal to userspace */
force_sig_fault(SIGTRAP, TRAP_HWBKPT, (void __user *)regs->dar);
} #endif/* CONFIG_PPC_ADV_DEBUG_REGS */
#ifdef CONFIG_PPC_ADV_DEBUG_REGS /* * Set the debug registers back to their default "safe" values.
*/ staticvoid set_debug_reg_defaults(struct thread_struct *thread)
{
thread->debug.iac1 = thread->debug.iac2 = 0; #if CONFIG_PPC_ADV_DEBUG_IACS > 2
thread->debug.iac3 = thread->debug.iac4 = 0; #endif
thread->debug.dac1 = thread->debug.dac2 = 0; #if CONFIG_PPC_ADV_DEBUG_DVCS > 0
thread->debug.dvc1 = thread->debug.dvc2 = 0; #endif
thread->debug.dbcr0 = 0; #ifdef CONFIG_BOOKE /* * Force User/Supervisor bits to b11 (user-only MSR[PR]=1)
*/
thread->debug.dbcr1 = DBCR1_IAC1US | DBCR1_IAC2US |
DBCR1_IAC3US | DBCR1_IAC4US; /* * Force Data Address Compare User/Supervisor bits to be User-only * (0b11 MSR[PR]=1) and set all other bits in DBCR2 register to be 0.
*/
thread->debug.dbcr2 = DBCR2_DAC1US | DBCR2_DAC2US; #else
thread->debug.dbcr1 = 0; #endif
}
staticvoid prime_debug_regs(struct debug_reg *debug)
{ /* * We could have inherited MSR_DE from userspace, since * it doesn't get cleared on exception entry. Make sure * MSR_DE is clear before we enable any debug events.
*/
mtmsr(mfmsr() & ~MSR_DE);
mtspr(SPRN_IAC1, debug->iac1);
mtspr(SPRN_IAC2, debug->iac2); #if CONFIG_PPC_ADV_DEBUG_IACS > 2
mtspr(SPRN_IAC3, debug->iac3);
mtspr(SPRN_IAC4, debug->iac4); #endif
mtspr(SPRN_DAC1, debug->dac1);
mtspr(SPRN_DAC2, debug->dac2); #if CONFIG_PPC_ADV_DEBUG_DVCS > 0
mtspr(SPRN_DVC1, debug->dvc1);
mtspr(SPRN_DVC2, debug->dvc2); #endif
mtspr(SPRN_DBCR0, debug->dbcr0);
mtspr(SPRN_DBCR1, debug->dbcr1); #ifdef CONFIG_BOOKE
mtspr(SPRN_DBCR2, debug->dbcr2); #endif
} /* * Unless neither the old or new thread are making use of the * debug registers, set the debug registers from the values * stored in the new thread.
*/ void switch_booke_debug_regs(struct debug_reg *new_debug)
{ if ((current->thread.debug.dbcr0 & DBCR0_IDM)
|| (new_debug->dbcr0 & DBCR0_IDM))
prime_debug_regs(new_debug);
}
EXPORT_SYMBOL_GPL(switch_booke_debug_regs); #else/* !CONFIG_PPC_ADV_DEBUG_REGS */ #ifndef CONFIG_HAVE_HW_BREAKPOINT staticvoid set_breakpoint(int i, struct arch_hw_breakpoint *brk)
{
preempt_disable();
__set_breakpoint(i, brk);
preempt_enable();
}
for (i = 0; i < nr_wp_slots(); i++) {
thread->hw_brk[i] = null_brk; if (ppc_breakpoint_available())
set_breakpoint(i, &thread->hw_brk[i]);
}
}
staticinlinebool hw_brk_match(struct arch_hw_breakpoint *a, struct arch_hw_breakpoint *b)
{ if (a->address != b->address) returnfalse; if (a->type != b->type) returnfalse; if (a->len != b->len) returnfalse; /* no need to check hw_len. it's calculated from address and len */ returntrue;
}
staticvoid switch_hw_breakpoint(struct task_struct *new)
{ int i;
for (i = 0; i < nr_wp_slots(); i++) { if (likely(hw_brk_match(this_cpu_ptr(¤t_brk[i]),
&new->thread.hw_brk[i]))) continue;
/* Check if we have DAWR or DABR hardware */ bool ppc_breakpoint_available(void)
{ if (dawr_enabled()) returntrue; /* POWER8 DAWR or POWER9 forced DAWR */ if (cpu_has_feature(CPU_FTR_ARCH_207S)) returnfalse; /* POWER9 with DAWR disabled */ /* DABR: Everything but POWER8 and POWER9 */ returntrue;
}
EXPORT_SYMBOL_GPL(ppc_breakpoint_available);
/* Disable the breakpoint in hardware without touching current_brk[] */ void suspend_breakpoints(void)
{ struct arch_hw_breakpoint brk = {0}; int i;
if (!ppc_breakpoint_available()) return;
for (i = 0; i < nr_wp_slots(); i++)
set_hw_breakpoint(i, &brk);
}
/* * Re-enable breakpoints suspended by suspend_breakpoints() in hardware * from current_brk[]
*/ void restore_breakpoints(void)
{ int i;
if (!ppc_breakpoint_available()) return;
for (i = 0; i < nr_wp_slots(); i++)
set_hw_breakpoint(i, this_cpu_ptr(¤t_brk[i]));
}
staticvoid tm_reclaim_thread(struct thread_struct *thr, uint8_t cause)
{ /* * Use the current MSR TM suspended bit to track if we have * checkpointed state outstanding. * On signal delivery, we'd normally reclaim the checkpointed * state to obtain stack pointer (see:get_tm_stackpointer()). * This will then directly return to userspace without going * through __switch_to(). However, if the stack frame is bad, * we need to exit this thread which calls __switch_to() which * will again attempt to reclaim the already saved tm state. * Hence we need to check that we've not already reclaimed * this state. * We do this using the current MSR, rather tracking it in * some specific thread_struct bit, as it has the additional * benefit of checking for a potential TM bad thing exception.
*/ if (!MSR_TM_SUSPENDED(mfmsr())) return;
/* * If we are in a transaction and FP is off then we can't have * used FP inside that transaction. Hence the checkpointed * state is the same as the live state. We need to copy the * live state to the checkpointed state so that when the * transaction is restored, the checkpointed state is correct * and the aborted transaction sees the correct state. We use * ckpt_regs.msr here as that's what tm_reclaim will use to * determine if it's going to write the checkpointed state or * not. So either this will write the checkpointed registers, * or reclaim will. Similarly for VMX.
*/ if ((thr->ckpt_regs.msr & MSR_FP) == 0)
memcpy(&thr->ckfp_state, &thr->fp_state, sizeof(struct thread_fp_state)); if ((thr->ckpt_regs.msr & MSR_VEC) == 0)
memcpy(&thr->ckvr_state, &thr->vr_state, sizeof(struct thread_vr_state));
}
staticinlinevoid tm_reclaim_task(struct task_struct *tsk)
{ /* We have to work out if we're switching from/to a task that's in the * middle of a transaction. * * In switching we need to maintain a 2nd register state as * oldtask->thread.ckpt_regs. We tm_reclaim(oldproc); this saves the * checkpointed (tbegin) state in ckpt_regs, ckfp_state and * ckvr_state * * We also context switch (save) TFHAR/TEXASR/TFIAR in here.
*/ struct thread_struct *thr = &tsk->thread;
if (!thr->regs) return;
if (!MSR_TM_ACTIVE(thr->regs->msr)) goto out_and_saveregs;
TM_DEBUG("---- tm_reclaim on pid %d complete\n",
tsk->pid);
out_and_saveregs: /* Always save the regs here, even if a transaction's not active. * This context-switches a thread's TM info SPRs. We do it here to * be consistent with the restore path (in recheckpoint) which * cannot happen later in _switch().
*/
tm_save_sprs(thr);
}
/* We really can't be interrupted here as the TEXASR registers can't * change and later in the trecheckpoint code, we have a userspace R1. * So let's hard disable over this region.
*/
local_irq_save(flags);
hard_irq_disable();
/* The TM SPRs are restored here, so that TEXASR.FS can be set * before the trecheckpoint and no explosion occurs.
*/
tm_restore_sprs(thread);
__tm_recheckpoint(thread);
local_irq_restore(flags);
}
staticinlinevoid tm_recheckpoint_new_task(struct task_struct *new)
{ if (!cpu_has_feature(CPU_FTR_TM)) return;
/* Recheckpoint the registers of the thread we're about to switch to. * * If the task was using FP, we non-lazily reload both the original and * the speculative FP register states. This is because the kernel * doesn't see if/when a TM rollback occurs, so if we take an FP * unavailable later, we are unable to determine which set of FP regs * need to be restored.
*/ if (!tm_enabled(new)) return;
if (!MSR_TM_ACTIVE(new->thread.regs->msr)){
tm_restore_sprs(&new->thread); return;
} /* Recheckpoint to restore original checkpointed register state. */
TM_DEBUG("*** tm_recheckpoint of pid %d (new->msr 0x%lx)\n",
new->pid, new->thread.regs->msr);
tm_recheckpoint(&new->thread);
/* * The checkpointed state has been restored but the live state has * not, ensure all the math functionality is turned off to trigger * restore_math() to reload.
*/
new->thread.regs->msr &= ~(MSR_FP | MSR_VEC | MSR_VSX);
staticinlinevoid __switch_to_tm(struct task_struct *prev, struct task_struct *new)
{ if (cpu_has_feature(CPU_FTR_TM)) { if (tm_enabled(prev) || tm_enabled(new))
tm_enable();
if (tm_enabled(prev)) {
prev->thread.load_tm++;
tm_reclaim_task(prev); if (!MSR_TM_ACTIVE(prev->thread.regs->msr) && prev->thread.load_tm == 0)
prev->thread.regs->msr &= ~MSR_TM;
}
tm_recheckpoint_new_task(new);
}
}
/* * This is called if we are on the way out to userspace and the * TIF_RESTORE_TM flag is set. It checks if we need to reload * FP and/or vector state and does so if necessary. * If userspace is inside a transaction (whether active or * suspended) and FP/VMX/VSX instructions have ever been enabled * inside that transaction, then we have to keep them enabled * and keep the FP/VMX/VSX state loaded while ever the transaction * continues. The reason is that if we didn't, and subsequently * got a FP/VMX/VSX unavailable interrupt inside a transaction, * we don't know whether it's the same transaction, and thus we * don't know which of the checkpointed state and the transactional * state to use.
*/ void restore_tm_state(struct pt_regs *regs)
{ unsignedlong msr_diff;
/* * This is the only moment we should clear TIF_RESTORE_TM as * it is here that ckpt_regs.msr and pt_regs.msr become the same * again, anything else could lead to an incorrect ckpt_msr being * saved and therefore incorrect signal contexts.
*/
clear_thread_flag(TIF_RESTORE_TM); if (!MSR_TM_ACTIVE(regs->msr)) return;
/* * Note that the TAR is not available for use in the kernel. * (To provide this, the TAR should be backed up/restored on * exception entry/exit instead, and be in pt_regs. FIXME, * this should be in pt_regs anyway (for debug).)
*/
t->tar = mfspr(SPRN_TAR);
}
if (cpu_has_feature(CPU_FTR_DEXCR_NPHIE))
t->hashkeyr = mfspr(SPRN_HASHKEYR);
if (cpu_has_feature(CPU_FTR_ARCH_31))
t->dexcr = mfspr(SPRN_DEXCR); #endif
}
#ifdef CONFIG_PPC_64S_HASH_MMU
batch = this_cpu_ptr(&ppc64_tlb_batch); if (batch->active) {
current_thread_info()->local_flags |= _TLF_LAZY_MMU; if (batch->index)
__flush_tlb_pending(batch);
batch->active = 0;
}
/* * On POWER9 the copy-paste buffer can only paste into * foreign real addresses, so unprivileged processes can not * see the data or use it in any way unless they have * foreign real mappings. If the new process has the foreign * real address mappings, we must issue a cp_abort to clear * any state and prevent snooping, corruption or a covert * channel. ISA v3.1 supports paste into local memory.
*/ if (new->mm && (cpu_has_feature(CPU_FTR_ARCH_31) ||
atomic_read(&new->mm->context.vas_windows))) asmvolatile(PPC_CP_ABORT); #endif/* CONFIG_PPC_BOOK3S_64 */
#ifdef CONFIG_PPC_ADV_DEBUG_REGS
switch_booke_debug_regs(&new->thread.debug); #else /* * For PPC_BOOK3S_64, we use the hw-breakpoint interfaces that would * schedule DABR
*/ #ifndef CONFIG_HAVE_HW_BREAKPOINT
switch_hw_breakpoint(new); #endif/* CONFIG_HAVE_HW_BREAKPOINT */ #endif
/* * We need to save SPRs before treclaim/trecheckpoint as these will * change a number of them.
*/
save_sprs(&prev->thread);
/* Save FPU, Altivec, VSX and SPE state */
giveup_all(prev);
__switch_to_tm(prev, new);
if (!radix_enabled()) { /* * We can't take a PMU exception inside _switch() since there * is a window where the kernel stack SLB and the kernel stack * are out of sync. Hard disable here.
*/
hard_irq_disable();
}
/* * Call restore_sprs() and set_return_regs_changed() before calling * _switch(). If we move it after _switch() then we miss out on calling * it for new tasks. The reason for this is we manually create a stack * frame for new tasks that directly returns through ret_from_fork() or * ret_from_kernel_thread(). See copy_thread() for details.
*/
restore_sprs(old_thread, new_thread);
if (!IS_ENABLED(CONFIG_PPC_BOOK3S_64))
kuap_assert_locked();
last = _switch(old_thread, new_thread);
/* * Nothing after _switch will be run for newly created tasks, * because they switch directly to ret_from_fork/ret_from_kernel_thread * etc. Code added here should have a comment explaining why that is * okay.
*/
#ifdef CONFIG_PPC_BOOK3S_64 #ifdef CONFIG_PPC_64S_HASH_MMU /* * This applies to a process that was context switched while inside * arch_enter_lazy_mmu_mode(), to re-activate the batch that was * deactivated above, before _switch(). This will never be the case * for new tasks.
*/ if (current_thread_info()->local_flags & _TLF_LAZY_MMU) {
current_thread_info()->local_flags &= ~_TLF_LAZY_MMU;
batch = this_cpu_ptr(&ppc64_tlb_batch);
batch->active = 1;
} #endif
/* * Math facilities are masked out of the child MSR in copy_thread. * A new task does not need to restore_math because it will * demand fault them.
*/ if (current->thread.regs)
restore_math(current->thread.regs); #endif/* CONFIG_PPC_BOOK3S_64 */
/* * If we were executing with the MMU off for instructions, adjust pc * rather than printing XXXXXXXX.
*/ if (!IS_ENABLED(CONFIG_BOOKE) && !(regs->msr & MSR_IR)) {
pc = (unsignedlong)phys_to_virt(pc);
nip = (unsignedlong)phys_to_virt(regs->nip);
}
for (i = 0; i < NR_INSN_TO_PRINT; i++) { int instr;
staticvoid print_tm_bits(unsignedlong val)
{ /* * This only prints something if at least one of the TM bit is set. * Inside the TM[], the output means: * E: Enabled (bit 32) * S: Suspended (bit 33) * T: Transactional (bit 34)
*/ if (val & (MSR_TM | MSR_TS_S | MSR_TS_T)) {
pr_cont(",TM[");
print_bits(val, msr_tm_bits, "");
pr_cont("]");
}
} #else staticvoid print_tm_bits(unsignedlong val) {} #endif
for (i = 0; i < 32; i++) { if ((i % REGS_PER_LINE) == 0)
pr_cont("\nGPR%02d: ", i);
pr_cont(REG " ", regs->gpr[i]);
}
pr_cont("\n"); /* * Lookup NIP late so we have the best change of getting the * above info out without failing
*/ if (IS_ENABLED(CONFIG_KALLSYMS)) {
printk("NIP ["REG"] %pS\n", regs->nip, (void *)regs->nip);
printk("LR ["REG"] %pS\n", regs->link, (void *)regs->link);
}
}
#ifdef CONFIG_PPC_BOOK3S_64 if (!radix_enabled())
hash__setup_new_exec(); #endif /* * If we exec out of a kernel thread then thread.regs will not be * set. Do it now.
*/ if (!current->thread.regs) { struct pt_regs *regs = task_stack_page(current) + THREAD_SIZE;
current->thread.regs = regs - 1;
}
#ifdef CONFIG_PPC64 /* * Assign a TIDR (thread ID) for task @t and set it in the thread * structure. For now, we only support setting TIDR for 'current' task. * * Since the TID value is a truncated form of it PID, it is possible * (but unlikely) for 2 threads to have the same TID. In the unlikely event * that 2 threads share the same TID and are waiting, one of the following * cases will happen: * * 1. The correct thread is running, the wrong thread is not * In this situation, the correct thread is woken and proceeds to pass its * condition check. * * 2. Neither threads are running * In this situation, neither thread will be woken. When scheduled, the waiting * threads will execute either a wait, which will return immediately, followed * by a condition check, which will pass for the correct thread and fail * for the wrong thread, or they will execute the condition check immediately. * * 3. The wrong thread is running, the correct thread is not * The wrong thread will be woken, but will fail its condition check and * re-execute wait. The correct thread, when scheduled, will execute either * its condition check (which will pass), or wait, which returns immediately * when called the first time after the thread is scheduled, followed by its * condition check (which will pass). * * 4. Both threads are running * Both threads will be woken. The wrong thread will fail its condition check * and execute another wait, while the correct thread will pass its condition * check. * * @t: the task to set the thread ID for
*/ int set_thread_tidr(struct task_struct *t)
{ if (!cpu_has_feature(CPU_FTR_P9_TIDR)) return -EINVAL;
/* * this gets called so that we can store coprocessor state into memory and * copy the current task into the new thread.
*/ int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
{
flush_all_to_thread(src); /* * Flush TM state out so we can copy it. __switch_to_tm() does this * flush but it removes the checkpointed state from the current CPU and * transitions the CPU out of TM mode. Hence we need to call * tm_recheckpoint_new_task() (on the same task) to restore the * checkpointed state back and the TM mode. * * Can't pass dst because it isn't ready. Doesn't matter, passing * dst is only important for __switch_to()
*/
__switch_to_tm(src, src);
f = start_kernel_thread;
p->thread.regs = NULL; /* no user register state */
clear_tsk_compat_task(p);
} else { /* user thread */ struct pt_regs *childregs;
if (unlikely(args->fn)) { /* * A user space thread, but it first runs a kernel * thread, and then returns as though it had called * execve rather than fork, so user regs will be * filled in (e.g., by kernel_execve()).
*/
((unsignedlong *)sp)[0] = 0;
memset(childregs, 0, sizeof(struct pt_regs)); #ifdef CONFIG_PPC64
childregs->softe = IRQS_ENABLED; #endif
f = ret_from_kernel_user_thread;
} else { struct pt_regs *regs = current_pt_regs();
u64 clone_flags = args->flags; unsignedlong usp = args->stack;
/* * The way this works is that at some point in the future * some task will call _switch to switch to the new task. * That will pop off the stack frame created below and start * the new task running at ret_from_fork. The new task will * do some house keeping and then return from the fork or clone * system call, using the stack frame created above.
*/
((unsignedlong *)sp)[STACK_FRAME_LR_SAVE] = (unsignedlong)f;
sp -= STACK_SWITCH_FRAME_SIZE;
((unsignedlong *)sp)[0] = sp + STACK_SWITCH_FRAME_SIZE;
kregs = (struct pt_regs *)(sp + STACK_SWITCH_FRAME_REGS);
kregs->nip = ppc_function_entry(f); if (unlikely(args->fn)) { /* * Put kthread fn, arg parameters in non-volatile GPRs in the * switch frame so they are loaded by _switch before it returns * to ret_from_kernel_thread.
*/
kregs->gpr[14] = ppc_function_entry((void *)args->fn);
kregs->gpr[15] = (unsignedlong)args->fn_arg;
}
p->thread.ksp = sp;
#ifdef CONFIG_HAVE_HW_BREAKPOINT for (i = 0; i < nr_wp_slots(); i++)
p->thread.ptrace_bps[i] = NULL; #endif
/* * Set up a thread for executing a new program
*/ void start_thread(struct pt_regs *regs, unsignedlong start, unsignedlong sp)
{ #ifdef CONFIG_PPC64 unsignedlong load_addr = regs->gpr[2]; /* saved by ELF_PLAT_INIT */
if (IS_ENABLED(CONFIG_PPC_BOOK3S_64) && !radix_enabled())
preload_new_slb_context(start, sp); #endif
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM /* * Clear any transactional state, we're exec()ing. The cause is * not important as there will never be a recheckpoint so it's not * user visible.
*/ if (MSR_TM_SUSPENDED(mfmsr()))
tm_reclaim_current(0); #endif
if (is_elf2_task()) { /* Look ma, no function descriptors! */
entry = start;
/* * Ulrich says: * The latest iteration of the ABI requires that when * calling a function (at its global entry point), * the caller must ensure r12 holds the entry point * address (so that the function can quickly * establish addressability).
*/
regs->gpr[12] = start; /* Make sure that's restored on entry to userspace. */
set_thread_flag(TIF_RESTOREALL);
} else { unsignedlong toc;
/* start is a relocated pointer to the function * descriptor for the elf _start routine. The first * entry in the function descriptor is the entry * address of _start and the second entry is the TOC * value we need to use.
*/
get_user(entry, (unsignedlong __user *)start);
get_user(toc, (unsignedlong __user *)start+1);
/* Check whether the e_entry function descriptor entries * need to be relocated before we can use them.
*/ if (load_addr != 0) {
entry += load_addr;
toc += load_addr;
}
regs->gpr[2] = toc;
}
regs_set_return_ip(regs, entry);
regs_set_return_msr(regs, MSR_USER64);
} else {
regs->gpr[2] = 0;
regs_set_return_ip(regs, start);
regs_set_return_msr(regs, MSR_USER32);
}
/* This is a bit hairy. If we are an SPE enabled processor * (have embedded fp) we store the IEEE exception enable flags in * fpexc_mode. fpexc_mode is also used for setting FP exception
* mode (asyn, precise, disabled) for 'Classic' FP. */ if (val & PR_FP_EXC_SW_ENABLE) { if (cpu_has_feature(CPU_FTR_SPE)) { /* * When the sticky exception bits are set * directly by userspace, it must call prctl * with PR_GET_FPEXC (with PR_FP_EXC_SW_ENABLE * in the existing prctl settings) or * PR_SET_FPEXC (with PR_FP_EXC_SW_ENABLE in * the bits being set). <fenv.h> functions * saving and restoring the whole * floating-point environment need to do so * anyway to restore the prctl settings from * the saved environment.
*/ #ifdef CONFIG_SPE
tsk->thread.spefscr_last = mfspr(SPRN_SPEFSCR);
tsk->thread.fpexc_mode = val &
(PR_FP_EXC_SW_ENABLE | PR_FP_ALL_EXCEPT); #endif return 0;
} else { return -EINVAL;
}
}
/* on a CONFIG_SPE this does not hurt us. The bits that * __pack_fe01 use do not overlap with bits used for * PR_FP_EXC_SW_ENABLE. Additionally, the MSR[FE0,FE1] bits * on CONFIG_SPE implementations are reserved so writing to
* them does not change anything */ if (val > PR_FP_EXC_PRECISE) return -EINVAL;
tsk->thread.fpexc_mode = __pack_fe01(val); if (regs != NULL && (regs->msr & MSR_FP) != 0) {
regs_set_return_msr(regs, (regs->msr & ~(MSR_FE0|MSR_FE1))
| tsk->thread.fpexc_mode);
} return 0;
}
int get_fpexc_mode(struct task_struct *tsk, unsignedlong adr)
{ unsignedint val = 0;
if (tsk->thread.fpexc_mode & PR_FP_EXC_SW_ENABLE) { if (cpu_has_feature(CPU_FTR_SPE)) { /* * When the sticky exception bits are set * directly by userspace, it must call prctl * with PR_GET_FPEXC (with PR_FP_EXC_SW_ENABLE * in the existing prctl settings) or * PR_SET_FPEXC (with PR_FP_EXC_SW_ENABLE in * the bits being set). <fenv.h> functions * saving and restoring the whole * floating-point environment need to do so * anyway to restore the prctl settings from * the saved environment.
*/ #ifdef CONFIG_SPE
tsk->thread.spefscr_last = mfspr(SPRN_SPEFSCR);
val = tsk->thread.fpexc_mode; #endif
} else return -EINVAL;
} else {
val = __unpack_fe01(tsk->thread.fpexc_mode);
} return put_user(val, (unsignedint __user *) adr);
}
/* * validate the stack frame of a particular minimum size, used for when we are * looking at a certain object in the stack beyond the minimum.
*/ int validate_sp_size(unsignedlong sp, struct task_struct *p, unsignedlong nbytes)
{ unsignedlong stack_page = (unsignedlong)task_stack_page(p);
// A non-empty pt_regs should never have a zero MSR or TRAP value. if (regs->msr || regs->trap) returnfalse;
// Check it sits at the very base of the stack
stack_page = (unsignedlong)task_stack_page(tsk); if ((unsignedlong)(regs + 1) != stack_page + THREAD_SIZE) returnfalse;
lr = 0;
printk("%sCall Trace:\n", loglvl); do { if (!validate_sp(sp, tsk)) break;
stack = (unsignedlong *) sp;
newsp = stack[0];
ip = stack[STACK_FRAME_LR_SAVE]; if (!firstframe || ip != lr) {
printk("%s["REG"] ["REG"] %pS",
loglvl, sp, ip, (void *)ip);
ret_addr = ftrace_graph_ret_addr(current,
&ftrace_idx, ip, stack); if (ret_addr != ip)
pr_cont(" (%pS)", (void *)ret_addr); if (firstframe)
pr_cont(" (unreliable)");
pr_cont("\n");
}
firstframe = 0;
/* * See if this is an exception frame. * We look for the "regs" marker in the current frame. * * STACK_SWITCH_FRAME_SIZE being the smallest frame that * could hold a pt_regs, if that does not fit then it can't * have regs.
*/ if (validate_sp_size(sp, tsk, STACK_SWITCH_FRAME_SIZE)
&& stack[STACK_INT_FRAME_MARKER_LONGS] == STACK_FRAME_REGS_MARKER) { struct pt_regs *regs = (struct pt_regs *)
(sp + STACK_INT_FRAME_REGS);
lr = regs->link;
printk("%s---- interrupt: %lx at %pS\n",
loglvl, regs->trap, (void *)regs->nip);
// Detect the case of an empty pt_regs at the very base // of the stack and suppress showing it in full. if (!empty_user_regs(regs, tsk)) {
__show_regs(regs);
printk("%s---- interrupt: %lx\n", loglvl, regs->trap);
}
firstframe = 1;
}
sp = newsp;
} while (count++ < kstack_depth_to_print);
put_task_stack(tsk);
}
#ifdef CONFIG_PPC64 /* Called with hard IRQs off */ void notrace __ppc64_runlatch_on(void)
{ struct thread_info *ti = current_thread_info();
if (cpu_has_feature(CPU_FTR_ARCH_206)) { /* * Least significant bit (RUN) is the only writable bit of * the CTRL register, so we can avoid mfspr. 2.06 is not the * earliest ISA where this is the case, but it's convenient.
*/
mtspr(SPRN_CTRLT, CTRL_RUNLATCH);
} else { unsignedlong ctrl;
/* * Some architectures (e.g., Cell) have writable fields other * than RUN, so do the read-modify-write.
*/
ctrl = mfspr(SPRN_CTRLF);
ctrl |= CTRL_RUNLATCH;
mtspr(SPRN_CTRLT, ctrl);
}
ti->local_flags |= _TLF_RUNLATCH;
}
/* Called with hard IRQs off */ void notrace __ppc64_runlatch_off(void)
{ struct thread_info *ti = current_thread_info();
ti->local_flags &= ~_TLF_RUNLATCH;
if (cpu_has_feature(CPU_FTR_ARCH_206)) {
mtspr(SPRN_CTRLT, 0);
} else { unsignedlong ctrl;
Die Informationen auf dieser Webseite wurden
nach bestem Wissen sorgfältig zusammengestellt. Es wird jedoch weder Vollständigkeit, noch Richtigkeit,
noch Qualität der bereit gestellten Informationen zugesichert.
Bemerkung:
Die farbliche Syntaxdarstellung und die Messung sind noch experimentell.