GIT 561f2c5913b49ace3d2a63434446941dc28ebdf4 git+ssh://master.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6.git#test commit c47953cfc6bad91c46272aad202ee33276671608 Author: Tony Luck Date: Fri May 11 15:18:45 2007 -0700 [IA64] s/scalibility/scalability/ Previous spelling patch from Simon Arlott broke one spot that didn't need fixing (reported by Simon within 35 minutes of the patch ... but not until after I'd applied to GIT and pushed :-( Signed-off-by: Tony Luck commit 311f594dec9b0c8693ec7df75b82c251b6b0e7c2 Author: Jay Lan Date: Tue Apr 3 17:53:42 2007 -0700 [IA64] kdump on INIT needs multi-nodes sync-up (v.2) The current implementation of kdump on INIT events would enter kdump processing on DIE_INIT_MONARCH_ENTER and DIE_INIT_SLAVE_ENTER events. Thus, the monarch cpu would go ahead and boot up the kdump On SN shub2 systems, this out-of-sync situation causes some slave cpus on different nodes to enter POD. This patch moves kdump entry points to DIE_INIT_MONARCH_LEAVE and DIE_INIT_SLAVE_LEAVE. It also sets kdump_in_progress variable in the DIE_INIT_MONARCH_PROCESS event to not dump all active stack traces to the console in the case of kdump. I have tested this patch on an SN machine and a HP RX2600. Signed-off-by: Jay Lan Acked-by: Zou Nan hai Signed-off-by: Tony Luck commit ae67e498a54259364f7211e10d9834575b340b21 Author: Tony Luck Date: Mon May 14 15:55:11 2007 -0700 [IA64] wire up {signal,timer,event}fd syscalls Signed-off-by: Tony Luck commit 3c3344ca815487568d617860894898737677600a Author: Tony Luck Date: Fri May 11 15:18:45 2007 -0700 [IA64] s/scalibility/scalability/ Previous spelling patch from Simon Arlott broke one spot that didn't need fixing (reported by Simon within 35 minutes of the patch ... but not until after I'd applied to GIT and pushed :-( Signed-off-by: Tony Luck commit 72fdbdce3d52282f8ea95f512e871791256754e6 Author: Simon Arlott Date: Fri May 11 14:55:43 2007 -0700 [IA64] spelling fixes: arch/ia64/ Spelling and apostrophe fixes in arch/ia64/. Signed-off-by: Simon Arlott Signed-off-by: Tony Luck commit a19d4e1a15544a12b811c161039212a2cf63e70c Author: Tony Luck Date: Tue Mar 6 15:41:24 2007 -0800 Revert "[IA64] add idle loop entry/exit notifier" By request from Stephane: This reverts commit 041be4f00312683016c49362f4575aa4cbc264b3. He has an alternative way to do this. Signed-off-by: Tony Luck commit 041be4f00312683016c49362f4575aa4cbc264b3 Author: Stephane Eranian Date: Wed Nov 15 14:23:34 2006 -0800 [IA64] add idle loop entry/exit notifier - add a notifier to the idle loop to get called when entering/leaving the lowest level of the idle loop - migrate SGI LED acitvity over to using the idle notifier Signed-off-by: stephane eranian Signed-off-by: Tony Luck commit 7ab62884069d6d300812fc2b07f1c2332e0882c6 Author: Maeda Naoaki Date: Thu Oct 26 14:50:43 2006 -0700 [IA64] Kdump don't clear vector in kexec_disable_iosapic Mask IOSAPIC will make the following EOI write to IOSAPIC fail because IOSAPIC will not be able to find RTE entry. However move EOI write before IOSAPIC mask will leave a small racy windows between them for interrupts. So mask IOSAPIC but leave the vector field. Signed-off-by: Zou Nan hai Signed-off-by: Tony Luck commit 1050658418ee3d0d385b8679edd1c86941860ab2 Author: Zou Nan hai Date: Wed Oct 25 12:25:59 2006 +0800 [IA64] Kdump Fix fc.i offset in relocate_kernel.S There is 8 byte wrong offset in icache flush instruction in relocate_kernel.S. Which may cause normal Kexec(not kdump) fail on Montecito Box. Signed-off-by: Zou Nan hai Signed-off-by: Tony Luck commit 40eee8ab3441cc4d4fcb7cfbdbd339e7b583f846 Author: Khalid Aziz Date: Thu Oct 12 13:19:45 2006 -0600 [IA64] Fix compile failure with CONFIG_KEXEC but not CONFIG_CRASH_DUMP 2.6.18 kernel patched with kexec/kdump patch from Tony's test tree fails to compile if CONFIG_KEXEC is turned on but CONFIG_CRASH_DUMP is not. Following patch fixes this. Signed-off-by: Khalid Aziz Acked-by: Simon Horman Signed-off-by: Tony Luck commit e1983ff22209e0de0595abe9b61d30078ef3ccb9 Author: Michal Piotrowski Date: Tue Oct 10 14:25:29 2006 -0700 [IA64] kill #include "linux/config.h" config.h is obsolete. This patch removes all #include "linux/config.h". (specifically the one in arch/ia64/kernel/relocate_kernel.S) Signed-off-by: Michal Piotrowski Cc: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Tony Luck commit dc52454505ea87f0e6d4dab534acdd29ef33e7df Author: Zou Nan hai Date: Mon Oct 2 13:51:30 2006 -0700 [IA64] Kexec/Kdump for Itanium Distillation of work largely by Zou Nan hai (who gets Author credit because 'git' only seems to support one author) but also inspired by Khalid Aziz, Simon Horman, and a few others whom I'm too lazy to look up in the mailing list archives, but whose support I appreciate. Signed-off-by: Tony Luck arch/ia64/kernel/acpi.c | 2 +- arch/ia64/kernel/crash.c | 24 ++++++++++++++++-------- arch/ia64/kernel/entry.S | 3 +++ arch/ia64/kernel/irq.c | 6 +++--- arch/ia64/kernel/irq_lsapic.c | 2 +- arch/ia64/kernel/kprobes.c | 6 +++--- arch/ia64/kernel/mca.c | 4 ++++ arch/ia64/kernel/mca_drv.c | 4 ++-- arch/ia64/kernel/module.c | 2 +- arch/ia64/kernel/perfmon.c | 18 +++++++++--------- arch/ia64/kernel/perfmon_mckinley.h | 2 +- arch/ia64/kernel/sal.c | 2 +- arch/ia64/kernel/salinfo.c | 2 +- arch/ia64/kernel/setup.c | 6 +++--- arch/ia64/kernel/smp.c | 12 ++++++------ arch/ia64/kernel/smpboot.c | 6 +++--- arch/ia64/kernel/traps.c | 2 +- arch/ia64/kernel/unwind.c | 2 +- arch/ia64/mm/discontig.c | 2 +- arch/ia64/sn/kernel/bte.c | 12 ++++++------ arch/ia64/sn/kernel/bte_error.c | 4 ++-- arch/ia64/sn/kernel/io_common.c | 2 +- arch/ia64/sn/kernel/setup.c | 2 +- arch/ia64/sn/kernel/sn2/sn2_smp.c | 2 +- arch/ia64/sn/kernel/xpc_channel.c | 8 ++++---- arch/ia64/sn/kernel/xpnet.c | 2 +- arch/ia64/sn/pci/pci_dma.c | 8 ++++---- arch/ia64/sn/pci/pcibr/pcibr_ate.c | 6 +++--- arch/ia64/sn/pci/pcibr/pcibr_dma.c | 2 +- arch/ia64/sn/pci/tioca_provider.c | 6 +++--- arch/ia64/sn/pci/tioce_provider.c | 16 ++++++++-------- include/asm-ia64/unistd.h | 5 ++++- 32 files changed, 100 insertions(+), 82 deletions(-) diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c index 3549c94..c478449 100644 --- a/arch/ia64/kernel/acpi.c +++ b/arch/ia64/kernel/acpi.c @@ -791,7 +791,7 @@ static __init int setup_additional_cpus( early_param("additional_cpus", setup_additional_cpus); /* - * cpu_possible_map should be static, it cannot change as cpu's + * cpu_possible_map should be static, it cannot change as CPUs * are onlined, or offlined. The reason is per-cpu data-structures * are allocated by some modules at init time, and dont expect to * do this dynamically on cpu arrival/departure. diff --git a/arch/ia64/kernel/crash.c b/arch/ia64/kernel/crash.c index aeb79fb..1d64ef4 100644 --- a/arch/ia64/kernel/crash.c +++ b/arch/ia64/kernel/crash.c @@ -156,24 +156,30 @@ kdump_init_notifier(struct notifier_bloc if (!kdump_on_init) return NOTIFY_DONE; - if (val != DIE_INIT_MONARCH_ENTER && - val != DIE_INIT_SLAVE_ENTER && + if (val != DIE_INIT_MONARCH_LEAVE && + val != DIE_INIT_SLAVE_LEAVE && + val != DIE_INIT_MONARCH_PROCESS && val != DIE_MCA_RENDZVOUS_LEAVE && val != DIE_MCA_MONARCH_LEAVE) return NOTIFY_DONE; nd = (struct ia64_mca_notify_die *)args->err; - /* Reason code 1 means machine check rendezous*/ - if ((val == DIE_INIT_MONARCH_ENTER || val == DIE_INIT_SLAVE_ENTER) && - nd->sos->rv_rc == 1) + /* Reason code 1 means machine check rendezvous*/ + if ((val == DIE_INIT_MONARCH_LEAVE || val == DIE_INIT_SLAVE_LEAVE + || val == DIE_INIT_MONARCH_PROCESS) && nd->sos->rv_rc == 1) return NOTIFY_DONE; switch (val) { - case DIE_INIT_MONARCH_ENTER: + case DIE_INIT_MONARCH_PROCESS: + atomic_set(&kdump_in_progress, 1); + *(nd->monarch_cpu) = -1; + break; + case DIE_INIT_MONARCH_LEAVE: machine_kdump_on_init(); break; - case DIE_INIT_SLAVE_ENTER: - unw_init_running(kdump_cpu_freeze, NULL); + case DIE_INIT_SLAVE_LEAVE: + if (atomic_read(&kdump_in_progress)) + unw_init_running(kdump_cpu_freeze, NULL); break; case DIE_MCA_RENDZVOUS_LEAVE: if (atomic_read(&kdump_in_progress)) @@ -215,8 +221,10 @@ #endif static int machine_crash_setup(void) { + /* be notified before default_monarch_init_process */ static struct notifier_block kdump_init_notifier_nb = { .notifier_call = kdump_init_notifier, + .priority = 1, }; int ret; if((ret = register_die_notifier(&kdump_init_notifier_nb)) != 0) diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S index 144b056..95f5175 100644 --- a/arch/ia64/kernel/entry.S +++ b/arch/ia64/kernel/entry.S @@ -1585,5 +1585,8 @@ sys_call_table: data8 sys_getcpu data8 sys_epoll_pwait // 1305 data8 sys_utimensat + data8 sys_signalfd + data8 sys_timerfd + data8 sys_eventfd .org sys_call_table + 8*NR_syscalls // guard against failures to increase NR_syscalls diff --git a/arch/ia64/kernel/irq.c b/arch/ia64/kernel/irq.c index b4c2396..407b458 100644 --- a/arch/ia64/kernel/irq.c +++ b/arch/ia64/kernel/irq.c @@ -4,7 +4,7 @@ * Copyright (C) 1992, 1998 Linus Torvalds, Ingo Molnar * * This file contains the code used by various IRQ handling routines: - * asking for different IRQ's should be done through these routines + * asking for different IRQs should be done through these routines * instead of just grabbing them. Thus setups with different IRQ numbers * shouldn't result in any weird surprises, and installing new handlers * should be easier. @@ -12,7 +12,7 @@ * Copyright (C) Ashok Raj, Intel Corporation 2004 * * 4/14/2004: Added code to handle cpu migration and do safe irq - * migration without lossing interrupts for iosapic + * migration without losing interrupts for iosapic * architecture. */ @@ -190,7 +190,7 @@ void fixup_irqs(void) } /* - * Phase 1: Locate irq's bound to this cpu and + * Phase 1: Locate IRQs bound to this cpu and * relocate them for cpu removal. */ migrate_irqs(); diff --git a/arch/ia64/kernel/irq_lsapic.c b/arch/ia64/kernel/irq_lsapic.c index c2f07be..e56a7a3 100644 --- a/arch/ia64/kernel/irq_lsapic.c +++ b/arch/ia64/kernel/irq_lsapic.c @@ -23,7 +23,7 @@ lsapic_noop_startup (unsigned int irq) static void lsapic_noop (unsigned int irq) { - /* nuthing to do... */ + /* nothing to do... */ } static int lsapic_retrigger(unsigned int irq) diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c index 72e593e..be81fe4 100644 --- a/arch/ia64/kernel/kprobes.c +++ b/arch/ia64/kernel/kprobes.c @@ -151,12 +151,12 @@ static uint __kprobes is_cmp_ctype_unc_i cmp_inst.l = kprobe_inst; if ((cmp_inst.f.x2 == 0) || (cmp_inst.f.x2 == 1)) { - /* Integere compare - Register Register (A6 type)*/ + /* Integer compare - Register Register (A6 type)*/ if ((cmp_inst.f.tb == 0) && (cmp_inst.f.ta == 0) &&(cmp_inst.f.c == 1)) ctype_unc = 1; } else if ((cmp_inst.f.x2 == 2)||(cmp_inst.f.x2 == 3)) { - /* Integere compare - Immediate Register (A8 type)*/ + /* Integer compare - Immediate Register (A8 type)*/ if ((cmp_inst.f.ta == 0) &&(cmp_inst.f.c == 1)) ctype_unc = 1; } @@ -954,7 +954,7 @@ int __kprobes setjmp_pre_handler(struct /* * Callee owns the argument space and could overwrite it, eg * tail call optimization. So to be absolutely safe - * we save the argument space before transfering the control + * we save the argument space before transferring the control * to instrumented jprobe function which runs in * the process context */ diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c index 26814de..93e82e4 100644 --- a/arch/ia64/kernel/mca.c +++ b/arch/ia64/kernel/mca.c @@ -1477,6 +1477,10 @@ default_monarch_init_process(struct noti struct task_struct *g, *t; if (val != DIE_INIT_MONARCH_PROCESS) return NOTIFY_DONE; +#ifdef CONFIG_KEXEC + if (atomic_read(&kdump_in_progress)) + return NOTIFY_DONE; +#endif /* * FIXME: mlogbuf will brim over with INIT stack dumps. diff --git a/arch/ia64/kernel/mca_drv.c b/arch/ia64/kernel/mca_drv.c index 70b8bdb..aba813c 100644 --- a/arch/ia64/kernel/mca_drv.c +++ b/arch/ia64/kernel/mca_drv.c @@ -438,7 +438,7 @@ is_mca_global(peidx_table_t *peidx, pal_ * @peidx: pointer of index of processor error section * * Return value: - * target address on Success / 0 on Failue + * target address on Success / 0 on Failure */ static u64 get_target_identifier(peidx_table_t *peidx) @@ -701,7 +701,7 @@ recover_from_processor_error(int platfor return fatal_mca("External bus check fatal status"); /* - * This is a local MCA and estimated as a recoverble error. + * This is a local MCA and estimated as a recoverable error. */ if (platform) return recover_from_platform_error(slidx, peidx, pbci, sos); diff --git a/arch/ia64/kernel/module.c b/arch/ia64/kernel/module.c index 158e3c5..1962879 100644 --- a/arch/ia64/kernel/module.c +++ b/arch/ia64/kernel/module.c @@ -861,7 +861,7 @@ apply_relocate (Elf64_Shdr *sechdrs, con /* * Modules contain a single unwind table which covers both the core and the init text * sections but since the two are not contiguous, we need to split this table up such that - * we can register (and unregister) each "segment" seperately. Fortunately, this sounds + * we can register (and unregister) each "segment" separately. Fortunately, this sounds * more complicated than it really is. */ static void diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c index e7191ca..b7133ca 100644 --- a/arch/ia64/kernel/perfmon.c +++ b/arch/ia64/kernel/perfmon.c @@ -1318,7 +1318,7 @@ pfm_reserve_session(struct task_struct * { unsigned long flags; /* - * validy checks on cpu_mask have been done upstream + * validity checks on cpu_mask have been done upstream */ LOCK_PFS(flags); @@ -1384,7 +1384,7 @@ pfm_unreserve_session(pfm_context_t *ctx { unsigned long flags; /* - * validy checks on cpu_mask have been done upstream + * validity checks on cpu_mask have been done upstream */ LOCK_PFS(flags); @@ -1835,7 +1835,7 @@ pfm_flush(struct file *filp, fl_owner_t /* * remove our file from the async queue, if we use this mode. * This can be done without the context being protected. We come - * here when the context has become unreacheable by other tasks. + * here when the context has become unreachable by other tasks. * * We may still have active monitoring at this point and we may * end up in pfm_overflow_handler(). However, fasync_helper() @@ -2132,7 +2132,7 @@ doit: filp->private_data = NULL; /* - * if we free on the spot, the context is now completely unreacheable + * if we free on the spot, the context is now completely unreachable * from the callers side. The monitored task side is also cut, so we * can freely cut. * @@ -2562,7 +2562,7 @@ pfm_reset_pmu_state(pfm_context_t *ctx) ctx->ctx_all_pmcs[0] = pmu_conf->impl_pmcs[0] & ~0x1; /* - * bitmask of all PMDs that are accesible to this context + * bitmask of all PMDs that are accessible to this context */ ctx->ctx_all_pmds[0] = pmu_conf->impl_pmds[0]; @@ -3395,7 +3395,7 @@ pfm_read_pmds(pfm_context_t *ctx, void * if (unlikely(!PMD_IS_IMPL(cnum))) goto error; /* * we can only read the register that we use. That includes - * the one we explicitely initialize AND the one we want included + * the one we explicitly initialize AND the one we want included * in the sampling buffer (smpl_regs). * * Having this restriction allows optimization in the ctxsw routine @@ -3715,7 +3715,7 @@ pfm_restart(pfm_context_t *ctx, void *ar * if non-blocking, then we ensure that the task will go into * pfm_handle_work() before returning to user mode. * - * We cannot explicitely reset another task, it MUST always + * We cannot explicitly reset another task, it MUST always * be done by the task itself. This works for system wide because * the tool that is controlling the session is logically doing * "self-monitoring". @@ -4644,7 +4644,7 @@ pfm_exit_thread(struct task_struct *task switch(state) { case PFM_CTX_UNLOADED: /* - * only comes to thios function if pfm_context is not NULL, i.e., cannot + * only comes to this function if pfm_context is not NULL, i.e., cannot * be in unloaded state */ printk(KERN_ERR "perfmon: pfm_exit_thread [%d] ctx unloaded\n", task->pid); @@ -5247,7 +5247,7 @@ pfm_end_notify_user(pfm_context_t *ctx) /* * main overflow processing routine. - * it can be called from the interrupt path or explicitely during the context switch code + * it can be called from the interrupt path or explicitly during the context switch code */ static void pfm_overflow_handler(struct task_struct *task, pfm_context_t *ctx, u64 pmc0, struct pt_regs *regs) diff --git a/arch/ia64/kernel/perfmon_mckinley.h b/arch/ia64/kernel/perfmon_mckinley.h index 9becccd..c4bec7a 100644 --- a/arch/ia64/kernel/perfmon_mckinley.h +++ b/arch/ia64/kernel/perfmon_mckinley.h @@ -181,7 +181,7 @@ static pmu_config_t pmu_conf_mck={ .pmc_desc = pfm_mck_pmc_desc, .num_ibrs = 8, .num_dbrs = 8, - .use_rr_dbregs = 1 /* debug register are use for range retrictions */ + .use_rr_dbregs = 1 /* debug register are use for range restrictions */ }; diff --git a/arch/ia64/kernel/sal.c b/arch/ia64/kernel/sal.c index 37c876f..27c2ef4 100644 --- a/arch/ia64/kernel/sal.c +++ b/arch/ia64/kernel/sal.c @@ -134,7 +134,7 @@ #else * interrupt redirection. The reason is this would require that * All interrupts be stopped and hard bind the irq to a cpu. * Later when the interrupt is fired we need to set the redir hint - * on again in the vector. This is combersome for something that the + * on again in the vector. This is cumbersome for something that the * user mode irq balancer will solve anyways. */ no_int_routing=1; diff --git a/arch/ia64/kernel/salinfo.c b/arch/ia64/kernel/salinfo.c index 89f6b13..25cd75f 100644 --- a/arch/ia64/kernel/salinfo.c +++ b/arch/ia64/kernel/salinfo.c @@ -162,7 +162,7 @@ static DEFINE_SPINLOCK(data_saved_lock); /** salinfo_platform_oemdata - optional callback to decode oemdata from an error * record. * @sect_header: pointer to the start of the section to decode. - * @oemdata: returns vmalloc area containing the decded output. + * @oemdata: returns vmalloc area containing the decoded output. * @oemdata_size: returns length of decoded output (strlen). * * Description: If user space asks for oem data to be decoded by the kernel diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c index 9df1efe..eaa6a24 100644 --- a/arch/ia64/kernel/setup.c +++ b/arch/ia64/kernel/setup.c @@ -576,7 +576,7 @@ #endif } /* - * Display cpu info for all cpu's. + * Display cpu info for all CPUs. */ static int show_cpuinfo (struct seq_file *m, void *v) @@ -761,7 +761,7 @@ #ifdef CONFIG_SMP c->cpu = smp_processor_id(); /* below default values will be overwritten by identify_siblings() - * for Multi-Threading/Multi-Core capable cpu's + * for Multi-Threading/Multi-Core capable CPUs */ c->threads_per_core = c->cores_per_socket = c->num_log = 1; c->socket_id = -1; @@ -947,7 +947,7 @@ #ifdef CONFIG_IA32_SUPPORT ia32_cpu_init(); #endif - /* Clear ITC to eliminiate sched_clock() overflows in human time. */ + /* Clear ITC to eliminate sched_clock() overflows in human time. */ ia64_set_itc(0); /* disable all local interrupt sources: */ diff --git a/arch/ia64/kernel/smp.c b/arch/ia64/kernel/smp.c index 221de38..b3a47f9 100644 --- a/arch/ia64/kernel/smp.c +++ b/arch/ia64/kernel/smp.c @@ -186,7 +186,7 @@ #endif } /* - * Called with preeemption disabled. + * Called with preemption disabled. */ static inline void send_IPI_single (int dest_cpu, int op) @@ -196,7 +196,7 @@ send_IPI_single (int dest_cpu, int op) } /* - * Called with preeemption disabled. + * Called with preemption disabled. */ static inline void send_IPI_allbutself (int op) @@ -210,7 +210,7 @@ send_IPI_allbutself (int op) } /* - * Called with preeemption disabled. + * Called with preemption disabled. */ static inline void send_IPI_all (int op) @@ -223,7 +223,7 @@ send_IPI_all (int op) } /* - * Called with preeemption disabled. + * Called with preemption disabled. */ static inline void send_IPI_self (int op) @@ -252,7 +252,7 @@ kdump_smp_send_init(void) } #endif /* - * Called with preeemption disabled. + * Called with preemption disabled. */ void smp_send_reschedule (int cpu) @@ -261,7 +261,7 @@ smp_send_reschedule (int cpu) } /* - * Called with preeemption disabled. + * Called with preemption disabled. */ static void smp_send_local_flush_tlb (int cpu) diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c index a44792d..5429580 100644 --- a/arch/ia64/kernel/smpboot.c +++ b/arch/ia64/kernel/smpboot.c @@ -694,7 +694,7 @@ int migrate_platform_irqs(unsigned int c set_cpei_target_cpu(new_cpei_cpu); desc = irq_desc + ia64_cpe_irq; /* - * Switch for now, immediatly, we need to do fake intr + * Switch for now, immediately, we need to do fake intr * as other interrupts, but need to study CPEI behaviour with * polling before making changes. */ @@ -840,7 +840,7 @@ __cpu_up (unsigned int cpu) } /* - * Assume that CPU's have been discovered by some platform-dependent interface. For + * Assume that CPUs have been discovered by some platform-dependent interface. For * SoftSDV/Lion, that would be ACPI. * * Setup of the IPI irq handler is done in irq.c:init_IRQ_SMP(). @@ -854,7 +854,7 @@ init_smp_config(void) } *ap_startup; long sal_ret; - /* Tell SAL where to drop the AP's. */ + /* Tell SAL where to drop the APs. */ ap_startup = (struct fptr *) start_ap; sal_ret = ia64_sal_set_vectors(SAL_VECTOR_OS_BOOT_RENDEZ, ia64_tpa(ap_startup->fp), ia64_tpa(ap_startup->gp), 0, 0, 0, 0); diff --git a/arch/ia64/kernel/traps.c b/arch/ia64/kernel/traps.c index b8e0d70..15ad85d 100644 --- a/arch/ia64/kernel/traps.c +++ b/arch/ia64/kernel/traps.c @@ -304,7 +304,7 @@ handle_fpu_swa (int fp_fault, struct pt_ * Lower 4 bits are used as a count. Upper bits are a sequence * number that is updated when count is reset. The cmpxchg will * fail is seqno has changed. This minimizes mutiple cpus - * reseting the count. + * resetting the count. */ if (current_jiffies > last.time) (void) cmpxchg_acq(&last.count, count, 16 + (count & ~15)); diff --git a/arch/ia64/kernel/unwind.c b/arch/ia64/kernel/unwind.c index fe14262..7d3dd6c 100644 --- a/arch/ia64/kernel/unwind.c +++ b/arch/ia64/kernel/unwind.c @@ -2,7 +2,7 @@ * Copyright (C) 1999-2004 Hewlett-Packard Co * David Mosberger-Tang * Copyright (C) 2003 Fenghua Yu - * - Change pt_regs_off() to make it less dependant on pt_regs structure. + * - Change pt_regs_off() to make it less dependent on pt_regs structure. */ /* * This file implements call frame unwind support for the Linux diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c index 38085ac..0dbf0e8 100644 --- a/arch/ia64/mm/discontig.c +++ b/arch/ia64/mm/discontig.c @@ -317,7 +317,7 @@ static void __meminit scatter_node_data( * node_online_map is not set for hot-added nodes at this time, * because we are halfway through initialization of the new node's * structures. If for_each_online_node() is used, a new node's - * pg_data_ptrs will be not initialized. Insted of using it, + * pg_data_ptrs will be not initialized. Instead of using it, * pgdat_list[] is checked. */ for_each_node(node) { diff --git a/arch/ia64/sn/kernel/bte.c b/arch/ia64/sn/kernel/bte.c index ff1c556..b362d6d 100644 --- a/arch/ia64/sn/kernel/bte.c +++ b/arch/ia64/sn/kernel/bte.c @@ -63,7 +63,7 @@ static inline void bte_start_transfer(st * Use the block transfer engine to move kernel memory from src to dest * using the assigned mode. * - * Paramaters: + * Parameters: * src - physical address of the transfer source. * dest - physical address of the transfer destination. * len - number of bytes to transfer from source to dest. @@ -247,7 +247,7 @@ EXPORT_SYMBOL(bte_copy); * use the block transfer engine to move kernel * memory from src to dest using the assigned mode. * - * Paramaters: + * Parameters: * src - physical address of the transfer source. * dest - physical address of the transfer destination. * len - number of bytes to transfer from source to dest. @@ -255,7 +255,7 @@ EXPORT_SYMBOL(bte_copy); * for IBCT0/1 in the SGI documentation. * * NOTE: If the source, dest, and len are all cache line aligned, - * then it would be _FAR_ preferrable to use bte_copy instead. + * then it would be _FAR_ preferable to use bte_copy instead. */ bte_result_t bte_unaligned_copy(u64 src, u64 dest, u64 len, u64 mode) { @@ -300,7 +300,7 @@ bte_result_t bte_unaligned_copy(u64 src, * a standard bte copy. * * One nasty exception to the above rule is when the - * source and destination are not symetrically + * source and destination are not symmetrically * mis-aligned. If the source offset from the first * cache line is different from the destination offset, * we make the first section be the entire transfer @@ -337,7 +337,7 @@ bte_result_t bte_unaligned_copy(u64 src, if (footBcopyDest == (headBcopyDest + headBcopyLen)) { /* - * We have two contigous bcopy + * We have two contiguous bcopy * blocks. Merge them. */ headBcopyLen += footBcopyLen; @@ -375,7 +375,7 @@ bte_result_t bte_unaligned_copy(u64 src, } else { /* - * The transfer is not symetric, we will + * The transfer is not symmetric, we will * allocate a buffer large enough for all the * data, bte_copy into that buffer and then * bcopy to the destination. diff --git a/arch/ia64/sn/kernel/bte_error.c b/arch/ia64/sn/kernel/bte_error.c index b6fcf81..27c5936 100644 --- a/arch/ia64/sn/kernel/bte_error.c +++ b/arch/ia64/sn/kernel/bte_error.c @@ -105,7 +105,7 @@ int shub1_bte_error_handler(unsigned lon } BTE_PRINTK(("eh:%p:%d Cleaning up\n", err_nodepda, smp_processor_id())); - /* Reenable both bte interfaces */ + /* Re-enable both bte interfaces */ imem.ii_imem_regval = REMOTE_HUB_L(nasid, IIO_IMEM); imem.ii_imem_fld_s.i_b0_esd = imem.ii_imem_fld_s.i_b1_esd = 1; REMOTE_HUB_S(nasid, IIO_IMEM, imem.ii_imem_regval); @@ -243,7 +243,7 @@ bte_crb_error_handler(cnodeid_t cnode, i /* * The caller has already figured out the error type, we save that - * in the bte handle structure for the thread excercising the + * in the bte handle structure for the thread exercising the * interface to consume. */ bte->bh_error = ioe->ie_errortype + BTEFAIL_OFFSET; diff --git a/arch/ia64/sn/kernel/io_common.c b/arch/ia64/sn/kernel/io_common.c index 7ed72d3..787ed64 100644 --- a/arch/ia64/sn/kernel/io_common.c +++ b/arch/ia64/sn/kernel/io_common.c @@ -479,7 +479,7 @@ sn_io_early_init(void) } /* - * prime sn_pci_provider[]. Individial provider init routines will + * prime sn_pci_provider[]. Individual provider init routines will * override their respective default entries. */ diff --git a/arch/ia64/sn/kernel/setup.c b/arch/ia64/sn/kernel/setup.c index a9bed5c..a574fcd 100644 --- a/arch/ia64/sn/kernel/setup.c +++ b/arch/ia64/sn/kernel/setup.c @@ -167,7 +167,7 @@ void __init early_sn_setup(void) * IO on SN2 is done via SAL calls, early_printk won't work without this. * * This code duplicates some of the ACPI table parsing that is in efi.c & sal.c. - * Any changes to those file may have to be made hereas well. + * Any changes to those file may have to be made here as well. */ efi_systab = (efi_system_table_t *) __va(ia64_boot_param->efi_systab); config_tables = __va(efi_systab->tables); diff --git a/arch/ia64/sn/kernel/sn2/sn2_smp.c b/arch/ia64/sn/kernel/sn2/sn2_smp.c index 5d318b5..033c8a9 100644 --- a/arch/ia64/sn/kernel/sn2/sn2_smp.c +++ b/arch/ia64/sn/kernel/sn2/sn2_smp.c @@ -104,7 +104,7 @@ static inline unsigned long wait_piowc(v * * SN2 PIO writes from separate CPUs are not guaranteed to arrive in order. * Context switching user threads which have memory-mapped MMIO may cause - * PIOs to issue from seperate CPUs, thus the PIO writes must be drained + * PIOs to issue from separate CPUs, thus the PIO writes must be drained * from the previous CPU's Shub before execution resumes on the new CPU. */ void sn_migrate(struct task_struct *task) diff --git a/arch/ia64/sn/kernel/xpc_channel.c b/arch/ia64/sn/kernel/xpc_channel.c index c08db9c..44ccc0d 100644 --- a/arch/ia64/sn/kernel/xpc_channel.c +++ b/arch/ia64/sn/kernel/xpc_channel.c @@ -293,7 +293,7 @@ xpc_pull_remote_cachelines(struct xpc_pa /* - * Pull the remote per partititon specific variables from the specified + * Pull the remote per partition specific variables from the specified * partition. */ enum xpc_retval @@ -461,7 +461,7 @@ xpc_allocate_local_msgqueue(struct xpc_c // >>> may want to check for ch->flags & XPC_C_DISCONNECTING between // >>> iterations of the for-loop, bail if set? - // >>> should we impose a minumum #of entries? like 4 or 8? + // >>> should we impose a minimum #of entries? like 4 or 8? for (nentries = ch->local_nentries; nentries > 0; nentries--) { nbytes = nentries * ch->msg_size; @@ -514,7 +514,7 @@ xpc_allocate_remote_msgqueue(struct xpc_ // >>> may want to check for ch->flags & XPC_C_DISCONNECTING between // >>> iterations of the for-loop, bail if set? - // >>> should we impose a minumum #of entries? like 4 or 8? + // >>> should we impose a minimum #of entries? like 4 or 8? for (nentries = ch->remote_nentries; nentries > 0; nentries--) { nbytes = nentries * ch->msg_size; @@ -1478,7 +1478,7 @@ xpc_teardown_infrastructure(struct xpc_p /* - * Before proceding with the teardown we have to wait until all + * Before proceeding with the teardown we have to wait until all * existing references cease. */ wait_event(part->teardown_wq, (atomic_read(&part->references) == 0)); diff --git a/arch/ia64/sn/kernel/xpnet.c b/arch/ia64/sn/kernel/xpnet.c index da72135..e58fcad 100644 --- a/arch/ia64/sn/kernel/xpnet.c +++ b/arch/ia64/sn/kernel/xpnet.c @@ -531,7 +531,7 @@ xpnet_dev_hard_start_xmit(struct sk_buff dev_dbg(xpnet, "destination Partitions mask (dp) = 0x%lx\n", dp); /* - * If we wanted to allow promiscous mode to work like an + * If we wanted to allow promiscuous mode to work like an * unswitched network, this would be a good point to OR in a * mask of partitions which should be receiving all packets. */ diff --git a/arch/ia64/sn/pci/pci_dma.c b/arch/ia64/sn/pci/pci_dma.c index 7a291a2..d79ddac 100644 --- a/arch/ia64/sn/pci/pci_dma.c +++ b/arch/ia64/sn/pci/pci_dma.c @@ -333,7 +333,7 @@ int sn_pci_legacy_read(struct pci_bus *b /* * First, try the SN_SAL_IOIF_PCI_SAFE SAL call which can work * around hw issues at the pci bus level. SGI proms older than - * 4.10 don't implment this. + * 4.10 don't implement this. */ SAL_CALL(isrv, SN_SAL_IOIF_PCI_SAFE, @@ -348,7 +348,7 @@ int sn_pci_legacy_read(struct pci_bus *b /* * If the above failed, retry using the SAL_PROBE call which should * be present in all proms (but which cannot work round PCI chipset - * bugs). This code is retained for compatability with old + * bugs). This code is retained for compatibility with old * pre-4.10 proms, and should be removed at some point in the future. */ @@ -379,7 +379,7 @@ int sn_pci_legacy_write(struct pci_bus * /* * First, try the SN_SAL_IOIF_PCI_SAFE SAL call which can work * around hw issues at the pci bus level. SGI proms older than - * 4.10 don't implment this. + * 4.10 don't implement this. */ SAL_CALL(isrv, SN_SAL_IOIF_PCI_SAFE, @@ -394,7 +394,7 @@ int sn_pci_legacy_write(struct pci_bus * /* * If the above failed, retry using the SAL_PROBE call which should * be present in all proms (but which cannot work round PCI chipset - * bugs). This code is retained for compatability with old + * bugs). This code is retained for compatibility with old * pre-4.10 proms, and should be removed at some point in the future. */ diff --git a/arch/ia64/sn/pci/pcibr/pcibr_ate.c b/arch/ia64/sn/pci/pcibr/pcibr_ate.c index 935029f..239b3ce 100644 --- a/arch/ia64/sn/pci/pcibr/pcibr_ate.c +++ b/arch/ia64/sn/pci/pcibr/pcibr_ate.c @@ -30,7 +30,7 @@ static void mark_ate(struct ate_resource /* * find_free_ate: Find the first free ate index starting from the given - * index for the desired consequtive count. + * index for the desired consecutive count. */ static int find_free_ate(struct ate_resource *ate_resource, int start, int count) @@ -88,7 +88,7 @@ static inline int alloc_ate_resource(str return -1; /* - * Find the required number of free consequtive ates. + * Find the required number of free consecutive ates. */ start_index = find_free_ate(ate_resource, ate_resource->lowest_free_index, @@ -105,7 +105,7 @@ static inline int alloc_ate_resource(str /* * Allocate "count" contiguous Bridge Address Translation Entries * on the specified bridge to be used for PCI to XTALK mappings. - * Indices in rm map range from 1..num_entries. Indicies returned + * Indices in rm map range from 1..num_entries. Indices returned * to caller range from 0..num_entries-1. * * Return the start index on success, -1 on failure. diff --git a/arch/ia64/sn/pci/pcibr/pcibr_dma.c b/arch/ia64/sn/pci/pcibr/pcibr_dma.c index 95af40c..e626e50 100644 --- a/arch/ia64/sn/pci/pcibr/pcibr_dma.c +++ b/arch/ia64/sn/pci/pcibr/pcibr_dma.c @@ -201,7 +201,7 @@ pcibr_dmatrans_direct32(struct pcidev_in } /* - * Wrapper routine for free'ing DMA maps + * Wrapper routine for freeing DMA maps * DMA mappings for Direct 64 and 32 do not have any DMA maps. */ void diff --git a/arch/ia64/sn/pci/tioca_provider.c b/arch/ia64/sn/pci/tioca_provider.c index 8a2cb4e..b9bedbd 100644 --- a/arch/ia64/sn/pci/tioca_provider.c +++ b/arch/ia64/sn/pci/tioca_provider.c @@ -223,7 +223,7 @@ tioca_fastwrite_enable(struct tioca_kern /* * Scan all vga controllers on this bus making sure they all - * suport FW. If not, return. + * support FW. If not, return. */ list_for_each_entry(pdev, tioca_kern->ca_devices, bus_list) { @@ -364,7 +364,7 @@ tioca_dma_d48(struct pci_dev *pdev, u64 * @req_size: len (bytes) to map * * Map @paddr into CA address space using the GART mechanism. The mapped - * dma_addr_t is guarenteed to be contiguous in CA bus space. + * dma_addr_t is guaranteed to be contiguous in CA bus space. */ static dma_addr_t tioca_dma_mapped(struct pci_dev *pdev, u64 paddr, size_t req_size) @@ -526,7 +526,7 @@ tioca_dma_map(struct pci_dev *pdev, u64 return 0; /* - * If card is 64 or 48 bit addresable, use a direct mapping. 32 + * If card is 64 or 48 bit addressable, use a direct mapping. 32 * bit direct is so restrictive w.r.t. where the memory resides that * we don't use it even though CA has some support. */ diff --git a/arch/ia64/sn/pci/tioce_provider.c b/arch/ia64/sn/pci/tioce_provider.c index 35f854f..f4c0b96 100644 --- a/arch/ia64/sn/pci/tioce_provider.c +++ b/arch/ia64/sn/pci/tioce_provider.c @@ -256,9 +256,9 @@ pcidev_to_tioce(struct pci_dev *pdev, st * @ct_addr: the coretalk address to map * @len: number of bytes to map * - * Given the addressing type, set up various paramaters that define the + * Given the addressing type, set up various parameters that define the * ATE pool to use. Search for a contiguous block of entries to cover the - * length, and if enough resources exist, fill in the ATE's and construct a + * length, and if enough resources exist, fill in the ATEs and construct a * tioce_dmamap struct to track the mapping. */ static u64 @@ -581,8 +581,8 @@ tioce_do_dma_map(struct pci_dev *pdev, u */ if (!mapaddr && !barrier && dma_mask >= 0xffffffffffUL) { /* - * We have two options for 40-bit mappings: 16GB "super" ATE's - * and 64MB "regular" ATE's. We'll try both if needed for a + * We have two options for 40-bit mappings: 16GB "super" ATEs + * and 64MB "regular" ATEs. We'll try both if needed for a * given mapping but which one we try first depends on the * size. For requests >64MB, prefer to use a super page with * regular as the fallback. Otherwise, try in the reverse order. @@ -687,8 +687,8 @@ tioce_error_intr_handler(int irq, void * } /** - * tioce_reserve_m32 - reserve M32 ate's for the indicated address range - * @tioce_kernel: TIOCE context to reserve ate's for + * tioce_reserve_m32 - reserve M32 ATEs for the indicated address range + * @tioce_kernel: TIOCE context to reserve ATEs for * @base: starting bus address to reserve * @limit: last bus address to reserve * @@ -763,7 +763,7 @@ tioce_kern_init(struct tioce_common *tio /* * Set PMU pagesize to the largest size available, and zero out - * the ate's. + * the ATEs. */ tioce_mmr = (struct tioce __iomem *)tioce_common->ce_pcibus.bs_base; @@ -784,7 +784,7 @@ tioce_kern_init(struct tioce_common *tio } /* - * Reserve ATE's corresponding to reserved address ranges. These + * Reserve ATEs corresponding to reserved address ranges. These * include: * * Memory space covered by each PPB mem base/limit register diff --git a/include/asm-ia64/unistd.h b/include/asm-ia64/unistd.h index f049bc4..d7781a2 100644 --- a/include/asm-ia64/unistd.h +++ b/include/asm-ia64/unistd.h @@ -296,11 +296,14 @@ #define __NR_vmsplice 1302 #define __NR_getcpu 1304 #define __NR_epoll_pwait 1305 #define __NR_utimensat 1306 +#define __NR_signalfd 1307 +#define __NR_timerfd 1308 +#define __NR_eventfd 1309 #ifdef __KERNEL__ -#define NR_syscalls 283 /* length of syscall table */ +#define NR_syscalls 286 /* length of syscall table */ #define __ARCH_WANT_SYS_RT_SIGACTION #define __ARCH_WANT_SYS_RT_SIGSUSPEND