From ff741906ad3cf4b8ca1a958acb013a97a6381ca2 Mon Sep 17 00:00:00 2001 From: Ashok Raj Date: Fri, 11 Nov 2005 14:32:40 -0800 Subject: [IA64] support for cpu0 removal here is the BSP removal support for IA64. Its pretty much the same thing that was released a while back, but has your feedback incorporated. - Removed CONFIG_BSP_REMOVE_WORKAROUND and associated cmdline param - Fixed compile issue with sn2/zx1 due to a undefined fix_b0_for_bsp - some formatting nits (whitespace etc) This has been tested on tiger and long back by alex on hp systems as well. Signed-off-by: Ashok Raj Signed-off-by: Tony Luck --- arch/ia64/Kconfig | 19 +++++++ arch/ia64/configs/tiger_defconfig | 2 + arch/ia64/kernel/acpi.c | 12 ++-- arch/ia64/kernel/iosapic.c | 6 ++ arch/ia64/kernel/irq.c | 13 ++++- arch/ia64/kernel/mca.c | 6 +- arch/ia64/kernel/perfmon.c | 5 +- arch/ia64/kernel/smpboot.c | 114 ++++++++++++++++++++++++++++++++++++-- arch/ia64/kernel/time.c | 9 ++- arch/ia64/mm/contig.c | 4 +- arch/ia64/mm/discontig.c | 9 ++- 11 files changed, 181 insertions(+), 18 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index 199eeaf0f4e..5e0f58e37c5 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -272,6 +272,25 @@ config SCHED_SMT Intel IA64 chips with MultiThreading at a cost of slightly increased overhead in some places. If unsure say N here. +config PERMIT_BSP_REMOVE + bool "Support removal of Bootstrap Processor" + depends on HOTPLUG_CPU + default n + ---help--- + Say Y here if your platform SAL will support removal of BSP with HOTPLUG_CPU + support. + +config FORCE_CPEI_RETARGET + bool "Force assumption that CPEI can be re-targetted" + depends on PERMIT_BSP_REMOVE + default n + ---help--- + Say Y if you need to force the assumption that CPEI can be re-targetted to + any cpu in the system. This hint is available via ACPI 3.0 specifications. + Tiger4 systems are capable of re-directing CPEI to any CPU other than BSP. + This option it useful to enable this feature on older BIOS's as well. + You can also enable this by using boot command line option force_cpei=1. + config PREEMPT bool "Preemptible Kernel" help diff --git a/arch/ia64/configs/tiger_defconfig b/arch/ia64/configs/tiger_defconfig index b1e8f09e9fd..aed034d3397 100644 --- a/arch/ia64/configs/tiger_defconfig +++ b/arch/ia64/configs/tiger_defconfig @@ -114,6 +114,8 @@ CONFIG_FORCE_MAX_ZONEORDER=17 CONFIG_SMP=y CONFIG_NR_CPUS=4 CONFIG_HOTPLUG_CPU=y +CONFIG_PERMIT_BSP_REMOVE=y +CONFIG_FORCE_CPEI_RETARGET=y # CONFIG_SCHED_SMT is not set # CONFIG_PREEMPT is not set CONFIG_SELECT_MEMORY_MODEL=y diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c index 9ad94ddf668..fe1d90b0c6e 100644 --- a/arch/ia64/kernel/acpi.c +++ b/arch/ia64/kernel/acpi.c @@ -287,16 +287,20 @@ acpi_parse_plat_int_src(acpi_table_entry_header * header, unsigned int can_cpei_retarget(void) { extern int cpe_vector; + extern unsigned int force_cpei_retarget; /* * Only if CPEI is supported and the override flag * is present, otherwise return that its re-targettable * if we are in polling mode. */ - if (cpe_vector > 0 && !acpi_cpei_override) - return 0; - else - return 1; + if (cpe_vector > 0) { + if (acpi_cpei_override || force_cpei_retarget) + return 1; + else + return 0; + } + return 1; } unsigned int is_cpu_cpei_target(unsigned int cpu) diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c index 574084f343f..37ac742da8e 100644 --- a/arch/ia64/kernel/iosapic.c +++ b/arch/ia64/kernel/iosapic.c @@ -631,6 +631,7 @@ get_target_cpu (unsigned int gsi, int vector) { #ifdef CONFIG_SMP static int cpu = -1; + extern int cpe_vector; /* * In case of vector shared by multiple RTEs, all RTEs that @@ -653,6 +654,11 @@ get_target_cpu (unsigned int gsi, int vector) if (!cpu_online(smp_processor_id())) return cpu_physical_id(smp_processor_id()); +#ifdef CONFIG_ACPI + if (cpe_vector > 0 && vector == IA64_CPEP_VECTOR) + return get_cpei_target_cpu(); +#endif + #ifdef CONFIG_NUMA { int num_cpus, cpu_index, iosapic_index, numa_cpu, i = 0; diff --git a/arch/ia64/kernel/irq.c b/arch/ia64/kernel/irq.c index d33244c3275..5ce908ef9c9 100644 --- a/arch/ia64/kernel/irq.c +++ b/arch/ia64/kernel/irq.c @@ -163,8 +163,19 @@ void fixup_irqs(void) { unsigned int irq; extern void ia64_process_pending_intr(void); + extern void ia64_disable_timer(void); + extern volatile int time_keeper_id; + + ia64_disable_timer(); + + /* + * Find a new timesync master + */ + if (smp_processor_id() == time_keeper_id) { + time_keeper_id = first_cpu(cpu_online_map); + printk ("CPU %d is now promoted to time-keeper master\n", time_keeper_id); + } - ia64_set_itv(1<<16); /* * Phase 1: Locate irq's bound to this cpu and * relocate them for cpu removal. diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c index 355af15287c..967571b466a 100644 --- a/arch/ia64/kernel/mca.c +++ b/arch/ia64/kernel/mca.c @@ -289,6 +289,7 @@ ia64_mca_log_sal_error_record(int sal_info_type) #ifdef CONFIG_ACPI int cpe_vector = -1; +int ia64_cpe_irq = -1; static irqreturn_t ia64_mca_cpe_int_handler (int cpe_irq, void *arg, struct pt_regs *ptregs) @@ -1444,11 +1445,13 @@ void __devinit ia64_mca_cpu_init(void *cpu_data) { void *pal_vaddr; + static int first_time = 1; - if (smp_processor_id() == 0) { + if (first_time) { void *mca_data; int cpu; + first_time = 0; mca_data = alloc_bootmem(sizeof(struct ia64_mca_cpu) * NR_CPUS + KERNEL_STACK_SIZE); mca_data = (void *)(((unsigned long)mca_data + @@ -1704,6 +1707,7 @@ ia64_mca_late_init(void) desc = irq_descp(irq); desc->status |= IRQ_PER_CPU; setup_irq(irq, &mca_cpe_irqaction); + ia64_cpe_irq = irq; } ia64_mca_register_cpev(cpe_vector); IA64_MCA_DEBUG("%s: CPEI/P setup and enabled.\n", __FUNCTION__); diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c index 410d4804fa6..18c51c37a9a 100644 --- a/arch/ia64/kernel/perfmon.c +++ b/arch/ia64/kernel/perfmon.c @@ -6718,6 +6718,7 @@ __initcall(pfm_init); void pfm_init_percpu (void) { + static int first_time=1; /* * make sure no measurement is active * (may inherit programmed PMCs from EFI). @@ -6730,8 +6731,10 @@ pfm_init_percpu (void) */ pfm_unfreeze_pmu(); - if (smp_processor_id() == 0) + if (first_time) { register_percpu_irq(IA64_PERFMON_VECTOR, &perfmon_irqaction); + first_time=0; + } ia64_setreg(_IA64_REG_CR_PMV, IA64_PERFMON_VECTOR); ia64_srlz_d(); diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c index 8f44e7d2df6..e9d37bf67d6 100644 --- a/arch/ia64/kernel/smpboot.c +++ b/arch/ia64/kernel/smpboot.c @@ -70,6 +70,12 @@ #endif #ifdef CONFIG_HOTPLUG_CPU +#ifdef CONFIG_PERMIT_BSP_REMOVE +#define bsp_remove_ok 1 +#else +#define bsp_remove_ok 0 +#endif + /* * Store all idle threads, this can be reused instead of creating * a new thread. Also avoids complicated thread destroy functionality @@ -104,7 +110,7 @@ struct sal_to_os_boot *sal_state_for_booting_cpu = &sal_boot_rendez_state[0]; /* * ITC synchronization related stuff: */ -#define MASTER 0 +#define MASTER (0) #define SLAVE (SMP_CACHE_BYTES/8) #define NUM_ROUNDS 64 /* magic value */ @@ -151,6 +157,27 @@ char __initdata no_int_routing; unsigned char smp_int_redirect; /* are INT and IPI redirectable by the chipset? */ +#ifdef CONFIG_FORCE_CPEI_RETARGET +#define CPEI_OVERRIDE_DEFAULT (1) +#else +#define CPEI_OVERRIDE_DEFAULT (0) +#endif + +unsigned int force_cpei_retarget = CPEI_OVERRIDE_DEFAULT; + +static int __init +cmdl_force_cpei(char *str) +{ + int value=0; + + get_option (&str, &value); + force_cpei_retarget = value; + + return 1; +} + +__setup("force_cpei=", cmdl_force_cpei); + static int __init nointroute (char *str) { @@ -161,6 +188,27 @@ nointroute (char *str) __setup("nointroute", nointroute); +static void fix_b0_for_bsp(void) +{ +#ifdef CONFIG_HOTPLUG_CPU + int cpuid; + static int fix_bsp_b0 = 1; + + cpuid = smp_processor_id(); + + /* + * Cache the b0 value on the first AP that comes up + */ + if (!(fix_bsp_b0 && cpuid)) + return; + + sal_boot_rendez_state[0].br[0] = sal_boot_rendez_state[cpuid].br[0]; + printk ("Fixed BSP b0 value from CPU %d\n", cpuid); + + fix_bsp_b0 = 0; +#endif +} + void sync_master (void *arg) { @@ -327,8 +375,9 @@ smp_setup_percpu_timer (void) static void __devinit smp_callin (void) { - int cpuid, phys_id; + int cpuid, phys_id, itc_master; extern void ia64_init_itm(void); + extern volatile int time_keeper_id; #ifdef CONFIG_PERFMON extern void pfm_init_percpu(void); @@ -336,6 +385,7 @@ smp_callin (void) cpuid = smp_processor_id(); phys_id = hard_smp_processor_id(); + itc_master = time_keeper_id; if (cpu_online(cpuid)) { printk(KERN_ERR "huh, phys CPU#0x%x, CPU#0x%x already present??\n", @@ -343,6 +393,8 @@ smp_callin (void) BUG(); } + fix_b0_for_bsp(); + lock_ipi_calllock(); cpu_set(cpuid, cpu_online_map); unlock_ipi_calllock(); @@ -365,8 +417,8 @@ smp_callin (void) * calls spin_unlock_bh(), which calls spin_unlock_bh(), which calls * local_bh_enable(), which bugs out if irqs are not enabled... */ - Dprintk("Going to syncup ITC with BP.\n"); - ia64_sync_itc(0); + Dprintk("Going to syncup ITC with ITC Master.\n"); + ia64_sync_itc(itc_master); } /* @@ -638,6 +690,47 @@ remove_siblinginfo(int cpu) } extern void fixup_irqs(void); + +int migrate_platform_irqs(unsigned int cpu) +{ + int new_cpei_cpu; + irq_desc_t *desc = NULL; + cpumask_t mask; + int retval = 0; + + /* + * dont permit CPEI target to removed. + */ + if (cpe_vector > 0 && is_cpu_cpei_target(cpu)) { + printk ("CPU (%d) is CPEI Target\n", cpu); + if (can_cpei_retarget()) { + /* + * Now re-target the CPEI to a different processor + */ + new_cpei_cpu = any_online_cpu(cpu_online_map); + mask = cpumask_of_cpu(new_cpei_cpu); + set_cpei_target_cpu(new_cpei_cpu); + desc = irq_descp(ia64_cpe_irq); + /* + * Switch for now, immediatly, we need to do fake intr + * as other interrupts, but need to study CPEI behaviour with + * polling before making changes. + */ + if (desc) { + desc->handler->disable(ia64_cpe_irq); + desc->handler->set_affinity(ia64_cpe_irq, mask); + desc->handler->enable(ia64_cpe_irq); + printk ("Re-targetting CPEI to cpu %d\n", new_cpei_cpu); + } + } + if (!desc) { + printk ("Unable to retarget CPEI, offline cpu [%d] failed\n", cpu); + retval = -EBUSY; + } + } + return retval; +} + /* must be called with cpucontrol mutex held */ int __cpu_disable(void) { @@ -646,8 +739,17 @@ int __cpu_disable(void) /* * dont permit boot processor for now */ - if (cpu == 0) - return -EBUSY; + if (cpu == 0 && !bsp_remove_ok) { + printk ("Your platform does not support removal of BSP\n"); + return (-EBUSY); + } + + cpu_clear(cpu, cpu_online_map); + + if (migrate_platform_irqs(cpu)) { + cpu_set(cpu, cpu_online_map); + return (-EBUSY); + } remove_siblinginfo(cpu); cpu_clear(cpu, cpu_online_map); diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c index 028a2b95936..1ca130a8385 100644 --- a/arch/ia64/kernel/time.c +++ b/arch/ia64/kernel/time.c @@ -32,7 +32,7 @@ extern unsigned long wall_jiffies; -#define TIME_KEEPER_ID 0 /* smp_processor_id() of time-keeper */ +volatile int time_keeper_id = 0; /* smp_processor_id() of time-keeper */ #ifdef CONFIG_IA64_DEBUG_IRQ @@ -71,7 +71,7 @@ timer_interrupt (int irq, void *dev_id, struct pt_regs *regs) new_itm += local_cpu_data->itm_delta; - if (smp_processor_id() == TIME_KEEPER_ID) { + if (smp_processor_id() == time_keeper_id) { /* * Here we are in the timer irq handler. We have irqs locally * disabled, but we don't know if the timer_bh is running on @@ -236,6 +236,11 @@ static struct irqaction timer_irqaction = { .name = "timer" }; +void __devinit ia64_disable_timer(void) +{ + ia64_set_itv(1 << 16); +} + void __init time_init (void) { diff --git a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c index acaaec4e468..9855ba31809 100644 --- a/arch/ia64/mm/contig.c +++ b/arch/ia64/mm/contig.c @@ -181,13 +181,15 @@ per_cpu_init (void) { void *cpu_data; int cpu; + static int first_time=1; /* * get_free_pages() cannot be used before cpu_init() done. BSP * allocates "NR_CPUS" pages for all CPUs to avoid that AP calls * get_zeroed_page(). */ - if (smp_processor_id() == 0) { + if (first_time) { + first_time=0; cpu_data = __alloc_bootmem(PERCPU_PAGE_SIZE * NR_CPUS, PERCPU_PAGE_SIZE, __pa(MAX_DMA_ADDRESS)); for (cpu = 0; cpu < NR_CPUS; cpu++) { diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c index c87d6d1d581..573d5cc63e2 100644 --- a/arch/ia64/mm/discontig.c +++ b/arch/ia64/mm/discontig.c @@ -528,12 +528,17 @@ void __init find_memory(void) void *per_cpu_init(void) { int cpu; + static int first_time = 1; + if (smp_processor_id() != 0) return __per_cpu_start + __per_cpu_offset[smp_processor_id()]; - for (cpu = 0; cpu < NR_CPUS; cpu++) - per_cpu(local_per_cpu_offset, cpu) = __per_cpu_offset[cpu]; + if (first_time) { + first_time = 0; + for (cpu = 0; cpu < NR_CPUS; cpu++) + per_cpu(local_per_cpu_offset, cpu) = __per_cpu_offset[cpu]; + } return __per_cpu_start + __per_cpu_offset[smp_processor_id()]; } -- cgit v1.2.3-70-g09d2 From b88e926584bf100bc23f5e76b7b674d4257edcb0 Mon Sep 17 00:00:00 2001 From: Ashok Raj Date: Thu, 19 Jan 2006 16:18:47 -0800 Subject: [IA64] Fix UP build with BSP removal support. Causes undefined force_cpei_retarget defined in arch/ia64/kernel/smpboot.c Push the unneeded code inside #ifdef CONFIG_HOTPLUG_CPU. Signed-off-by: Ashok Raj Signed-off-by: Tony Luck --- arch/ia64/kernel/acpi.c | 2 ++ arch/ia64/kernel/iosapic.c | 4 ++-- arch/ia64/kernel/topology.c | 2 +- 3 files changed, 5 insertions(+), 3 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c index fe1d90b0c6e..8d350b33a20 100644 --- a/arch/ia64/kernel/acpi.c +++ b/arch/ia64/kernel/acpi.c @@ -284,6 +284,7 @@ acpi_parse_plat_int_src(acpi_table_entry_header * header, return 0; } +#ifdef CONFIG_HOTPLUG_CPU unsigned int can_cpei_retarget(void) { extern int cpe_vector; @@ -319,6 +320,7 @@ void set_cpei_target_cpu(unsigned int cpu) { acpi_cpei_phys_cpuid = cpu_physical_id(cpu); } +#endif unsigned int get_cpei_target_cpu(void) { diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c index 37ac742da8e..8832c553230 100644 --- a/arch/ia64/kernel/iosapic.c +++ b/arch/ia64/kernel/iosapic.c @@ -655,8 +655,8 @@ get_target_cpu (unsigned int gsi, int vector) return cpu_physical_id(smp_processor_id()); #ifdef CONFIG_ACPI - if (cpe_vector > 0 && vector == IA64_CPEP_VECTOR) - return get_cpei_target_cpu(); + if (cpe_vector > 0 && vector == IA64_CPEP_VECTOR) + return get_cpei_target_cpu(); #endif #ifdef CONFIG_NUMA diff --git a/arch/ia64/kernel/topology.c b/arch/ia64/kernel/topology.c index 706b7734e19..c9562d94b9c 100644 --- a/arch/ia64/kernel/topology.c +++ b/arch/ia64/kernel/topology.c @@ -36,7 +36,7 @@ int arch_register_cpu(int num) parent = &sysfs_nodes[cpu_to_node(num)]; #endif /* CONFIG_NUMA */ -#ifdef CONFIG_ACPI +#if defined (CONFIG_ACPI) && defined (CONFIG_HOTPLUG_CPU) /* * If CPEI cannot be re-targetted, and this is * CPEI target, then dont create the control file -- cgit v1.2.3-70-g09d2 From b0a06623dc4caf6dfb6a84419507643471676d20 Mon Sep 17 00:00:00 2001 From: Keith Owens Date: Sun, 22 Jan 2006 10:55:25 +1100 Subject: [IA64] Delete MCA/INIT sigdelayed code The only user of the MCA/INIT sigdelayed code (SGI's I/O probing) has moved from the kernel into SAL. Delete the MCA/INIT sigdelayed code. Signed-off-by: Keith Owens Signed-off-by: Tony Luck --- arch/ia64/kernel/entry.S | 14 ------ arch/ia64/kernel/signal.c | 101 ----------------------------------------- include/asm-ia64/signal.h | 2 - include/asm-ia64/thread_info.h | 11 +---- 4 files changed, 1 insertion(+), 127 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S index 7a6ffd61378..9dda7a36d1e 100644 --- a/arch/ia64/kernel/entry.S +++ b/arch/ia64/kernel/entry.S @@ -1100,9 +1100,6 @@ skip_rbs_switch: st8 [r2]=r8 st8 [r3]=r10 .work_pending: - tbit.nz p6,p0=r31,TIF_SIGDELAYED // signal delayed from MCA/INIT/NMI/PMI context? -(p6) br.cond.sptk.few .sigdelayed - ;; tbit.z p6,p0=r31,TIF_NEED_RESCHED // current_thread_info()->need_resched==0? (p6) br.cond.sptk.few .notify #ifdef CONFIG_PREEMPT @@ -1129,17 +1126,6 @@ skip_rbs_switch: (pLvSys)br.cond.sptk.few .work_pending_syscall_end br.cond.sptk.many .work_processed_kernel // don't re-check -// There is a delayed signal that was detected in MCA/INIT/NMI/PMI context where -// it could not be delivered. Deliver it now. The signal might be for us and -// may set TIF_SIGPENDING, so redrive ia64_leave_* after processing the delayed -// signal. - -.sigdelayed: - br.call.sptk.many rp=do_sigdelayed - cmp.eq p6,p0=r0,r0 // p6 <- 1, always re-check -(pLvSys)br.cond.sptk.few .work_pending_syscall_end - br.cond.sptk.many .work_processed_kernel // re-check - .work_pending_syscall_end: adds r2=PT(R8)+16,r12 adds r3=PT(R10)+16,r12 diff --git a/arch/ia64/kernel/signal.c b/arch/ia64/kernel/signal.c index 463f6bb44d0..1d7903ee212 100644 --- a/arch/ia64/kernel/signal.c +++ b/arch/ia64/kernel/signal.c @@ -588,104 +588,3 @@ ia64_do_signal (sigset_t *oldset, struct sigscratch *scr, long in_syscall) } return 0; } - -/* Set a delayed signal that was detected in MCA/INIT/NMI/PMI context where it - * could not be delivered. It is important that the target process is not - * allowed to do any more work in user space. Possible cases for the target - * process: - * - * - It is sleeping and will wake up soon. Store the data in the current task, - * the signal will be sent when the current task returns from the next - * interrupt. - * - * - It is running in user context. Store the data in the current task, the - * signal will be sent when the current task returns from the next interrupt. - * - * - It is running in kernel context on this or another cpu and will return to - * user context. Store the data in the target task, the signal will be sent - * to itself when the target task returns to user space. - * - * - It is running in kernel context on this cpu and will sleep before - * returning to user context. Because this is also the current task, the - * signal will not get delivered and the task could sleep indefinitely. - * Store the data in the idle task for this cpu, the signal will be sent - * after the idle task processes its next interrupt. - * - * To cover all cases, store the data in the target task, the current task and - * the idle task on this cpu. Whatever happens, the signal will be delivered - * to the target task before it can do any useful user space work. Multiple - * deliveries have no unwanted side effects. - * - * Note: This code is executed in MCA/INIT/NMI/PMI context, with interrupts - * disabled. It must not take any locks nor use kernel structures or services - * that require locks. - */ - -/* To ensure that we get the right pid, check its start time. To avoid extra - * include files in thread_info.h, convert the task start_time to unsigned long, - * giving us a cycle time of > 580 years. - */ -static inline unsigned long -start_time_ul(const struct task_struct *t) -{ - return t->start_time.tv_sec * NSEC_PER_SEC + t->start_time.tv_nsec; -} - -void -set_sigdelayed(pid_t pid, int signo, int code, void __user *addr) -{ - struct task_struct *t; - unsigned long start_time = 0; - int i; - - for (i = 1; i <= 3; ++i) { - switch (i) { - case 1: - t = find_task_by_pid(pid); - if (t) - start_time = start_time_ul(t); - break; - case 2: - t = current; - break; - default: - t = idle_task(smp_processor_id()); - break; - } - - if (!t) - return; - task_thread_info(t)->sigdelayed.signo = signo; - task_thread_info(t)->sigdelayed.code = code; - task_thread_info(t)->sigdelayed.addr = addr; - task_thread_info(t)->sigdelayed.start_time = start_time; - task_thread_info(t)->sigdelayed.pid = pid; - wmb(); - set_tsk_thread_flag(t, TIF_SIGDELAYED); - } -} - -/* Called from entry.S when it detects TIF_SIGDELAYED, a delayed signal that - * was detected in MCA/INIT/NMI/PMI context where it could not be delivered. - */ - -void -do_sigdelayed(void) -{ - struct siginfo siginfo; - pid_t pid; - struct task_struct *t; - - clear_thread_flag(TIF_SIGDELAYED); - memset(&siginfo, 0, sizeof(siginfo)); - siginfo.si_signo = current_thread_info()->sigdelayed.signo; - siginfo.si_code = current_thread_info()->sigdelayed.code; - siginfo.si_addr = current_thread_info()->sigdelayed.addr; - pid = current_thread_info()->sigdelayed.pid; - t = find_task_by_pid(pid); - if (!t) - return; - if (current_thread_info()->sigdelayed.start_time != start_time_ul(t)) - return; - force_sig_info(siginfo.si_signo, &siginfo, t); -} diff --git a/include/asm-ia64/signal.h b/include/asm-ia64/signal.h index 608168d713d..5e328ed5d01 100644 --- a/include/asm-ia64/signal.h +++ b/include/asm-ia64/signal.h @@ -158,8 +158,6 @@ struct k_sigaction { #define ptrace_signal_deliver(regs, cookie) do { } while (0) -void set_sigdelayed(pid_t pid, int signo, int code, void __user *addr); - #endif /* __KERNEL__ */ # endif /* !__ASSEMBLY__ */ diff --git a/include/asm-ia64/thread_info.h b/include/asm-ia64/thread_info.h index 1d6518fe1f0..a6ee27343a4 100644 --- a/include/asm-ia64/thread_info.h +++ b/include/asm-ia64/thread_info.h @@ -29,13 +29,6 @@ struct thread_info { mm_segment_t addr_limit; /* user-level address space limit */ int preempt_count; /* 0=premptable, <0=BUG; will also serve as bh-counter */ struct restart_block restart_block; - struct { - int signo; - int code; - void __user *addr; - unsigned long start_time; - pid_t pid; - } sigdelayed; /* Saved information for TIF_SIGDELAYED */ }; #define THREAD_SIZE KERNEL_STACK_SIZE @@ -89,7 +82,6 @@ struct thread_info { #define TIF_NEED_RESCHED 2 /* rescheduling necessary */ #define TIF_SYSCALL_TRACE 3 /* syscall trace active */ #define TIF_SYSCALL_AUDIT 4 /* syscall auditing active */ -#define TIF_SIGDELAYED 5 /* signal delayed from MCA/INIT/NMI/PMI context */ #define TIF_POLLING_NRFLAG 16 /* true if poll_idle() is polling TIF_NEED_RESCHED */ #define TIF_MEMDIE 17 #define TIF_MCA_INIT 18 /* this task is processing MCA or INIT */ @@ -101,13 +93,12 @@ struct thread_info { #define _TIF_NOTIFY_RESUME (1 << TIF_NOTIFY_RESUME) #define _TIF_SIGPENDING (1 << TIF_SIGPENDING) #define _TIF_NEED_RESCHED (1 << TIF_NEED_RESCHED) -#define _TIF_SIGDELAYED (1 << TIF_SIGDELAYED) #define _TIF_POLLING_NRFLAG (1 << TIF_POLLING_NRFLAG) #define _TIF_MCA_INIT (1 << TIF_MCA_INIT) #define _TIF_DB_DISABLED (1 << TIF_DB_DISABLED) /* "work to do on user-return" bits */ -#define TIF_ALLWORK_MASK (_TIF_NOTIFY_RESUME|_TIF_SIGPENDING|_TIF_NEED_RESCHED|_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SIGDELAYED) +#define TIF_ALLWORK_MASK (_TIF_NOTIFY_RESUME|_TIF_SIGPENDING|_TIF_NEED_RESCHED|_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT) /* like TIF_ALLWORK_BITS but sans TIF_SYSCALL_TRACE or TIF_SYSCALL_AUDIT */ #define TIF_WORK_MASK (TIF_ALLWORK_MASK&~(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT)) -- cgit v1.2.3-70-g09d2 From 13938ca7a1ad9a4788cf73309f187d99c97ddfde Mon Sep 17 00:00:00 2001 From: Mark Maule Date: Thu, 26 Jan 2006 14:46:39 -0600 Subject: [IA64-SGI] driver bugfixes and hardware workarounds for CE1.0 asic Various bugfixes and hardware bug workarounds necessary for the rev 1.0 version of the altix TIO CE asic. Signed-off-by: Mark Maule Signed-off-by: Tony Luck --- arch/ia64/sn/pci/tioce_provider.c | 326 +++++++++++++++++++++++++++++++++++--- include/asm-ia64/sn/addrs.h | 8 + include/asm-ia64/sn/tioce.h | 36 ++++- 3 files changed, 345 insertions(+), 25 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/sn/pci/tioce_provider.c b/arch/ia64/sn/pci/tioce_provider.c index e52831ed93e..fa073cc4b56 100644 --- a/arch/ia64/sn/pci/tioce_provider.c +++ b/arch/ia64/sn/pci/tioce_provider.c @@ -15,6 +15,124 @@ #include #include #include +#include + +/* + * 1/26/2006 + * + * WAR for SGI PV 944642. For revA TIOCE, need to use the following recipe + * (taken from the above PV) before and after accessing tioce internal MMR's + * to avoid tioce lockups. + * + * The recipe as taken from the PV: + * + * if(mmr address < 0x45000) { + * if(mmr address == 0 or 0x80) + * mmr wrt or read address 0xc0 + * else if(mmr address == 0x148 or 0x200) + * mmr wrt or read address 0x28 + * else + * mmr wrt or read address 0x158 + * + * do desired mmr access (rd or wrt) + * + * if(mmr address == 0x100) + * mmr wrt or read address 0x38 + * mmr wrt or read address 0xb050 + * } else + * do desired mmr access + * + * According to hw, we can use reads instead of writes to the above addres + * + * Note this WAR can only to be used for accessing internal MMR's in the + * TIOCE Coretalk Address Range 0x0 - 0x07ff_ffff. This includes the + * "Local CE Registers and Memories" and "PCI Compatible Config Space" address + * spaces from table 2-1 of the "CE Programmer's Reference Overview" document. + * + * All registers defined in struct tioce will meet that criteria. + */ + +static void inline +tioce_mmr_war_pre(struct tioce_kernel *kern, void *mmr_addr) +{ + u64 mmr_base; + u64 mmr_offset; + + if (kern->ce_common->ce_rev != TIOCE_REV_A) + return; + + mmr_base = kern->ce_common->ce_pcibus.bs_base; + mmr_offset = (u64)mmr_addr - mmr_base; + + if (mmr_offset < 0x45000) { + u64 mmr_war_offset; + + if (mmr_offset == 0 || mmr_offset == 0x80) + mmr_war_offset = 0xc0; + else if (mmr_offset == 0x148 || mmr_offset == 0x200) + mmr_war_offset = 0x28; + else + mmr_war_offset = 0x158; + + readq_relaxed((void *)(mmr_base + mmr_war_offset)); + } +} + +static void inline +tioce_mmr_war_post(struct tioce_kernel *kern, void *mmr_addr) +{ + u64 mmr_base; + u64 mmr_offset; + + if (kern->ce_common->ce_rev != TIOCE_REV_A) + return; + + mmr_base = kern->ce_common->ce_pcibus.bs_base; + mmr_offset = (u64)mmr_addr - mmr_base; + + if (mmr_offset < 0x45000) { + if (mmr_offset == 0x100) + readq_relaxed((void *)(mmr_base + 0x38)); + readq_relaxed((void *)(mmr_base + 0xb050)); + } +} + +/* load mmr contents into a variable */ +#define tioce_mmr_load(kern, mmrp, varp) do {\ + tioce_mmr_war_pre(kern, mmrp); \ + *(varp) = readq_relaxed(mmrp); \ + tioce_mmr_war_post(kern, mmrp); \ +} while (0) + +/* store variable contents into mmr */ +#define tioce_mmr_store(kern, mmrp, varp) do {\ + tioce_mmr_war_pre(kern, mmrp); \ + writeq(*varp, mmrp); \ + tioce_mmr_war_post(kern, mmrp); \ +} while (0) + +/* store immediate value into mmr */ +#define tioce_mmr_storei(kern, mmrp, val) do {\ + tioce_mmr_war_pre(kern, mmrp); \ + writeq(val, mmrp); \ + tioce_mmr_war_post(kern, mmrp); \ +} while (0) + +/* set bits (immediate value) into mmr */ +#define tioce_mmr_seti(kern, mmrp, bits) do {\ + u64 tmp; \ + tioce_mmr_load(kern, mmrp, &tmp); \ + tmp |= (bits); \ + tioce_mmr_store(kern, mmrp, &tmp); \ +} while (0) + +/* clear bits (immediate value) into mmr */ +#define tioce_mmr_clri(kern, mmrp, bits) do { \ + u64 tmp; \ + tioce_mmr_load(kern, mmrp, &tmp); \ + tmp &= ~(bits); \ + tioce_mmr_store(kern, mmrp, &tmp); \ +} while (0) /** * Bus address ranges for the 5 flavors of TIOCE DMA @@ -62,9 +180,9 @@ #define TIOCE_ATE_M40 2 #define TIOCE_ATE_M40S 3 -#define KB(x) ((x) << 10) -#define MB(x) ((x) << 20) -#define GB(x) ((x) << 30) +#define KB(x) ((u64)(x) << 10) +#define MB(x) ((u64)(x) << 20) +#define GB(x) ((u64)(x) << 30) /** * tioce_dma_d64 - create a DMA mapping using 64-bit direct mode @@ -151,7 +269,7 @@ tioce_alloc_map(struct tioce_kernel *ce_kern, int type, int port, int last; int entries; int nates; - int pagesize; + u64 pagesize; u64 *ate_shadow; u64 *ate_reg; u64 addr; @@ -228,7 +346,7 @@ tioce_alloc_map(struct tioce_kernel *ce_kern, int type, int port, ate = ATE_MAKE(addr, pagesize); ate_shadow[i + j] = ate; - writeq(ate, &ate_reg[i + j]); + tioce_mmr_storei(ce_kern, &ate_reg[i + j], ate); addr += pagesize; } @@ -272,7 +390,8 @@ tioce_dma_d32(struct pci_dev *pdev, u64 ct_addr) u64 tmp; ce_kern->ce_port[port].dirmap_shadow = ct_upper; - writeq(ct_upper, &ce_mmr->ce_ure_dir_map[port]); + tioce_mmr_storei(ce_kern, &ce_mmr->ce_ure_dir_map[port], + ct_upper); tmp = ce_mmr->ce_ure_dir_map[port]; dma_ok = 1; } else @@ -344,7 +463,8 @@ tioce_dma_unmap(struct pci_dev *pdev, dma_addr_t bus_addr, int dir) if (TIOCE_D32_ADDR(bus_addr)) { if (--ce_kern->ce_port[port].dirmap_refcnt == 0) { ce_kern->ce_port[port].dirmap_shadow = 0; - writeq(0, &ce_mmr->ce_ure_dir_map[port]); + tioce_mmr_storei(ce_kern, &ce_mmr->ce_ure_dir_map[port], + 0); } } else { struct tioce_dmamap *map; @@ -365,7 +485,7 @@ tioce_dma_unmap(struct pci_dev *pdev, dma_addr_t bus_addr, int dir) } else if (--map->refcnt == 0) { for (i = 0; i < map->ate_count; i++) { map->ate_shadow[i] = 0; - map->ate_hw[i] = 0; + tioce_mmr_storei(ce_kern, &map->ate_hw[i], 0); } list_del(&map->ce_dmamap_list); @@ -486,7 +606,7 @@ tioce_do_dma_map(struct pci_dev *pdev, u64 paddr, size_t byte_count, spin_unlock_irqrestore(&ce_kern->ce_lock, flags); dma_map_done: - if (mapaddr & barrier) + if (mapaddr && barrier) mapaddr = tioce_dma_barrier(mapaddr, 1); return mapaddr; @@ -541,17 +661,61 @@ tioce_error_intr_handler(int irq, void *arg, struct pt_regs *pt) soft->ce_pcibus.bs_persist_segment, soft->ce_pcibus.bs_persist_busnum, 0, 0, 0, 0, 0); + if (ret_stuff.v0) + panic("tioce_error_intr_handler: Fatal TIOCE error"); + return IRQ_HANDLED; } +/** + * tioce_reserve_m32 - reserve M32 ate's for the indicated address range + * @tioce_kernel: TIOCE context to reserve ate's for + * @base: starting bus address to reserve + * @limit: last bus address to reserve + * + * If base/limit falls within the range of bus space mapped through the + * M32 space, reserve the resources corresponding to the range. + */ +static void +tioce_reserve_m32(struct tioce_kernel *ce_kern, u64 base, u64 limit) +{ + int ate_index, last_ate, ps; + struct tioce *ce_mmr; + + if (!TIOCE_M32_ADDR(base)) + return; + + ce_mmr = (struct tioce *)ce_kern->ce_common->ce_pcibus.bs_base; + ps = ce_kern->ce_ate3240_pagesize; + ate_index = ATE_PAGE(base, ps); + last_ate = ate_index + ATE_NPAGES(base, limit-base+1, ps) - 1; + + if (ate_index < 64) + ate_index = 64; + + while (ate_index <= last_ate) { + u64 ate; + + ate = ATE_MAKE(0xdeadbeef, ps); + ce_kern->ce_ate3240_shadow[ate_index] = ate; + tioce_mmr_storei(ce_kern, &ce_mmr->ce_ure_ate3240[ate_index], + ate); + ate_index++; + } +} + /** * tioce_kern_init - init kernel structures related to a given TIOCE * @tioce_common: ptr to a cached tioce_common struct that originated in prom - */ static struct tioce_kernel * + */ +static struct tioce_kernel * tioce_kern_init(struct tioce_common *tioce_common) { int i; + int ps; + int dev; u32 tmp; + unsigned int seg, bus; struct tioce *tioce_mmr; struct tioce_kernel *tioce_kern; @@ -572,9 +736,10 @@ tioce_kern_init(struct tioce_common *tioce_common) * here to use pci_read_config_xxx() so use the raw_pci_ops vector. */ - raw_pci_ops->read(tioce_common->ce_pcibus.bs_persist_segment, - tioce_common->ce_pcibus.bs_persist_busnum, - PCI_DEVFN(2, 0), PCI_SECONDARY_BUS, 1, &tmp); + seg = tioce_common->ce_pcibus.bs_persist_segment; + bus = tioce_common->ce_pcibus.bs_persist_busnum; + + raw_pci_ops->read(seg, bus, PCI_DEVFN(2, 0), PCI_SECONDARY_BUS, 1,&tmp); tioce_kern->ce_port1_secondary = (u8) tmp; /* @@ -583,18 +748,76 @@ tioce_kern_init(struct tioce_common *tioce_common) */ tioce_mmr = (struct tioce *)tioce_common->ce_pcibus.bs_base; - __sn_clrq_relaxed(&tioce_mmr->ce_ure_page_map, CE_URE_PAGESIZE_MASK); - __sn_setq_relaxed(&tioce_mmr->ce_ure_page_map, CE_URE_256K_PAGESIZE); - tioce_kern->ce_ate3240_pagesize = KB(256); + tioce_mmr_clri(tioce_kern, &tioce_mmr->ce_ure_page_map, + CE_URE_PAGESIZE_MASK); + tioce_mmr_seti(tioce_kern, &tioce_mmr->ce_ure_page_map, + CE_URE_256K_PAGESIZE); + ps = tioce_kern->ce_ate3240_pagesize = KB(256); for (i = 0; i < TIOCE_NUM_M40_ATES; i++) { tioce_kern->ce_ate40_shadow[i] = 0; - writeq(0, &tioce_mmr->ce_ure_ate40[i]); + tioce_mmr_storei(tioce_kern, &tioce_mmr->ce_ure_ate40[i], 0); } for (i = 0; i < TIOCE_NUM_M3240_ATES; i++) { tioce_kern->ce_ate3240_shadow[i] = 0; - writeq(0, &tioce_mmr->ce_ure_ate3240[i]); + tioce_mmr_storei(tioce_kern, &tioce_mmr->ce_ure_ate3240[i], 0); + } + + /* + * Reserve ATE's corresponding to reserved address ranges. These + * include: + * + * Memory space covered by each PPB mem base/limit register + * Memory space covered by each PPB prefetch base/limit register + * + * These bus ranges are for pio (downstream) traffic only, and so + * cannot be used for DMA. + */ + + for (dev = 1; dev <= 2; dev++) { + u64 base, limit; + + /* mem base/limit */ + + raw_pci_ops->read(seg, bus, PCI_DEVFN(dev, 0), + PCI_MEMORY_BASE, 2, &tmp); + base = (u64)tmp << 16; + + raw_pci_ops->read(seg, bus, PCI_DEVFN(dev, 0), + PCI_MEMORY_LIMIT, 2, &tmp); + limit = (u64)tmp << 16; + limit |= 0xfffffUL; + + if (base < limit) + tioce_reserve_m32(tioce_kern, base, limit); + + /* + * prefetch mem base/limit. The tioce ppb's have 64-bit + * decoders, so read the upper portions w/o checking the + * attributes. + */ + + raw_pci_ops->read(seg, bus, PCI_DEVFN(dev, 0), + PCI_PREF_MEMORY_BASE, 2, &tmp); + base = ((u64)tmp & PCI_PREF_RANGE_MASK) << 16; + + raw_pci_ops->read(seg, bus, PCI_DEVFN(dev, 0), + PCI_PREF_BASE_UPPER32, 4, &tmp); + base |= (u64)tmp << 32; + + raw_pci_ops->read(seg, bus, PCI_DEVFN(dev, 0), + PCI_PREF_MEMORY_LIMIT, 2, &tmp); + + limit = ((u64)tmp & PCI_PREF_RANGE_MASK) << 16; + limit |= 0xfffffUL; + + raw_pci_ops->read(seg, bus, PCI_DEVFN(dev, 0), + PCI_PREF_LIMIT_UPPER32, 4, &tmp); + limit |= (u64)tmp << 32; + + if ((base < limit) && TIOCE_M32_ADDR(base)) + tioce_reserve_m32(tioce_kern, base, limit); } return tioce_kern; @@ -614,6 +837,7 @@ tioce_force_interrupt(struct sn_irq_info *sn_irq_info) { struct pcidev_info *pcidev_info; struct tioce_common *ce_common; + struct tioce_kernel *ce_kern; struct tioce *ce_mmr; u64 force_int_val; @@ -629,6 +853,29 @@ tioce_force_interrupt(struct sn_irq_info *sn_irq_info) ce_common = (struct tioce_common *)pcidev_info->pdi_pcibus_info; ce_mmr = (struct tioce *)ce_common->ce_pcibus.bs_base; + ce_kern = (struct tioce_kernel *)ce_common->ce_kernel_private; + + /* + * TIOCE Rev A workaround (PV 945826), force an interrupt by writing + * the TIO_INTx register directly (1/26/2006) + */ + if (ce_common->ce_rev == TIOCE_REV_A) { + u64 int_bit_mask = (1ULL << sn_irq_info->irq_int_bit); + u64 status; + + tioce_mmr_load(ce_kern, &ce_mmr->ce_adm_int_status, &status); + if (status & int_bit_mask) { + u64 force_irq = (1 << 8) | sn_irq_info->irq_irq; + u64 ctalk = sn_irq_info->irq_xtalkaddr; + u64 nasid, offset; + + nasid = (ctalk & CTALK_NASID_MASK) >> CTALK_NASID_SHFT; + offset = (ctalk & CTALK_NODE_OFFSET); + HUB_S(TIO_IOSPACE_ADDR(nasid, offset), force_irq); + } + + return; + } /* * irq_int_bit is originally set up by prom, and holds the interrupt @@ -666,7 +913,7 @@ tioce_force_interrupt(struct sn_irq_info *sn_irq_info) default: return; } - writeq(force_int_val, &ce_mmr->ce_adm_force_int); + tioce_mmr_storei(ce_kern, &ce_mmr->ce_adm_force_int, force_int_val); } /** @@ -685,6 +932,7 @@ tioce_target_interrupt(struct sn_irq_info *sn_irq_info) { struct pcidev_info *pcidev_info; struct tioce_common *ce_common; + struct tioce_kernel *ce_kern; struct tioce *ce_mmr; int bit; u64 vector; @@ -695,14 +943,15 @@ tioce_target_interrupt(struct sn_irq_info *sn_irq_info) ce_common = (struct tioce_common *)pcidev_info->pdi_pcibus_info; ce_mmr = (struct tioce *)ce_common->ce_pcibus.bs_base; + ce_kern = (struct tioce_kernel *)ce_common->ce_kernel_private; bit = sn_irq_info->irq_int_bit; - __sn_setq_relaxed(&ce_mmr->ce_adm_int_mask, (1UL << bit)); + tioce_mmr_seti(ce_kern, &ce_mmr->ce_adm_int_mask, (1UL << bit)); vector = (u64)sn_irq_info->irq_irq << INTR_VECTOR_SHFT; vector |= sn_irq_info->irq_xtalkaddr; - writeq(vector, &ce_mmr->ce_adm_int_dest[bit]); - __sn_clrq_relaxed(&ce_mmr->ce_adm_int_mask, (1UL << bit)); + tioce_mmr_storei(ce_kern, &ce_mmr->ce_adm_int_dest[bit], vector); + tioce_mmr_clri(ce_kern, &ce_mmr->ce_adm_int_mask, (1UL << bit)); tioce_force_interrupt(sn_irq_info); } @@ -721,7 +970,11 @@ tioce_target_interrupt(struct sn_irq_info *sn_irq_info) static void * tioce_bus_fixup(struct pcibus_bussoft *prom_bussoft, struct pci_controller *controller) { + int my_nasid; + cnodeid_t my_cnode, mem_cnode; struct tioce_common *tioce_common; + struct tioce_kernel *tioce_kern; + struct tioce *tioce_mmr; /* * Allocate kernel bus soft and copy from prom. @@ -734,11 +987,23 @@ tioce_bus_fixup(struct pcibus_bussoft *prom_bussoft, struct pci_controller *cont memcpy(tioce_common, prom_bussoft, sizeof(struct tioce_common)); tioce_common->ce_pcibus.bs_base |= __IA64_UNCACHED_OFFSET; - if (tioce_kern_init(tioce_common) == NULL) { + tioce_kern = tioce_kern_init(tioce_common); + if (tioce_kern == NULL) { kfree(tioce_common); return NULL; } + /* + * Clear out any transient errors before registering the error + * interrupt handler. + */ + + tioce_mmr = (struct tioce *)tioce_common->ce_pcibus.bs_base; + tioce_mmr_seti(tioce_kern, &tioce_mmr->ce_adm_int_status_alias, ~0ULL); + tioce_mmr_seti(tioce_kern, &tioce_mmr->ce_adm_error_summary_alias, + ~0ULL); + tioce_mmr_seti(tioce_kern, &tioce_mmr->ce_dre_comp_err_addr, ~0ULL); + if (request_irq(SGI_PCIASIC_ERROR, tioce_error_intr_handler, SA_SHIRQ, "TIOCE error", (void *)tioce_common)) @@ -750,6 +1015,21 @@ tioce_bus_fixup(struct pcibus_bussoft *prom_bussoft, struct pci_controller *cont tioce_common->ce_pcibus.bs_persist_segment, tioce_common->ce_pcibus.bs_persist_busnum); + /* + * identify closest nasid for memory allocations + */ + + my_nasid = NASID_GET(tioce_common->ce_pcibus.bs_base); + my_cnode = nasid_to_cnodeid(my_nasid); + + if (sn_hwperf_get_nearest_node(my_cnode, &mem_cnode, NULL) < 0) { + printk(KERN_WARNING "tioce_bus_fixup: failed to find " + "closest node with MEM to TIO node %d\n", my_cnode); + mem_cnode = (cnodeid_t)-1; /* use any node */ + } + + controller->node = mem_cnode; + return tioce_common; } diff --git a/include/asm-ia64/sn/addrs.h b/include/asm-ia64/sn/addrs.h index 2c32e4b77b5..1d9efe54166 100644 --- a/include/asm-ia64/sn/addrs.h +++ b/include/asm-ia64/sn/addrs.h @@ -283,5 +283,13 @@ #define REMOTE_HUB_L(n, a) HUB_L(REMOTE_HUB_ADDR((n), (a))) #define REMOTE_HUB_S(n, a, d) HUB_S(REMOTE_HUB_ADDR((n), (a)), (d)) +/* + * Coretalk address breakdown + */ +#define CTALK_NASID_SHFT 40 +#define CTALK_NASID_MASK (0x3FFFULL << CTALK_NASID_SHFT) +#define CTALK_CID_SHFT 38 +#define CTALK_CID_MASK (0x3ULL << CTALK_CID_SHFT) +#define CTALK_NODE_OFFSET 0x3FFFFFFFFF #endif /* _ASM_IA64_SN_ADDRS_H */ diff --git a/include/asm-ia64/sn/tioce.h b/include/asm-ia64/sn/tioce.h index d4c990712ea..893468e1b41 100644 --- a/include/asm-ia64/sn/tioce.h +++ b/include/asm-ia64/sn/tioce.h @@ -11,7 +11,7 @@ /* CE ASIC part & mfgr information */ #define TIOCE_PART_NUM 0xCE00 -#define TIOCE_MFGR_NUM 0x36 +#define TIOCE_SRC_ID 0x01 #define TIOCE_REV_A 0x1 /* CE Virtual PPB Vendor/Device IDs */ @@ -20,7 +20,7 @@ /* CE Host Bridge Vendor/Device IDs */ #define CE_HOST_BRIDGE_VENDOR_ID 0x10a9 -#define CE_HOST_BRIDGE_DEVICE_ID 0x4003 +#define CE_HOST_BRIDGE_DEVICE_ID 0x4001 #define TIOCE_NUM_M40_ATES 4096 @@ -463,6 +463,25 @@ typedef volatile struct tioce { u64 ce_end_of_struct; /* 0x044400 */ } tioce_t; +/* ce_lsiX_gb_cfg1 register bit masks & shifts */ +#define CE_LSI_GB_CFG1_RXL0S_THS_SHFT 0 +#define CE_LSI_GB_CFG1_RXL0S_THS_MASK (0xffULL << 0) +#define CE_LSI_GB_CFG1_RXL0S_SMP_SHFT 8 +#define CE_LSI_GB_CFG1_RXL0S_SMP_MASK (0xfULL << 8); +#define CE_LSI_GB_CFG1_RXL0S_ADJ_SHFT 12 +#define CE_LSI_GB_CFG1_RXL0S_ADJ_MASK (0x7ULL << 12) +#define CE_LSI_GB_CFG1_RXL0S_FLT_SHFT 15 +#define CE_LSI_GB_CFG1_RXL0S_FLT_MASK (0x1ULL << 15) +#define CE_LSI_GB_CFG1_LPBK_SEL_SHFT 16 +#define CE_LSI_GB_CFG1_LPBK_SEL_MASK (0x3ULL << 16) +#define CE_LSI_GB_CFG1_LPBK_EN_SHFT 18 +#define CE_LSI_GB_CFG1_LPBK_EN_MASK (0x1ULL << 18) +#define CE_LSI_GB_CFG1_RVRS_LB_SHFT 19 +#define CE_LSI_GB_CFG1_RVRS_LB_MASK (0x1ULL << 19) +#define CE_LSI_GB_CFG1_RVRS_CLK_SHFT 20 +#define CE_LSI_GB_CFG1_RVRS_CLK_MASK (0x3ULL << 20) +#define CE_LSI_GB_CFG1_SLF_TS_SHFT 24 +#define CE_LSI_GB_CFG1_SLF_TS_MASK (0xfULL << 24) /* ce_adm_int_mask/ce_adm_int_status register bit defines */ #define CE_ADM_INT_CE_ERROR_SHFT 0 @@ -592,6 +611,11 @@ typedef volatile struct tioce { #define CE_URE_RD_MRG_ENABLE (0x1ULL << 0) #define CE_URE_WRT_MRG_ENABLE1 (0x1ULL << 4) #define CE_URE_WRT_MRG_ENABLE2 (0x1ULL << 5) +#define CE_URE_WRT_MRG_TIMER_SHFT 12 +#define CE_URE_WRT_MRG_TIMER_MASK (0x7FFULL << CE_URE_WRT_MRG_TIMER_SHFT) +#define CE_URE_WRT_MRG_TIMER(x) (((u64)(x) << \ + CE_URE_WRT_MRG_TIMER_SHFT) & \ + CE_URE_WRT_MRG_TIMER_MASK) #define CE_URE_RSPQ_BYPASS_DISABLE (0x1ULL << 24) #define CE_URE_UPS_DAT1_PAR_DISABLE (0x1ULL << 32) #define CE_URE_UPS_HDR1_PAR_DISABLE (0x1ULL << 33) @@ -653,8 +677,12 @@ typedef volatile struct tioce { #define CE_URE_SI (0x1ULL << 0) #define CE_URE_ELAL_SHFT 4 #define CE_URE_ELAL_MASK (0x7ULL << CE_URE_ELAL_SHFT) +#define CE_URE_ELAL_SET(n) (((u64)(n) << CE_URE_ELAL_SHFT) & \ + CE_URE_ELAL_MASK) #define CE_URE_ELAL1_SHFT 8 #define CE_URE_ELAL1_MASK (0x7ULL << CE_URE_ELAL1_SHFT) +#define CE_URE_ELAL1_SET(n) (((u64)(n) << CE_URE_ELAL1_SHFT) & \ + CE_URE_ELAL1_MASK) #define CE_URE_SCC (0x1ULL << 12) #define CE_URE_PN1_SHFT 16 #define CE_URE_PN1_MASK (0xFFULL << CE_URE_PN1_SHFT) @@ -675,8 +703,12 @@ typedef volatile struct tioce { #define CE_URE_HPC (0x1ULL << 6) #define CE_URE_SPLV_SHFT 7 #define CE_URE_SPLV_MASK (0xFFULL << CE_URE_SPLV_SHFT) +#define CE_URE_SPLV_SET(n) (((u64)(n) << CE_URE_SPLV_SHFT) & \ + CE_URE_SPLV_MASK) #define CE_URE_SPLS_SHFT 15 #define CE_URE_SPLS_MASK (0x3ULL << CE_URE_SPLS_SHFT) +#define CE_URE_SPLS_SET(n) (((u64)(n) << CE_URE_SPLS_SHFT) & \ + CE_URE_SPLS_MASK) #define CE_URE_PSN1_SHFT 19 #define CE_URE_PSN1_MASK (0x1FFFULL << CE_URE_PSN1_SHFT) #define CE_URE_PSN2_SHFT 32 -- cgit v1.2.3-70-g09d2 From e08e6c521355cd33e647b2f739885bc3050eead6 Mon Sep 17 00:00:00 2001 From: Brent Casavant Date: Thu, 26 Jan 2006 15:55:52 -0800 Subject: [IA64] hooks to wait for mmio writes to drain when migrating processes On SN2, MMIO writes which are issued from separate processors are not guaranteed to arrive in any particular order at the IO hardware. When performing such writes from the kernel this is not a problem, as a kernel thread will not migrate to another CPU during execution, and mmiowb() calls can guarantee write ordering when control of the IO resource is allowed to move between threads. However, when MMIO writes can be performed from user space (e.g. DRM) there are no such guarantees and mechanisms, as the process may context-switch at any time, and may migrate to a different CPU as part of the switch. For such programs/hardware to operate correctly, it is required that the MMIO writes from the old CPU be accepted by the IO hardware before subsequent writes from the new CPU can be issued. The following patch implements this behavior on SN2 by waiting for a Shub register to indicate that these writes have been accepted. This is placed in the context switch-in path, and only performs the wait when the newly scheduled task changes CPUs. Signed-off-by: Prarit Bhargava Signed-off-by: Brent Casavant --- arch/ia64/sn/kernel/setup.c | 6 ++++-- arch/ia64/sn/kernel/sn2/sn2_smp.c | 23 ++++++++++++++++++++++- include/asm-ia64/machvec.h | 13 +++++++++++++ include/asm-ia64/machvec_sn2.h | 4 +++- include/asm-ia64/processor.h | 3 ++- include/asm-ia64/system.h | 7 +++++++ include/asm-ia64/thread_info.h | 1 + 7 files changed, 52 insertions(+), 5 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/sn/kernel/setup.c b/arch/ia64/sn/kernel/setup.c index e510dce9971..f1c1338b10b 100644 --- a/arch/ia64/sn/kernel/setup.c +++ b/arch/ia64/sn/kernel/setup.c @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (C) 1999,2001-2005 Silicon Graphics, Inc. All rights reserved. + * Copyright (C) 1999,2001-2006 Silicon Graphics, Inc. All rights reserved. */ #include @@ -496,6 +496,7 @@ void __init sn_setup(char **cmdline_p) * for sn. */ pm_power_off = ia64_sn_power_down; + current->thread.flags |= IA64_THREAD_MIGRATION; } /** @@ -654,7 +655,8 @@ void __init sn_cpu_init(void) SH2_PIO_WRITE_STATUS_1, SH2_PIO_WRITE_STATUS_3}; u64 *pio; pio = is_shub1() ? pio1 : pio2; - pda->pio_write_status_addr = (volatile unsigned long *) LOCAL_MMR_ADDR(pio[slice]); + pda->pio_write_status_addr = + (volatile unsigned long *)GLOBAL_MMR_ADDR(nasid, pio[slice]); pda->pio_write_status_val = is_shub1() ? SH_PIO_WRITE_STATUS_PENDING_WRITE_COUNT_MASK : 0; } diff --git a/arch/ia64/sn/kernel/sn2/sn2_smp.c b/arch/ia64/sn/kernel/sn2/sn2_smp.c index 471bbaa65d1..1b33fd5e4e3 100644 --- a/arch/ia64/sn/kernel/sn2/sn2_smp.c +++ b/arch/ia64/sn/kernel/sn2/sn2_smp.c @@ -5,7 +5,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (C) 2000-2005 Silicon Graphics, Inc. All rights reserved. + * Copyright (C) 2000-2006 Silicon Graphics, Inc. All rights reserved. */ #include @@ -169,6 +169,27 @@ static inline unsigned long wait_piowc(void) return ws; } +/** + * sn_migrate - SN-specific task migration actions + * @task: Task being migrated to new CPU + * + * 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 + * from the previous CPU's Shub before execution resumes on the new CPU. + */ +void sn_migrate(struct task_struct *task) +{ + pda_t *last_pda = pdacpu(task_thread_info(task)->last_cpu); + volatile unsigned long *adr = last_pda->pio_write_status_addr; + unsigned long val = last_pda->pio_write_status_val; + + /* Drain PIO writes from old CPU's Shub */ + while (unlikely((*adr & SH_PIO_WRITE_STATUS_PENDING_WRITE_COUNT_MASK) + != val)) + cpu_relax(); +} + void sn_tlb_migrate_finish(struct mm_struct *mm) { if (mm == current->mm) diff --git a/include/asm-ia64/machvec.h b/include/asm-ia64/machvec.h index ca5ea994d68..c3e4ed8a3e1 100644 --- a/include/asm-ia64/machvec.h +++ b/include/asm-ia64/machvec.h @@ -20,6 +20,7 @@ struct scatterlist; struct page; struct mm_struct; struct pci_bus; +struct task_struct; typedef void ia64_mv_setup_t (char **); typedef void ia64_mv_cpu_init_t (void); @@ -34,6 +35,7 @@ typedef int ia64_mv_pci_legacy_read_t (struct pci_bus *, u16 port, u32 *val, u8 size); typedef int ia64_mv_pci_legacy_write_t (struct pci_bus *, u16 port, u32 val, u8 size); +typedef void ia64_mv_migrate_t(struct task_struct * task); /* DMA-mapping interface: */ typedef void ia64_mv_dma_init (void); @@ -85,6 +87,11 @@ machvec_noop_mm (struct mm_struct *mm) { } +static inline void +machvec_noop_task (struct task_struct *task) +{ +} + extern void machvec_setup (char **); extern void machvec_timer_interrupt (int, void *, struct pt_regs *); extern void machvec_dma_sync_single (struct device *, dma_addr_t, size_t, int); @@ -146,6 +153,7 @@ extern void machvec_tlb_migrate_finish (struct mm_struct *); # define platform_readw_relaxed ia64_mv.readw_relaxed # define platform_readl_relaxed ia64_mv.readl_relaxed # define platform_readq_relaxed ia64_mv.readq_relaxed +# define platform_migrate ia64_mv.migrate # endif /* __attribute__((__aligned__(16))) is required to make size of the @@ -194,6 +202,7 @@ struct ia64_machine_vector { ia64_mv_readw_relaxed_t *readw_relaxed; ia64_mv_readl_relaxed_t *readl_relaxed; ia64_mv_readq_relaxed_t *readq_relaxed; + ia64_mv_migrate_t *migrate; } __attribute__((__aligned__(16))); /* align attrib? see above comment */ #define MACHVEC_INIT(name) \ @@ -238,6 +247,7 @@ struct ia64_machine_vector { platform_readw_relaxed, \ platform_readl_relaxed, \ platform_readq_relaxed, \ + platform_migrate, \ } extern struct ia64_machine_vector ia64_mv; @@ -386,5 +396,8 @@ extern ia64_mv_dma_supported swiotlb_dma_supported; #ifndef platform_readq_relaxed # define platform_readq_relaxed __ia64_readq_relaxed #endif +#ifndef platform_migrate +# define platform_migrate machvec_noop_task +#endif #endif /* _ASM_IA64_MACHVEC_H */ diff --git a/include/asm-ia64/machvec_sn2.h b/include/asm-ia64/machvec_sn2.h index e1b6cd63f49..6f0021bb387 100644 --- a/include/asm-ia64/machvec_sn2.h +++ b/include/asm-ia64/machvec_sn2.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2002-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2002-2003,2006 Silicon Graphics, Inc. All Rights Reserved. * * This program is free software; you can redistribute it and/or modify it * under the terms of version 2 of the GNU General Public License @@ -71,6 +71,7 @@ extern ia64_mv_dma_sync_single_for_device sn_dma_sync_single_for_device; extern ia64_mv_dma_sync_sg_for_device sn_dma_sync_sg_for_device; extern ia64_mv_dma_mapping_error sn_dma_mapping_error; extern ia64_mv_dma_supported sn_dma_supported; +extern ia64_mv_migrate_t sn_migrate; /* * This stuff has dual use! @@ -120,6 +121,7 @@ extern ia64_mv_dma_supported sn_dma_supported; #define platform_dma_sync_sg_for_device sn_dma_sync_sg_for_device #define platform_dma_mapping_error sn_dma_mapping_error #define platform_dma_supported sn_dma_supported +#define platform_migrate sn_migrate #include diff --git a/include/asm-ia64/processor.h b/include/asm-ia64/processor.h index 09b99029ac1..29d5574d437 100644 --- a/include/asm-ia64/processor.h +++ b/include/asm-ia64/processor.h @@ -50,7 +50,8 @@ #define IA64_THREAD_PM_VALID (__IA64_UL(1) << 2) /* performance registers valid? */ #define IA64_THREAD_UAC_NOPRINT (__IA64_UL(1) << 3) /* don't log unaligned accesses */ #define IA64_THREAD_UAC_SIGBUS (__IA64_UL(1) << 4) /* generate SIGBUS on unaligned acc. */ - /* bit 5 is currently unused */ +#define IA64_THREAD_MIGRATION (__IA64_UL(1) << 5) /* require migration + sync at ctx sw */ #define IA64_THREAD_FPEMU_NOPRINT (__IA64_UL(1) << 6) /* don't log any fpswa faults */ #define IA64_THREAD_FPEMU_SIGFPE (__IA64_UL(1) << 7) /* send a SIGFPE for fpswa faults */ diff --git a/include/asm-ia64/system.h b/include/asm-ia64/system.h index 80c5a234e25..99b6f307e94 100644 --- a/include/asm-ia64/system.h +++ b/include/asm-ia64/system.h @@ -244,6 +244,13 @@ extern void ia64_load_extra (struct task_struct *task); __ia64_save_fpu((prev)->thread.fph); \ } \ __switch_to(prev, next, last); \ + /* "next" in old context is "current" in new context */ \ + if (unlikely((current->thread.flags & IA64_THREAD_MIGRATION) && \ + (task_cpu(current) != \ + task_thread_info(current)->last_cpu))) { \ + platform_migrate(current); \ + task_thread_info(current)->last_cpu = task_cpu(current); \ + } \ } while (0) #else # define switch_to(prev,next,last) __switch_to(prev, next, last) diff --git a/include/asm-ia64/thread_info.h b/include/asm-ia64/thread_info.h index 1d6518fe1f0..81641a6905d 100644 --- a/include/asm-ia64/thread_info.h +++ b/include/asm-ia64/thread_info.h @@ -26,6 +26,7 @@ struct thread_info { struct exec_domain *exec_domain;/* execution domain */ __u32 flags; /* thread_info flags (see TIF_*) */ __u32 cpu; /* current CPU */ + __u32 last_cpu; /* Last CPU thread ran on */ mm_segment_t addr_limit; /* user-level address space limit */ int preempt_count; /* 0=premptable, <0=BUG; will also serve as bh-counter */ struct restart_block restart_block; -- cgit v1.2.3-70-g09d2 From dcc1dd2366a7c355fd8b6543c52685b864a2044f Mon Sep 17 00:00:00 2001 From: Jack Steiner Date: Tue, 7 Feb 2006 09:24:14 -0800 Subject: [IA64-SGI] - Eliminate SN pio_phys_xxx macros. Move to assembly Rewrite the SN pio_phys_xxx macros in assembly language. This avoids issues with the Intel icc compiler. Function call overhead is not an issue - the functions reference PIOs and take 100's nsec to complete. In addition, the functions should likely be in assembly language anyway - they reference memory using physical addressing mode. One function executes with psr.ic disabled. Signed-off-by: Jack Steiner Signed-off-by: Tony Luck --- arch/ia64/sn/kernel/Makefile | 3 +- arch/ia64/sn/kernel/pio_phys.S | 71 ++++++++++++++++++++++++++++++++++++++++++ include/asm-ia64/sn/rw_mmr.h | 56 +++------------------------------ 3 files changed, 78 insertions(+), 52 deletions(-) create mode 100644 arch/ia64/sn/kernel/pio_phys.S (limited to 'arch/ia64') diff --git a/arch/ia64/sn/kernel/Makefile b/arch/ia64/sn/kernel/Makefile index 3e9b4eea741..ab9c48c8801 100644 --- a/arch/ia64/sn/kernel/Makefile +++ b/arch/ia64/sn/kernel/Makefile @@ -10,7 +10,8 @@ CPPFLAGS += -I$(srctree)/arch/ia64/sn/include obj-y += setup.o bte.o bte_error.o irq.o mca.o idle.o \ - huberror.o io_init.o iomv.o klconflib.o sn2/ + huberror.o io_init.o iomv.o klconflib.o pio_phys.o \ + sn2/ obj-$(CONFIG_IA64_GENERIC) += machvec.o obj-$(CONFIG_SGI_TIOCX) += tiocx.o obj-$(CONFIG_IA64_SGI_SN_XP) += xp.o diff --git a/arch/ia64/sn/kernel/pio_phys.S b/arch/ia64/sn/kernel/pio_phys.S new file mode 100644 index 00000000000..3c7d48d6ecb --- /dev/null +++ b/arch/ia64/sn/kernel/pio_phys.S @@ -0,0 +1,71 @@ +/* + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (C) 2000-2005 Silicon Graphics, Inc. All rights reserved. + * + * This file contains macros used to access MMR registers via + * uncached physical addresses. + * pio_phys_read_mmr - read an MMR + * pio_phys_write_mmr - write an MMR + * pio_atomic_phys_write_mmrs - atomically write 1 or 2 MMRs with psr.ic=0 + * Second MMR will be skipped if address is NULL + * + * Addresses passed to these routines should be uncached physical addresses + * ie., 0x80000.... + */ + + + +#include +#include + +GLOBAL_ENTRY(pio_phys_read_mmr) + .prologue + .regstk 1,0,0,0 + .body + mov r2=psr + rsm psr.i | psr.dt + ;; + srlz.d + ld8.acq r8=[r32] + ;; + mov psr.l=r2;; + srlz.d + br.ret.sptk.many rp +END(pio_phys_read_mmr) + +GLOBAL_ENTRY(pio_phys_write_mmr) + .prologue + .regstk 2,0,0,0 + .body + mov r2=psr + rsm psr.i | psr.dt + ;; + srlz.d + st8.rel [r32]=r33 + ;; + mov psr.l=r2;; + srlz.d + br.ret.sptk.many rp +END(pio_phys_write_mmr) + +GLOBAL_ENTRY(pio_atomic_phys_write_mmrs) + .prologue + .regstk 4,0,0,0 + .body + mov r2=psr + cmp.ne p9,p0=r34,r0; + rsm psr.i | psr.dt | psr.ic + ;; + srlz.d + st8.rel [r32]=r33 +(p9) st8.rel [r34]=r35 + ;; + mov psr.l=r2;; + srlz.d + br.ret.sptk.many rp +END(pio_atomic_phys_write_mmrs) + + diff --git a/include/asm-ia64/sn/rw_mmr.h b/include/asm-ia64/sn/rw_mmr.h index f40fd1a5510..2d78f4c5a45 100644 --- a/include/asm-ia64/sn/rw_mmr.h +++ b/include/asm-ia64/sn/rw_mmr.h @@ -3,15 +3,14 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (C) 2002-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (C) 2002-2006 Silicon Graphics, Inc. All Rights Reserved. */ #ifndef _ASM_IA64_SN_RW_MMR_H #define _ASM_IA64_SN_RW_MMR_H /* - * This file contains macros used to access MMR registers via - * uncached physical addresses. + * This file that access MMRs via uncached physical addresses. * pio_phys_read_mmr - read an MMR * pio_phys_write_mmr - write an MMR * pio_atomic_phys_write_mmrs - atomically write 1 or 2 MMRs with psr.ic=0 @@ -22,53 +21,8 @@ */ -extern inline long -pio_phys_read_mmr(volatile long *mmr) -{ - long val; - asm volatile - ("mov r2=psr;;" - "rsm psr.i | psr.dt;;" - "srlz.i;;" - "ld8.acq %0=[%1];;" - "mov psr.l=r2;;" - "srlz.i;;" - : "=r"(val) - : "r"(mmr) - : "r2"); - return val; -} - - - -extern inline void -pio_phys_write_mmr(volatile long *mmr, long val) -{ - asm volatile - ("mov r2=psr;;" - "rsm psr.i | psr.dt;;" - "srlz.i;;" - "st8.rel [%0]=%1;;" - "mov psr.l=r2;;" - "srlz.i;;" - :: "r"(mmr), "r"(val) - : "r2", "memory"); -} - -extern inline void -pio_atomic_phys_write_mmrs(volatile long *mmr1, long val1, volatile long *mmr2, long val2) -{ - asm volatile - ("mov r2=psr;;" - "rsm psr.i | psr.dt | psr.ic;;" - "cmp.ne p9,p0=%2,r0;" - "srlz.i;;" - "st8.rel [%0]=%1;" - "(p9) st8.rel [%2]=%3;;" - "mov psr.l=r2;;" - "srlz.i;;" - :: "r"(mmr1), "r"(val1), "r"(mmr2), "r"(val2) - : "p9", "r2", "memory"); -} +extern long pio_phys_read_mmr(volatile long *mmr); +extern void pio_phys_write_mmr(volatile long *mmr, long val); +extern void pio_atomic_phys_write_mmrs(volatile long *mmr1, long val1, volatile long *mmr2, long val2); #endif /* _ASM_IA64_SN_RW_MMR_H */ -- cgit v1.2.3-70-g09d2 From 9336b0836bf789136b51caf9ddd49dcbf1726cf4 Mon Sep 17 00:00:00 2001 From: Keith Owens Date: Wed, 8 Feb 2006 13:40:59 +1100 Subject: [IA64] MCA: print messages in MCA handler Print a message identifying the monarch MCA handler. Print a summary of the status of the slave MCA cpus. Signed-off-by: Keith Owens Signed-off-by: Tony Luck --- arch/ia64/kernel/mca.c | 25 +++++++++++++++++++++++-- 1 file changed, 23 insertions(+), 2 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c index ee7eec9ee57..23d54413c00 100644 --- a/arch/ia64/kernel/mca.c +++ b/arch/ia64/kernel/mca.c @@ -908,7 +908,7 @@ no_mod: static void ia64_wait_for_slaves(int monarch) { - int c, wait = 0; + int c, wait = 0, missing = 0; for_each_online_cpu(c) { if (c == monarch) continue; @@ -919,15 +919,32 @@ ia64_wait_for_slaves(int monarch) } } if (!wait) - return; + goto all_in; for_each_online_cpu(c) { if (c == monarch) continue; if (ia64_mc_info.imi_rendez_checkin[c] == IA64_MCA_RENDEZ_CHECKIN_NOTDONE) { udelay(5*1000000); /* wait 5 seconds for slaves (arbitrary) */ + if (ia64_mc_info.imi_rendez_checkin[c] == IA64_MCA_RENDEZ_CHECKIN_NOTDONE) + missing = 1; break; } } + if (!missing) + goto all_in; + printk(KERN_INFO "OS MCA slave did not rendezvous on cpu"); + for_each_online_cpu(c) { + if (c == monarch) + continue; + if (ia64_mc_info.imi_rendez_checkin[c] == IA64_MCA_RENDEZ_CHECKIN_NOTDONE) + printk(" %d", c); + } + printk("\n"); + return; + +all_in: + printk(KERN_INFO "All OS MCA slaves have reached rendezvous\n"); + return; } /* @@ -953,6 +970,10 @@ ia64_mca_handler(struct pt_regs *regs, struct switch_stack *sw, task_t *previous_current; oops_in_progress = 1; /* FIXME: make printk NMI/MCA/INIT safe */ + console_loglevel = 15; /* make sure printks make it to console */ + printk(KERN_INFO "Entered OS MCA handler. PSP=%lx cpu=%d monarch=%ld\n", + sos->proc_state_param, cpu, sos->monarch); + previous_current = ia64_mca_modify_original_stack(regs, sw, sos, "MCA"); monarch_cpu = cpu; if (notify_die(DIE_MCA_MONARCH_ENTER, "MCA", regs, 0, 0, 0) -- cgit v1.2.3-70-g09d2 From e9ac054daaecf8a11f2113b60f2b6ce381c4f131 Mon Sep 17 00:00:00 2001 From: Keith Owens Date: Wed, 8 Feb 2006 13:41:04 +1100 Subject: [IA64] MCA: update MCA comm field for user space tasks Update the comm field on the MCA handler for user tasks as well as for verified kernel tasks. This helps to identify the task that was running when the MCA occurred. Signed-off-by: Keith Owens Signed-off-by: Tony Luck --- arch/ia64/kernel/mca.c | 52 +++++++++++++++++++++++++++++++------------------- 1 file changed, 32 insertions(+), 20 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c index 23d54413c00..4f8464ead63 100644 --- a/arch/ia64/kernel/mca.c +++ b/arch/ia64/kernel/mca.c @@ -630,6 +630,32 @@ copy_reg(const u64 *fr, u64 fnat, u64 *tr, u64 *tnat) *tnat |= (nat << tslot); } +/* Change the comm field on the MCA/INT task to include the pid that + * was interrupted, it makes for easier debugging. If that pid was 0 + * (swapper or nested MCA/INIT) then use the start of the previous comm + * field suffixed with its cpu. + */ + +static void +ia64_mca_modify_comm(const task_t *previous_current) +{ + char *p, comm[sizeof(current->comm)]; + if (previous_current->pid) + snprintf(comm, sizeof(comm), "%s %d", + current->comm, previous_current->pid); + else { + int l; + if ((p = strchr(previous_current->comm, ' '))) + l = p - previous_current->comm; + else + l = strlen(previous_current->comm); + snprintf(comm, sizeof(comm), "%s %*s %d", + current->comm, l, previous_current->comm, + task_thread_info(previous_current)->cpu); + } + memcpy(current->comm, comm, sizeof(current->comm)); +} + /* On entry to this routine, we are running on the per cpu stack, see * mca_asm.h. The original stack has not been touched by this event. Some of * the original stack's registers will be in the RBS on this stack. This stack @@ -648,7 +674,7 @@ ia64_mca_modify_original_stack(struct pt_regs *regs, struct ia64_sal_os_state *sos, const char *type) { - char *p, comm[sizeof(current->comm)]; + char *p; ia64_va va; extern char ia64_leave_kernel[]; /* Need asm address, not function descriptor */ const pal_min_state_area_t *ms = sos->pal_min_state; @@ -721,6 +747,10 @@ ia64_mca_modify_original_stack(struct pt_regs *regs, /* Verify the previous stack state before we change it */ if (user_mode(regs)) { msg = "occurred in user space"; + /* previous_current is guaranteed to be valid when the task was + * in user space, so ... + */ + ia64_mca_modify_comm(previous_current); goto no_mod; } if (r13 != sos->prev_IA64_KR_CURRENT) { @@ -750,25 +780,7 @@ ia64_mca_modify_original_stack(struct pt_regs *regs, goto no_mod; } - /* Change the comm field on the MCA/INT task to include the pid that - * was interrupted, it makes for easier debugging. If that pid was 0 - * (swapper or nested MCA/INIT) then use the start of the previous comm - * field suffixed with its cpu. - */ - if (previous_current->pid) - snprintf(comm, sizeof(comm), "%s %d", - current->comm, previous_current->pid); - else { - int l; - if ((p = strchr(previous_current->comm, ' '))) - l = p - previous_current->comm; - else - l = strlen(previous_current->comm); - snprintf(comm, sizeof(comm), "%s %*s %d", - current->comm, l, previous_current->comm, - task_thread_info(previous_current)->cpu); - } - memcpy(current->comm, comm, sizeof(current->comm)); + ia64_mca_modify_comm(previous_current); /* Make the original task look blocked. First stack a struct pt_regs, * describing the state at the time of interrupt. mca_asm.S built a -- cgit v1.2.3-70-g09d2 From 2730c9295a9a797a22b800d3befd6a64fdc56b02 Mon Sep 17 00:00:00 2001 From: Keith Owens Date: Wed, 8 Feb 2006 13:41:10 +1100 Subject: [IA64] MCA: remove obsolete ifdef No platform in the community tree uses PLATFORM_MCA_HANDLERS, remove the references. Signed-off-by: Keith Owens Signed-off-by: Tony Luck --- arch/ia64/kernel/mca.c | 7 ------- 1 file changed, 7 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c index 4f8464ead63..909fed2c249 100644 --- a/arch/ia64/kernel/mca.c +++ b/arch/ia64/kernel/mca.c @@ -281,11 +281,6 @@ ia64_mca_log_sal_error_record(int sal_info_type) ia64_sal_clear_state_info(sal_info_type); } -/* - * platform dependent error handling - */ -#ifndef PLATFORM_MCA_HANDLERS - #ifdef CONFIG_ACPI int cpe_vector = -1; @@ -377,8 +372,6 @@ ia64_mca_register_cpev (int cpev) } #endif /* CONFIG_ACPI */ -#endif /* PLATFORM_MCA_HANDLERS */ - /* * ia64_mca_cmc_vector_setup * -- cgit v1.2.3-70-g09d2 From 7aa6ba41362a7f888ad11fdcfe51ca8d92226cd3 Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Fri, 17 Feb 2006 05:18:43 -0500 Subject: [IA64-SGI] SN2-XP reduce kmalloc wrapper inlining Take advantage of kzalloc() as well as reduce the size of code generated for the error returns in xpc_setup_infrastructure(). Signed-off-by: Jes Sorensen Acked-by: Dean Nelson Signed-off-by: Tony Luck --- arch/ia64/sn/kernel/xpc_channel.c | 102 +++++++++++++++++++++--------------- arch/ia64/sn/kernel/xpc_main.c | 1 - arch/ia64/sn/kernel/xpc_partition.c | 28 +++++++++- include/asm-ia64/sn/xpc.h | 22 -------- 4 files changed, 85 insertions(+), 68 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/sn/kernel/xpc_channel.c b/arch/ia64/sn/kernel/xpc_channel.c index cdf6856ce08..d0abddd9ffe 100644 --- a/arch/ia64/sn/kernel/xpc_channel.c +++ b/arch/ia64/sn/kernel/xpc_channel.c @@ -21,7 +21,6 @@ #include #include #include -#include #include #include #include @@ -29,6 +28,31 @@ #include +/* + * Guarantee that the kzalloc'd memory is cacheline aligned. + */ +static void * +xpc_kzalloc_cacheline_aligned(size_t size, gfp_t flags, void **base) +{ + /* see if kzalloc will give us cachline aligned memory by default */ + *base = kzalloc(size, flags); + if (*base == NULL) { + return NULL; + } + if ((u64) *base == L1_CACHE_ALIGN((u64) *base)) { + return *base; + } + kfree(*base); + + /* nope, we'll have to do it ourselves */ + *base = kzalloc(size + L1_CACHE_BYTES, flags); + if (*base == NULL) { + return NULL; + } + return (void *) L1_CACHE_ALIGN((u64) *base); +} + + /* * Set up the initial values for the XPartition Communication channels. */ @@ -93,20 +117,19 @@ xpc_setup_infrastructure(struct xpc_partition *part) * Allocate all of the channel structures as a contiguous chunk of * memory. */ - part->channels = kmalloc(sizeof(struct xpc_channel) * XPC_NCHANNELS, + part->channels = kzalloc(sizeof(struct xpc_channel) * XPC_NCHANNELS, GFP_KERNEL); if (part->channels == NULL) { dev_err(xpc_chan, "can't get memory for channels\n"); return xpcNoMemory; } - memset(part->channels, 0, sizeof(struct xpc_channel) * XPC_NCHANNELS); part->nchannels = XPC_NCHANNELS; /* allocate all the required GET/PUT values */ - part->local_GPs = xpc_kmalloc_cacheline_aligned(XPC_GP_SIZE, + part->local_GPs = xpc_kzalloc_cacheline_aligned(XPC_GP_SIZE, GFP_KERNEL, &part->local_GPs_base); if (part->local_GPs == NULL) { kfree(part->channels); @@ -115,55 +138,51 @@ xpc_setup_infrastructure(struct xpc_partition *part) "values\n"); return xpcNoMemory; } - memset(part->local_GPs, 0, XPC_GP_SIZE); - part->remote_GPs = xpc_kmalloc_cacheline_aligned(XPC_GP_SIZE, + part->remote_GPs = xpc_kzalloc_cacheline_aligned(XPC_GP_SIZE, GFP_KERNEL, &part->remote_GPs_base); if (part->remote_GPs == NULL) { - kfree(part->channels); - part->channels = NULL; - kfree(part->local_GPs_base); - part->local_GPs = NULL; dev_err(xpc_chan, "can't get memory for remote get/put " "values\n"); + kfree(part->local_GPs_base); + part->local_GPs = NULL; + kfree(part->channels); + part->channels = NULL; return xpcNoMemory; } - memset(part->remote_GPs, 0, XPC_GP_SIZE); /* allocate all the required open and close args */ - part->local_openclose_args = xpc_kmalloc_cacheline_aligned( + part->local_openclose_args = xpc_kzalloc_cacheline_aligned( XPC_OPENCLOSE_ARGS_SIZE, GFP_KERNEL, &part->local_openclose_args_base); if (part->local_openclose_args == NULL) { - kfree(part->channels); - part->channels = NULL; - kfree(part->local_GPs_base); - part->local_GPs = NULL; + dev_err(xpc_chan, "can't get memory for local connect args\n"); kfree(part->remote_GPs_base); part->remote_GPs = NULL; - dev_err(xpc_chan, "can't get memory for local connect args\n"); + kfree(part->local_GPs_base); + part->local_GPs = NULL; + kfree(part->channels); + part->channels = NULL; return xpcNoMemory; } - memset(part->local_openclose_args, 0, XPC_OPENCLOSE_ARGS_SIZE); - part->remote_openclose_args = xpc_kmalloc_cacheline_aligned( + part->remote_openclose_args = xpc_kzalloc_cacheline_aligned( XPC_OPENCLOSE_ARGS_SIZE, GFP_KERNEL, &part->remote_openclose_args_base); if (part->remote_openclose_args == NULL) { - kfree(part->channels); - part->channels = NULL; - kfree(part->local_GPs_base); - part->local_GPs = NULL; - kfree(part->remote_GPs_base); - part->remote_GPs = NULL; + dev_err(xpc_chan, "can't get memory for remote connect args\n"); kfree(part->local_openclose_args_base); part->local_openclose_args = NULL; - dev_err(xpc_chan, "can't get memory for remote connect args\n"); + kfree(part->remote_GPs_base); + part->remote_GPs = NULL; + kfree(part->local_GPs_base); + part->local_GPs = NULL; + kfree(part->channels); + part->channels = NULL; return xpcNoMemory; } - memset(part->remote_openclose_args, 0, XPC_OPENCLOSE_ARGS_SIZE); xpc_initialize_channels(part, partid); @@ -186,18 +205,18 @@ xpc_setup_infrastructure(struct xpc_partition *part) ret = request_irq(SGI_XPC_NOTIFY, xpc_notify_IRQ_handler, SA_SHIRQ, part->IPI_owner, (void *) (u64) partid); if (ret != 0) { - kfree(part->channels); - part->channels = NULL; - kfree(part->local_GPs_base); - part->local_GPs = NULL; - kfree(part->remote_GPs_base); - part->remote_GPs = NULL; - kfree(part->local_openclose_args_base); - part->local_openclose_args = NULL; - kfree(part->remote_openclose_args_base); - part->remote_openclose_args = NULL; dev_err(xpc_chan, "can't register NOTIFY IRQ handler, " "errno=%d\n", -ret); + kfree(part->remote_openclose_args_base); + part->remote_openclose_args = NULL; + kfree(part->local_openclose_args_base); + part->local_openclose_args = NULL; + kfree(part->remote_GPs_base); + part->remote_GPs = NULL; + kfree(part->local_GPs_base); + part->local_GPs = NULL; + kfree(part->channels); + part->channels = NULL; return xpcLackOfResources; } @@ -446,22 +465,20 @@ xpc_allocate_local_msgqueue(struct xpc_channel *ch) for (nentries = ch->local_nentries; nentries > 0; nentries--) { nbytes = nentries * ch->msg_size; - ch->local_msgqueue = xpc_kmalloc_cacheline_aligned(nbytes, + ch->local_msgqueue = xpc_kzalloc_cacheline_aligned(nbytes, GFP_KERNEL, &ch->local_msgqueue_base); if (ch->local_msgqueue == NULL) { continue; } - memset(ch->local_msgqueue, 0, nbytes); nbytes = nentries * sizeof(struct xpc_notify); - ch->notify_queue = kmalloc(nbytes, GFP_KERNEL); + ch->notify_queue = kzalloc(nbytes, GFP_KERNEL); if (ch->notify_queue == NULL) { kfree(ch->local_msgqueue_base); ch->local_msgqueue = NULL; continue; } - memset(ch->notify_queue, 0, nbytes); spin_lock_irqsave(&ch->lock, irq_flags); if (nentries < ch->local_nentries) { @@ -501,13 +518,12 @@ xpc_allocate_remote_msgqueue(struct xpc_channel *ch) for (nentries = ch->remote_nentries; nentries > 0; nentries--) { nbytes = nentries * ch->msg_size; - ch->remote_msgqueue = xpc_kmalloc_cacheline_aligned(nbytes, + ch->remote_msgqueue = xpc_kzalloc_cacheline_aligned(nbytes, GFP_KERNEL, &ch->remote_msgqueue_base); if (ch->remote_msgqueue == NULL) { continue; } - memset(ch->remote_msgqueue, 0, nbytes); spin_lock_irqsave(&ch->lock, irq_flags); if (nentries < ch->remote_nentries) { diff --git a/arch/ia64/sn/kernel/xpc_main.c b/arch/ia64/sn/kernel/xpc_main.c index 8cbf1643257..99b123a6421 100644 --- a/arch/ia64/sn/kernel/xpc_main.c +++ b/arch/ia64/sn/kernel/xpc_main.c @@ -52,7 +52,6 @@ #include #include #include -#include #include #include #include diff --git a/arch/ia64/sn/kernel/xpc_partition.c b/arch/ia64/sn/kernel/xpc_partition.c index 88a730e6cfd..94211429fd0 100644 --- a/arch/ia64/sn/kernel/xpc_partition.c +++ b/arch/ia64/sn/kernel/xpc_partition.c @@ -80,6 +80,31 @@ char ____cacheline_aligned xpc_remote_copy_buffer[XPC_RP_HEADER_SIZE + XP_NASID_MASK_BYTES]; +/* + * Guarantee that the kmalloc'd memory is cacheline aligned. + */ +static void * +xpc_kmalloc_cacheline_aligned(size_t size, gfp_t flags, void **base) +{ + /* see if kmalloc will give us cachline aligned memory by default */ + *base = kmalloc(size, flags); + if (*base == NULL) { + return NULL; + } + if ((u64) *base == L1_CACHE_ALIGN((u64) *base)) { + return *base; + } + kfree(*base); + + /* nope, we'll have to do it ourselves */ + *base = kmalloc(size + L1_CACHE_BYTES, flags); + if (*base == NULL) { + return NULL; + } + return (void *) L1_CACHE_ALIGN((u64) *base); +} + + /* * Given a nasid, get the physical address of the partition's reserved page * for that nasid. This function returns 0 on any error. @@ -1038,13 +1063,12 @@ xpc_discovery(void) remote_vars = (struct xpc_vars *) remote_rp; - discovered_nasids = kmalloc(sizeof(u64) * xp_nasid_mask_words, + discovered_nasids = kzalloc(sizeof(u64) * xp_nasid_mask_words, GFP_KERNEL); if (discovered_nasids == NULL) { kfree(remote_rp_base); return; } - memset(discovered_nasids, 0, sizeof(u64) * xp_nasid_mask_words); rp = (struct xpc_rsvd_page *) xpc_rsvd_page; diff --git a/include/asm-ia64/sn/xpc.h b/include/asm-ia64/sn/xpc.h index df7f5f4f3cd..aa3b8ace903 100644 --- a/include/asm-ia64/sn/xpc.h +++ b/include/asm-ia64/sn/xpc.h @@ -1227,28 +1227,6 @@ xpc_map_bte_errors(bte_result_t error) -static inline void * -xpc_kmalloc_cacheline_aligned(size_t size, gfp_t flags, void **base) -{ - /* see if kmalloc will give us cachline aligned memory by default */ - *base = kmalloc(size, flags); - if (*base == NULL) { - return NULL; - } - if ((u64) *base == L1_CACHE_ALIGN((u64) *base)) { - return *base; - } - kfree(*base); - - /* nope, we'll have to do it ourselves */ - *base = kmalloc(size + L1_CACHE_BYTES, flags); - if (*base == NULL) { - return NULL; - } - return (void *) L1_CACHE_ALIGN((u64) *base); -} - - /* * Check to see if there is any channel activity to/from the specified * partition. -- cgit v1.2.3-70-g09d2 From 4f1933620f57145212cdbb1ac6ce099eeeb21c5a Mon Sep 17 00:00:00 2001 From: Paul Smith Date: Sun, 5 Mar 2006 17:14:10 -0500 Subject: kbuild: change kbuild to not rely on incorrect GNU make behavior The kbuild system takes advantage of an incorrect behavior in GNU make. Once this behavior is fixed, all files in the kernel rebuild every time, even if nothing has changed. This patch ensures kbuild works with both the incorrect and correct behaviors of GNU make. For more details on the incorrect behavior, see: http://lists.gnu.org/archive/html/bug-make/2006-03/msg00003.html Changes in this patch: - Keep all targets that are to be marked .PHONY in a variable, PHONY. - Add .PHONY: $(PHONY) to mark them properly. - Remove any $(PHONY) files from the $? list when determining whether targets are up-to-date or not. Signed-off-by: Paul Smith Signed-off-by: Sam Ravnborg --- Documentation/DocBook/Makefile | 8 ++++- Makefile | 64 ++++++++++++++++++++----------------- arch/arm/Makefile | 5 ++- arch/arm/boot/Makefile | 5 ++- arch/arm/boot/bootp/Makefile | 5 ++- arch/arm26/Makefile | 7 ++-- arch/arm26/boot/Makefile | 5 ++- arch/i386/Makefile | 4 +-- arch/ia64/Makefile | 5 ++- arch/m32r/Makefile | 5 ++- arch/powerpc/Makefile | 2 +- arch/ppc/Makefile | 2 +- arch/ppc/boot/Makefile | 5 ++- arch/ppc/boot/openfirmware/Makefile | 7 ++-- arch/sh/Makefile | 2 +- arch/um/Makefile | 7 ++-- arch/x86_64/Makefile | 4 +-- scripts/Kbuild.include | 13 +++++--- scripts/Makefile.build | 12 +++++-- scripts/Makefile.clean | 10 ++++-- scripts/Makefile.modinst | 10 ++++-- scripts/Makefile.modpost | 12 +++++-- scripts/kconfig/Makefile | 4 +-- scripts/kconfig/lxdialog/Makefile | 6 ++-- scripts/package/Makefile | 10 +++--- 25 files changed, 144 insertions(+), 75 deletions(-) (limited to 'arch/ia64') diff --git a/Documentation/DocBook/Makefile b/Documentation/DocBook/Makefile index 1c955883cf5..2c6f66dea65 100644 --- a/Documentation/DocBook/Makefile +++ b/Documentation/DocBook/Makefile @@ -28,7 +28,7 @@ PS_METHOD = $(prefer-db2x) ### # The targets that may be used. -.PHONY: xmldocs sgmldocs psdocs pdfdocs htmldocs mandocs installmandocs +PHONY += xmldocs sgmldocs psdocs pdfdocs htmldocs mandocs installmandocs BOOKS := $(addprefix $(obj)/,$(DOCBOOKS)) xmldocs: $(BOOKS) @@ -211,3 +211,9 @@ clean-dirs := $(patsubst %.xml,%,$(DOCBOOKS)) #man put files in man subdir - traverse down subdir- := man/ + + +# Declare the contents of the .PHONY variable as phony. We keep that +# information in a variable se we can use it in if_changed and friends. + +.PHONY: $(PHONY) diff --git a/Makefile b/Makefile index 12c8d714777..a59c1e2d0c2 100644 --- a/Makefile +++ b/Makefile @@ -95,7 +95,7 @@ ifdef O endif # That's our default target when none is given on the command line -.PHONY: _all +PHONY := _all _all: ifneq ($(KBUILD_OUTPUT),) @@ -106,7 +106,7 @@ KBUILD_OUTPUT := $(shell cd $(KBUILD_OUTPUT) && /bin/pwd) $(if $(KBUILD_OUTPUT),, \ $(error output directory "$(saved-output)" does not exist)) -.PHONY: $(MAKECMDGOALS) +PHONY += $(MAKECMDGOALS) $(filter-out _all,$(MAKECMDGOALS)) _all: $(if $(KBUILD_VERBOSE:1=),@)$(MAKE) -C $(KBUILD_OUTPUT) \ @@ -123,7 +123,7 @@ ifeq ($(skip-makefile),) # If building an external module we do not care about the all: rule # but instead _all depend on modules -.PHONY: all +PHONY += all ifeq ($(KBUILD_EXTMOD),) _all: all else @@ -337,14 +337,14 @@ export RCS_TAR_IGNORE := --exclude SCCS --exclude BitKeeper --exclude .svn --exc # Rules shared between *config targets and build targets # Basic helpers built in scripts/ -.PHONY: scripts_basic +PHONY += scripts_basic scripts_basic: $(Q)$(MAKE) $(build)=scripts/basic # To avoid any implicit rule to kick in, define an empty command. scripts/basic/%: scripts_basic ; -.PHONY: outputmakefile +PHONY += outputmakefile # outputmakefile generate a Makefile to be placed in output directory, if # using a seperate output directory. This allows convinient use # of make in output directory @@ -420,7 +420,7 @@ ifeq ($(KBUILD_EXTMOD),) # Additional helpers built in scripts/ # Carefully list dependencies so we do not try to build scripts twice # in parrallel -.PHONY: scripts +PHONY += scripts scripts: scripts_basic include/config/MARKER $(Q)$(MAKE) $(build)=$(@) @@ -720,7 +720,7 @@ $(sort $(vmlinux-init) $(vmlinux-main)) $(vmlinux-lds): $(vmlinux-dirs) ; # make menuconfig etc. # Error messages still appears in the original language -.PHONY: $(vmlinux-dirs) +PHONY += $(vmlinux-dirs) $(vmlinux-dirs): prepare scripts $(Q)$(MAKE) $(build)=$@ @@ -773,10 +773,10 @@ kernelrelease = $(KERNELVERSION)$(localver-full) # version.h and scripts_basic is processed / created. # Listed in dependency order -.PHONY: prepare archprepare prepare0 prepare1 prepare2 prepare3 +PHONY += prepare archprepare prepare0 prepare1 prepare2 prepare3 # prepare-all is deprecated, use prepare as valid replacement -.PHONY: prepare-all +PHONY += prepare-all # prepare3 is used to check if we are building in a separate output directory, # and if so do: @@ -857,7 +857,7 @@ include/linux/version.h: $(srctree)/Makefile .config .kernelrelease FORCE # --------------------------------------------------------------------------- -.PHONY: depend dep +PHONY += depend dep depend dep: @echo '*** Warning: make $@ is unnecessary now.' @@ -872,21 +872,21 @@ all: modules # Build modules -.PHONY: modules +PHONY += modules modules: $(vmlinux-dirs) $(if $(KBUILD_BUILTIN),vmlinux) @echo ' Building modules, stage 2.'; $(Q)$(MAKE) -rR -f $(srctree)/scripts/Makefile.modpost # Target to prepare building external modules -.PHONY: modules_prepare +PHONY += modules_prepare modules_prepare: prepare scripts # Target to install modules -.PHONY: modules_install +PHONY += modules_install modules_install: _modinst_ _modinst_post -.PHONY: _modinst_ +PHONY += _modinst_ _modinst_: @if [ -z "`$(DEPMOD) -V 2>/dev/null | grep module-init-tools`" ]; then \ echo "Warning: you may need to install module-init-tools"; \ @@ -913,7 +913,7 @@ depmod_opts := else depmod_opts := -b $(INSTALL_MOD_PATH) -r endif -.PHONY: _modinst_post +PHONY += _modinst_post _modinst_post: _modinst_ if [ -r System.map -a -x $(DEPMOD) ]; then $(DEPMOD) -ae -F System.map $(depmod_opts) $(KERNELRELEASE); fi @@ -956,7 +956,7 @@ clean: rm-dirs := $(CLEAN_DIRS) clean: rm-files := $(CLEAN_FILES) clean-dirs := $(addprefix _clean_,$(srctree) $(vmlinux-alldirs)) -.PHONY: $(clean-dirs) clean archclean +PHONY += $(clean-dirs) clean archclean $(clean-dirs): $(Q)$(MAKE) $(clean)=$(patsubst _clean_%,%,$@) @@ -974,7 +974,7 @@ mrproper: rm-dirs := $(wildcard $(MRPROPER_DIRS)) mrproper: rm-files := $(wildcard $(MRPROPER_FILES)) mrproper-dirs := $(addprefix _mrproper_,Documentation/DocBook scripts) -.PHONY: $(mrproper-dirs) mrproper archmrproper +PHONY += $(mrproper-dirs) mrproper archmrproper $(mrproper-dirs): $(Q)$(MAKE) $(clean)=$(patsubst _mrproper_%,%,$@) @@ -984,7 +984,7 @@ mrproper: clean archmrproper $(mrproper-dirs) # distclean # -.PHONY: distclean +PHONY += distclean distclean: mrproper @find $(srctree) $(RCS_FIND_IGNORE) \ @@ -1000,7 +1000,7 @@ distclean: mrproper # rpm target kept for backward compatibility package-dir := $(srctree)/scripts/package -.PHONY: %-pkg rpm +PHONY += %-pkg rpm %pkg: FORCE $(Q)$(MAKE) -f $(package-dir)/Makefile $@ @@ -1092,12 +1092,12 @@ else # KBUILD_EXTMOD # We are always building modules KBUILD_MODULES := 1 -.PHONY: crmodverdir +PHONY += crmodverdir crmodverdir: $(Q)rm -rf $(MODVERDIR) $(Q)mkdir -p $(MODVERDIR) -.PHONY: $(objtree)/Module.symvers +PHONY += $(objtree)/Module.symvers $(objtree)/Module.symvers: @test -e $(objtree)/Module.symvers || ( \ echo; \ @@ -1106,7 +1106,7 @@ $(objtree)/Module.symvers: echo ) module-dirs := $(addprefix _module_,$(KBUILD_EXTMOD)) -.PHONY: $(module-dirs) modules +PHONY += $(module-dirs) modules $(module-dirs): crmodverdir $(objtree)/Module.symvers $(Q)$(MAKE) $(build)=$(patsubst _module_%,%,$@) @@ -1114,11 +1114,11 @@ modules: $(module-dirs) @echo ' Building modules, stage 2.'; $(Q)$(MAKE) -rR -f $(srctree)/scripts/Makefile.modpost -.PHONY: modules_install +PHONY += modules_install modules_install: _emodinst_ _emodinst_post -install-dir := $(if $(INSTALL_MOD_DIR),$(INSTALL_MOD_DIR),extra) -.PHONY: _emodinst_ +install-dir := $(if $(INSTALL_MOD_DIR),$(INSTALL_MOD_DIR),extra) +PHONY += _emodinst_ _emodinst_: $(Q)rm -rf $(MODLIB)/$(install-dir) $(Q)mkdir -p $(MODLIB)/$(install-dir) @@ -1133,13 +1133,13 @@ quiet_cmd_depmod = DEPMOD $(KERNELRELEASE) $(KERNELRELEASE); \ fi -.PHONY: _emodinst_post +PHONY += _emodinst_post _emodinst_post: _emodinst_ $(call cmd,depmod) clean-dirs := $(addprefix _clean_,$(KBUILD_EXTMOD)) -.PHONY: $(clean-dirs) clean +PHONY += $(clean-dirs) clean $(clean-dirs): $(Q)$(MAKE) $(clean)=$(patsubst _clean_%,%,$@) @@ -1161,7 +1161,7 @@ help: @echo '' # Dummies... -.PHONY: prepare scripts +PHONY += prepare scripts prepare: ; scripts: ; endif # KBUILD_EXTMOD @@ -1274,7 +1274,7 @@ namespacecheck: endif #ifeq ($(config-targets),1) endif #ifeq ($(mixed-targets),1) -.PHONY: checkstack +PHONY += checkstack checkstack: $(OBJDUMP) -d vmlinux $$(find . -name '*.ko') | \ $(PERL) $(src)/scripts/checkstack.pl $(ARCH) @@ -1357,4 +1357,10 @@ clean := -f $(if $(KBUILD_SRC),$(srctree)/)scripts/Makefile.clean obj endif # skip-makefile +PHONY += FORCE FORCE: + + +# Declare the contents of the .PHONY variable as phony. We keep that +# information in a variable se we can use it in if_changed and friends. +.PHONY: $(PHONY) diff --git a/arch/arm/Makefile b/arch/arm/Makefile index fbfc14a56b9..585d334234f 100644 --- a/arch/arm/Makefile +++ b/arch/arm/Makefile @@ -1,6 +1,9 @@ # # arch/arm/Makefile # +# This file is included by the global makefile so that you can add your own +# architecture-specific flags and dependencies. +# # This file is subject to the terms and conditions of the GNU General Public # License. See the file "COPYING" in the main directory of this archive # for more details. @@ -176,7 +179,7 @@ endif archprepare: maketools -.PHONY: maketools FORCE +PHONY += maketools FORCE maketools: include/linux/version.h include/asm-arm/.arch FORCE $(Q)$(MAKE) $(build)=arch/arm/tools include/asm-arm/mach-types.h diff --git a/arch/arm/boot/Makefile b/arch/arm/boot/Makefile index a174d63395e..ec9c400c7f8 100644 --- a/arch/arm/boot/Makefile +++ b/arch/arm/boot/Makefile @@ -1,6 +1,9 @@ # # arch/arm/boot/Makefile # +# This file is included by the global makefile so that you can add your own +# architecture-specific flags and dependencies. +# # This file is subject to the terms and conditions of the GNU General Public # License. See the file "COPYING" in the main directory of this archive # for more details. @@ -73,7 +76,7 @@ $(obj)/bootpImage: $(obj)/bootp/bootp FORCE $(call if_changed,objcopy) @echo ' Kernel: $@ is ready' -.PHONY: initrd FORCE +PHONY += initrd FORCE initrd: @test "$(INITRD_PHYS)" != "" || \ (echo This machine does not support INITRD; exit -1) diff --git a/arch/arm/boot/bootp/Makefile b/arch/arm/boot/bootp/Makefile index 8e8879b6b3d..c394e305447 100644 --- a/arch/arm/boot/bootp/Makefile +++ b/arch/arm/boot/bootp/Makefile @@ -1,6 +1,9 @@ # # linux/arch/arm/boot/bootp/Makefile # +# This file is included by the global makefile so that you can add your own +# architecture-specific flags and dependencies. +# LDFLAGS_bootp :=-p --no-undefined -X \ --defsym initrd_phys=$(INITRD_PHYS) \ @@ -21,4 +24,4 @@ $(obj)/kernel.o: arch/arm/boot/zImage FORCE $(obj)/initrd.o: $(INITRD) FORCE -.PHONY: $(INITRD) FORCE +PHONY += $(INITRD) FORCE diff --git a/arch/arm26/Makefile b/arch/arm26/Makefile index 844a9e46886..fe91eda98a9 100644 --- a/arch/arm26/Makefile +++ b/arch/arm26/Makefile @@ -1,6 +1,9 @@ # # arch/arm26/Makefile # +# This file is included by the global makefile so that you can add your own +# architecture-specific flags and dependencies. +# # This file is subject to the terms and conditions of the GNU General Public # License. See the file "COPYING" in the main directory of this archive # for more details. @@ -49,9 +52,9 @@ all: zImage boot := arch/arm26/boot -.PHONY: maketools FORCE +PHONY += maketools FORCE maketools: FORCE - + # Convert bzImage to zImage bzImage: vmlinux diff --git a/arch/arm26/boot/Makefile b/arch/arm26/boot/Makefile index b5c2277654d..68acb7b0d47 100644 --- a/arch/arm26/boot/Makefile +++ b/arch/arm26/boot/Makefile @@ -1,6 +1,9 @@ # # arch/arm26/boot/Makefile # +# This file is included by the global makefile so that you can add your own +# architecture-specific flags and dependencies. +# # This file is subject to the terms and conditions of the GNU General Public # License. See the file "COPYING" in the main directory of this archive # for more details. @@ -60,7 +63,7 @@ $(obj)/xipImage: vmlinux FORCE @echo ' Kernel: $@ is ready' endif -.PHONY: initrd +PHONY += initrd initrd: @test "$(INITRD_PHYS)" != "" || \ (echo This machine does not support INITRD; exit -1) diff --git a/arch/i386/Makefile b/arch/i386/Makefile index 36bef6543ac..ff6973a85c8 100644 --- a/arch/i386/Makefile +++ b/arch/i386/Makefile @@ -99,8 +99,8 @@ AFLAGS += $(mflags-y) boot := arch/i386/boot -.PHONY: zImage bzImage compressed zlilo bzlilo \ - zdisk bzdisk fdimage fdimage144 fdimage288 install +PHONY += zImage bzImage compressed zlilo bzlilo \ + zdisk bzdisk fdimage fdimage144 fdimage288 install all: bzImage diff --git a/arch/ia64/Makefile b/arch/ia64/Makefile index f722e1a2594..80ea7506fa1 100644 --- a/arch/ia64/Makefile +++ b/arch/ia64/Makefile @@ -1,6 +1,9 @@ # # ia64/Makefile # +# This file is included by the global makefile so that you can add your own +# architecture-specific flags and dependencies. +# # This file is subject to the terms and conditions of the GNU General Public # License. See the file "COPYING" in the main directory of this archive # for more details. @@ -62,7 +65,7 @@ drivers-$(CONFIG_OPROFILE) += arch/ia64/oprofile/ boot := arch/ia64/hp/sim/boot -.PHONY: boot compressed check +PHONY += boot compressed check all: compressed unwcheck diff --git a/arch/m32r/Makefile b/arch/m32r/Makefile index 983d438b14b..229f66fc8e5 100644 --- a/arch/m32r/Makefile +++ b/arch/m32r/Makefile @@ -1,6 +1,9 @@ # # m32r/Makefile # +# This file is included by the global makefile so that you can add your own +# architecture-specific flags and dependencies. +# LDFLAGS := OBJCOPYFLAGS := -O binary -R .note -R .comment -S @@ -39,7 +42,7 @@ drivers-$(CONFIG_OPROFILE) += arch/m32r/oprofile/ boot := arch/m32r/boot -.PHONY: zImage +PHONY += zImage all: zImage diff --git a/arch/powerpc/Makefile b/arch/powerpc/Makefile index 5500ab55d04..5787d55a20b 100644 --- a/arch/powerpc/Makefile +++ b/arch/powerpc/Makefile @@ -150,7 +150,7 @@ CPPFLAGS_vmlinux.lds := -Upowerpc BOOT_TARGETS = zImage zImage.initrd znetboot znetboot.initrd vmlinux.sm uImage -.PHONY: $(BOOT_TARGETS) +PHONY += $(BOOT_TARGETS) boot := arch/$(ARCH)/boot diff --git a/arch/ppc/Makefile b/arch/ppc/Makefile index 98e940beeb3..9fbdf54ba2b 100644 --- a/arch/ppc/Makefile +++ b/arch/ppc/Makefile @@ -82,7 +82,7 @@ drivers-$(CONFIG_OPROFILE) += arch/powerpc/oprofile/ BOOT_TARGETS = zImage zImage.initrd znetboot znetboot.initrd vmlinux.sm -.PHONY: $(BOOT_TARGETS) +PHONY += $(BOOT_TARGETS) all: uImage zImage diff --git a/arch/ppc/boot/Makefile b/arch/ppc/boot/Makefile index efd8ce515d5..84eec0bef93 100644 --- a/arch/ppc/boot/Makefile +++ b/arch/ppc/boot/Makefile @@ -1,6 +1,9 @@ # # arch/ppc/boot/Makefile # +# This file is included by the global makefile so that you can add your own +# architecture-specific flags and dependencies. +# # This file is subject to the terms and conditions of the GNU General Public # License. See the file "COPYING" in the main directory of this archive # for more details. @@ -25,7 +28,7 @@ subdir- += simple openfirmware hostprogs-y := $(addprefix utils/, addnote mknote hack-coff mkprep mkbugboot mktree) -.PHONY: $(BOOT_TARGETS) $(bootdir-y) +PHONY += $(BOOT_TARGETS) $(bootdir-y) $(BOOT_TARGETS): $(bootdir-y) diff --git a/arch/ppc/boot/openfirmware/Makefile b/arch/ppc/boot/openfirmware/Makefile index 2a411ec2e65..66b73974375 100644 --- a/arch/ppc/boot/openfirmware/Makefile +++ b/arch/ppc/boot/openfirmware/Makefile @@ -1,5 +1,8 @@ # Makefile for making bootable images on various OpenFirmware machines. # +# This file is included by the global makefile so that you can add your own +# architecture-specific flags and dependencies. +# # Paul Mackerras January 1997 # XCOFF bootable images for PowerMacs # Geert Uytterhoeven September 1997 @@ -86,7 +89,7 @@ $(images)/zImage.chrp-rs6k $(images)/zImage.initrd.chrp-rs6k: \ # The targets used on the make command-line -.PHONY: zImage zImage.initrd +PHONY += zImage zImage.initrd zImage: $(images)/zImage.chrp \ $(images)/zImage.chrp-rs6k @echo ' kernel: $@ is ready ($<)' @@ -96,7 +99,7 @@ zImage.initrd: $(images)/zImage.initrd.chrp \ TFTPIMAGE := /tftpboot/zImage -.PHONY: znetboot znetboot.initrd +PHONY += znetboot znetboot.initrd znetboot: $(images)/zImage.chrp cp $(images)/zImage.chrp $(TFTPIMAGE).chrp$(END) @echo ' kernel: $@ is ready ($<)' diff --git a/arch/sh/Makefile b/arch/sh/Makefile index 08c9515c480..c72e17a96ee 100644 --- a/arch/sh/Makefile +++ b/arch/sh/Makefile @@ -172,7 +172,7 @@ include/asm-sh/.mach: $(wildcard include/config/sh/*.h) include/config/MARKER archprepare: maketools include/asm-sh/.cpu include/asm-sh/.mach -.PHONY: maketools FORCE +PHONY += maketools FORCE maketools: include/linux/version.h FORCE $(Q)$(MAKE) $(build)=arch/sh/tools include/asm-sh/machtypes.h diff --git a/arch/um/Makefile b/arch/um/Makefile index c58b657f009..8d14c7a831b 100644 --- a/arch/um/Makefile +++ b/arch/um/Makefile @@ -1,4 +1,7 @@ -# +# +# This file is included by the global makefile so that you can add your own +# architecture-specific flags and dependencies. +# # Copyright (C) 2002 Jeff Dike (jdike@karaya.com) # Licensed under the GPL # @@ -88,7 +91,7 @@ CONFIG_KERNEL_HALF_GIGS ?= 0 SIZE = (($(CONFIG_NEST_LEVEL) + $(CONFIG_KERNEL_HALF_GIGS)) * 0x20000000) -.PHONY: linux +PHONY += linux all: linux diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile index d7fd46479c5..7405dfd6522 100644 --- a/arch/x86_64/Makefile +++ b/arch/x86_64/Makefile @@ -67,8 +67,8 @@ drivers-$(CONFIG_OPROFILE) += arch/x86_64/oprofile/ boot := arch/x86_64/boot -.PHONY: bzImage bzlilo install archmrproper \ - fdimage fdimage144 fdimage288 archclean +PHONY += bzImage bzlilo install archmrproper \ + fdimage fdimage144 fdimage288 archclean #Default target when executing "make" all: bzImage diff --git a/scripts/Kbuild.include b/scripts/Kbuild.include index c3d2e4e068c..59620b1554e 100644 --- a/scripts/Kbuild.include +++ b/scripts/Kbuild.include @@ -116,16 +116,18 @@ make-cmd = $(subst \#,\\\#,$(subst $$,$$$$,$(call escsq,$(cmd_$(1))))) # function to only execute the passed command if necessary # >'< substitution is for echo to work, >$< substitution to preserve $ when reloading .cmd file # note: when using inline perl scripts [perl -e '...$$t=1;...'] in $(cmd_xxx) double $$ your perl vars -# -if_changed = $(if $(strip $? $(call arg-check, $(cmd_$(1)), $(cmd_$@)) ), \ +# +if_changed = $(if $(strip $(filter-out $(PHONY),$?) \ + $(call arg-check, $(cmd_$(1)), $(cmd_$@)) ), \ @set -e; \ $(echo-cmd) $(cmd_$(1)); \ echo 'cmd_$@ := $(make-cmd)' > $(@D)/.$(@F).cmd) # execute the command and also postprocess generated .d dependencies # file -if_changed_dep = $(if $(strip $? $(filter-out FORCE $(wildcard $^),$^)\ - $(call arg-check, $(cmd_$(1)), $(cmd_$@)) ), \ +if_changed_dep = $(if $(strip $(filter-out $(PHONY),$?) \ + $(filter-out FORCE $(wildcard $^),$^) \ + $(call arg-check, $(cmd_$(1)), $(cmd_$@)) ), \ @set -e; \ $(echo-cmd) $(cmd_$(1)); \ scripts/basic/fixdep $(depfile) $@ '$(make-cmd)' > $(@D)/.$(@F).tmp; \ @@ -135,6 +137,7 @@ if_changed_dep = $(if $(strip $? $(filter-out FORCE $(wildcard $^),$^)\ # Usage: $(call if_changed_rule,foo) # will check if $(cmd_foo) changed, or any of the prequisites changed, # and if so will execute $(rule_foo) -if_changed_rule = $(if $(strip $? $(call arg-check, $(cmd_$(1)), $(cmd_$@)) ),\ +if_changed_rule = $(if $(strip $(filter-out $(PHONY),$?) \ + $(call arg-check, $(cmd_$(1)), $(cmd_$@)) ),\ @set -e; \ $(rule_$(1))) diff --git a/scripts/Makefile.build b/scripts/Makefile.build index 6ac96ea92bf..7afe3e76cb5 100644 --- a/scripts/Makefile.build +++ b/scripts/Makefile.build @@ -4,7 +4,7 @@ src := $(obj) -.PHONY: __build +PHONY := __build __build: # Read .config if it exist, otherwise ignore @@ -308,14 +308,14 @@ targets += $(multi-used-y) $(multi-used-m) # Descending # --------------------------------------------------------------------------- -.PHONY: $(subdir-ym) +PHONY += $(subdir-ym) $(subdir-ym): $(Q)$(MAKE) $(build)=$@ # Add FORCE to the prequisites of a target to force it to be always rebuilt. # --------------------------------------------------------------------------- -.PHONY: FORCE +PHONY += FORCE FORCE: @@ -330,3 +330,9 @@ cmd_files := $(wildcard $(foreach f,$(targets),$(dir $(f)).$(notdir $(f)).cmd)) ifneq ($(cmd_files),) include $(cmd_files) endif + + +# Declare the contents of the .PHONY variable as phony. We keep that +# information in a variable se we can use it in if_changed and friends. + +.PHONY: $(PHONY) diff --git a/scripts/Makefile.clean b/scripts/Makefile.clean index 8974ea5fc87..cff33498fa1 100644 --- a/scripts/Makefile.clean +++ b/scripts/Makefile.clean @@ -4,7 +4,7 @@ src := $(obj) -.PHONY: __clean +PHONY := __clean __clean: # Shorthand for $(Q)$(MAKE) scripts/Makefile.clean obj=dir @@ -87,10 +87,16 @@ endif # Descending # --------------------------------------------------------------------------- -.PHONY: $(subdir-ymn) +PHONY += $(subdir-ymn) $(subdir-ymn): $(Q)$(MAKE) $(clean)=$@ # If quiet is set, only print short version of command cmd = @$(if $($(quiet)cmd_$(1)),echo ' $($(quiet)cmd_$(1))' &&) $(cmd_$(1)) + + +# Declare the contents of the .PHONY variable as phony. We keep that +# information in a variable se we can use it in if_changed and friends. + +.PHONY: $(PHONY) diff --git a/scripts/Makefile.modinst b/scripts/Makefile.modinst index 23fd1bdc25c..2686dd5dce8 100644 --- a/scripts/Makefile.modinst +++ b/scripts/Makefile.modinst @@ -2,7 +2,7 @@ # Installing modules # ========================================================================== -.PHONY: __modinst +PHONY := __modinst __modinst: include scripts/Kbuild.include @@ -12,7 +12,7 @@ include scripts/Kbuild.include __modules := $(sort $(shell grep -h '\.ko' /dev/null $(wildcard $(MODVERDIR)/*.mod))) modules := $(patsubst %.o,%.ko,$(wildcard $(__modules:.ko=.o))) -.PHONY: $(modules) +PHONY += $(modules) __modinst: $(modules) @: @@ -27,3 +27,9 @@ modinst_dir = $(if $(KBUILD_EXTMOD),$(ext-mod-dir),kernel/$(@D)) $(modules): $(call cmd,modules_install,$(MODLIB)/$(modinst_dir)) + + +# Declare the contents of the .PHONY variable as phony. We keep that +# information in a variable se we can use it in if_changed and friends. + +.PHONY: $(PHONY) diff --git a/scripts/Makefile.modpost b/scripts/Makefile.modpost index 563e3c5bd8d..0cfbe1cf243 100644 --- a/scripts/Makefile.modpost +++ b/scripts/Makefile.modpost @@ -32,7 +32,7 @@ # Step 4 is solely used to allow module versioning in external modules, # where the CRC of each module is retrieved from the Module.symers file. -.PHONY: _modpost +PHONY := _modpost _modpost: __modpost include .config @@ -60,7 +60,7 @@ quiet_cmd_modpost = MODPOST $(if $(KBUILD_EXTMOD),-o $(modulesymfile)) \ $(filter-out FORCE,$^) -.PHONY: __modpost +PHONY += __modpost __modpost: $(wildcard vmlinux) $(modules:.ko=.o) FORCE $(call cmd,modpost) @@ -97,7 +97,7 @@ targets += $(modules) # Add FORCE to the prequisites of a target to force it to be always rebuilt. # --------------------------------------------------------------------------- -.PHONY: FORCE +PHONY += FORCE FORCE: @@ -112,3 +112,9 @@ cmd_files := $(wildcard $(foreach f,$(targets),$(dir $(f)).$(notdir $(f)).cmd)) ifneq ($(cmd_files),) include $(cmd_files) endif + + +# Declare the contents of the .PHONY variable as phony. We keep that +# information in a variable se we can use it in if_changed and friends. + +.PHONY: $(PHONY) diff --git a/scripts/kconfig/Makefile b/scripts/kconfig/Makefile index 52809450bee..e6499db4c8c 100644 --- a/scripts/kconfig/Makefile +++ b/scripts/kconfig/Makefile @@ -2,7 +2,7 @@ # Kernel configuration targets # These targets are used from top-level makefile -.PHONY: oldconfig xconfig gconfig menuconfig config silentoldconfig update-po-config +PHONY += oldconfig xconfig gconfig menuconfig config silentoldconfig update-po-config xconfig: $(obj)/qconf $< arch/$(ARCH)/Kconfig @@ -42,7 +42,7 @@ update-po-config: $(obj)/kxgettext $(Q)rm -f arch/um/Kconfig_arch $(Q)rm -f scripts/kconfig/linux_*.pot scripts/kconfig/config.pot -.PHONY: randconfig allyesconfig allnoconfig allmodconfig defconfig +PHONY += randconfig allyesconfig allnoconfig allmodconfig defconfig randconfig: $(obj)/conf $< -r arch/$(ARCH)/Kconfig diff --git a/scripts/kconfig/lxdialog/Makefile b/scripts/kconfig/lxdialog/Makefile index bbf4887cff7..a8b02632624 100644 --- a/scripts/kconfig/lxdialog/Makefile +++ b/scripts/kconfig/lxdialog/Makefile @@ -7,10 +7,10 @@ check-lxdialog := $(srctree)/$(src)/check-lxdialog.sh # we really need to do so. (Do not call gcc as part of make mrproper) HOST_EXTRACFLAGS = $(shell $(CONFIG_SHELL) $(check-lxdialog) -ccflags) HOST_LOADLIBES = $(shell $(CONFIG_SHELL) $(check-lxdialog) -ldflags $(HOSTCC)) - -HOST_EXTRACFLAGS += -DLOCALE -.PHONY: dochecklxdialog +HOST_EXTRACFLAGS += -DLOCALE + +PHONY += dochecklxdialog $(obj)/dochecklxdialog: $(Q)$(CONFIG_SHELL) $(check-lxdialog) -check $(HOSTCC) $(HOST_LOADLIBES) diff --git a/scripts/package/Makefile b/scripts/package/Makefile index c201ef001f0..d3038b7643a 100644 --- a/scripts/package/Makefile +++ b/scripts/package/Makefile @@ -32,7 +32,7 @@ MKSPEC := $(srctree)/scripts/package/mkspec PREV := set -e; cd ..; # rpm-pkg -.PHONY: rpm-pkg rpm +PHONY += rpm-pkg rpm $(objtree)/kernel.spec: $(MKSPEC) $(srctree)/Makefile $(CONFIG_SHELL) $(MKSPEC) > $@ @@ -54,10 +54,10 @@ rpm-pkg rpm: $(objtree)/kernel.spec clean-files := $(objtree)/kernel.spec # binrpm-pkg -.PHONY: binrpm-pkg +PHONY += binrpm-pkg $(objtree)/binkernel.spec: $(MKSPEC) $(srctree)/Makefile $(CONFIG_SHELL) $(MKSPEC) prebuilt > $@ - + binrpm-pkg: $(objtree)/binkernel.spec $(MAKE) KBUILD_SRC= set -e; \ @@ -72,7 +72,7 @@ clean-files += $(objtree)/binkernel.spec # Deb target # --------------------------------------------------------------------------- # -.PHONY: deb-pkg +PHONY += deb-pkg deb-pkg: $(MAKE) KBUILD_SRC= $(CONFIG_SHELL) $(srctree)/scripts/package/builddeb @@ -82,7 +82,7 @@ clean-dirs += $(objtree)/debian/ # tarball targets # --------------------------------------------------------------------------- -.PHONY: tar%pkg +PHONY += tar%pkg tar%pkg: $(MAKE) KBUILD_SRC= $(CONFIG_SHELL) $(srctree)/scripts/package/buildtar $@ -- cgit v1.2.3-70-g09d2 From ee436dc46a762f430e37952d375a23d87735f73f Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Fri, 18 Nov 2005 14:43:54 +0000 Subject: [PATCH] Fix IA64 success/failure indication in syscall auditing. Original 2.6.9 patch and explanation from somewhere within HP via bugzilla... ia64 stores a success/failure code in r10, and the return value (normal return, or *positive* errno) in r8. The patch also sets the exit code to negative errno if it's a failure result for consistency with other architectures. Signed-off-by: David Woodhouse --- arch/ia64/kernel/ptrace.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/ptrace.c b/arch/ia64/kernel/ptrace.c index eaed14aac6a..9887c8787e7 100644 --- a/arch/ia64/kernel/ptrace.c +++ b/arch/ia64/kernel/ptrace.c @@ -1656,8 +1656,14 @@ syscall_trace_leave (long arg0, long arg1, long arg2, long arg3, long arg4, long arg5, long arg6, long arg7, struct pt_regs regs) { - if (unlikely(current->audit_context)) - audit_syscall_exit(current, AUDITSC_RESULT(regs.r10), regs.r8); + if (unlikely(current->audit_context)) { + int success = AUDITSC_RESULT(regs.r10); + long result = regs.r8; + + if (success != AUDITSC_SUCCESS) + result = -result; + audit_syscall_exit(current, success, result); + } if (test_thread_flag(TIF_SYSCALL_TRACE) && (current->ptrace & PT_PTRACED)) -- cgit v1.2.3-70-g09d2 From 7835e98b2e3c66dba79cb0ff8ebb90a2fe030c29 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Wed, 22 Mar 2006 00:08:40 -0800 Subject: [PATCH] remove set_page_count() outside mm/ set_page_count usage outside mm/ is limited to setting the refcount to 1. Remove set_page_count from outside mm/, and replace those users with init_page_count() and set_page_refcounted(). This allows more debug checking, and tighter control on how code is allowed to play around with page->_count. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/alpha/mm/init.c | 2 +- arch/arm/mm/init.c | 2 +- arch/arm26/mm/init.c | 2 +- arch/cris/mm/init.c | 2 +- arch/frv/mm/init.c | 6 +++--- arch/h8300/mm/init.c | 4 ++-- arch/i386/mm/init.c | 6 +++--- arch/ia64/mm/init.c | 6 +++--- arch/m32r/mm/init.c | 4 ++-- arch/m68k/mm/init.c | 2 +- arch/m68k/mm/memory.c | 2 +- arch/m68k/mm/motorola.c | 2 +- arch/m68knommu/mm/init.c | 4 ++-- arch/mips/arc/memory.c | 2 +- arch/mips/dec/prom/memory.c | 2 +- arch/mips/mips-boards/generic/memory.c | 2 +- arch/mips/mips-boards/sim/sim_mem.c | 2 +- arch/mips/mm/init.c | 6 +++--- arch/mips/sgi-ip27/ip27-memory.c | 2 +- arch/parisc/mm/init.c | 4 ++-- arch/powerpc/mm/init_32.c | 4 ++-- arch/powerpc/mm/init_64.c | 4 ++-- arch/powerpc/mm/mem.c | 4 ++-- arch/powerpc/platforms/cell/setup.c | 2 +- arch/ppc/mm/init.c | 6 +++--- arch/s390/mm/init.c | 4 ++-- arch/sh/mm/init.c | 4 ++-- arch/sh64/mm/init.c | 4 ++-- arch/sparc/kernel/sun4d_smp.c | 6 +++--- arch/sparc/kernel/sun4m_smp.c | 6 +++--- arch/sparc/mm/init.c | 6 +++--- arch/sparc64/mm/init.c | 4 ++-- arch/um/kernel/mem.c | 4 ++-- arch/x86_64/mm/init.c | 6 +++--- arch/xtensa/mm/init.c | 2 +- drivers/video/acornfb.c | 2 +- include/linux/mm.h | 11 +++++++++-- mm/hugetlb.c | 5 +++-- mm/internal.h | 13 ++++++++++++- mm/page_alloc.c | 14 ++++++-------- 40 files changed, 96 insertions(+), 79 deletions(-) (limited to 'arch/ia64') diff --git a/arch/alpha/mm/init.c b/arch/alpha/mm/init.c index 486d7945583..544ac5dc09e 100644 --- a/arch/alpha/mm/init.c +++ b/arch/alpha/mm/init.c @@ -357,7 +357,7 @@ free_reserved_mem(void *start, void *end) void *__start = start; for (; __start < end; __start += PAGE_SIZE) { ClearPageReserved(virt_to_page(__start)); - set_page_count(virt_to_page(__start), 1); + init_page_count(virt_to_page(__start)); free_page((long)__start); totalram_pages++; } diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c index 8b276ee38ac..b0321e943b7 100644 --- a/arch/arm/mm/init.c +++ b/arch/arm/mm/init.c @@ -531,7 +531,7 @@ static inline void free_area(unsigned long addr, unsigned long end, char *s) for (; addr < end; addr += PAGE_SIZE) { struct page *page = virt_to_page(addr); ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); free_page(addr); totalram_pages++; } diff --git a/arch/arm26/mm/init.c b/arch/arm26/mm/init.c index 1f09a9d0fb8..e3ecaa45374 100644 --- a/arch/arm26/mm/init.c +++ b/arch/arm26/mm/init.c @@ -324,7 +324,7 @@ static inline void free_area(unsigned long addr, unsigned long end, char *s) for (; addr < end; addr += PAGE_SIZE) { struct page *page = virt_to_page(addr); ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); free_page(addr); totalram_pages++; } diff --git a/arch/cris/mm/init.c b/arch/cris/mm/init.c index 31a0018b525..b7842ff213a 100644 --- a/arch/cris/mm/init.c +++ b/arch/cris/mm/init.c @@ -216,7 +216,7 @@ free_initmem(void) addr = (unsigned long)(&__init_begin); for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } diff --git a/arch/frv/mm/init.c b/arch/frv/mm/init.c index 765088ea8a5..8899aa1a4f0 100644 --- a/arch/frv/mm/init.c +++ b/arch/frv/mm/init.c @@ -169,7 +169,7 @@ void __init mem_init(void) struct page *page = &mem_map[pfn]; ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); __free_page(page); totalram_pages++; } @@ -210,7 +210,7 @@ void __init free_initmem(void) /* next to check that the page we free is not a partial page */ for (addr = start; addr < end; addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } @@ -230,7 +230,7 @@ void __init free_initrd_mem(unsigned long start, unsigned long end) int pages = 0; for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; pages++; diff --git a/arch/h8300/mm/init.c b/arch/h8300/mm/init.c index 1e0929ddc8c..09efc4b1f03 100644 --- a/arch/h8300/mm/init.c +++ b/arch/h8300/mm/init.c @@ -196,7 +196,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) int pages = 0; for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; pages++; @@ -219,7 +219,7 @@ free_initmem() /* next to check that the page we free is not a partial page */ for (; addr + PAGE_SIZE < (unsigned long)(&__init_end); addr +=PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c index 2700f01994b..7ba55a6e2db 100644 --- a/arch/i386/mm/init.c +++ b/arch/i386/mm/init.c @@ -270,7 +270,7 @@ static void __init permanent_kmaps_init(pgd_t *pgd_base) static void __meminit free_new_highpage(struct page *page) { - set_page_count(page, 1); + init_page_count(page); __free_page(page); totalhigh_pages++; } @@ -727,7 +727,7 @@ void free_initmem(void) addr = (unsigned long)(&__init_begin); for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); memset((void *)addr, 0xcc, PAGE_SIZE); free_page(addr); totalram_pages++; @@ -766,7 +766,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) printk (KERN_INFO "Freeing initrd memory: %ldk freed\n", (end - start) >> 10); for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; } diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c index b38b6d213c1..08d94e6bfa1 100644 --- a/arch/ia64/mm/init.c +++ b/arch/ia64/mm/init.c @@ -197,7 +197,7 @@ free_initmem (void) eaddr = (unsigned long) ia64_imva(__init_end); while (addr < eaddr) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); ++totalram_pages; addr += PAGE_SIZE; @@ -252,7 +252,7 @@ free_initrd_mem (unsigned long start, unsigned long end) continue; page = virt_to_page(start); ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); free_page(start); ++totalram_pages; } @@ -640,7 +640,7 @@ mem_init (void) void online_page(struct page *page) { ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); __free_page(page); totalram_pages++; num_physpages++; diff --git a/arch/m32r/mm/init.c b/arch/m32r/mm/init.c index 6facf15b04f..c9e7dad860b 100644 --- a/arch/m32r/mm/init.c +++ b/arch/m32r/mm/init.c @@ -226,7 +226,7 @@ void free_initmem(void) addr = (unsigned long)(&__init_begin); for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } @@ -244,7 +244,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) unsigned long p; for (p = start; p < end; p += PAGE_SIZE) { ClearPageReserved(virt_to_page(p)); - set_page_count(virt_to_page(p), 1); + init_page_count(virt_to_page(p)); free_page(p); totalram_pages++; } diff --git a/arch/m68k/mm/init.c b/arch/m68k/mm/init.c index c45beb95594..a190e39c907 100644 --- a/arch/m68k/mm/init.c +++ b/arch/m68k/mm/init.c @@ -137,7 +137,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) int pages = 0; for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; pages++; diff --git a/arch/m68k/mm/memory.c b/arch/m68k/mm/memory.c index 559942ce0e1..d6d582a5abb 100644 --- a/arch/m68k/mm/memory.c +++ b/arch/m68k/mm/memory.c @@ -54,7 +54,7 @@ void __init init_pointer_table(unsigned long ptable) /* unreserve the page so it's possible to free that page */ PD_PAGE(dp)->flags &= ~(1 << PG_reserved); - set_page_count(PD_PAGE(dp), 1); + init_page_count(PD_PAGE(dp)); return; } diff --git a/arch/m68k/mm/motorola.c b/arch/m68k/mm/motorola.c index d855fec2631..afb57eeafdc 100644 --- a/arch/m68k/mm/motorola.c +++ b/arch/m68k/mm/motorola.c @@ -276,7 +276,7 @@ void free_initmem(void) addr = (unsigned long)&__init_begin; for (; addr < (unsigned long)&__init_end; addr += PAGE_SIZE) { virt_to_page(addr)->flags &= ~(1 << PG_reserved); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } diff --git a/arch/m68knommu/mm/init.c b/arch/m68knommu/mm/init.c index 89f0b554ffb..d79503fe6e4 100644 --- a/arch/m68knommu/mm/init.c +++ b/arch/m68knommu/mm/init.c @@ -195,7 +195,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) int pages = 0; for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; pages++; @@ -218,7 +218,7 @@ free_initmem() /* next to check that the page we free is not a partial page */ for (; addr + PAGE_SIZE < (unsigned long)(&__init_end); addr +=PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } diff --git a/arch/mips/arc/memory.c b/arch/mips/arc/memory.c index 958d2eb7886..8a9ef58cc39 100644 --- a/arch/mips/arc/memory.c +++ b/arch/mips/arc/memory.c @@ -158,7 +158,7 @@ unsigned long __init prom_free_prom_memory(void) while (addr < boot_mem_map.map[i].addr + boot_mem_map.map[i].size) { ClearPageReserved(virt_to_page(__va(addr))); - set_page_count(virt_to_page(__va(addr)), 1); + init_page_count(virt_to_page(__va(addr))); free_page((unsigned long)__va(addr)); addr += PAGE_SIZE; freed += PAGE_SIZE; diff --git a/arch/mips/dec/prom/memory.c b/arch/mips/dec/prom/memory.c index 81cb5a76cfb..1edaf3074ee 100644 --- a/arch/mips/dec/prom/memory.c +++ b/arch/mips/dec/prom/memory.c @@ -118,7 +118,7 @@ unsigned long __init prom_free_prom_memory(void) addr = PAGE_SIZE; while (addr < end) { ClearPageReserved(virt_to_page(__va(addr))); - set_page_count(virt_to_page(__va(addr)), 1); + init_page_count(virt_to_page(__va(addr))); free_page((unsigned long)__va(addr)); addr += PAGE_SIZE; } diff --git a/arch/mips/mips-boards/generic/memory.c b/arch/mips/mips-boards/generic/memory.c index 2c8afd77a20..ee5e70c95cf 100644 --- a/arch/mips/mips-boards/generic/memory.c +++ b/arch/mips/mips-boards/generic/memory.c @@ -174,7 +174,7 @@ unsigned long __init prom_free_prom_memory(void) while (addr < boot_mem_map.map[i].addr + boot_mem_map.map[i].size) { ClearPageReserved(virt_to_page(__va(addr))); - set_page_count(virt_to_page(__va(addr)), 1); + init_page_count(virt_to_page(__va(addr))); free_page((unsigned long)__va(addr)); addr += PAGE_SIZE; freed += PAGE_SIZE; diff --git a/arch/mips/mips-boards/sim/sim_mem.c b/arch/mips/mips-boards/sim/sim_mem.c index 0dbd7435bb2..1ec4e75656b 100644 --- a/arch/mips/mips-boards/sim/sim_mem.c +++ b/arch/mips/mips-boards/sim/sim_mem.c @@ -117,7 +117,7 @@ unsigned long __init prom_free_prom_memory(void) while (addr < boot_mem_map.map[i].addr + boot_mem_map.map[i].size) { ClearPageReserved(virt_to_page(__va(addr))); - set_page_count(virt_to_page(__va(addr)), 1); + init_page_count(virt_to_page(__va(addr))); free_page((unsigned long)__va(addr)); addr += PAGE_SIZE; freed += PAGE_SIZE; diff --git a/arch/mips/mm/init.c b/arch/mips/mm/init.c index a140da9732d..52f7d59fe61 100644 --- a/arch/mips/mm/init.c +++ b/arch/mips/mm/init.c @@ -245,7 +245,7 @@ void __init mem_init(void) #ifdef CONFIG_LIMITED_DMA set_page_address(page, lowmem_page_address(page)); #endif - set_page_count(page, 1); + init_page_count(page); __free_page(page); totalhigh_pages++; } @@ -292,7 +292,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; } @@ -315,7 +315,7 @@ void free_initmem(void) page = addr; #endif ClearPageReserved(virt_to_page(page)); - set_page_count(virt_to_page(page), 1); + init_page_count(virt_to_page(page)); free_page(page); totalram_pages++; freed += PAGE_SIZE; diff --git a/arch/mips/sgi-ip27/ip27-memory.c b/arch/mips/sgi-ip27/ip27-memory.c index ed93a979295..e0d095daa5e 100644 --- a/arch/mips/sgi-ip27/ip27-memory.c +++ b/arch/mips/sgi-ip27/ip27-memory.c @@ -559,7 +559,7 @@ void __init mem_init(void) /* if (!page_is_ram(pgnr)) continue; */ /* commented out until page_is_ram works */ ClearPageReserved(p); - set_page_count(p, 1); + init_page_count(p); __free_page(p); totalram_pages++; } diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c index 7847ca13d6c..852eda3953d 100644 --- a/arch/parisc/mm/init.c +++ b/arch/parisc/mm/init.c @@ -398,7 +398,7 @@ void free_initmem(void) addr = (unsigned long)(&__init_begin); for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); num_physpages++; totalram_pages++; @@ -1018,7 +1018,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) printk(KERN_INFO "Freeing initrd memory: %ldk freed\n", (end - start) >> 10); for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); num_physpages++; totalram_pages++; diff --git a/arch/powerpc/mm/init_32.c b/arch/powerpc/mm/init_32.c index 7d0d75c1184..b57fb3a2b7b 100644 --- a/arch/powerpc/mm/init_32.c +++ b/arch/powerpc/mm/init_32.c @@ -216,7 +216,7 @@ static void free_sec(unsigned long start, unsigned long end, const char *name) while (start < end) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); cnt++; start += PAGE_SIZE; @@ -248,7 +248,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) printk ("Freeing initrd memory: %ldk freed\n", (end - start) >> 10); for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; } diff --git a/arch/powerpc/mm/init_64.c b/arch/powerpc/mm/init_64.c index 81cfb0c2ec5..bacb71c8981 100644 --- a/arch/powerpc/mm/init_64.c +++ b/arch/powerpc/mm/init_64.c @@ -140,7 +140,7 @@ void free_initmem(void) for (; addr < (unsigned long)__init_end; addr += PAGE_SIZE) { memset((void *)addr, 0xcc, PAGE_SIZE); ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } @@ -155,7 +155,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) printk ("Freeing initrd memory: %ldk freed\n", (end - start) >> 10); for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; } diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c index 6ae5c130d0d..454cac01d8c 100644 --- a/arch/powerpc/mm/mem.c +++ b/arch/powerpc/mm/mem.c @@ -108,7 +108,7 @@ EXPORT_SYMBOL(phys_mem_access_prot); void online_page(struct page *page) { ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); __free_page(page); totalram_pages++; num_physpages++; @@ -376,7 +376,7 @@ void __init mem_init(void) struct page *page = pfn_to_page(pfn); ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); __free_page(page); totalhigh_pages++; } diff --git a/arch/powerpc/platforms/cell/setup.c b/arch/powerpc/platforms/cell/setup.c index b33a4443f5a..fec8e65b36e 100644 --- a/arch/powerpc/platforms/cell/setup.c +++ b/arch/powerpc/platforms/cell/setup.c @@ -115,7 +115,7 @@ static void __init cell_spuprop_present(struct device_node *spe, for (pfn = start_pfn; pfn < end_pfn; pfn++) { struct page *page = pfn_to_page(pfn); set_page_links(page, ZONE_DMA, node_id, pfn); - set_page_count(page, 1); + init_page_count(page); reset_page_mapcount(page); SetPageReserved(page); INIT_LIST_HEAD(&page->lru); diff --git a/arch/ppc/mm/init.c b/arch/ppc/mm/init.c index 134db5c0420..cb1c294fb93 100644 --- a/arch/ppc/mm/init.c +++ b/arch/ppc/mm/init.c @@ -140,7 +140,7 @@ static void free_sec(unsigned long start, unsigned long end, const char *name) while (start < end) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); cnt++; start += PAGE_SIZE; @@ -172,7 +172,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; } @@ -441,7 +441,7 @@ void __init mem_init(void) struct page *page = mem_map + pfn; ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); __free_page(page); totalhigh_pages++; } diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c index df953383724..a055894f3bd 100644 --- a/arch/s390/mm/init.c +++ b/arch/s390/mm/init.c @@ -292,7 +292,7 @@ void free_initmem(void) addr = (unsigned long)(&__init_begin); for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } @@ -307,7 +307,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) printk ("Freeing initrd memory: %ldk freed\n", (end - start) >> 10); for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; } diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c index e342565f75f..77b4a838fe1 100644 --- a/arch/sh/mm/init.c +++ b/arch/sh/mm/init.c @@ -273,7 +273,7 @@ void free_initmem(void) addr = (unsigned long)(&__init_begin); for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } @@ -286,7 +286,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) unsigned long p; for (p = start; p < end; p += PAGE_SIZE) { ClearPageReserved(virt_to_page(p)); - set_page_count(virt_to_page(p), 1); + init_page_count(virt_to_page(p)); free_page(p); totalram_pages++; } diff --git a/arch/sh64/mm/init.c b/arch/sh64/mm/init.c index a65e8bb2c3c..1169757fb38 100644 --- a/arch/sh64/mm/init.c +++ b/arch/sh64/mm/init.c @@ -173,7 +173,7 @@ void free_initmem(void) addr = (unsigned long)(&__init_begin); for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); free_page(addr); totalram_pages++; } @@ -186,7 +186,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) unsigned long p; for (p = start; p < end; p += PAGE_SIZE) { ClearPageReserved(virt_to_page(p)); - set_page_count(virt_to_page(p), 1); + init_page_count(virt_to_page(p)); free_page(p); totalram_pages++; } diff --git a/arch/sparc/kernel/sun4d_smp.c b/arch/sparc/kernel/sun4d_smp.c index 40d426cce82..4219dd2ce3a 100644 --- a/arch/sparc/kernel/sun4d_smp.c +++ b/arch/sparc/kernel/sun4d_smp.c @@ -266,19 +266,19 @@ void __init smp4d_boot_cpus(void) /* Free unneeded trap tables */ ClearPageReserved(virt_to_page(trapbase_cpu1)); - set_page_count(virt_to_page(trapbase_cpu1), 1); + init_page_count(virt_to_page(trapbase_cpu1)); free_page((unsigned long)trapbase_cpu1); totalram_pages++; num_physpages++; ClearPageReserved(virt_to_page(trapbase_cpu2)); - set_page_count(virt_to_page(trapbase_cpu2), 1); + init_page_count(virt_to_page(trapbase_cpu2)); free_page((unsigned long)trapbase_cpu2); totalram_pages++; num_physpages++; ClearPageReserved(virt_to_page(trapbase_cpu3)); - set_page_count(virt_to_page(trapbase_cpu3), 1); + init_page_count(virt_to_page(trapbase_cpu3)); free_page((unsigned long)trapbase_cpu3); totalram_pages++; num_physpages++; diff --git a/arch/sparc/kernel/sun4m_smp.c b/arch/sparc/kernel/sun4m_smp.c index a21f27d10e5..fbbd8a474c4 100644 --- a/arch/sparc/kernel/sun4m_smp.c +++ b/arch/sparc/kernel/sun4m_smp.c @@ -233,21 +233,21 @@ void __init smp4m_boot_cpus(void) /* Free unneeded trap tables */ if (!cpu_isset(i, cpu_present_map)) { ClearPageReserved(virt_to_page(trapbase_cpu1)); - set_page_count(virt_to_page(trapbase_cpu1), 1); + init_page_count(virt_to_page(trapbase_cpu1)); free_page((unsigned long)trapbase_cpu1); totalram_pages++; num_physpages++; } if (!cpu_isset(2, cpu_present_map)) { ClearPageReserved(virt_to_page(trapbase_cpu2)); - set_page_count(virt_to_page(trapbase_cpu2), 1); + init_page_count(virt_to_page(trapbase_cpu2)); free_page((unsigned long)trapbase_cpu2); totalram_pages++; num_physpages++; } if (!cpu_isset(3, cpu_present_map)) { ClearPageReserved(virt_to_page(trapbase_cpu3)); - set_page_count(virt_to_page(trapbase_cpu3), 1); + init_page_count(virt_to_page(trapbase_cpu3)); free_page((unsigned long)trapbase_cpu3); totalram_pages++; num_physpages++; diff --git a/arch/sparc/mm/init.c b/arch/sparc/mm/init.c index c03babaa049..89866973246 100644 --- a/arch/sparc/mm/init.c +++ b/arch/sparc/mm/init.c @@ -383,7 +383,7 @@ void map_high_region(unsigned long start_pfn, unsigned long end_pfn) struct page *page = pfn_to_page(tmp); ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); __free_page(page); totalhigh_pages++; } @@ -480,7 +480,7 @@ void free_initmem (void) p = virt_to_page(addr); ClearPageReserved(p); - set_page_count(p, 1); + init_page_count(p); __free_page(p); totalram_pages++; num_physpages++; @@ -497,7 +497,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) struct page *p = virt_to_page(start); ClearPageReserved(p); - set_page_count(p, 1); + init_page_count(p); __free_page(p); num_physpages++; } diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c index c2b556106fc..2ae143ba50d 100644 --- a/arch/sparc64/mm/init.c +++ b/arch/sparc64/mm/init.c @@ -1461,7 +1461,7 @@ void free_initmem(void) p = virt_to_page(page); ClearPageReserved(p); - set_page_count(p, 1); + init_page_count(p); __free_page(p); num_physpages++; totalram_pages++; @@ -1477,7 +1477,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) struct page *p = virt_to_page(start); ClearPageReserved(p); - set_page_count(p, 1); + init_page_count(p); __free_page(p); num_physpages++; totalram_pages++; diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c index fa4f915be5c..92cce96b5e2 100644 --- a/arch/um/kernel/mem.c +++ b/arch/um/kernel/mem.c @@ -57,7 +57,7 @@ static void setup_highmem(unsigned long highmem_start, for(i = 0; i < highmem_len >> PAGE_SHIFT; i++){ page = &mem_map[highmem_pfn + i]; ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); __free_page(page); } } @@ -296,7 +296,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) (end - start) >> 10); for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; } diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c index 7af1742aa95..40ed13d263c 100644 --- a/arch/x86_64/mm/init.c +++ b/arch/x86_64/mm/init.c @@ -486,7 +486,7 @@ void __init clear_kernel_mapping(unsigned long address, unsigned long size) void online_page(struct page *page) { ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); __free_page(page); totalram_pages++; num_physpages++; @@ -592,7 +592,7 @@ void free_initmem(void) addr = (unsigned long)(&__init_begin); for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) { ClearPageReserved(virt_to_page(addr)); - set_page_count(virt_to_page(addr), 1); + init_page_count(virt_to_page(addr)); memset((void *)(addr & ~(PAGE_SIZE-1)), 0xcc, PAGE_SIZE); free_page(addr); totalram_pages++; @@ -632,7 +632,7 @@ void free_initrd_mem(unsigned long start, unsigned long end) printk ("Freeing initrd memory: %ldk freed\n", (end - start) >> 10); for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page(start); totalram_pages++; } diff --git a/arch/xtensa/mm/init.c b/arch/xtensa/mm/init.c index 5a91d6c9e66..e1be4235f36 100644 --- a/arch/xtensa/mm/init.c +++ b/arch/xtensa/mm/init.c @@ -272,7 +272,7 @@ free_reserved_mem(void *start, void *end) { for (; start < end; start += PAGE_SIZE) { ClearPageReserved(virt_to_page(start)); - set_page_count(virt_to_page(start), 1); + init_page_count(virt_to_page(start)); free_page((unsigned long)start); totalram_pages++; } diff --git a/drivers/video/acornfb.c b/drivers/video/acornfb.c index b058273527b..76448d6ae89 100644 --- a/drivers/video/acornfb.c +++ b/drivers/video/acornfb.c @@ -1269,7 +1269,7 @@ free_unused_pages(unsigned int virtual_start, unsigned int virtual_end) */ page = virt_to_page(virtual_start); ClearPageReserved(page); - set_page_count(page, 1); + init_page_count(page); free_page(virtual_start); virtual_start += PAGE_SIZE; diff --git a/include/linux/mm.h b/include/linux/mm.h index 3d84b7a35e0..7d8c127daad 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -307,8 +307,6 @@ static inline int get_page_unless_zero(struct page *page) return atomic_inc_not_zero(&page->_count); } -#define set_page_count(p,v) atomic_set(&(p)->_count, (v)) - extern void FASTCALL(__page_cache_release(struct page *)); static inline int page_count(struct page *page) @@ -325,6 +323,15 @@ static inline void get_page(struct page *page) atomic_inc(&page->_count); } +/* + * Setup the page count before being freed into the page allocator for + * the first time (boot or memory hotplug) + */ +static inline void init_page_count(struct page *page) +{ + atomic_set(&page->_count, 1); +} + void put_page(struct page *page); void split_page(struct page *page, unsigned int order); diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 39d49ecea8e..20117a4b8ab 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -18,6 +18,7 @@ #include #include +#include "internal.h" const unsigned long hugetlb_zero = 0, hugetlb_infinity = ~0UL; static unsigned long nr_huge_pages, free_huge_pages; @@ -106,7 +107,7 @@ struct page *alloc_huge_page(struct vm_area_struct *vma, unsigned long addr) return NULL; } spin_unlock(&hugetlb_lock); - set_page_count(page, 1); + set_page_refcounted(page); for (i = 0; i < (HPAGE_SIZE/PAGE_SIZE); ++i) clear_user_highpage(&page[i], addr); return page; @@ -152,7 +153,7 @@ static void update_and_free_page(struct page *page) 1 << PG_private | 1<< PG_writeback); } page[1].lru.next = NULL; - set_page_count(page, 1); + set_page_refcounted(page); __free_pages(page, HUGETLB_PAGE_ORDER); } diff --git a/mm/internal.h b/mm/internal.h index 7bb33977981..d20e3cc4aef 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -13,8 +13,19 @@ #include -static inline void set_page_refs(struct page *page, int order) +static inline void set_page_count(struct page *page, int v) { + atomic_set(&page->_count, v); +} + +/* + * Turn a non-refcounted page (->_count == 0) into refcounted with + * a count of one. + */ +static inline void set_page_refcounted(struct page *page) +{ + BUG_ON(PageCompound(page) && page_private(page) != (unsigned long)page); + BUG_ON(atomic_read(&page->_count)); set_page_count(page, 1); } diff --git a/mm/page_alloc.c b/mm/page_alloc.c index e197818a7cf..7f65b5a63bb 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -442,7 +442,7 @@ void fastcall __init __free_pages_bootmem(struct page *page, unsigned int order) if (order == 0) { __ClearPageReserved(page); set_page_count(page, 0); - set_page_refs(page, 0); + set_page_refcounted(page); __free_page(page); } else { int loop; @@ -457,7 +457,7 @@ void fastcall __init __free_pages_bootmem(struct page *page, unsigned int order) set_page_count(p, 0); } - set_page_refs(page, order); + set_page_refcounted(page); __free_pages(page, order); } } @@ -525,7 +525,7 @@ static int prep_new_page(struct page *page, int order) 1 << PG_referenced | 1 << PG_arch_1 | 1 << PG_checked | 1 << PG_mappedtodisk); set_page_private(page, 0); - set_page_refs(page, order); + set_page_refcounted(page); kernel_map_pages(page, 1 << order, 1); return 0; } @@ -755,10 +755,8 @@ void split_page(struct page *page, unsigned int order) BUG_ON(PageCompound(page)); BUG_ON(!page_count(page)); - for (i = 1; i < (1 << order); i++) { - BUG_ON(page_count(page + i)); - set_page_count(page + i, 1); - } + for (i = 1; i < (1 << order); i++) + set_page_refcounted(page + i); } /* @@ -1771,7 +1769,7 @@ void __meminit memmap_init_zone(unsigned long size, int nid, unsigned long zone, continue; page = pfn_to_page(pfn); set_page_links(page, zone, nid, pfn); - set_page_count(page, 1); + init_page_count(page); reset_page_mapcount(page); SetPageReserved(page); INIT_LIST_HEAD(&page->lru); -- cgit v1.2.3-70-g09d2 From 42b88befd6e0dae1a5fe04c03925037fa890e1f3 Mon Sep 17 00:00:00 2001 From: David Gibson Date: Wed, 22 Mar 2006 00:09:01 -0800 Subject: [PATCH] hugepage: is_aligned_hugepage_range() cleanup Quite a long time back, prepare_hugepage_range() replaced is_aligned_hugepage_range() as the callback from mm/mmap.c to arch code to verify if an address range is suitable for a hugepage mapping. is_aligned_hugepage_range() stuck around, but only to implement prepare_hugepage_range() on archs which didn't implement their own. Most archs (everything except ia64 and powerpc) used the same implementation of is_aligned_hugepage_range(). On powerpc, which implements its own prepare_hugepage_range(), the custom version was never used. In addition, "is_aligned_hugepage_range()" was a bad name, because it suggests it returns true iff the given range is a good hugepage range, whereas in fact it returns 0-or-error (so the sense is reversed). This patch cleans up by abolishing is_aligned_hugepage_range(). Instead prepare_hugepage_range() is defined directly. Most archs use the default version, which simply checks the given region is aligned to the size of a hugepage. ia64 and powerpc define custom versions. The ia64 one simply checks that the range is in the correct address space region in addition to being suitably aligned. The powerpc version (just as previously) checks for suitable addresses, and if necessary performs low-level MMU frobbing to set up new areas for use by hugepages. No libhugetlbfs testsuite regressions on ppc64 (POWER5 LPAR). Signed-off-by: David Gibson Signed-off-by: Zhang Yanmin Cc: "David S. Miller" Cc: Benjamin Herrenschmidt Cc: Paul Mackerras Cc: William Lee Irwin III Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/mm/hugetlbpage.c | 12 ------------ arch/ia64/mm/hugetlbpage.c | 5 +++-- arch/powerpc/mm/hugetlbpage.c | 15 --------------- arch/sh/mm/hugetlbpage.c | 12 ------------ arch/sh64/mm/hugetlbpage.c | 12 ------------ arch/sparc64/mm/hugetlbpage.c | 12 ------------ include/asm-ia64/page.h | 1 + include/linux/hugetlb.h | 16 ++++++++++++---- 8 files changed, 16 insertions(+), 69 deletions(-) (limited to 'arch/ia64') diff --git a/arch/i386/mm/hugetlbpage.c b/arch/i386/mm/hugetlbpage.c index d524127c9af..a7d89158541 100644 --- a/arch/i386/mm/hugetlbpage.c +++ b/arch/i386/mm/hugetlbpage.c @@ -48,18 +48,6 @@ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr) return (pte_t *) pmd; } -/* - * This function checks for proper alignment of input addr and len parameters. - */ -int is_aligned_hugepage_range(unsigned long addr, unsigned long len) -{ - if (len & ~HPAGE_MASK) - return -EINVAL; - if (addr & ~HPAGE_MASK) - return -EINVAL; - return 0; -} - #if 0 /* This is just for testing */ struct page * follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c index 2d13889d0a9..9dbc7dadd16 100644 --- a/arch/ia64/mm/hugetlbpage.c +++ b/arch/ia64/mm/hugetlbpage.c @@ -68,9 +68,10 @@ huge_pte_offset (struct mm_struct *mm, unsigned long addr) #define mk_pte_huge(entry) { pte_val(entry) |= _PAGE_P; } /* - * This function checks for proper alignment of input addr and len parameters. + * Don't actually need to do any preparation, but need to make sure + * the address is in the right region. */ -int is_aligned_hugepage_range(unsigned long addr, unsigned long len) +int prepare_hugepage_range(unsigned long addr, unsigned long len) { if (len & ~HPAGE_MASK) return -EINVAL; diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index b51bb28c054..7370f9f33e2 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -133,21 +133,6 @@ pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, return __pte(old); } -/* - * This function checks for proper alignment of input addr and len parameters. - */ -int is_aligned_hugepage_range(unsigned long addr, unsigned long len) -{ - if (len & ~HPAGE_MASK) - return -EINVAL; - if (addr & ~HPAGE_MASK) - return -EINVAL; - if (! (within_hugepage_low_range(addr, len) - || within_hugepage_high_range(addr, len)) ) - return -EINVAL; - return 0; -} - struct slb_flush_info { struct mm_struct *mm; u16 newareas; diff --git a/arch/sh/mm/hugetlbpage.c b/arch/sh/mm/hugetlbpage.c index 6b7a7688c98..a3568fd5150 100644 --- a/arch/sh/mm/hugetlbpage.c +++ b/arch/sh/mm/hugetlbpage.c @@ -84,18 +84,6 @@ pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, return entry; } -/* - * This function checks for proper alignment of input addr and len parameters. - */ -int is_aligned_hugepage_range(unsigned long addr, unsigned long len) -{ - if (len & ~HPAGE_MASK) - return -EINVAL; - if (addr & ~HPAGE_MASK) - return -EINVAL; - return 0; -} - struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) { diff --git a/arch/sh64/mm/hugetlbpage.c b/arch/sh64/mm/hugetlbpage.c index ed6a505b3ee..3d89f2a6c78 100644 --- a/arch/sh64/mm/hugetlbpage.c +++ b/arch/sh64/mm/hugetlbpage.c @@ -84,18 +84,6 @@ pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, return entry; } -/* - * This function checks for proper alignment of input addr and len parameters. - */ -int is_aligned_hugepage_range(unsigned long addr, unsigned long len) -{ - if (len & ~HPAGE_MASK) - return -EINVAL; - if (addr & ~HPAGE_MASK) - return -EINVAL; - return 0; -} - struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) { diff --git a/arch/sparc64/mm/hugetlbpage.c b/arch/sparc64/mm/hugetlbpage.c index a7a24869d04..280dc7958a1 100644 --- a/arch/sparc64/mm/hugetlbpage.c +++ b/arch/sparc64/mm/hugetlbpage.c @@ -263,18 +263,6 @@ pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, return entry; } -/* - * This function checks for proper alignment of input addr and len parameters. - */ -int is_aligned_hugepage_range(unsigned long addr, unsigned long len) -{ - if (len & ~HPAGE_MASK) - return -EINVAL; - if (addr & ~HPAGE_MASK) - return -EINVAL; - return 0; -} - struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, int write) { diff --git a/include/asm-ia64/page.h b/include/asm-ia64/page.h index 732cf308674..3ab27333dae 100644 --- a/include/asm-ia64/page.h +++ b/include/asm-ia64/page.h @@ -57,6 +57,7 @@ # define HAVE_ARCH_HUGETLB_UNMAPPED_AREA # define ARCH_HAS_HUGEPAGE_ONLY_RANGE +# define ARCH_HAS_PREPARE_HUGEPAGE_RANGE # define ARCH_HAS_HUGETLB_FREE_PGD_RANGE #endif /* CONFIG_HUGETLB_PAGE */ diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index 5db25ffdb3e..d6f1019625a 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -36,7 +36,6 @@ struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address, int write); struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address, pmd_t *pmd, int write); -int is_aligned_hugepage_range(unsigned long addr, unsigned long len); int pmd_huge(pmd_t pmd); void hugetlb_change_protection(struct vm_area_struct *vma, unsigned long address, unsigned long end, pgprot_t newprot); @@ -54,8 +53,18 @@ void hugetlb_free_pgd_range(struct mmu_gather **tlb, unsigned long addr, #endif #ifndef ARCH_HAS_PREPARE_HUGEPAGE_RANGE -#define prepare_hugepage_range(addr, len) \ - is_aligned_hugepage_range(addr, len) +/* + * If the arch doesn't supply something else, assume that hugepage + * size aligned regions are ok without further preparation. + */ +static inline int prepare_hugepage_range(unsigned long addr, unsigned long len) +{ + if (len & ~HPAGE_MASK) + return -EINVAL; + if (addr & ~HPAGE_MASK) + return -EINVAL; + return 0; +} #else int prepare_hugepage_range(unsigned long addr, unsigned long len); #endif @@ -95,7 +104,6 @@ static inline unsigned long hugetlb_total_pages(void) #define hugetlb_report_meminfo(buf) 0 #define hugetlb_report_node_meminfo(n, buf) 0 #define follow_huge_pmd(mm, addr, pmd, write) NULL -#define is_aligned_hugepage_range(addr, len) 0 #define prepare_hugepage_range(addr, len) (-EINVAL) #define pmd_huge(x) 0 #define is_hugepage_only_range(mm, addr, len) 0 -- cgit v1.2.3-70-g09d2 From 2332c9ae7911618575241e0c843cd686968db8e3 Mon Sep 17 00:00:00 2001 From: "Chen, Kenneth W" Date: Wed, 22 Mar 2006 10:49:00 -0800 Subject: [IA64] fix ia64 is_hugepage_only_range fix is_hugepage_only_range() definition to be "overlaps" instead of "within architectural restricted hugetlb address range". Simplify the ia64 specific code that used to use is_hugepage_only_range() to just check which region the address is in. Signed-off-by: Ken Chen Signed-off-by: Tony Luck --- arch/ia64/mm/hugetlbpage.c | 7 +++---- include/asm-ia64/page.h | 2 +- 2 files changed, 4 insertions(+), 5 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c index 9dbc7dadd16..8d506710fdb 100644 --- a/arch/ia64/mm/hugetlbpage.c +++ b/arch/ia64/mm/hugetlbpage.c @@ -113,8 +113,7 @@ void hugetlb_free_pgd_range(struct mmu_gather **tlb, unsigned long floor, unsigned long ceiling) { /* - * This is called only when is_hugepage_only_range(addr,), - * and it follows that is_hugepage_only_range(end,) also. + * This is called to free hugetlb page tables. * * The offset of these addresses from the base of the hugetlb * region must be scaled down by HPAGE_SIZE/PAGE_SIZE so that @@ -126,9 +125,9 @@ void hugetlb_free_pgd_range(struct mmu_gather **tlb, addr = htlbpage_to_page(addr); end = htlbpage_to_page(end); - if (is_hugepage_only_range(tlb->mm, floor, HPAGE_SIZE)) + if (REGION_NUMBER(floor) == RGN_HPAGE) floor = htlbpage_to_page(floor); - if (is_hugepage_only_range(tlb->mm, ceiling, HPAGE_SIZE)) + if (REGION_NUMBER(ceiling) == RGN_HPAGE) ceiling = htlbpage_to_page(ceiling); free_pgd_range(tlb, addr, end, floor, ceiling); diff --git a/include/asm-ia64/page.h b/include/asm-ia64/page.h index 3ab27333dae..6e9aa23250c 100644 --- a/include/asm-ia64/page.h +++ b/include/asm-ia64/page.h @@ -149,7 +149,7 @@ typedef union ia64_va { | (REGION_OFFSET(x) >> (HPAGE_SHIFT-PAGE_SHIFT))) # define HUGETLB_PAGE_ORDER (HPAGE_SHIFT - PAGE_SHIFT) # define is_hugepage_only_range(mm, addr, len) \ - (REGION_NUMBER(addr) == RGN_HPAGE && \ + (REGION_NUMBER(addr) == RGN_HPAGE || \ REGION_NUMBER((addr)+(len)-1) == RGN_HPAGE) extern unsigned int hpage_shift; #endif -- cgit v1.2.3-70-g09d2 From 9a4e5549b7cb5d0ef187184641d8defe8be875f5 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Tue, 21 Mar 2006 10:44:07 -0700 Subject: [IA64] sn_check_intr: use ia64_get_irr() Use the recently-added ia64_get_irr() rather than duplicating the code. Signed-off-by: Bjorn Helgaas Acked-by: Jes Sorensen Signed-off-by: Tony Luck --- arch/ia64/sn/kernel/irq.c | 21 +-------------------- 1 file changed, 1 insertion(+), 20 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/sn/kernel/irq.c b/arch/ia64/sn/kernel/irq.c index c373113d073..c265e02f503 100644 --- a/arch/ia64/sn/kernel/irq.c +++ b/arch/ia64/sn/kernel/irq.c @@ -350,9 +350,6 @@ static void force_interrupt(int irq) static void sn_check_intr(int irq, struct sn_irq_info *sn_irq_info) { u64 regval; - int irr_reg_num; - int irr_bit; - u64 irr_reg; struct pcidev_info *pcidev_info; struct pcibus_info *pcibus_info; @@ -373,23 +370,7 @@ static void sn_check_intr(int irq, struct sn_irq_info *sn_irq_info) pdi_pcibus_info; regval = pcireg_intr_status_get(pcibus_info); - irr_reg_num = irq_to_vector(irq) / 64; - irr_bit = irq_to_vector(irq) % 64; - switch (irr_reg_num) { - case 0: - irr_reg = ia64_getreg(_IA64_REG_CR_IRR0); - break; - case 1: - irr_reg = ia64_getreg(_IA64_REG_CR_IRR1); - break; - case 2: - irr_reg = ia64_getreg(_IA64_REG_CR_IRR2); - break; - case 3: - irr_reg = ia64_getreg(_IA64_REG_CR_IRR3); - break; - } - if (!test_bit(irr_bit, &irr_reg)) { + if (!ia64_get_irr(irq_to_vector(irq))) { if (!test_bit(irq, pda->sn_in_service_ivecs)) { regval &= 0xff; if (sn_irq_info->irq_int_bit & regval & -- cgit v1.2.3-70-g09d2 From d903cea3811994d4c4122a0e3ca6d3a197e812e2 Mon Sep 17 00:00:00 2001 From: Kenji Kaneshige Date: Wed, 15 Mar 2006 14:45:11 +0900 Subject: [IA64] Ignore disabled Local SAPIC Affinity Structure in SRAT According to the ACPI spec, the OSPM must ignore the contents of the Processor Local APIC/SAPIC Affinity Structure in System Resource Affinity Table (SRAT), if its enable flag is cleared. However, ia64 linux refers all of the Processor Local APIC/SAPIC Affinity Structures in SRAT regardless of the enable flag. This is obviously against the ACPI spec. This patch fixes this bug. Signed-off-by: Kenji Kaneshige Signed-off-by: Tony Luck --- arch/ia64/kernel/acpi.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c index 4722ec51c70..24fe7c81e10 100644 --- a/arch/ia64/kernel/acpi.c +++ b/arch/ia64/kernel/acpi.c @@ -443,6 +443,9 @@ void __init acpi_numa_slit_init(struct acpi_table_slit *slit) void __init acpi_numa_processor_affinity_init(struct acpi_table_processor_affinity *pa) { + if (!pa->flags.enabled) + return; + /* record this node in proximity bitmap */ pxm_bit_set(pa->proximity_domain); -- cgit v1.2.3-70-g09d2 From 0881fc8df2e2bf1a7900a0ab3860109beb46fbab Mon Sep 17 00:00:00 2001 From: "Chen, Kenneth W" Date: Sun, 12 Mar 2006 08:52:20 -0800 Subject: [IA64] add __init declaration to mca functions Mark init related variable and functions with appropriate __init* declaration to mca functions. Signed-off-by: Ken Chen Signed-off-by: Tony Luck --- arch/ia64/kernel/mca.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c index b57e723f194..cedcae713e9 100644 --- a/arch/ia64/kernel/mca.c +++ b/arch/ia64/kernel/mca.c @@ -133,7 +133,7 @@ static int cpe_poll_enabled = 1; extern void salinfo_log_wakeup(int type, u8 *buffer, u64 size, int irqsafe); -static int mca_init; +static int mca_init __initdata; static void inline @@ -184,7 +184,7 @@ static ia64_state_log_t ia64_state_log[IA64_MAX_LOG_TYPES]; * Inputs : info_type (SAL_INFO_TYPE_{MCA,INIT,CMC,CPE}) * Outputs : None */ -static void +static void __init ia64_log_init(int sal_info_type) { u64 max_size = 0; @@ -355,7 +355,7 @@ ia64_mca_cpe_int_handler (int cpe_irq, void *arg, struct pt_regs *ptregs) * Outputs * None */ -static void +static void __init ia64_mca_register_cpev (int cpev) { /* Register the CPE interrupt vector with SAL */ @@ -386,7 +386,7 @@ ia64_mca_register_cpev (int cpev) * Outputs * None */ -void +void __cpuinit ia64_mca_cmc_vector_setup (void) { cmcv_reg_t cmcv; @@ -1443,7 +1443,7 @@ static struct irqaction mca_cpep_irqaction = { * format most of the fields. */ -static void +static void __cpuinit format_mca_init_stack(void *mca_data, unsigned long offset, const char *type, int cpu) { @@ -1467,7 +1467,7 @@ format_mca_init_stack(void *mca_data, unsigned long offset, /* Do per-CPU MCA-related initialization. */ -void __devinit +void __cpuinit ia64_mca_cpu_init(void *cpu_data) { void *pal_vaddr; -- cgit v1.2.3-70-g09d2 From 244fd54540806a5e3391d117794105a35815cbb2 Mon Sep 17 00:00:00 2001 From: "Chen, Kenneth W" Date: Sun, 12 Mar 2006 09:00:13 -0800 Subject: [IA64] add init declaration to cpu initialization functions Add init declaration to cpu initialization functions. Signed-off-by: Ken Chen Signed-off-by: Tony Luck --- arch/ia64/kernel/setup.c | 12 ++++++------ arch/ia64/mm/contig.c | 2 +- arch/ia64/mm/discontig.c | 2 +- include/asm-ia64/processor.h | 1 - 4 files changed, 8 insertions(+), 9 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c index 3258e09278d..340dc0362fe 100644 --- a/arch/ia64/kernel/setup.c +++ b/arch/ia64/kernel/setup.c @@ -362,7 +362,7 @@ mark_bsp_online (void) } #ifdef CONFIG_SMP -static void +static void __init check_for_logical_procs (void) { pal_logical_to_physical_t info; @@ -623,7 +623,7 @@ struct seq_operations cpuinfo_op = { .show = show_cpuinfo }; -void +static void __cpuinit identify_cpu (struct cpuinfo_ia64 *c) { union { @@ -700,7 +700,7 @@ setup_per_cpu_areas (void) * In addition, the minimum of the i-cache stride sizes is calculated for * "flush_icache_range()". */ -static void +static void __cpuinit get_max_cacheline_size (void) { unsigned long line_size, max = 1; @@ -763,10 +763,10 @@ get_max_cacheline_size (void) * cpu_init() initializes state that is per-CPU. This function acts * as a 'CPU state barrier', nothing should get across. */ -void +void __cpuinit cpu_init (void) { - extern void __devinit ia64_mmu_init (void *); + extern void __cpuinit ia64_mmu_init (void *); unsigned long num_phys_stacked; pal_vm_info_2_u_t vmi; unsigned int max_ctx; @@ -894,7 +894,7 @@ void sched_cacheflush(void) ia64_sal_cache_flush(3); } -void +void __init check_bugs (void) { ia64_patch_mckinley_e9((unsigned long) __start___mckinley_e9_bundles, diff --git a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c index 9855ba31809..8abb9e86d8a 100644 --- a/arch/ia64/mm/contig.c +++ b/arch/ia64/mm/contig.c @@ -176,7 +176,7 @@ find_memory (void) * * Allocate and setup per-cpu data areas. */ -void * +void * __cpuinit per_cpu_init (void) { void *cpu_data; diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c index 573d5cc63e2..2f5e44862e9 100644 --- a/arch/ia64/mm/discontig.c +++ b/arch/ia64/mm/discontig.c @@ -525,7 +525,7 @@ void __init find_memory(void) * find_pernode_space() does most of this already, we just need to set * local_per_cpu_offset */ -void *per_cpu_init(void) +void __cpuinit *per_cpu_init(void) { int cpu; static int first_time = 1; diff --git a/include/asm-ia64/processor.h b/include/asm-ia64/processor.h index 128fefd8056..b3bd58e8069 100644 --- a/include/asm-ia64/processor.h +++ b/include/asm-ia64/processor.h @@ -181,7 +181,6 @@ DECLARE_PER_CPU(struct cpuinfo_ia64, cpu_info); #define local_cpu_data (&__ia64_per_cpu_var(cpu_info)) #define cpu_data(cpu) (&per_cpu(cpu_info, cpu)) -extern void identify_cpu (struct cpuinfo_ia64 *); extern void print_cpu_info (struct cpuinfo_ia64 *); typedef struct { -- cgit v1.2.3-70-g09d2 From dae2806615c91cd7afa7c4dee65518435e45854a Mon Sep 17 00:00:00 2001 From: "Chen, Kenneth W" Date: Wed, 22 Mar 2006 16:54:15 -0800 Subject: [IA64] add init declaration to memory initialization functions Add init declaration to variables/functions used for memory initialization. I don't think they would clash with memory hotplug. If they do, please yell. Signed-off-by: Ken Chen Signed-off-by: Tony Luck --- arch/ia64/kernel/setup.c | 12 ++++++------ arch/ia64/mm/contig.c | 6 +++--- arch/ia64/mm/init.c | 12 ++++++------ 3 files changed, 15 insertions(+), 15 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c index 340dc0362fe..eb0793628b8 100644 --- a/arch/ia64/kernel/setup.c +++ b/arch/ia64/kernel/setup.c @@ -131,8 +131,8 @@ EXPORT_SYMBOL(ia64_max_iommu_merge_mask); /* * We use a special marker for the end of memory and it uses the extra (+1) slot */ -struct rsvd_region rsvd_region[IA64_MAX_RSVD_REGIONS + 1]; -int num_rsvd_regions; +struct rsvd_region rsvd_region[IA64_MAX_RSVD_REGIONS + 1] __initdata; +int num_rsvd_regions __initdata; /* @@ -141,7 +141,7 @@ int num_rsvd_regions; * caller-specified function is called with the memory ranges that remain after filtering. * This routine does not assume the incoming segments are sorted. */ -int +int __init filter_rsvd_memory (unsigned long start, unsigned long end, void *arg) { unsigned long range_start, range_end, prev_start; @@ -177,7 +177,7 @@ filter_rsvd_memory (unsigned long start, unsigned long end, void *arg) return 0; } -static void +static void __init sort_regions (struct rsvd_region *rsvd_region, int max) { int j; @@ -218,7 +218,7 @@ __initcall(register_memory); * initrd, etc. There are currently %IA64_MAX_RSVD_REGIONS defined, * see include/asm-ia64/meminit.h if you need to define more. */ -void +void __init reserve_memory (void) { int n = 0; @@ -270,7 +270,7 @@ reserve_memory (void) * Grab the initrd start and end from the boot parameter struct given us by * the boot loader. */ -void +void __init find_initrd (void) { #ifdef CONFIG_BLK_DEV_INITRD diff --git a/arch/ia64/mm/contig.c b/arch/ia64/mm/contig.c index 8abb9e86d8a..84fd1c14c8a 100644 --- a/arch/ia64/mm/contig.c +++ b/arch/ia64/mm/contig.c @@ -97,7 +97,7 @@ find_max_pfn (unsigned long start, unsigned long end, void *arg) * Find a place to put the bootmap and return its starting address in * bootmap_start. This address must be page-aligned. */ -int +static int __init find_bootmap_location (unsigned long start, unsigned long end, void *arg) { unsigned long needed = *(unsigned long *)arg; @@ -141,7 +141,7 @@ find_bootmap_location (unsigned long start, unsigned long end, void *arg) * Walk the EFI memory map and find usable memory for the system, taking * into account reserved areas. */ -void +void __init find_memory (void) { unsigned long bootmap_size; @@ -228,7 +228,7 @@ count_dma_pages (u64 start, u64 end, void *arg) * Set up the page tables. */ -void +void __init paging_init (void) { unsigned long max_dma; diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c index 08d94e6bfa1..3748768c787 100644 --- a/arch/ia64/mm/init.c +++ b/arch/ia64/mm/init.c @@ -206,7 +206,7 @@ free_initmem (void) (__init_end - __init_begin) >> 10); } -void +void __init free_initrd_mem (unsigned long start, unsigned long end) { struct page *page; @@ -261,7 +261,7 @@ free_initrd_mem (unsigned long start, unsigned long end) /* * This installs a clean page in the kernel's page table. */ -struct page * +static struct page * __init put_kernel_page (struct page *page, unsigned long address, pgprot_t pgprot) { pgd_t *pgd; @@ -411,7 +411,7 @@ ia64_mmu_init (void *my_cpu_data) #ifdef CONFIG_VIRTUAL_MEM_MAP -int +int __init create_mem_map_page_table (u64 start, u64 end, void *arg) { unsigned long address, start_page, end_page; @@ -519,7 +519,7 @@ ia64_pfn_valid (unsigned long pfn) } EXPORT_SYMBOL(ia64_pfn_valid); -int +int __init find_largest_hole (u64 start, u64 end, void *arg) { u64 *max_gap = arg; @@ -535,7 +535,7 @@ find_largest_hole (u64 start, u64 end, void *arg) } #endif /* CONFIG_VIRTUAL_MEM_MAP */ -static int +static int __init count_reserved_pages (u64 start, u64 end, void *arg) { unsigned long num_reserved = 0; @@ -567,7 +567,7 @@ nolwsys_setup (char *s) __setup("nolwsys", nolwsys_setup); -void +void __init mem_init (void) { long reserved_pages, codesize, datasize, initsize; -- cgit v1.2.3-70-g09d2 From 914a4ea44169c8420d2767862755f34f76772819 Mon Sep 17 00:00:00 2001 From: "Chen, Kenneth W" Date: Sun, 12 Mar 2006 09:08:26 -0800 Subject: [IA64] add init declaration - gate page functions Add init declaration to bunch of patch functions and gate page setup function. Signed-off-by: Ken Chen Signed-off-by: Tony Luck --- arch/ia64/kernel/patch.c | 8 ++++---- arch/ia64/mm/init.c | 2 +- 2 files changed, 5 insertions(+), 5 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/patch.c b/arch/ia64/kernel/patch.c index 6a4ac7d70b3..bc11bb096f5 100644 --- a/arch/ia64/kernel/patch.c +++ b/arch/ia64/kernel/patch.c @@ -115,7 +115,7 @@ ia64_patch_vtop (unsigned long start, unsigned long end) ia64_srlz_i(); } -void +void __init ia64_patch_mckinley_e9 (unsigned long start, unsigned long end) { static int first_time = 1; @@ -149,7 +149,7 @@ ia64_patch_mckinley_e9 (unsigned long start, unsigned long end) ia64_srlz_i(); } -static void +static void __init patch_fsyscall_table (unsigned long start, unsigned long end) { extern unsigned long fsyscall_table[NR_syscalls]; @@ -166,7 +166,7 @@ patch_fsyscall_table (unsigned long start, unsigned long end) ia64_srlz_i(); } -static void +static void __init patch_brl_fsys_bubble_down (unsigned long start, unsigned long end) { extern char fsys_bubble_down[]; @@ -184,7 +184,7 @@ patch_brl_fsys_bubble_down (unsigned long start, unsigned long end) ia64_srlz_i(); } -void +void __init ia64_patch_gate (void) { # define START(name) ((unsigned long) __start_gate_##name##_patchlist) diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c index 3748768c787..e3a10002d64 100644 --- a/arch/ia64/mm/init.c +++ b/arch/ia64/mm/init.c @@ -294,7 +294,7 @@ put_kernel_page (struct page *page, unsigned long address, pgprot_t pgprot) return page; } -static void +static void __init setup_gate (void) { struct page *page; -- cgit v1.2.3-70-g09d2 From 03906ea0345d3ae6e906495f86846afa4b8b69b6 Mon Sep 17 00:00:00 2001 From: "Chen, Kenneth W" Date: Sun, 12 Mar 2006 09:10:59 -0800 Subject: [IA64] add init declaration - nolwsys Add __initdata to nolwsys. Signed-off-by: Ken Chen Signed-off-by: Tony Luck --- arch/ia64/mm/init.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/ia64') diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c index e3a10002d64..ff4f31fcd33 100644 --- a/arch/ia64/mm/init.c +++ b/arch/ia64/mm/init.c @@ -556,7 +556,7 @@ count_reserved_pages (u64 start, u64 end, void *arg) * purposes. */ -static int nolwsys; +static int nolwsys __initdata; static int __init nolwsys_setup (char *s) -- cgit v1.2.3-70-g09d2 From 39e18de810049e836390830174a248ad241b156a Mon Sep 17 00:00:00 2001 From: "Chen, Kenneth W" Date: Sun, 12 Mar 2006 09:20:27 -0800 Subject: [IA64] move patchlist and machvec into init section ia64_mv is initialized based on platform detected or specified. However, there is one instantiation of each platform type. We don't expect to switch platform vector during run time. Move those platform specific type into init section since a copy is made into global ia64_mv at initialization. Also move instruction patch list into init section as well. Signed-off-by: Ken Chen Signed-off-by: Tony Luck --- arch/ia64/kernel/vmlinux.lds.S | 51 +++++++++++++++++++++--------------------- 1 file changed, 26 insertions(+), 25 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/vmlinux.lds.S b/arch/ia64/kernel/vmlinux.lds.S index 73af6267d2e..632d65cc068 100644 --- a/arch/ia64/kernel/vmlinux.lds.S +++ b/arch/ia64/kernel/vmlinux.lds.S @@ -70,34 +70,9 @@ SECTIONS __stop___ex_table = .; } - .data.patch.vtop : AT(ADDR(.data.patch.vtop) - LOAD_OFFSET) - { - __start___vtop_patchlist = .; - *(.data.patch.vtop) - __end___vtop_patchlist = .; - } - - .data.patch.mckinley_e9 : AT(ADDR(.data.patch.mckinley_e9) - LOAD_OFFSET) - { - __start___mckinley_e9_bundles = .; - *(.data.patch.mckinley_e9) - __end___mckinley_e9_bundles = .; - } - /* Global data */ _data = .; -#if defined(CONFIG_IA64_GENERIC) - /* Machine Vector */ - . = ALIGN(16); - .machvec : AT(ADDR(.machvec) - LOAD_OFFSET) - { - machvec_start = .; - *(.machvec) - machvec_end = .; - } -#endif - /* Unwind info & table: */ . = ALIGN(8); .IA_64.unwind_info : AT(ADDR(.IA_64.unwind_info) - LOAD_OFFSET) @@ -154,6 +129,32 @@ SECTIONS *(.initcall7.init) __initcall_end = .; } + + .data.patch.vtop : AT(ADDR(.data.patch.vtop) - LOAD_OFFSET) + { + __start___vtop_patchlist = .; + *(.data.patch.vtop) + __end___vtop_patchlist = .; + } + + .data.patch.mckinley_e9 : AT(ADDR(.data.patch.mckinley_e9) - LOAD_OFFSET) + { + __start___mckinley_e9_bundles = .; + *(.data.patch.mckinley_e9) + __end___mckinley_e9_bundles = .; + } + +#if defined(CONFIG_IA64_GENERIC) + /* Machine Vector */ + . = ALIGN(16); + .machvec : AT(ADDR(.machvec) - LOAD_OFFSET) + { + machvec_start = .; + *(.machvec) + machvec_end = .; + } +#endif + __con_initcall_start = .; .con_initcall.init : AT(ADDR(.con_initcall.init) - LOAD_OFFSET) { *(.con_initcall.init) } -- cgit v1.2.3-70-g09d2 From 41c28ff1635e71af072c4711ff5fadd5855d48e7 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 23 Mar 2006 03:00:56 -0800 Subject: [PATCH] kill _INLINE_ This patch removes all occurances of _INLINE_ in the kernel. With the exception of tty_flip.h, I've simply removed the inline's since gcc should know best which functions to be inlined. Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/hp/sim/simserial.c | 7 +--- arch/xtensa/platform-iss/console.c | 4 --- drivers/char/amiserial.c | 18 ++++------ drivers/isdn/hisax/config.c | 1 - drivers/isdn/hisax/elsa.c | 1 - drivers/serial/68328serial.c | 9 +++-- drivers/serial/au1x00_uart.c | 11 +++--- drivers/serial/crisv10.c | 68 ++++++++++++-------------------------- drivers/serial/m32r_sio.c | 15 ++++----- drivers/serial/sunsu.c | 13 +++----- drivers/tc/zs.c | 9 ++--- include/linux/tty_flip.h | 12 ++----- 12 files changed, 56 insertions(+), 112 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/hp/sim/simserial.c b/arch/ia64/hp/sim/simserial.c index 626cdc83668..0e5c6ae5022 100644 --- a/arch/ia64/hp/sim/simserial.c +++ b/arch/ia64/hp/sim/simserial.c @@ -46,11 +46,6 @@ #define KEYBOARD_INTR 3 /* must match with simulator! */ #define NR_PORTS 1 /* only one port for now */ -#define SERIAL_INLINE 1 - -#ifdef SERIAL_INLINE -#define _INLINE_ inline -#endif #define IRQ_T(info) ((info->flags & ASYNC_SHARE_IRQ) ? SA_SHIRQ : SA_INTERRUPT) @@ -237,7 +232,7 @@ static void rs_put_char(struct tty_struct *tty, unsigned char ch) local_irq_restore(flags); } -static _INLINE_ void transmit_chars(struct async_struct *info, int *intr_done) +static void transmit_chars(struct async_struct *info, int *intr_done) { int count; unsigned long flags; diff --git a/arch/xtensa/platform-iss/console.c b/arch/xtensa/platform-iss/console.c index 94fdfe474ac..2a580efb58e 100644 --- a/arch/xtensa/platform-iss/console.c +++ b/arch/xtensa/platform-iss/console.c @@ -31,10 +31,6 @@ #include #include -#ifdef SERIAL_INLINE -#define _INLINE_ inline -#endif - #define SERIAL_MAX_NUM_LINES 1 #define SERIAL_TIMER_VALUE (20 * HZ) diff --git a/drivers/char/amiserial.c b/drivers/char/amiserial.c index 7ac365b5d9e..6602b3156df 100644 --- a/drivers/char/amiserial.c +++ b/drivers/char/amiserial.c @@ -46,8 +46,6 @@ /* Sanity checks */ -#define SERIAL_INLINE - #if defined(MODULE) && defined(SERIAL_DEBUG_MCOUNT) #define DBG_CNT(s) printk("(%s): [%x] refc=%d, serc=%d, ttyc=%d -> %s\n", \ tty->name, (info->flags), serial_driver->refcount,info->count,tty->count,s) @@ -95,10 +93,6 @@ static char *serial_version = "4.30"; #include #include -#ifdef SERIAL_INLINE -#define _INLINE_ inline -#endif - #define custom amiga_custom static char *serial_name = "Amiga-builtin serial driver"; @@ -253,14 +247,14 @@ static void rs_start(struct tty_struct *tty) * This routine is used by the interrupt handler to schedule * processing in the software interrupt portion of the driver. */ -static _INLINE_ void rs_sched_event(struct async_struct *info, - int event) +static void rs_sched_event(struct async_struct *info, + int event) { info->event |= 1 << event; tasklet_schedule(&info->tlet); } -static _INLINE_ void receive_chars(struct async_struct *info) +static void receive_chars(struct async_struct *info) { int status; int serdatr; @@ -349,7 +343,7 @@ out: return; } -static _INLINE_ void transmit_chars(struct async_struct *info) +static void transmit_chars(struct async_struct *info) { custom.intreq = IF_TBE; mb(); @@ -389,7 +383,7 @@ static _INLINE_ void transmit_chars(struct async_struct *info) } } -static _INLINE_ void check_modem_status(struct async_struct *info) +static void check_modem_status(struct async_struct *info) { unsigned char status = ciab.pra & (SER_DCD | SER_CTS | SER_DSR); unsigned char dstatus; @@ -1959,7 +1953,7 @@ done: * number, and identifies which options were configured into this * driver. */ -static _INLINE_ void show_serial_version(void) +static void show_serial_version(void) { printk(KERN_INFO "%s version %s\n", serial_name, serial_version); } diff --git a/drivers/isdn/hisax/config.c b/drivers/isdn/hisax/config.c index df9d6520181..27332506f9f 100644 --- a/drivers/isdn/hisax/config.c +++ b/drivers/isdn/hisax/config.c @@ -25,7 +25,6 @@ #include #include #define HISAX_STATUS_BUFSIZE 4096 -#define INCLUDE_INLINE_FUNCS /* * This structure array contains one entry per card. An entry looks diff --git a/drivers/isdn/hisax/elsa.c b/drivers/isdn/hisax/elsa.c index 110e9fd669c..f8ca4b32333 100644 --- a/drivers/isdn/hisax/elsa.c +++ b/drivers/isdn/hisax/elsa.c @@ -108,7 +108,6 @@ static const char *ITACVer[] = #define ELSA_ASSIGN 4 #define RS_ISR_PASS_LIMIT 256 -#define _INLINE_ inline #define FLG_MODEM_ACTIVE 1 /* IPAC AUX */ #define ELSA_IPAC_LINE_LED 0x40 /* Bit 6 Gelbe LED */ diff --git a/drivers/serial/68328serial.c b/drivers/serial/68328serial.c index 7f0f35a05dc..b88a7c1158a 100644 --- a/drivers/serial/68328serial.c +++ b/drivers/serial/68328serial.c @@ -101,8 +101,6 @@ struct tty_driver *serial_driver; #define RS_ISR_PASS_LIMIT 256 -#define _INLINE_ inline - static void change_speed(struct m68k_serial *info); /* @@ -262,7 +260,7 @@ static void batten_down_hatches(void) /* Drop into the debugger */ } -static _INLINE_ void status_handle(struct m68k_serial *info, unsigned short status) +static void status_handle(struct m68k_serial *info, unsigned short status) { #if 0 if(status & DCD) { @@ -289,7 +287,8 @@ static _INLINE_ void status_handle(struct m68k_serial *info, unsigned short stat return; } -static _INLINE_ void receive_chars(struct m68k_serial *info, struct pt_regs *regs, unsigned short rx) +static void receive_chars(struct m68k_serial *info, struct pt_regs *regs, + unsigned short rx) { struct tty_struct *tty = info->tty; m68328_uart *uart = &uart_addr[info->line]; @@ -359,7 +358,7 @@ clear_and_exit: return; } -static _INLINE_ void transmit_chars(struct m68k_serial *info) +static void transmit_chars(struct m68k_serial *info) { m68328_uart *uart = &uart_addr[info->line]; diff --git a/drivers/serial/au1x00_uart.c b/drivers/serial/au1x00_uart.c index 29f94bbb79b..948880ac587 100644 --- a/drivers/serial/au1x00_uart.c +++ b/drivers/serial/au1x00_uart.c @@ -133,13 +133,12 @@ static const struct serial_uart_config uart_config[PORT_MAX_8250+1] = { { "AU1X00_UART",16, UART_CLEAR_FIFO | UART_USE_FIFO }, }; -static _INLINE_ unsigned int serial_in(struct uart_8250_port *up, int offset) +static unsigned int serial_in(struct uart_8250_port *up, int offset) { return au_readl((unsigned long)up->port.membase + offset); } -static _INLINE_ void -serial_out(struct uart_8250_port *up, int offset, int value) +static void serial_out(struct uart_8250_port *up, int offset, int value) { au_writel(value, (unsigned long)up->port.membase + offset); } @@ -237,7 +236,7 @@ static void serial8250_enable_ms(struct uart_port *port) serial_out(up, UART_IER, up->ier); } -static _INLINE_ void +static void receive_chars(struct uart_8250_port *up, int *status, struct pt_regs *regs) { struct tty_struct *tty = up->port.info->tty; @@ -312,7 +311,7 @@ receive_chars(struct uart_8250_port *up, int *status, struct pt_regs *regs) spin_lock(&up->port.lock); } -static _INLINE_ void transmit_chars(struct uart_8250_port *up) +static void transmit_chars(struct uart_8250_port *up) { struct circ_buf *xmit = &up->port.info->xmit; int count; @@ -346,7 +345,7 @@ static _INLINE_ void transmit_chars(struct uart_8250_port *up) serial8250_stop_tx(&up->port); } -static _INLINE_ void check_modem_status(struct uart_8250_port *up) +static void check_modem_status(struct uart_8250_port *up) { int status; diff --git a/drivers/serial/crisv10.c b/drivers/serial/crisv10.c index be12623d854..89700141f87 100644 --- a/drivers/serial/crisv10.c +++ b/drivers/serial/crisv10.c @@ -481,8 +481,6 @@ static char *serial_version = "$Revision: 1.25 $"; #include "serial_compat.h" #endif -#define _INLINE_ inline - struct tty_driver *serial_driver; /* serial subtype definitions */ @@ -591,8 +589,6 @@ static void rs_throttle(struct tty_struct * tty); static void rs_wait_until_sent(struct tty_struct *tty, int timeout); static int rs_write(struct tty_struct * tty, int from_user, const unsigned char *buf, int count); -extern _INLINE_ int rs_raw_write(struct tty_struct * tty, int from_user, - const unsigned char *buf, int count); #ifdef CONFIG_ETRAX_RS485 static int e100_write_rs485(struct tty_struct * tty, int from_user, const unsigned char *buf, int count); @@ -1538,8 +1534,7 @@ e100_enable_rxdma_irq(struct e100_serial *info) /* the tx DMA uses only dma_descr interrupt */ -static _INLINE_ void -e100_disable_txdma_irq(struct e100_serial *info) +static void e100_disable_txdma_irq(struct e100_serial *info) { #ifdef SERIAL_DEBUG_INTR printk("txdma_irq(%d): 0\n",info->line); @@ -1548,8 +1543,7 @@ e100_disable_txdma_irq(struct e100_serial *info) *R_IRQ_MASK2_CLR = info->irq; } -static _INLINE_ void -e100_enable_txdma_irq(struct e100_serial *info) +static void e100_enable_txdma_irq(struct e100_serial *info) { #ifdef SERIAL_DEBUG_INTR printk("txdma_irq(%d): 1\n",info->line); @@ -1558,8 +1552,7 @@ e100_enable_txdma_irq(struct e100_serial *info) *R_IRQ_MASK2_SET = info->irq; } -static _INLINE_ void -e100_disable_txdma_channel(struct e100_serial *info) +static void e100_disable_txdma_channel(struct e100_serial *info) { unsigned long flags; @@ -1599,8 +1592,7 @@ e100_disable_txdma_channel(struct e100_serial *info) } -static _INLINE_ void -e100_enable_txdma_channel(struct e100_serial *info) +static void e100_enable_txdma_channel(struct e100_serial *info) { unsigned long flags; @@ -1625,8 +1617,7 @@ e100_enable_txdma_channel(struct e100_serial *info) restore_flags(flags); } -static _INLINE_ void -e100_disable_rxdma_channel(struct e100_serial *info) +static void e100_disable_rxdma_channel(struct e100_serial *info) { unsigned long flags; @@ -1665,8 +1656,7 @@ e100_disable_rxdma_channel(struct e100_serial *info) } -static _INLINE_ void -e100_enable_rxdma_channel(struct e100_serial *info) +static void e100_enable_rxdma_channel(struct e100_serial *info) { unsigned long flags; @@ -1913,9 +1903,7 @@ rs_start(struct tty_struct *tty) * This routine is used by the interrupt handler to schedule * processing in the software interrupt portion of the driver. */ -static _INLINE_ void -rs_sched_event(struct e100_serial *info, - int event) +static void rs_sched_event(struct e100_serial *info, int event) { if (info->event & (1 << event)) return; @@ -2155,8 +2143,9 @@ add_char_and_flag(struct e100_serial *info, unsigned char data, unsigned char fl return 1; } -extern _INLINE_ unsigned int -handle_descr_data(struct e100_serial *info, struct etrax_dma_descr *descr, unsigned int recvl) +static unsigned int handle_descr_data(struct e100_serial *info, + struct etrax_dma_descr *descr, + unsigned int recvl) { struct etrax_recv_buffer *buffer = phys_to_virt(descr->buf) - sizeof *buffer; @@ -2182,8 +2171,7 @@ handle_descr_data(struct e100_serial *info, struct etrax_dma_descr *descr, unsig return recvl; } -static _INLINE_ unsigned int -handle_all_descr_data(struct e100_serial *info) +static unsigned int handle_all_descr_data(struct e100_serial *info) { struct etrax_dma_descr *descr; unsigned int recvl; @@ -2230,8 +2218,7 @@ handle_all_descr_data(struct e100_serial *info) return ret; } -static _INLINE_ void -receive_chars_dma(struct e100_serial *info) +static void receive_chars_dma(struct e100_serial *info) { struct tty_struct *tty; unsigned char rstat; @@ -2292,8 +2279,7 @@ receive_chars_dma(struct e100_serial *info) *info->icmdadr = IO_STATE(R_DMA_CH6_CMD, cmd, restart); } -static _INLINE_ int -start_recv_dma(struct e100_serial *info) +static int start_recv_dma(struct e100_serial *info) { struct etrax_dma_descr *descr = info->rec_descr; struct etrax_recv_buffer *buffer; @@ -2348,11 +2334,6 @@ start_receive(struct e100_serial *info) } -static _INLINE_ void -status_handle(struct e100_serial *info, unsigned short status) -{ -} - /* the bits in the MASK2 register are laid out like this: DMAI_EOP DMAI_DESCR DMAO_EOP DMAO_DESCR where I is the input channel and O is the output channel for the port. @@ -2454,8 +2435,7 @@ rec_interrupt(int irq, void *dev_id, struct pt_regs * regs) return IRQ_RETVAL(handled); } /* rec_interrupt */ -static _INLINE_ int -force_eop_if_needed(struct e100_serial *info) +static int force_eop_if_needed(struct e100_serial *info) { /* We check data_avail bit to determine if data has * arrived since last time @@ -2499,8 +2479,7 @@ force_eop_if_needed(struct e100_serial *info) return 1; } -extern _INLINE_ void -flush_to_flip_buffer(struct e100_serial *info) +static void flush_to_flip_buffer(struct e100_serial *info) { struct tty_struct *tty; struct etrax_recv_buffer *buffer; @@ -2611,8 +2590,7 @@ flush_to_flip_buffer(struct e100_serial *info) tty_flip_buffer_push(tty); } -static _INLINE_ void -check_flush_timeout(struct e100_serial *info) +static void check_flush_timeout(struct e100_serial *info) { /* Flip what we've got (if we can) */ flush_to_flip_buffer(info); @@ -2741,7 +2719,7 @@ TODO: The break will be delayed until an F or V character is received. */ -extern _INLINE_ +static struct e100_serial * handle_ser_rx_interrupt_no_dma(struct e100_serial *info) { unsigned long data_read; @@ -2875,8 +2853,7 @@ more_data: return info; } -extern _INLINE_ -struct e100_serial* handle_ser_rx_interrupt(struct e100_serial *info) +static struct e100_serial* handle_ser_rx_interrupt(struct e100_serial *info) { unsigned char rstat; @@ -2995,7 +2972,7 @@ struct e100_serial* handle_ser_rx_interrupt(struct e100_serial *info) return info; } /* handle_ser_rx_interrupt */ -extern _INLINE_ void handle_ser_tx_interrupt(struct e100_serial *info) +static void handle_ser_tx_interrupt(struct e100_serial *info) { unsigned long flags; @@ -3621,9 +3598,8 @@ rs_flush_chars(struct tty_struct *tty) restore_flags(flags); } -extern _INLINE_ int -rs_raw_write(struct tty_struct * tty, int from_user, - const unsigned char *buf, int count) +static int rs_raw_write(struct tty_struct * tty, int from_user, + const unsigned char *buf, int count) { int c, ret = 0; struct e100_serial *info = (struct e100_serial *)tty->driver_data; @@ -4710,7 +4686,7 @@ rs_open(struct tty_struct *tty, struct file * filp) * /proc fs routines.... */ -extern _INLINE_ int line_info(char *buf, struct e100_serial *info) +static int line_info(char *buf, struct e100_serial *info) { char stat_buf[30]; int ret; diff --git a/drivers/serial/m32r_sio.c b/drivers/serial/m32r_sio.c index 876bc5e027b..e9c10c0a30f 100644 --- a/drivers/serial/m32r_sio.c +++ b/drivers/serial/m32r_sio.c @@ -248,17 +248,17 @@ static void sio_error(int *status) #endif /* CONFIG_SERIAL_M32R_PLDSIO */ -static _INLINE_ unsigned int sio_in(struct uart_sio_port *up, int offset) +static unsigned int sio_in(struct uart_sio_port *up, int offset) { return __sio_in(up->port.iobase + offset); } -static _INLINE_ void sio_out(struct uart_sio_port *up, int offset, int value) +static void sio_out(struct uart_sio_port *up, int offset, int value) { __sio_out(value, up->port.iobase + offset); } -static _INLINE_ unsigned int serial_in(struct uart_sio_port *up, int offset) +static unsigned int serial_in(struct uart_sio_port *up, int offset) { if (!offset) return 0; @@ -266,8 +266,7 @@ static _INLINE_ unsigned int serial_in(struct uart_sio_port *up, int offset) return __sio_in(offset); } -static _INLINE_ void -serial_out(struct uart_sio_port *up, int offset, int value) +static void serial_out(struct uart_sio_port *up, int offset, int value) { if (!offset) return; @@ -326,8 +325,8 @@ static void m32r_sio_enable_ms(struct uart_port *port) serial_out(up, UART_IER, up->ier); } -static _INLINE_ void receive_chars(struct uart_sio_port *up, int *status, - struct pt_regs *regs) +static void receive_chars(struct uart_sio_port *up, int *status, + struct pt_regs *regs) { struct tty_struct *tty = up->port.info->tty; unsigned char ch; @@ -400,7 +399,7 @@ static _INLINE_ void receive_chars(struct uart_sio_port *up, int *status, tty_flip_buffer_push(tty); } -static _INLINE_ void transmit_chars(struct uart_sio_port *up) +static void transmit_chars(struct uart_sio_port *up) { struct circ_buf *xmit = &up->port.info->xmit; int count; diff --git a/drivers/serial/sunsu.c b/drivers/serial/sunsu.c index 7fc3d3b41d1..9fe2283d91e 100644 --- a/drivers/serial/sunsu.c +++ b/drivers/serial/sunsu.c @@ -102,9 +102,7 @@ struct uart_sunsu_port { #endif }; -#define _INLINE_ - -static _INLINE_ unsigned int serial_in(struct uart_sunsu_port *up, int offset) +static unsigned int serial_in(struct uart_sunsu_port *up, int offset) { offset <<= up->port.regshift; @@ -121,8 +119,7 @@ static _INLINE_ unsigned int serial_in(struct uart_sunsu_port *up, int offset) } } -static _INLINE_ void -serial_out(struct uart_sunsu_port *up, int offset, int value) +static void serial_out(struct uart_sunsu_port *up, int offset, int value) { #ifndef CONFIG_SPARC64 /* @@ -316,7 +313,7 @@ static void sunsu_enable_ms(struct uart_port *port) spin_unlock_irqrestore(&up->port.lock, flags); } -static _INLINE_ struct tty_struct * +static struct tty_struct * receive_chars(struct uart_sunsu_port *up, unsigned char *status, struct pt_regs *regs) { struct tty_struct *tty = up->port.info->tty; @@ -395,7 +392,7 @@ receive_chars(struct uart_sunsu_port *up, unsigned char *status, struct pt_regs return tty; } -static _INLINE_ void transmit_chars(struct uart_sunsu_port *up) +static void transmit_chars(struct uart_sunsu_port *up) { struct circ_buf *xmit = &up->port.info->xmit; int count; @@ -431,7 +428,7 @@ static _INLINE_ void transmit_chars(struct uart_sunsu_port *up) __stop_tx(up); } -static _INLINE_ void check_modem_status(struct uart_sunsu_port *up) +static void check_modem_status(struct uart_sunsu_port *up) { int status; diff --git a/drivers/tc/zs.c b/drivers/tc/zs.c index 6756d0fab6f..2dffa8e303b 100644 --- a/drivers/tc/zs.c +++ b/drivers/tc/zs.c @@ -186,8 +186,6 @@ static struct tty_driver *serial_driver; #define RS_STROBE_TIME 10 #define RS_ISR_PASS_LIMIT 256 -#define _INLINE_ inline - static void probe_sccs(void); static void change_speed(struct dec_serial *info); static void rs_wait_until_sent(struct tty_struct *tty, int timeout); @@ -344,14 +342,13 @@ static inline void rs_recv_clear(struct dec_zschannel *zsc) * This routine is used by the interrupt handler to schedule * processing in the software interrupt portion of the driver. */ -static _INLINE_ void rs_sched_event(struct dec_serial *info, int event) +static void rs_sched_event(struct dec_serial *info, int event) { info->event |= 1 << event; tasklet_schedule(&info->tlet); } -static _INLINE_ void receive_chars(struct dec_serial *info, - struct pt_regs *regs) +static void receive_chars(struct dec_serial *info, struct pt_regs *regs) { struct tty_struct *tty = info->tty; unsigned char ch, stat, flag; @@ -441,7 +438,7 @@ static void transmit_chars(struct dec_serial *info) rs_sched_event(info, RS_EVENT_WRITE_WAKEUP); } -static _INLINE_ void status_handle(struct dec_serial *info) +static void status_handle(struct dec_serial *info) { unsigned char stat; diff --git a/include/linux/tty_flip.h b/include/linux/tty_flip.h index 222faf97d5f..0c6169fff36 100644 --- a/include/linux/tty_flip.h +++ b/include/linux/tty_flip.h @@ -7,14 +7,8 @@ extern int tty_insert_flip_string_flags(struct tty_struct *tty, unsigned char *c extern int tty_prepare_flip_string(struct tty_struct *tty, unsigned char **chars, size_t size); extern int tty_prepare_flip_string_flags(struct tty_struct *tty, unsigned char **chars, char **flags, size_t size); -#ifdef INCLUDE_INLINE_FUNCS -#define _INLINE_ extern -#else -#define _INLINE_ static __inline__ -#endif - -_INLINE_ int tty_insert_flip_char(struct tty_struct *tty, - unsigned char ch, char flag) +static inline int tty_insert_flip_char(struct tty_struct *tty, + unsigned char ch, char flag) { struct tty_buffer *tb = tty->buf.tail; if (tb && tb->active && tb->used < tb->size) { @@ -25,7 +19,7 @@ _INLINE_ int tty_insert_flip_char(struct tty_struct *tty, return tty_insert_flip_string_flags(tty, &ch, &flag, 1); } -_INLINE_ void tty_schedule_flip(struct tty_struct *tty) +static inline void tty_schedule_flip(struct tty_struct *tty) { unsigned long flags; spin_lock_irqsave(&tty->buf.lock, flags); -- cgit v1.2.3-70-g09d2 From a5b00bb4fe60796c791238cf5653b82110031c93 Mon Sep 17 00:00:00 2001 From: Horms Date: Thu, 23 Mar 2006 14:27:12 -0800 Subject: IA64: Use early_parm to handle mvec_name and nomca I'm not sure of the worthiness of this idea, so please consider it an RFC. Its key merits are: * Reuse existing infrastructure * Greatly tightens up the parsing of nomca * Greatly simplifies the parsing of machvec Addition cleanup (moving setup_mvec() to machvec.c) by Ken Chen. Signed-Off-By: Horms Signed-Off-By: Tony Luck --- arch/ia64/kernel/machvec.c | 19 +++++++++++++++---- arch/ia64/kernel/setup.c | 36 ++++++++++++------------------------ 2 files changed, 27 insertions(+), 28 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/machvec.c b/arch/ia64/kernel/machvec.c index c3a04ee7f4f..4b0b71d5aef 100644 --- a/arch/ia64/kernel/machvec.c +++ b/arch/ia64/kernel/machvec.c @@ -14,7 +14,15 @@ struct ia64_machine_vector ia64_mv; EXPORT_SYMBOL(ia64_mv); -static struct ia64_machine_vector * +static __initdata const char *mvec_name; +static __init int setup_mvec(char *s) +{ + mvec_name = s; + return 0; +} +early_param("machvec", setup_mvec); + +static struct ia64_machine_vector * __init lookup_machvec (const char *name) { extern struct ia64_machine_vector machvec_start[]; @@ -33,10 +41,13 @@ machvec_init (const char *name) { struct ia64_machine_vector *mv; + if (!name) + name = mvec_name ? mvec_name : acpi_get_sysname(); mv = lookup_machvec(name); - if (!mv) { - panic("generic kernel failed to find machine vector for platform %s!", name); - } + if (!mv) + panic("generic kernel failed to find machine vector for" + " platform %s!", name); + ia64_mv = *mv; printk(KERN_INFO "booting generic kernel on platform %s\n", name); } diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c index eb0793628b8..fc860778fa4 100644 --- a/arch/ia64/kernel/setup.c +++ b/arch/ia64/kernel/setup.c @@ -389,6 +389,14 @@ check_for_logical_procs (void) } #endif +static __initdata int nomca; +static __init int setup_nomca(char *s) +{ + nomca = 1; + return 0; +} +early_param("nomca", setup_nomca); + void __init setup_arch (char **cmdline_p) { @@ -402,35 +410,15 @@ setup_arch (char **cmdline_p) efi_init(); io_port_init(); + parse_early_param(); + #ifdef CONFIG_IA64_GENERIC - { - const char *mvec_name = strstr (*cmdline_p, "machvec="); - char str[64]; - - if (mvec_name) { - const char *end; - size_t len; - - mvec_name += 8; - end = strchr (mvec_name, ' '); - if (end) - len = end - mvec_name; - else - len = strlen (mvec_name); - len = min(len, sizeof (str) - 1); - strncpy (str, mvec_name, len); - str[len] = '\0'; - mvec_name = str; - } else - mvec_name = acpi_get_sysname(); - machvec_init(mvec_name); - } + machvec_init(NULL); #endif if (early_console_setup(*cmdline_p) == 0) mark_bsp_online(); - parse_early_param(); #ifdef CONFIG_ACPI /* Initialize the ACPI boot-time table parser */ acpi_table_init(); @@ -493,7 +481,7 @@ setup_arch (char **cmdline_p) #endif /* enable IA-64 Machine Check Abort Handling unless disabled */ - if (!strstr(saved_command_line, "nomca")) + if (!nomca) ia64_mca_init(); platform_setup(cmdline_p); -- cgit v1.2.3-70-g09d2 From cdb0452789d365695b5b173542af9c7e3d24f185 Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Fri, 24 Mar 2006 03:15:57 -0800 Subject: [PATCH] kill include/linux/platform.h, default_idle() cleanup include/linux/platform.h contained nothing that was actually used except the default_idle() prototype, and is therefore removed by this patch. This patch does the following with the platform specific default_idle() functions on different architectures: - remove the unused function: - parisc - sparc64 - make the needlessly global function static: - arm - h8300 - m68k - m68knommu - s390 - v850 - x86_64 - add a prototype in asm/system.h: - cris - i386 - ia64 Signed-off-by: Adrian Bunk Acked-by: Patrick Mochel Acked-by: Kyle McMartin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/cris/kernel/process.c | 3 +-- arch/h8300/kernel/process.c | 4 ++-- arch/i386/kernel/apm.c | 2 -- arch/i386/mach-visws/reboot.c | 1 - arch/ia64/kernel/setup.c | 1 - arch/m68k/kernel/process.c | 2 +- arch/m68knommu/kernel/process.c | 2 +- arch/parisc/kernel/process.c | 5 ----- arch/s390/kernel/process.c | 2 +- arch/sh/kernel/process.c | 1 - arch/v850/kernel/process.c | 2 +- arch/x86_64/kernel/process.c | 2 +- include/asm-cris/system.h | 2 ++ include/asm-i386/system.h | 2 ++ include/asm-ia64/system.h | 2 ++ include/linux/platform.h | 43 ----------------------------------------- 16 files changed, 14 insertions(+), 62 deletions(-) delete mode 100644 include/linux/platform.h (limited to 'arch/ia64') diff --git a/arch/cris/kernel/process.c b/arch/cris/kernel/process.c index 4ab3e87115b..123451c4415 100644 --- a/arch/cris/kernel/process.c +++ b/arch/cris/kernel/process.c @@ -116,6 +116,7 @@ #include #include #include +#include #include #include #include @@ -194,8 +195,6 @@ EXPORT_SYMBOL(enable_hlt); */ void (*pm_idle)(void); -extern void default_idle(void); - /* * The idle thread. There's no useful work to be * done, so just try to conserve power and have a diff --git a/arch/h8300/kernel/process.c b/arch/h8300/kernel/process.c index dd344f112cf..16ccddc69c2 100644 --- a/arch/h8300/kernel/process.c +++ b/arch/h8300/kernel/process.c @@ -54,7 +54,7 @@ asmlinkage void ret_from_fork(void); * The idle loop on an H8/300.. */ #if !defined(CONFIG_H8300H_SIM) && !defined(CONFIG_H8S_SIM) -void default_idle(void) +static void default_idle(void) { local_irq_disable(); if (!need_resched()) { @@ -65,7 +65,7 @@ void default_idle(void) local_irq_enable(); } #else -void default_idle(void) +static void default_idle(void) { cpu_relax(); } diff --git a/arch/i386/kernel/apm.c b/arch/i386/kernel/apm.c index 05312a8abb8..da30a374dd4 100644 --- a/arch/i386/kernel/apm.c +++ b/arch/i386/kernel/apm.c @@ -824,8 +824,6 @@ static void apm_do_busy(void) static void (*original_pm_idle)(void); -extern void default_idle(void); - /** * apm_cpu_idle - cpu idling for APM capable Linux * diff --git a/arch/i386/mach-visws/reboot.c b/arch/i386/mach-visws/reboot.c index 5d73e042ed0..99332abfad4 100644 --- a/arch/i386/mach-visws/reboot.c +++ b/arch/i386/mach-visws/reboot.c @@ -1,7 +1,6 @@ #include #include #include -#include #include #include "piix4.h" diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c index 3258e09278d..958c1508036 100644 --- a/arch/ia64/kernel/setup.c +++ b/arch/ia64/kernel/setup.c @@ -41,7 +41,6 @@ #include #include #include -#include #include #include diff --git a/arch/m68k/kernel/process.c b/arch/m68k/kernel/process.c index 2d8ad0727b6..33648efb772 100644 --- a/arch/m68k/kernel/process.c +++ b/arch/m68k/kernel/process.c @@ -77,7 +77,7 @@ unsigned long thread_saved_pc(struct task_struct *tsk) /* * The idle loop on an m68k.. */ -void default_idle(void) +static void default_idle(void) { if (!need_resched()) #if defined(MACH_ATARI_ONLY) && !defined(CONFIG_HADES) diff --git a/arch/m68knommu/kernel/process.c b/arch/m68knommu/kernel/process.c index 63c117dae0c..f861755ec88 100644 --- a/arch/m68knommu/kernel/process.c +++ b/arch/m68knommu/kernel/process.c @@ -51,7 +51,7 @@ EXPORT_SYMBOL(pm_power_off); /* * The idle loop on an m68knommu.. */ -void default_idle(void) +static void default_idle(void) { local_irq_disable(); while (!need_resched()) { diff --git a/arch/parisc/kernel/process.c b/arch/parisc/kernel/process.c index e8dea417711..0b485ef4be8 100644 --- a/arch/parisc/kernel/process.c +++ b/arch/parisc/kernel/process.c @@ -54,11 +54,6 @@ #include #include -void default_idle(void) -{ - barrier(); -} - /* * The idle thread. There's no useful work to be * done, so just try to conserve power and have a diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c index da6fbae8df9..99182a415fe 100644 --- a/arch/s390/kernel/process.c +++ b/arch/s390/kernel/process.c @@ -103,7 +103,7 @@ extern void s390_handle_mcck(void); /* * The idle loop on a S390... */ -void default_idle(void) +static void default_idle(void) { int cpu, rc; diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c index 9fd1723e621..22dc9c21201 100644 --- a/arch/sh/kernel/process.c +++ b/arch/sh/kernel/process.c @@ -19,7 +19,6 @@ #include #include #include -#include #include #include diff --git a/arch/v850/kernel/process.c b/arch/v850/kernel/process.c index 621111ddf90..57218c76925 100644 --- a/arch/v850/kernel/process.c +++ b/arch/v850/kernel/process.c @@ -37,7 +37,7 @@ extern void ret_from_fork (void); /* The idle loop. */ -void default_idle (void) +static void default_idle (void) { while (! need_resched ()) asm ("halt; nop; nop; nop; nop; nop" ::: "cc"); diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c index 22a05dec81a..80a8f307917 100644 --- a/arch/x86_64/kernel/process.c +++ b/arch/x86_64/kernel/process.c @@ -114,7 +114,7 @@ void exit_idle(void) * We use this if we don't have any better * idle routine.. */ -void default_idle(void) +static void default_idle(void) { local_irq_enable(); diff --git a/include/asm-cris/system.h b/include/asm-cris/system.h index d48670107a8..1d63c2aa8ec 100644 --- a/include/asm-cris/system.h +++ b/include/asm-cris/system.h @@ -71,4 +71,6 @@ static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int siz #define arch_align_stack(x) (x) +void default_idle(void); + #endif diff --git a/include/asm-i386/system.h b/include/asm-i386/system.h index d0d8d7448d8..19cc79c9a35 100644 --- a/include/asm-i386/system.h +++ b/include/asm-i386/system.h @@ -499,4 +499,6 @@ static inline void sched_cacheflush(void) extern unsigned long arch_align_stack(unsigned long sp); extern void free_init_pages(char *what, unsigned long begin, unsigned long end); +void default_idle(void); + #endif diff --git a/include/asm-ia64/system.h b/include/asm-ia64/system.h index cd4233d66f1..2f362059368 100644 --- a/include/asm-ia64/system.h +++ b/include/asm-ia64/system.h @@ -265,6 +265,8 @@ void sched_cacheflush(void); #define arch_align_stack(x) (x) +void default_idle(void); + #endif /* __KERNEL__ */ #endif /* __ASSEMBLY__ */ diff --git a/include/linux/platform.h b/include/linux/platform.h deleted file mode 100644 index 3c33084a6ec..00000000000 --- a/include/linux/platform.h +++ /dev/null @@ -1,43 +0,0 @@ -/* - * include/linux/platform.h - platform driver definitions - * - * Because of the prolific consumerism of the average American, - * and the dominant marketing budgets of PC OEMs, we have been - * blessed with frequent updates of the PC architecture. - * - * While most of these calls are singular per architecture, they - * require an extra layer of abstraction on the x86 so the right - * subsystem gets the right call. - * - * Basically, this consolidates the power off and reboot callbacks - * into one structure, as well as adding power management hooks. - * - * When adding a platform driver, please make sure all callbacks are - * filled. There are defaults defined below that do nothing; use those - * if you do not support that callback. - */ - -#ifndef _PLATFORM_H_ -#define _PLATFORM_H_ -#ifdef __KERNEL__ - -#include - -struct platform_t { - char * name; - u32 suspend_states; - void (*reboot)(char * cmd); - void (*halt)(void); - void (*power_off)(void); - int (*suspend)(int state, int flags); - void (*idle)(void); -}; - -extern struct platform_t * platform; -extern void default_reboot(char * cmd); -extern void default_halt(void); -extern int default_suspend(int state, int flags); -extern void default_idle(void); - -#endif /* __KERNEL__ */ -#endif /* _PLATFORM_H */ -- cgit v1.2.3-70-g09d2 From 53b3531bbbf70ac7551b32d1acc229d94de52658 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Fri, 24 Mar 2006 03:16:13 -0800 Subject: [PATCH] s/;;/;/g Signed-off-by: Alexey Dobriyan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/mach-pxa/leds-mainstone.c | 6 +++--- arch/arm/mach-s3c2410/cpu.c | 2 +- arch/frv/kernel/gdb-stub.c | 2 +- arch/ia64/sn/kernel/bte.c | 2 +- arch/ia64/sn/pci/tioca_provider.c | 2 +- arch/mips/mm/dma-ip32.c | 6 +++--- arch/ppc/syslib/ppc85xx_setup.c | 2 +- arch/sparc64/kernel/irq.c | 2 +- drivers/char/pcmcia/synclink_cs.c | 2 +- drivers/char/synclink.c | 2 +- drivers/char/synclink_gt.c | 2 +- drivers/char/synclinkmp.c | 2 +- drivers/hwmon/gl520sm.c | 2 +- drivers/i2c/chips/rtc8564.c | 2 +- drivers/ide/ide-dma.c | 2 +- drivers/net/sk98lin/skge.c | 2 +- drivers/net/sky2.h | 2 +- drivers/net/wireless/prism54/oid_mgt.c | 4 ++-- drivers/net/wireless/spectrum_cs.c | 2 +- drivers/scsi/megaraid/megaraid_mbox.c | 2 +- drivers/usb/image/microtek.c | 2 +- drivers/usb/input/hid-core.c | 2 +- drivers/video/aty/radeon_pm.c | 2 +- fs/cifs/cifssmb.c | 2 +- fs/pnode.c | 2 +- include/acpi/acpi_bus.h | 2 +- include/asm-ia64/sn/sn_sal.h | 2 +- net/ipv4/inet_hashtables.c | 2 +- net/ipv4/netfilter/ip_nat_standalone.c | 2 +- net/tipc/link.c | 2 +- sound/ppc/toonie.c | 4 ++-- 31 files changed, 37 insertions(+), 37 deletions(-) (limited to 'arch/ia64') diff --git a/arch/arm/mach-pxa/leds-mainstone.c b/arch/arm/mach-pxa/leds-mainstone.c index bbd3f87a9fc..c06d3d7a8dd 100644 --- a/arch/arm/mach-pxa/leds-mainstone.c +++ b/arch/arm/mach-pxa/leds-mainstone.c @@ -85,7 +85,7 @@ void mainstone_leds_event(led_event_t evt) break; case led_green_on: - hw_led_state |= D21;; + hw_led_state |= D21; break; case led_green_off: @@ -93,7 +93,7 @@ void mainstone_leds_event(led_event_t evt) break; case led_amber_on: - hw_led_state |= D22;; + hw_led_state |= D22; break; case led_amber_off: @@ -101,7 +101,7 @@ void mainstone_leds_event(led_event_t evt) break; case led_red_on: - hw_led_state |= D23;; + hw_led_state |= D23; break; case led_red_off: diff --git a/arch/arm/mach-s3c2410/cpu.c b/arch/arm/mach-s3c2410/cpu.c index 00a379334b6..70c34fcf785 100644 --- a/arch/arm/mach-s3c2410/cpu.c +++ b/arch/arm/mach-s3c2410/cpu.c @@ -146,7 +146,7 @@ void s3c24xx_set_board(struct s3c24xx_board *b) board = b; if (b->clocks_count != 0) { - struct clk **ptr = b->clocks;; + struct clk **ptr = b->clocks; for (i = b->clocks_count; i > 0; i--, ptr++) s3c24xx_register_clock(*ptr); diff --git a/arch/frv/kernel/gdb-stub.c b/arch/frv/kernel/gdb-stub.c index 8f860d9c494..508601fad07 100644 --- a/arch/frv/kernel/gdb-stub.c +++ b/arch/frv/kernel/gdb-stub.c @@ -1406,7 +1406,7 @@ void gdbstub(int sigval) __debug_frame->psr |= PSR_S; __debug_regs->brr = (__debug_frame->tbr & TBR_TT) << 12; __debug_regs->brr |= BRR_EB; - sigval = SIGXCPU;; + sigval = SIGXCPU; } LEDS(0x5002); diff --git a/arch/ia64/sn/kernel/bte.c b/arch/ia64/sn/kernel/bte.c index 1f11db470d9..e952ef4f6d9 100644 --- a/arch/ia64/sn/kernel/bte.c +++ b/arch/ia64/sn/kernel/bte.c @@ -36,7 +36,7 @@ static struct bteinfo_s *bte_if_on_node(nasid_t nasid, int interface) nodepda_t *tmp_nodepda; if (nasid_to_cnodeid(nasid) == -1) - return (struct bteinfo_s *)NULL;; + return (struct bteinfo_s *)NULL; tmp_nodepda = NODEPDA(nasid_to_cnodeid(nasid)); return &tmp_nodepda->bte_if[interface]; diff --git a/arch/ia64/sn/pci/tioca_provider.c b/arch/ia64/sn/pci/tioca_provider.c index 7571a402552..be017691296 100644 --- a/arch/ia64/sn/pci/tioca_provider.c +++ b/arch/ia64/sn/pci/tioca_provider.c @@ -377,7 +377,7 @@ tioca_dma_mapped(struct pci_dev *pdev, u64 paddr, size_t req_size) struct tioca_dmamap *ca_dmamap; void *map; unsigned long flags; - struct pcidev_info *pcidev_info = SN_PCIDEV_INFO(pdev);; + struct pcidev_info *pcidev_info = SN_PCIDEV_INFO(pdev); tioca_common = (struct tioca_common *)pcidev_info->pdi_pcibus_info; tioca_kern = (struct tioca_kernel *)tioca_common->ca_kernel_private; diff --git a/arch/mips/mm/dma-ip32.c b/arch/mips/mm/dma-ip32.c index a7e3072ff78..ec54ed0d26f 100644 --- a/arch/mips/mm/dma-ip32.c +++ b/arch/mips/mm/dma-ip32.c @@ -138,7 +138,7 @@ dma_addr_t dma_map_single(struct device *dev, void *ptr, size_t size, BUG(); } - addr = virt_to_phys(ptr)&RAM_OFFSET_MASK;; + addr = virt_to_phys(ptr)&RAM_OFFSET_MASK; if(dev == NULL) addr+=CRIME_HI_MEM_BASE; return (dma_addr_t)addr; @@ -179,7 +179,7 @@ int dma_map_sg(struct device *dev, struct scatterlist *sg, int nents, addr = (unsigned long) page_address(sg->page)+sg->offset; if (addr) __dma_sync(addr, sg->length, direction); - addr = __pa(addr)&RAM_OFFSET_MASK;; + addr = __pa(addr)&RAM_OFFSET_MASK; if(dev == NULL) addr += CRIME_HI_MEM_BASE; sg->dma_address = (dma_addr_t)addr; @@ -199,7 +199,7 @@ dma_addr_t dma_map_page(struct device *dev, struct page *page, addr = (unsigned long) page_address(page) + offset; dma_cache_wback_inv(addr, size); - addr = __pa(addr)&RAM_OFFSET_MASK;; + addr = __pa(addr)&RAM_OFFSET_MASK; if(dev == NULL) addr += CRIME_HI_MEM_BASE; diff --git a/arch/ppc/syslib/ppc85xx_setup.c b/arch/ppc/syslib/ppc85xx_setup.c index e70b34ee627..79b7089d750 100644 --- a/arch/ppc/syslib/ppc85xx_setup.c +++ b/arch/ppc/syslib/ppc85xx_setup.c @@ -235,7 +235,7 @@ mpc85xx_setup_pci2(struct pci_controller *hose) (__ilog2(MPC85XX_PCI2_UPPER_MEM - MPC85XX_PCI2_LOWER_MEM + 1) - 1); /* Setup outbound IO windows @ MPC85XX_PCI2_IO_BASE */ - pci->potar2 = (MPC85XX_PCI2_LOWER_IO >> 12) & 0x000fffff;; + pci->potar2 = (MPC85XX_PCI2_LOWER_IO >> 12) & 0x000fffff; pci->potear2 = 0x00000000; pci->powbar2 = (MPC85XX_PCI2_IO_BASE >> 12) & 0x000fffff; /* Enable, IO R/W */ diff --git a/arch/sparc64/kernel/irq.c b/arch/sparc64/kernel/irq.c index e505a4125e3..11e645c9ec5 100644 --- a/arch/sparc64/kernel/irq.c +++ b/arch/sparc64/kernel/irq.c @@ -727,7 +727,7 @@ void handler_irq(int irq, struct pt_regs *regs) } #ifdef CONFIG_BLK_DEV_FD -extern irqreturn_t floppy_interrupt(int, void *, struct pt_regs *);; +extern irqreturn_t floppy_interrupt(int, void *, struct pt_regs *); /* XXX No easy way to include asm/floppy.h XXX */ extern unsigned char *pdma_vaddr; diff --git a/drivers/char/pcmcia/synclink_cs.c b/drivers/char/pcmcia/synclink_cs.c index 8a8ca32822b..e6b714b6390 100644 --- a/drivers/char/pcmcia/synclink_cs.c +++ b/drivers/char/pcmcia/synclink_cs.c @@ -4181,7 +4181,7 @@ static int hdlcdev_attach(struct net_device *dev, unsigned short encoding, } info->params.encoding = new_encoding; - info->params.crc_type = new_crctype;; + info->params.crc_type = new_crctype; /* if network interface up, reprogram hardware */ if (info->netcount) diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c index ede688a4e14..d68be61f0a4 100644 --- a/drivers/char/synclink.c +++ b/drivers/char/synclink.c @@ -7770,7 +7770,7 @@ static int hdlcdev_attach(struct net_device *dev, unsigned short encoding, } info->params.encoding = new_encoding; - info->params.crc_type = new_crctype;; + info->params.crc_type = new_crctype; /* if network interface up, reprogram hardware */ if (info->netcount) diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c index b046390cd25..738ec2f4e56 100644 --- a/drivers/char/synclink_gt.c +++ b/drivers/char/synclink_gt.c @@ -1365,7 +1365,7 @@ static int hdlcdev_attach(struct net_device *dev, unsigned short encoding, } info->params.encoding = new_encoding; - info->params.crc_type = new_crctype;; + info->params.crc_type = new_crctype; /* if network interface up, reprogram hardware */ if (info->netcount) diff --git a/drivers/char/synclinkmp.c b/drivers/char/synclinkmp.c index 960adb256fb..85874013111 100644 --- a/drivers/char/synclinkmp.c +++ b/drivers/char/synclinkmp.c @@ -1650,7 +1650,7 @@ static int hdlcdev_attach(struct net_device *dev, unsigned short encoding, } info->params.encoding = new_encoding; - info->params.crc_type = new_crctype;; + info->params.crc_type = new_crctype; /* if network interface up, reprogram hardware */ if (info->netcount) diff --git a/drivers/hwmon/gl520sm.c b/drivers/hwmon/gl520sm.c index 47b4d49f75c..14e810f3c2c 100644 --- a/drivers/hwmon/gl520sm.c +++ b/drivers/hwmon/gl520sm.c @@ -456,7 +456,7 @@ static ssize_t set_temp_max(struct i2c_client *client, struct gl520_data *data, long v = simple_strtol(buf, NULL, 10); mutex_lock(&data->update_lock); - data->temp_max[n - 1] = TEMP_TO_REG(v);; + data->temp_max[n - 1] = TEMP_TO_REG(v); gl520_write_value(client, reg, data->temp_max[n - 1]); mutex_unlock(&data->update_lock); return count; diff --git a/drivers/i2c/chips/rtc8564.c b/drivers/i2c/chips/rtc8564.c index ceaa6b0bdfd..0d8699b3f48 100644 --- a/drivers/i2c/chips/rtc8564.c +++ b/drivers/i2c/chips/rtc8564.c @@ -53,7 +53,7 @@ static inline u8 _rtc8564_ctrl2(struct i2c_client *client) #define CTRL1(c) _rtc8564_ctrl1(c) #define CTRL2(c) _rtc8564_ctrl2(c) -static int debug;; +static int debug; module_param(debug, int, S_IRUGO | S_IWUSR); static struct i2c_driver rtc8564_driver; diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c index 0523da77425..c481be8b807 100644 --- a/drivers/ide/ide-dma.c +++ b/drivers/ide/ide-dma.c @@ -175,7 +175,7 @@ ide_startstop_t ide_dma_intr (ide_drive_t *drive) if (rq->rq_disk) { ide_driver_t *drv; - drv = *(ide_driver_t **)rq->rq_disk->private_data;; + drv = *(ide_driver_t **)rq->rq_disk->private_data; drv->end_request(drive, 1, rq->nr_sectors); } else ide_end_request(drive, 1, rq->nr_sectors); diff --git a/drivers/net/sk98lin/skge.c b/drivers/net/sk98lin/skge.c index a5f2b1ee075..38a26df4095 100644 --- a/drivers/net/sk98lin/skge.c +++ b/drivers/net/sk98lin/skge.c @@ -1727,7 +1727,7 @@ struct sk_buff *pMessage) /* pointer to send-message */ pTxd->VDataHigh = (SK_U32) (PhysAddr >> 32); pTxd->pMBuf = pMessage; - pTxd->TBControl = Control | BMU_OWN | sk_frag->size;; + pTxd->TBControl = Control | BMU_OWN | sk_frag->size; /* ** Do we have the last fragment? diff --git a/drivers/net/sky2.h b/drivers/net/sky2.h index 2838f661b39..62532b4e45c 100644 --- a/drivers/net/sky2.h +++ b/drivers/net/sky2.h @@ -1804,7 +1804,7 @@ struct sky2_rx_le { __le16 length; u8 ctrl; u8 opcode; -} __attribute((packed));; +} __attribute((packed)); struct sky2_status_le { __le32 status; /* also checksum */ diff --git a/drivers/net/wireless/prism54/oid_mgt.c b/drivers/net/wireless/prism54/oid_mgt.c index eea2f04c8c6..ebb23878583 100644 --- a/drivers/net/wireless/prism54/oid_mgt.c +++ b/drivers/net/wireless/prism54/oid_mgt.c @@ -332,7 +332,7 @@ mgt_le_to_cpu(int type, void *data) case OID_TYPE_ATTACH:{ struct obj_attachment *attach = data; attach->id = le16_to_cpu(attach->id); - attach->size = le16_to_cpu(attach->size);; + attach->size = le16_to_cpu(attach->size); break; } case OID_TYPE_SSID: @@ -401,7 +401,7 @@ mgt_cpu_to_le(int type, void *data) case OID_TYPE_ATTACH:{ struct obj_attachment *attach = data; attach->id = cpu_to_le16(attach->id); - attach->size = cpu_to_le16(attach->size);; + attach->size = cpu_to_le16(attach->size); break; } case OID_TYPE_SSID: diff --git a/drivers/net/wireless/spectrum_cs.c b/drivers/net/wireless/spectrum_cs.c index fee4be1ce81..5fa6fbe35bb 100644 --- a/drivers/net/wireless/spectrum_cs.c +++ b/drivers/net/wireless/spectrum_cs.c @@ -147,7 +147,7 @@ struct pdi { __le16 _len; /* length of ID and data, in words */ __le16 _id; /* record ID */ char data[0]; /* plug data */ -} __attribute__ ((packed));; +} __attribute__ ((packed)); /* Functions for access to little-endian data */ diff --git a/drivers/scsi/megaraid/megaraid_mbox.c b/drivers/scsi/megaraid/megaraid_mbox.c index bf9f7f7ba35..c11e5ce6865 100644 --- a/drivers/scsi/megaraid/megaraid_mbox.c +++ b/drivers/scsi/megaraid/megaraid_mbox.c @@ -2797,7 +2797,7 @@ mbox_post_sync_cmd(adapter_t *adapter, uint8_t raw_mbox[]) // available within 1 second, assume FW is initializing and wait // for an extended amount of time if (mbox->numstatus == 0xFF) { // status not yet available - udelay(25);; + udelay(25); for (i = 0; mbox->numstatus == 0xFF && i < 1000; i++) { rmb(); diff --git a/drivers/usb/image/microtek.c b/drivers/usb/image/microtek.c index 28538db9eaf..2a0e18a4874 100644 --- a/drivers/usb/image/microtek.c +++ b/drivers/usb/image/microtek.c @@ -360,7 +360,7 @@ static int mts_scsi_host_reset (Scsi_Cmnd *srb) rc = usb_lock_device_for_reset(desc->usb_dev, desc->usb_intf); if (rc < 0) return FAILED; - result = usb_reset_device(desc->usb_dev);; + result = usb_reset_device(desc->usb_dev); if (rc) usb_unlock_device(desc->usb_dev); return result ? FAILED : SUCCESS; diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c index 58b59f6e988..d4bf1701046 100644 --- a/drivers/usb/input/hid-core.c +++ b/drivers/usb/input/hid-core.c @@ -1215,7 +1215,7 @@ static void hid_irq_out(struct urb *urb, struct pt_regs *regs) if (hid->outhead != hid->outtail) { if (hid_submit_out(hid)) { - clear_bit(HID_OUT_RUNNING, &hid->iofl);; + clear_bit(HID_OUT_RUNNING, &hid->iofl); wake_up(&hid->wait); } spin_unlock_irqrestore(&hid->outlock, flags); diff --git a/drivers/video/aty/radeon_pm.c b/drivers/video/aty/radeon_pm.c index 1f8d805c61e..5886a2f1323 100644 --- a/drivers/video/aty/radeon_pm.c +++ b/drivers/video/aty/radeon_pm.c @@ -2080,7 +2080,7 @@ static void radeon_reinitialize_M9P(struct radeonfb_info *rinfo) OUTREG(0x2ec, 0x6332a3f0); mdelay(17); - OUTPLL(pllPPLL_REF_DIV, rinfo->pll.ref_div);; + OUTPLL(pllPPLL_REF_DIV, rinfo->pll.ref_div); OUTPLL(pllPPLL_DIV_0, rinfo->save_regs[92]); mdelay(40); diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index b41e8b37965..a243fe2792d 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -4908,7 +4908,7 @@ SetEARetry: parm_data->list_len = cpu_to_le32(count); parm_data->list[0].EA_flags = 0; /* we checked above that name len is less than 255 */ - parm_data->list[0].name_len = (__u8)name_len;; + parm_data->list[0].name_len = (__u8)name_len; /* EA names are always ASCII */ if(ea_name) strncpy(parm_data->list[0].name,ea_name,name_len); diff --git a/fs/pnode.c b/fs/pnode.c index f1871f773f6..37b568ed0e0 100644 --- a/fs/pnode.c +++ b/fs/pnode.c @@ -130,7 +130,7 @@ static struct vfsmount *get_source(struct vfsmount *dest, { struct vfsmount *p_last_src = NULL; struct vfsmount *p_last_dest = NULL; - *type = CL_PROPAGATION;; + *type = CL_PROPAGATION; if (IS_MNT_SHARED(dest)) *type |= CL_MAKE_SHARED; diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h index e496fac860a..6dca3d54208 100644 --- a/include/acpi/acpi_bus.h +++ b/include/acpi/acpi_bus.h @@ -269,7 +269,7 @@ struct acpi_device_wakeup_state { struct acpi_device_wakeup { acpi_handle gpe_device; - acpi_integer gpe_number;; + acpi_integer gpe_number; acpi_integer sleep_state; struct acpi_handle_list resources; struct acpi_device_wakeup_state state; diff --git a/include/asm-ia64/sn/sn_sal.h b/include/asm-ia64/sn/sn_sal.h index e77f0c9b7d3..b546de2fdce 100644 --- a/include/asm-ia64/sn/sn_sal.h +++ b/include/asm-ia64/sn/sn_sal.h @@ -1037,7 +1037,7 @@ ia64_sn_get_sn_info(int fc, u8 *shubtype, u16 *nasid_bitmask, u8 *nasid_shift, /***** BEGIN HACK - temp til old proms no longer supported ********/ if (ret_stuff.status == SALRET_NOT_IMPLEMENTED) { - int nasid = get_sapicid() & 0xfff;; + int nasid = get_sapicid() & 0xfff; #define SH_SHUB_ID_NODES_PER_BIT_MASK 0x001f000000000000UL #define SH_SHUB_ID_NODES_PER_BIT_SHFT 48 if (shubtype) *shubtype = 0; diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c index 33228115cda..ef7366fc132 100644 --- a/net/ipv4/inet_hashtables.c +++ b/net/ipv4/inet_hashtables.c @@ -315,7 +315,7 @@ ok: spin_unlock(&head->lock); if (tw) { - inet_twsk_deschedule(tw, death_row);; + inet_twsk_deschedule(tw, death_row); inet_twsk_put(tw); } diff --git a/net/ipv4/netfilter/ip_nat_standalone.c b/net/ipv4/netfilter/ip_nat_standalone.c index ab1f88fa21e..380aef3d786 100644 --- a/net/ipv4/netfilter/ip_nat_standalone.c +++ b/net/ipv4/netfilter/ip_nat_standalone.c @@ -394,7 +394,7 @@ static int init_or_cleanup(int init) ret = nf_register_hook(&ip_nat_local_out_ops); if (ret < 0) { printk("ip_nat_init: can't register local out hook.\n"); - goto cleanup_adjustout_ops;; + goto cleanup_adjustout_ops; } ret = nf_register_hook(&ip_nat_local_in_ops); if (ret < 0) { diff --git a/net/tipc/link.c b/net/tipc/link.c index 910b37e5083..784b24b6d10 100644 --- a/net/tipc/link.c +++ b/net/tipc/link.c @@ -1629,7 +1629,7 @@ void tipc_link_retransmit(struct link *l_ptr, struct sk_buff *buf, tipc_msg_print(TIPC_CONS, buf_msg(buf), ">RETR>"); info("...Retransmitted %u times\n", l_ptr->stale_count); - link_print(l_ptr, TIPC_CONS, "Resetting Link\n");; + link_print(l_ptr, TIPC_CONS, "Resetting Link\n"); tipc_link_reset(l_ptr); break; } diff --git a/sound/ppc/toonie.c b/sound/ppc/toonie.c index 210be20dc27..4e595172e42 100644 --- a/sound/ppc/toonie.c +++ b/sound/ppc/toonie.c @@ -117,7 +117,7 @@ static int toonie_get_mute_switch(struct snd_kcontrol *kcontrol, gp = &mix->amp_mute_gpio; break; default: - return -EINVAL;; + return -EINVAL; } ucontrol->value.integer.value[0] = !check_audio_gpio(gp); return 0; @@ -145,7 +145,7 @@ static int toonie_put_mute_switch(struct snd_kcontrol *kcontrol, gp = &mix->amp_mute_gpio; break; default: - return -EINVAL;; + return -EINVAL; } val = ! check_audio_gpio(gp); if (val != ucontrol->value.integer.value[0]) { -- cgit v1.2.3-70-g09d2 From d2a28ad9fa7bf16761d070d8a3338375e1574b32 Mon Sep 17 00:00:00 2001 From: Russ Anderson Date: Fri, 24 Mar 2006 09:49:52 -0800 Subject: [IA64] MCA recovery: kernel context recovery table Memory errors encountered by user applications may surface when the CPU is running in kernel context. The current code will not attempt recovery if the MCA surfaces in kernel context (privilage mode 0). This patch adds a check for cases where the user initiated the load that surfaces in kernel interrupt code. An example is a user process lauching a load from memory and the data in memory had bad ECC. Before the bad data gets to the CPU register, and interrupt comes in. The code jumps to the IVT interrupt entry point and begins execution in kernel context. The process of saving the user registers (SAVE_REST) causes the bad data to be loaded into a CPU register, triggering the MCA. The MCA surfaces in kernel context, even though the load was initiated from user context. As suggested by David and Tony, this patch uses an exception table like approach, puting the tagged recovery addresses in a searchable table. One difference from the exception table is that MCAs do not surface in precise places (such as with a TLB miss), so instead of tagging specific instructions, address ranges are registers. A single macro is used to do the tagging, with the input parameter being the label of the starting address and the macro being the ending address. This limits clutter in the code. This patch only tags one spot, the interrupt ivt entry. Testing showed that spot to be a "heavy hitter" with MCAs surfacing while saving user registers. Other spots can be added as needed by adding a single macro. Signed-off-by: Russ Anderson (rja@sgi.com) Signed-off-by: Tony Luck --- arch/ia64/kernel/ivt.S | 1 + arch/ia64/kernel/mca.c | 98 +++++++++++++++++++++++++++++++----------- arch/ia64/kernel/mca_drv.c | 22 +++++++--- arch/ia64/kernel/mca_drv.h | 7 +++ arch/ia64/kernel/mca_drv_asm.S | 13 ++---- arch/ia64/kernel/vmlinux.lds.S | 9 ++++ include/asm-ia64/asmmacro.h | 11 +++++ 7 files changed, 120 insertions(+), 41 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/ivt.S b/arch/ia64/kernel/ivt.S index dcd906fe574..829a43cab79 100644 --- a/arch/ia64/kernel/ivt.S +++ b/arch/ia64/kernel/ivt.S @@ -865,6 +865,7 @@ ENTRY(interrupt) ;; SAVE_REST ;; + MCA_RECOVER_RANGE(interrupt) alloc r14=ar.pfs,0,0,2,0 // must be first in an insn group mov out0=cr.ivr // pass cr.ivr as first arg add out1=16,sp // pass pointer to pt_regs as second arg diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c index cedcae713e9..87ff7fe33cf 100644 --- a/arch/ia64/kernel/mca.c +++ b/arch/ia64/kernel/mca.c @@ -83,6 +83,7 @@ #include #include +#include "mca_drv.h" #include "entry.h" #if defined(IA64_MCA_DEBUG_INFO) @@ -281,6 +282,50 @@ ia64_mca_log_sal_error_record(int sal_info_type) ia64_sal_clear_state_info(sal_info_type); } +/* + * search_mca_table + * See if the MCA surfaced in an instruction range + * that has been tagged as recoverable. + * + * Inputs + * first First address range to check + * last Last address range to check + * ip Instruction pointer, address we are looking for + * + * Return value: + * 1 on Success (in the table)/ 0 on Failure (not in the table) + */ +int +search_mca_table (const struct mca_table_entry *first, + const struct mca_table_entry *last, + unsigned long ip) +{ + const struct mca_table_entry *curr; + u64 curr_start, curr_end; + + curr = first; + while (curr <= last) { + curr_start = (u64) &curr->start_addr + curr->start_addr; + curr_end = (u64) &curr->end_addr + curr->end_addr; + + if ((ip >= curr_start) && (ip <= curr_end)) { + return 1; + } + curr++; + } + return 0; +} + +/* Given an address, look for it in the mca tables. */ +int mca_recover_range(unsigned long addr) +{ + extern struct mca_table_entry __start___mca_table[]; + extern struct mca_table_entry __stop___mca_table[]; + + return search_mca_table(__start___mca_table, __stop___mca_table-1, addr); +} +EXPORT_SYMBOL_GPL(mca_recover_range); + #ifdef CONFIG_ACPI int cpe_vector = -1; @@ -747,31 +792,34 @@ ia64_mca_modify_original_stack(struct pt_regs *regs, ia64_mca_modify_comm(previous_current); goto no_mod; } - if (r13 != sos->prev_IA64_KR_CURRENT) { - msg = "inconsistent previous current and r13"; - goto no_mod; - } - if ((r12 - r13) >= KERNEL_STACK_SIZE) { - msg = "inconsistent r12 and r13"; - goto no_mod; - } - if ((ar_bspstore - r13) >= KERNEL_STACK_SIZE) { - msg = "inconsistent ar.bspstore and r13"; - goto no_mod; - } - va.p = old_bspstore; - if (va.f.reg < 5) { - msg = "old_bspstore is in the wrong region"; - goto no_mod; - } - if ((ar_bsp - r13) >= KERNEL_STACK_SIZE) { - msg = "inconsistent ar.bsp and r13"; - goto no_mod; - } - size += (ia64_rse_skip_regs(old_bspstore, slots) - old_bspstore) * 8; - if (ar_bspstore + size > r12) { - msg = "no room for blocked state"; - goto no_mod; + + if (!mca_recover_range(ms->pmsa_iip)) { + if (r13 != sos->prev_IA64_KR_CURRENT) { + msg = "inconsistent previous current and r13"; + goto no_mod; + } + if ((r12 - r13) >= KERNEL_STACK_SIZE) { + msg = "inconsistent r12 and r13"; + goto no_mod; + } + if ((ar_bspstore - r13) >= KERNEL_STACK_SIZE) { + msg = "inconsistent ar.bspstore and r13"; + goto no_mod; + } + va.p = old_bspstore; + if (va.f.reg < 5) { + msg = "old_bspstore is in the wrong region"; + goto no_mod; + } + if ((ar_bsp - r13) >= KERNEL_STACK_SIZE) { + msg = "inconsistent ar.bsp and r13"; + goto no_mod; + } + size += (ia64_rse_skip_regs(old_bspstore, slots) - old_bspstore) * 8; + if (ar_bspstore + size > r12) { + msg = "no room for blocked state"; + goto no_mod; + } } ia64_mca_modify_comm(previous_current); diff --git a/arch/ia64/kernel/mca_drv.c b/arch/ia64/kernel/mca_drv.c index e883d85906d..37c88eb5587 100644 --- a/arch/ia64/kernel/mca_drv.c +++ b/arch/ia64/kernel/mca_drv.c @@ -6,6 +6,7 @@ * Copyright (C) Hidetoshi Seto (seto.hidetoshi@jp.fujitsu.com) * Copyright (C) 2005 Silicon Graphics, Inc * Copyright (C) 2005 Keith Owens + * Copyright (C) 2006 Russ Anderson */ #include #include @@ -121,11 +122,12 @@ mca_page_isolate(unsigned long paddr) */ void -mca_handler_bh(unsigned long paddr) +mca_handler_bh(unsigned long paddr, void *iip, unsigned long ipsr) { - printk(KERN_ERR - "OS_MCA: process [pid: %d](%s) encounters MCA (paddr=%lx)\n", - current->pid, current->comm, paddr); + printk(KERN_ERR "OS_MCA: process [cpu %d, pid: %d, uid: %d, " + "iip: %p, psr: 0x%lx,paddr: 0x%lx](%s) encounters MCA.\n", + raw_smp_processor_id(), current->pid, current->uid, + iip, ipsr, paddr, current->comm); spin_lock(&mca_bh_lock); switch (mca_page_isolate(paddr)) { @@ -442,21 +444,26 @@ recover_from_read_error(slidx_table_t *slidx, if (!peidx_bottom(peidx) || !(peidx_bottom(peidx)->valid.minstate)) return 0; psr1 =(struct ia64_psr *)&(peidx_minstate_area(peidx)->pmsa_ipsr); + psr2 =(struct ia64_psr *)&(peidx_minstate_area(peidx)->pmsa_xpsr); /* * Check the privilege level of interrupted context. * If it is user-mode, then terminate affected process. */ - if (psr1->cpl != 0) { + + pmsa = sos->pal_min_state; + if (psr1->cpl != 0 || + ((psr2->cpl != 0) && mca_recover_range(pmsa->pmsa_iip))) { smei = peidx_bus_check(peidx, 0); if (smei->valid.target_identifier) { /* * setup for resume to bottom half of MCA, * "mca_handler_bhhook" */ - pmsa = sos->pal_min_state; - /* pass to bhhook as 1st argument (gr8) */ + /* pass to bhhook as argument (gr8, ...) */ pmsa->pmsa_gr[8-1] = smei->target_identifier; + pmsa->pmsa_gr[9-1] = pmsa->pmsa_iip; + pmsa->pmsa_gr[10-1] = pmsa->pmsa_ipsr; /* set interrupted return address (but no use) */ pmsa->pmsa_br0 = pmsa->pmsa_iip; /* change resume address to bottom half */ @@ -466,6 +473,7 @@ recover_from_read_error(slidx_table_t *slidx, psr2 = (struct ia64_psr *)&pmsa->pmsa_ipsr; psr2->cpl = 0; psr2->ri = 0; + psr2->bn = 1; psr2->i = 0; return 1; diff --git a/arch/ia64/kernel/mca_drv.h b/arch/ia64/kernel/mca_drv.h index e2f6fa1e0ef..31a2e52bb16 100644 --- a/arch/ia64/kernel/mca_drv.h +++ b/arch/ia64/kernel/mca_drv.h @@ -111,3 +111,10 @@ typedef struct slidx_table { slidx_foreach_entry(__pos, &((slidx)->sec)) { __count++; }\ __count; }) +struct mca_table_entry { + int start_addr; /* location-relative starting address of MCA recoverable range */ + int end_addr; /* location-relative ending address of MCA recoverable range */ +}; + +extern const struct mca_table_entry *search_mca_tables (unsigned long addr); +extern int mca_recover_range(unsigned long); diff --git a/arch/ia64/kernel/mca_drv_asm.S b/arch/ia64/kernel/mca_drv_asm.S index 3f298ee4d00..e6a580d354b 100644 --- a/arch/ia64/kernel/mca_drv_asm.S +++ b/arch/ia64/kernel/mca_drv_asm.S @@ -14,15 +14,12 @@ GLOBAL_ENTRY(mca_handler_bhhook) invala // clear RSE ? - ;; cover ;; clrrrb ;; - alloc r16=ar.pfs,0,2,1,0 // make a new frame - ;; + alloc r16=ar.pfs,0,2,3,0 // make a new frame mov ar.rsc=0 - ;; mov r13=IA64_KR(CURRENT) // current task pointer ;; mov r2=r13 @@ -30,7 +27,6 @@ GLOBAL_ENTRY(mca_handler_bhhook) addl r22=IA64_RBS_OFFSET,r2 ;; mov ar.bspstore=r22 - ;; addl sp=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r2 ;; adds r2=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13 @@ -40,12 +36,12 @@ GLOBAL_ENTRY(mca_handler_bhhook) movl loc1=mca_handler_bh // recovery C function ;; mov out0=r8 // poisoned address + mov out1=r9 // iip + mov out2=r10 // psr mov b6=loc1 ;; mov loc1=rp - ;; - ssm psr.i - ;; + ssm psr.i | psr.ic br.call.sptk.many rp=b6 // does not return ... ;; mov ar.pfs=loc0 @@ -53,5 +49,4 @@ GLOBAL_ENTRY(mca_handler_bhhook) ;; mov r8=r0 br.ret.sptk.many rp - ;; END(mca_handler_bhhook) diff --git a/arch/ia64/kernel/vmlinux.lds.S b/arch/ia64/kernel/vmlinux.lds.S index 632d65cc068..0b9e56dd7f0 100644 --- a/arch/ia64/kernel/vmlinux.lds.S +++ b/arch/ia64/kernel/vmlinux.lds.S @@ -130,6 +130,15 @@ SECTIONS __initcall_end = .; } + /* MCA table */ + . = ALIGN(16); + __mca_table : AT(ADDR(__mca_table) - LOAD_OFFSET) + { + __start___mca_table = .; + *(__mca_table) + __stop___mca_table = .; + } + .data.patch.vtop : AT(ADDR(.data.patch.vtop) - LOAD_OFFSET) { __start___vtop_patchlist = .; diff --git a/include/asm-ia64/asmmacro.h b/include/asm-ia64/asmmacro.h index 77af457f4ad..d4cec32083d 100644 --- a/include/asm-ia64/asmmacro.h +++ b/include/asm-ia64/asmmacro.h @@ -50,6 +50,17 @@ name: .xdata4 "__ex_table", 99f-., y-.+4; \ [99:] x +/* + * Tag MCA recoverable instruction ranges. + */ + + .section "__mca_table", "a" // declare section & section attributes + .previous + +# define MCA_RECOVER_RANGE(y) \ + .xdata4 "__mca_table", y-., 99f-.; \ + [99:] + /* * Mark instructions that need a load of a virtual address patched to be * a load of a physical address. We use this either in critical performance -- cgit v1.2.3-70-g09d2 From b17ea91a43ea0c746ab4cabb698275e1771ed23d Mon Sep 17 00:00:00 2001 From: "Chen, Kenneth W" Date: Sun, 12 Mar 2006 11:10:00 -0800 Subject: [IA64] cleanup dig_irq_init dig_irq_init is equivalent to machvec_noop, no need to define another empty function. Signed-off-by: Ken Chen Signed-off-by: Tony Luck --- arch/ia64/dig/setup.c | 5 ----- include/asm-ia64/machvec_dig.h | 2 -- 2 files changed, 7 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/dig/setup.c b/arch/ia64/dig/setup.c index c9104bfff66..38aa9c10885 100644 --- a/arch/ia64/dig/setup.c +++ b/arch/ia64/dig/setup.c @@ -69,8 +69,3 @@ dig_setup (char **cmdline_p) screen_info.orig_video_isVGA = 1; /* XXX fake */ screen_info.orig_video_ega_bx = 3; /* XXX fake */ } - -void __init -dig_irq_init (void) -{ -} diff --git a/include/asm-ia64/machvec_dig.h b/include/asm-ia64/machvec_dig.h index 4dc8522c974..8a0752f4098 100644 --- a/include/asm-ia64/machvec_dig.h +++ b/include/asm-ia64/machvec_dig.h @@ -2,7 +2,6 @@ #define _ASM_IA64_MACHVEC_DIG_h extern ia64_mv_setup_t dig_setup; -extern ia64_mv_irq_init_t dig_irq_init; /* * This stuff has dual use! @@ -13,6 +12,5 @@ extern ia64_mv_irq_init_t dig_irq_init; */ #define platform_name "dig" #define platform_setup dig_setup -#define platform_irq_init dig_irq_init #endif /* _ASM_IA64_MACHVEC_DIG_h */ -- cgit v1.2.3-70-g09d2 From f90aa8c4febb306e1266e1ad34fd8464e201aa7f Mon Sep 17 00:00:00 2001 From: Prarit Bhargava Date: Wed, 8 Mar 2006 13:30:18 -0500 Subject: [IA64] Tollhouse HP: IA64 arch changes arch/ia64/sn and include/asm-ia64/sn changes required to support Tollhouse system PCI hotplug, fixes the ia64_sn_sysctl_ioboard_get call, and introduces the PRF_HOTPLUG_SUPPORT feature bit. Signed-off-by: Prarit Bhargava Signed-off-by: Tony Luck --- arch/ia64/sn/kernel/io_init.c | 29 +++++++++++++++++++++++++++++ arch/ia64/sn/kernel/tiocx.c | 10 ++++++++-- arch/ia64/sn/pci/pcibr/pcibr_provider.c | 17 +++++++++++++++++ include/asm-ia64/sn/l1.h | 3 ++- include/asm-ia64/sn/pcibr_provider.h | 1 + include/asm-ia64/sn/pcidev.h | 1 + include/asm-ia64/sn/sn_feature_sets.h | 3 +-- include/asm-ia64/sn/sn_sal.h | 26 +++++++++++++++----------- 8 files changed, 74 insertions(+), 16 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/sn/kernel/io_init.c b/arch/ia64/sn/kernel/io_init.c index dfb3f290237..5101ac46264 100644 --- a/arch/ia64/sn/kernel/io_init.c +++ b/arch/ia64/sn/kernel/io_init.c @@ -13,6 +13,8 @@ #include #include #include +#include +#include #include #include #include @@ -710,9 +712,36 @@ cnodeid_get_geoid(cnodeid_t cnode) return hubdev->hdi_geoid; } +void sn_generate_path(struct pci_bus *pci_bus, char *address) +{ + nasid_t nasid; + cnodeid_t cnode; + geoid_t geoid; + moduleid_t moduleid; + u16 bricktype; + + nasid = NASID_GET(SN_PCIBUS_BUSSOFT(pci_bus)->bs_base); + cnode = nasid_to_cnodeid(nasid); + geoid = cnodeid_get_geoid(cnode); + moduleid = geo_module(geoid); + + sprintf(address, "module_%c%c%c%c%.2d", + '0'+RACK_GET_CLASS(MODULE_GET_RACK(moduleid)), + '0'+RACK_GET_GROUP(MODULE_GET_RACK(moduleid)), + '0'+RACK_GET_NUM(MODULE_GET_RACK(moduleid)), + MODULE_GET_BTCHAR(moduleid), MODULE_GET_BPOS(moduleid)); + + /* Tollhouse requires slot id to be displayed */ + bricktype = MODULE_GET_BTYPE(moduleid); + if ((bricktype == L1_BRICKTYPE_191010) || + (bricktype == L1_BRICKTYPE_1932)) + sprintf(address, "%s^%d", address, geo_slot(geoid)); +} + subsys_initcall(sn_pci_init); EXPORT_SYMBOL(sn_pci_fixup_slot); EXPORT_SYMBOL(sn_pci_unfixup_slot); EXPORT_SYMBOL(sn_pci_controller_fixup); EXPORT_SYMBOL(sn_bus_store_sysdata); EXPORT_SYMBOL(sn_bus_free_sysdata); +EXPORT_SYMBOL(sn_generate_path); diff --git a/arch/ia64/sn/kernel/tiocx.c b/arch/ia64/sn/kernel/tiocx.c index 99cb28e7429..feaf1a6e810 100644 --- a/arch/ia64/sn/kernel/tiocx.c +++ b/arch/ia64/sn/kernel/tiocx.c @@ -369,9 +369,15 @@ static void tio_corelet_reset(nasid_t nasid, int corelet) static int is_fpga_tio(int nasid, int *bt) { - int ioboard_type; + u16 ioboard_type; + s64 rc; - ioboard_type = ia64_sn_sysctl_ioboard_get(nasid); + rc = ia64_sn_sysctl_ioboard_get(nasid, &ioboard_type); + if (rc) { + printk(KERN_WARNING "ia64_sn_sysctl_ioboard_get failed: %ld\n", + rc); + return 0; + } switch (ioboard_type) { case L1_BRICKTYPE_SA: diff --git a/arch/ia64/sn/pci/pcibr/pcibr_provider.c b/arch/ia64/sn/pci/pcibr/pcibr_provider.c index 98f716bd92f..ab1211ef017 100644 --- a/arch/ia64/sn/pci/pcibr/pcibr_provider.c +++ b/arch/ia64/sn/pci/pcibr/pcibr_provider.c @@ -74,6 +74,22 @@ static int sal_pcibr_error_interrupt(struct pcibus_info *soft) return (int)ret_stuff.v0; } +u16 sn_ioboard_to_pci_bus(struct pci_bus *pci_bus) +{ + s64 rc; + u16 ioboard; + nasid_t nasid = NASID_GET(SN_PCIBUS_BUSSOFT(pci_bus)->bs_base); + + rc = ia64_sn_sysctl_ioboard_get(nasid, &ioboard); + if (rc) { + printk(KERN_WARNING "ia64_sn_sysctl_ioboard_get failed: %ld\n", + rc); + return 0; + } + + return ioboard; +} + /* * PCI Bridge Error interrupt handler. Gets invoked whenever a PCI * bridge sends an error interrupt. @@ -255,3 +271,4 @@ pcibr_init_provider(void) EXPORT_SYMBOL_GPL(sal_pcibr_slot_enable); EXPORT_SYMBOL_GPL(sal_pcibr_slot_disable); +EXPORT_SYMBOL_GPL(sn_ioboard_to_pci_bus); diff --git a/include/asm-ia64/sn/l1.h b/include/asm-ia64/sn/l1.h index e3b819110d4..344bf44bb35 100644 --- a/include/asm-ia64/sn/l1.h +++ b/include/asm-ia64/sn/l1.h @@ -34,6 +34,8 @@ #define L1_BRICKTYPE_IA 0x6b /* k */ #define L1_BRICKTYPE_ATHENA 0x2b /* + */ #define L1_BRICKTYPE_DAYTONA 0x7a /* z */ +#define L1_BRICKTYPE_1932 0x2c /* . */ +#define L1_BRICKTYPE_191010 0x2e /* , */ /* board type response codes */ #define L1_BOARDTYPE_IP69 0x0100 /* CA */ @@ -46,5 +48,4 @@ #define L1_BOARDTYPE_DAYTONA 0x0800 /* AD */ #define L1_BOARDTYPE_INVAL (-1) /* invalid brick type */ - #endif /* _ASM_IA64_SN_L1_H */ diff --git a/include/asm-ia64/sn/pcibr_provider.h b/include/asm-ia64/sn/pcibr_provider.h index a601d3af39b..51260ab70d9 100644 --- a/include/asm-ia64/sn/pcibr_provider.h +++ b/include/asm-ia64/sn/pcibr_provider.h @@ -144,4 +144,5 @@ extern int sal_pcibr_slot_enable(struct pcibus_info *soft, int device, void *resp); extern int sal_pcibr_slot_disable(struct pcibus_info *soft, int device, int action, void *resp); +extern u16 sn_ioboard_to_pci_bus(struct pci_bus *pci_bus); #endif diff --git a/include/asm-ia64/sn/pcidev.h b/include/asm-ia64/sn/pcidev.h index 38cdffbc4c7..eac3561574b 100644 --- a/include/asm-ia64/sn/pcidev.h +++ b/include/asm-ia64/sn/pcidev.h @@ -76,6 +76,7 @@ extern void sn_pci_controller_fixup(int segment, int busnum, struct pci_bus *bus); extern void sn_bus_store_sysdata(struct pci_dev *dev); extern void sn_bus_free_sysdata(void); +extern void sn_generate_path(struct pci_bus *pci_bus, char *address); extern void sn_pci_fixup_slot(struct pci_dev *dev); extern void sn_pci_unfixup_slot(struct pci_dev *dev); extern void sn_irq_lh_init(void); diff --git a/include/asm-ia64/sn/sn_feature_sets.h b/include/asm-ia64/sn/sn_feature_sets.h index ff33e3bd3f8..30dcfa442e5 100644 --- a/include/asm-ia64/sn/sn_feature_sets.h +++ b/include/asm-ia64/sn/sn_feature_sets.h @@ -30,8 +30,7 @@ extern int sn_prom_feature_available(int id); #define PRF_PAL_CACHE_FLUSH_SAFE 0 #define PRF_DEVICE_FLUSH_LIST 1 - - +#define PRF_HOTPLUG_SUPPORT 2 /* --------------------- OS Features -------------------------------*/ diff --git a/include/asm-ia64/sn/sn_sal.h b/include/asm-ia64/sn/sn_sal.h index e77f0c9b7d3..246f43a796d 100644 --- a/include/asm-ia64/sn/sn_sal.h +++ b/include/asm-ia64/sn/sn_sal.h @@ -907,18 +907,22 @@ ia64_sn_sysctl_tio_clock_reset(nasid_t nasid) /* * Get the associated ioboard type for a given nasid. */ -static inline int -ia64_sn_sysctl_ioboard_get(nasid_t nasid) +static inline s64 +ia64_sn_sysctl_ioboard_get(nasid_t nasid, u16 *ioboard) { - struct ia64_sal_retval rv; - SAL_CALL_REENTRANT(rv, SN_SAL_SYSCTL_OP, SAL_SYSCTL_OP_IOBOARD, - nasid, 0, 0, 0, 0, 0); - if (rv.v0 != 0) - return (int)rv.v0; - if (rv.v1 != 0) - return (int)rv.v1; - - return 0; + struct ia64_sal_retval isrv; + SAL_CALL_REENTRANT(isrv, SN_SAL_SYSCTL_OP, SAL_SYSCTL_OP_IOBOARD, + nasid, 0, 0, 0, 0, 0); + if (isrv.v0 != 0) { + *ioboard = isrv.v0; + return isrv.status; + } + if (isrv.v1 != 0) { + *ioboard = isrv.v1; + return isrv.status; + } + + return isrv.status; } /** -- cgit v1.2.3-70-g09d2 From b354a8388891adc5dc5e5fb0130f000152f3fb94 Mon Sep 17 00:00:00 2001 From: Jack Steiner Date: Thu, 2 Mar 2006 16:02:21 -0600 Subject: [IA64] Increase max node count on SN platforms Add a configuration option to allow the maximum number of nodes to be configurable for GENERIC or SN kernels. Signed-off-by: Jack Steiner Signed-off-by: Tony Luck --- arch/ia64/Kconfig | 9 +++++++++ include/asm-ia64/numnodes.h | 13 +++++++++---- 2 files changed, 18 insertions(+), 4 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index ff7ae6b664e..10b6b9e7716 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -252,6 +252,15 @@ config NR_CPUS than 64 will cause the use of a CPU mask array, causing a small performance hit. +config IA64_NR_NODES + int "Maximum number of NODEs (256-1024)" if (IA64_SGI_SN2 || IA64_GENERIC) + range 256 1024 + depends on IA64_SGI_SN2 || IA64_GENERIC + default "256" + help + This option specifies the maximum number of nodes in your SSI system. + If in doubt, use the default. + config HOTPLUG_CPU bool "Support for hot-pluggable CPUs (EXPERIMENTAL)" depends on SMP && EXPERIMENTAL diff --git a/include/asm-ia64/numnodes.h b/include/asm-ia64/numnodes.h index 21cff4da548..e9d356f549d 100644 --- a/include/asm-ia64/numnodes.h +++ b/include/asm-ia64/numnodes.h @@ -3,13 +3,18 @@ #ifdef CONFIG_IA64_DIG /* Max 8 Nodes */ -#define NODES_SHIFT 3 +# define NODES_SHIFT 3 #elif defined(CONFIG_IA64_HP_ZX1) || defined(CONFIG_IA64_HP_ZX1_SWIOTLB) /* Max 32 Nodes */ -#define NODES_SHIFT 5 +# define NODES_SHIFT 5 #elif defined(CONFIG_IA64_SGI_SN2) || defined(CONFIG_IA64_GENERIC) -/* Max 256 Nodes */ -#define NODES_SHIFT 8 +# if CONFIG_IA64_NR_NODES == 256 +# define NODES_SHIFT 8 +# elif CONFIG_IA64_NR_NODES <= 512 +# define NODES_SHIFT 9 +# elif CONFIG_IA64_NR_NODES <= 1024 +# define NODES_SHIFT 10 +# endif #endif #endif /* _ASM_MAX_NUMNODES_H */ -- cgit v1.2.3-70-g09d2 From 3ad5ef8b9d0d0cc2d4b2c63e766ef903d482dfc7 Mon Sep 17 00:00:00 2001 From: Jack Steiner Date: Thu, 2 Mar 2006 16:02:25 -0600 Subject: [IA64] Increase max node count on SN platforms Add support in IA64 acpi for platforms that support more than 256 nodes. Currently, ACPI is limited to 256 nodes because the proximity domain number is 8-bits. Long term, we expect to use ACPI3.0 to support >256 nodes. This patch is an interim solution that works with platforms that pass the high order bits of the proximity domain in "reserved" fields of the ACPI tables. This code is enabled ONLY on SN platforms. Signed-off-by: Jack Steiner Signed-off-by: Tony Luck --- arch/ia64/kernel/acpi.c | 32 ++++++++++++++++++++++++++++---- include/asm-ia64/acpi.h | 4 ++++ 2 files changed, 32 insertions(+), 4 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c index 24fe7c81e10..a4e218ce2ed 100644 --- a/arch/ia64/kernel/acpi.c +++ b/arch/ia64/kernel/acpi.c @@ -420,6 +420,26 @@ int __devinitdata pxm_to_nid_map[MAX_PXM_DOMAINS]; int __initdata nid_to_pxm_map[MAX_NUMNODES]; static struct acpi_table_slit __initdata *slit_table; +static int get_processor_proximity_domain(struct acpi_table_processor_affinity *pa) +{ + int pxm; + + pxm = pa->proximity_domain; + if (ia64_platform_is("sn2")) + pxm += pa->reserved[0] << 8; + return pxm; +} + +static int get_memory_proximity_domain(struct acpi_table_memory_affinity *ma) +{ + int pxm; + + pxm = ma->proximity_domain; + if (ia64_platform_is("sn2")) + pxm += ma->reserved1[0] << 8; + return pxm; +} + /* * ACPI 2.0 SLIT (System Locality Information Table) * http://devresource.hp.com/devresource/Docs/TechPapers/IA64/slit.pdf @@ -443,16 +463,20 @@ void __init acpi_numa_slit_init(struct acpi_table_slit *slit) void __init acpi_numa_processor_affinity_init(struct acpi_table_processor_affinity *pa) { + int pxm; + if (!pa->flags.enabled) return; + pxm = get_processor_proximity_domain(pa); + /* record this node in proximity bitmap */ - pxm_bit_set(pa->proximity_domain); + pxm_bit_set(pxm); node_cpuid[srat_num_cpus].phys_id = (pa->apic_id << 8) | (pa->lsapic_eid); /* nid should be overridden as logical node id later */ - node_cpuid[srat_num_cpus].nid = pa->proximity_domain; + node_cpuid[srat_num_cpus].nid = pxm; srat_num_cpus++; } @@ -460,10 +484,10 @@ void __init acpi_numa_memory_affinity_init(struct acpi_table_memory_affinity *ma) { unsigned long paddr, size; - u8 pxm; + int pxm; struct node_memblk_s *p, *q, *pend; - pxm = ma->proximity_domain; + pxm = get_memory_proximity_domain(ma); /* fill node memory chunk structure */ paddr = ma->base_addr_hi; diff --git a/include/asm-ia64/acpi.h b/include/asm-ia64/acpi.h index f7a51765430..d734585a23c 100644 --- a/include/asm-ia64/acpi.h +++ b/include/asm-ia64/acpi.h @@ -111,7 +111,11 @@ extern int additional_cpus; #ifdef CONFIG_ACPI_NUMA /* Proximity bitmap length; _PXM is at most 255 (8 bit)*/ +#ifdef CONFIG_IA64_NR_NODES +#define MAX_PXM_DOMAINS CONFIG_IA64_NR_NODES +#else #define MAX_PXM_DOMAINS (256) +#endif extern int __devinitdata pxm_to_nid_map[MAX_PXM_DOMAINS]; extern int __initdata nid_to_pxm_map[MAX_NUMNODES]; #endif -- cgit v1.2.3-70-g09d2 From a9de98351436b25b3c2f234addb6d66a6a6f42f8 Mon Sep 17 00:00:00 2001 From: Jack Steiner Date: Thu, 2 Mar 2006 16:02:28 -0600 Subject: [IA64] Increase max node count on SN platforms Node number are kept in the cpu_to_node_map which is currently defined as u8. Change to u16 to accomodate larger node numbers. Signed-off-by: Jack Steiner Signed-off-by: Tony Luck --- arch/ia64/kernel/numa.c | 2 +- include/asm-ia64/numa.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/numa.c b/arch/ia64/kernel/numa.c index a68ce667809..0766493d4d0 100644 --- a/arch/ia64/kernel/numa.c +++ b/arch/ia64/kernel/numa.c @@ -25,7 +25,7 @@ #include #include -u8 cpu_to_node_map[NR_CPUS] __cacheline_aligned; +u16 cpu_to_node_map[NR_CPUS] __cacheline_aligned; EXPORT_SYMBOL(cpu_to_node_map); cpumask_t node_to_cpu_mask[MAX_NUMNODES] __cacheline_aligned; diff --git a/include/asm-ia64/numa.h b/include/asm-ia64/numa.h index 3ae128fe082..dae6aeb7b11 100644 --- a/include/asm-ia64/numa.h +++ b/include/asm-ia64/numa.h @@ -23,7 +23,7 @@ #include -extern u8 cpu_to_node_map[NR_CPUS] __cacheline_aligned; +extern u16 cpu_to_node_map[NR_CPUS] __cacheline_aligned; extern cpumask_t node_to_cpu_mask[MAX_NUMNODES] __cacheline_aligned; /* Stuff below this line could be architecture independent */ -- cgit v1.2.3-70-g09d2 From 4d357acadd7a5e60767c748ed7807e11c4387bdf Mon Sep 17 00:00:00 2001 From: Jack Steiner Date: Thu, 2 Mar 2006 16:02:32 -0600 Subject: [IA64] Increase max node count on SN platforms Update configuration files with new CONFIG option. Signed-off-by: Jack Steiner Signed-off-by: Tony Luck --- arch/ia64/configs/gensparse_defconfig | 1 + arch/ia64/configs/sn2_defconfig | 1 + arch/ia64/defconfig | 1 + 3 files changed, 3 insertions(+) (limited to 'arch/ia64') diff --git a/arch/ia64/configs/gensparse_defconfig b/arch/ia64/configs/gensparse_defconfig index 744fd2f79f6..0d29aa2066b 100644 --- a/arch/ia64/configs/gensparse_defconfig +++ b/arch/ia64/configs/gensparse_defconfig @@ -116,6 +116,7 @@ CONFIG_IOSAPIC=y CONFIG_FORCE_MAX_ZONEORDER=17 CONFIG_SMP=y CONFIG_NR_CPUS=512 +CONFIG_IA64_NR_NODES=256 CONFIG_HOTPLUG_CPU=y # CONFIG_SCHED_SMT is not set # CONFIG_PREEMPT is not set diff --git a/arch/ia64/configs/sn2_defconfig b/arch/ia64/configs/sn2_defconfig index 8206752161b..a718034d68d 100644 --- a/arch/ia64/configs/sn2_defconfig +++ b/arch/ia64/configs/sn2_defconfig @@ -116,6 +116,7 @@ CONFIG_IA64_SGI_SN_XP=m CONFIG_FORCE_MAX_ZONEORDER=17 CONFIG_SMP=y CONFIG_NR_CPUS=1024 +CONFIG_IA64_NR_NODES=256 # CONFIG_HOTPLUG_CPU is not set CONFIG_SCHED_SMT=y CONFIG_PREEMPT=y diff --git a/arch/ia64/defconfig b/arch/ia64/defconfig index 3e767288a74..6cba55da572 100644 --- a/arch/ia64/defconfig +++ b/arch/ia64/defconfig @@ -116,6 +116,7 @@ CONFIG_IOSAPIC=y CONFIG_FORCE_MAX_ZONEORDER=17 CONFIG_SMP=y CONFIG_NR_CPUS=512 +CONFIG_IA64_NR_NODES=256 CONFIG_HOTPLUG_CPU=y # CONFIG_SCHED_SMT is not set # CONFIG_PREEMPT is not set -- cgit v1.2.3-70-g09d2 From 4129a953ad4db379d8e07b0dd2157998653a1325 Mon Sep 17 00:00:00 2001 From: Fenghua Yu Date: Mon, 27 Feb 2006 16:16:22 -0800 Subject: [IA64] New IA64 core/thread detection patch IPF SDM 2.2 changes definition of PAL_LOGICAL_TO_PHYSICAL to add proc_number=-1 to get core/thread mapping info on the running processer. Based on this change, we had better to update existing core/thread detection in IA64 kernel correspondingly. The attached patch implements this change. It simplifies detection code and eliminates potential race condition. It also runs a bit faster and has better scalability especially when cores and threads number grows up in one package. Signed-off-by: Fenghua Yu Signed-off-by: Tony Luck --- arch/ia64/kernel/smpboot.c | 109 ++------------------------------------------- include/asm-ia64/pal.h | 3 +- 2 files changed, 5 insertions(+), 107 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c index c4b633b36da..44e9547878a 100644 --- a/arch/ia64/kernel/smpboot.c +++ b/arch/ia64/kernel/smpboot.c @@ -624,31 +624,7 @@ void __devinit smp_prepare_boot_cpu(void) per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE; } -/* - * mt_info[] is a temporary store for all info returned by - * PAL_LOGICAL_TO_PHYSICAL, to be copied into cpuinfo_ia64 when the - * specific cpu comes. - */ -static struct { - __u32 socket_id; - __u16 core_id; - __u16 thread_id; - __u16 proc_fixed_addr; - __u8 valid; -} mt_info[NR_CPUS] __devinitdata; - #ifdef CONFIG_HOTPLUG_CPU -static inline void -remove_from_mtinfo(int cpu) -{ - int i; - - for_each_cpu(i) - if (mt_info[i].valid && mt_info[i].socket_id == - cpu_data(cpu)->socket_id) - mt_info[i].valid = 0; -} - static inline void clear_cpu_sibling_map(int cpu) { @@ -678,12 +654,6 @@ remove_siblinginfo(int cpu) /* remove it from all sibling map's */ clear_cpu_sibling_map(cpu); - - /* if this cpu is the last in the core group, remove all its info - * from mt_info structure - */ - if (last) - remove_from_mtinfo(cpu); } extern void fixup_irqs(void); @@ -878,40 +848,6 @@ init_smp_config(void) ia64_sal_strerror(sal_ret)); } -static inline int __devinit -check_for_mtinfo_index(void) -{ - int i; - - for_each_cpu(i) - if (!mt_info[i].valid) - return i; - - return -1; -} - -/* - * Search the mt_info to find out if this socket's cid/tid information is - * cached or not. If the socket exists, fill in the core_id and thread_id - * in cpuinfo - */ -static int __devinit -check_for_new_socket(__u16 logical_address, struct cpuinfo_ia64 *c) -{ - int i; - __u32 sid = c->socket_id; - - for_each_cpu(i) { - if (mt_info[i].valid && mt_info[i].proc_fixed_addr == logical_address - && mt_info[i].socket_id == sid) { - c->core_id = mt_info[i].core_id; - c->thread_id = mt_info[i].thread_id; - return 1; /* not a new socket */ - } - } - return 0; -} - /* * identify_siblings(cpu) gets called from identify_cpu. This populates the * information related to logical execution units in per_cpu_data structure. @@ -921,14 +857,12 @@ identify_siblings(struct cpuinfo_ia64 *c) { s64 status; u16 pltid; - u64 proc_fixed_addr; - int count, i; pal_logical_to_physical_t info; if (smp_num_cpucores == 1 && smp_num_siblings == 1) return; - if ((status = ia64_pal_logical_to_phys(0, &info)) != PAL_STATUS_SUCCESS) { + if ((status = ia64_pal_logical_to_phys(-1, &info)) != PAL_STATUS_SUCCESS) { printk(KERN_ERR "ia64_pal_logical_to_phys failed with %ld\n", status); return; @@ -937,47 +871,12 @@ identify_siblings(struct cpuinfo_ia64 *c) printk(KERN_ERR "ia64_sal_pltid failed with %ld\n", status); return; } - if ((status = ia64_pal_fixed_addr(&proc_fixed_addr)) != PAL_STATUS_SUCCESS) { - printk(KERN_ERR "ia64_pal_fixed_addr failed with %ld\n", status); - return; - } c->socket_id = (pltid << 8) | info.overview_ppid; c->cores_per_socket = info.overview_cpp; c->threads_per_core = info.overview_tpc; - count = c->num_log = info.overview_num_log; + c->num_log = info.overview_num_log; - /* If the thread and core id information is already cached, then - * we will simply update cpu_info and return. Otherwise, we will - * do the PAL calls and cache core and thread id's of all the siblings. - */ - if (check_for_new_socket(proc_fixed_addr, c)) - return; - - for (i = 0; i < count; i++) { - int index; - - if (i && (status = ia64_pal_logical_to_phys(i, &info)) - != PAL_STATUS_SUCCESS) { - printk(KERN_ERR "ia64_pal_logical_to_phys failed" - " with %ld\n", status); - return; - } - if (info.log2_la == proc_fixed_addr) { - c->core_id = info.log1_cid; - c->thread_id = info.log1_tid; - } - - index = check_for_mtinfo_index(); - /* We will not do the mt_info caching optimization in this case. - */ - if (index < 0) - continue; - - mt_info[index].valid = 1; - mt_info[index].socket_id = c->socket_id; - mt_info[index].core_id = info.log1_cid; - mt_info[index].thread_id = info.log1_tid; - mt_info[index].proc_fixed_addr = info.log2_la; - } + c->core_id = info.log1_cid; + c->thread_id = info.log1_tid; } diff --git a/include/asm-ia64/pal.h b/include/asm-ia64/pal.h index 7708ec669a3..4e7e6f23b08 100644 --- a/include/asm-ia64/pal.h +++ b/include/asm-ia64/pal.h @@ -1640,8 +1640,7 @@ ia64_pal_logical_to_phys(u64 proc_number, pal_logical_to_physical_t *mapping) if (iprv.status == PAL_STATUS_SUCCESS) { - if (proc_number == 0) - mapping->overview.overview_data = iprv.v0; + mapping->overview.overview_data = iprv.v0; mapping->ppli1.ppli1_data = iprv.v1; mapping->ppli2.ppli2_data = iprv.v2; } -- cgit v1.2.3-70-g09d2 From c08b8a49100715b20e6f7c997e992428b5e06078 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sat, 25 Mar 2006 03:06:33 -0800 Subject: [PATCH] sys_alarm() unsigned signed conversion fixup alarm() calls the kernel with an unsigend int timeout in seconds. The value is stored in the tv_sec field of a struct timeval to setup the itimer. The tv_sec field of struct timeval is of type long, which causes the tv_sec value to be negative on 32 bit machines if seconds > INT_MAX. Before the hrtimer merge (pre 2.6.16) such a negative value was converted to the maximum jiffies timeout by the timeval_to_jiffies conversion. It's not clear whether this was intended or just happened to be done by the timeval_to_jiffies code. hrtimers expect a timeval in canonical form and treat a negative timeout as already expired. This breaks the legitimate usage of alarm() with a timeout value > INT_MAX seconds. For 32 bit machines it is therefor necessary to limit the internal seconds value to avoid API breakage. Instead of doing this in all implementations of sys_alarm the duplicated sys_alarm code is moved into a common function in itimer.c Signed-off-by: Thomas Gleixner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/ia32/sys_ia32.c | 14 +------------- arch/mips/kernel/sysirix.c | 22 +--------------------- arch/x86_64/ia32/sys_ia32.c | 16 ++-------------- include/linux/time.h | 1 + kernel/itimer.c | 37 +++++++++++++++++++++++++++++++++++++ kernel/timer.c | 14 +------------- 6 files changed, 43 insertions(+), 61 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/ia32/sys_ia32.c b/arch/ia64/ia32/sys_ia32.c index 70dba1f0e2e..13e739e4c84 100644 --- a/arch/ia64/ia32/sys_ia32.c +++ b/arch/ia64/ia32/sys_ia32.c @@ -1166,19 +1166,7 @@ put_tv32 (struct compat_timeval __user *o, struct timeval *i) asmlinkage unsigned long sys32_alarm (unsigned int seconds) { - struct itimerval it_new, it_old; - unsigned int oldalarm; - - it_new.it_interval.tv_sec = it_new.it_interval.tv_usec = 0; - it_new.it_value.tv_sec = seconds; - it_new.it_value.tv_usec = 0; - do_setitimer(ITIMER_REAL, &it_new, &it_old); - oldalarm = it_old.it_value.tv_sec; - /* ehhh.. We can't return 0 if we have an alarm pending.. */ - /* And we'd better return too much than too little anyway */ - if (it_old.it_value.tv_usec) - oldalarm++; - return oldalarm; + return alarm_setitimer(seconds); } /* Translations due to time_t size differences. Which affects all diff --git a/arch/mips/kernel/sysirix.c b/arch/mips/kernel/sysirix.c index 0fc3730a294..5407b784cd0 100644 --- a/arch/mips/kernel/sysirix.c +++ b/arch/mips/kernel/sysirix.c @@ -645,27 +645,7 @@ static inline void getitimer_real(struct itimerval *value) asmlinkage unsigned int irix_alarm(unsigned int seconds) { - struct itimerval it_new, it_old; - unsigned int oldalarm; - - if (!seconds) { - getitimer_real(&it_old); - del_timer(¤t->real_timer); - } else { - it_new.it_interval.tv_sec = it_new.it_interval.tv_usec = 0; - it_new.it_value.tv_sec = seconds; - it_new.it_value.tv_usec = 0; - do_setitimer(ITIMER_REAL, &it_new, &it_old); - } - oldalarm = it_old.it_value.tv_sec; - /* - * ehhh.. We can't return 0 if we have an alarm pending ... - * And we'd better return too much than too little anyway - */ - if (it_old.it_value.tv_usec) - oldalarm++; - - return oldalarm; + return alarm_setitimer(seconds); } asmlinkage int irix_pause(void) diff --git a/arch/x86_64/ia32/sys_ia32.c b/arch/x86_64/ia32/sys_ia32.c index 2bc55af9541..2b2d029f477 100644 --- a/arch/x86_64/ia32/sys_ia32.c +++ b/arch/x86_64/ia32/sys_ia32.c @@ -430,24 +430,12 @@ put_tv32(struct compat_timeval __user *o, struct timeval *i) return err; } -extern int do_setitimer(int which, struct itimerval *, struct itimerval *); +extern unsigned int alarm_setitimer(unsigned int seconds); asmlinkage long sys32_alarm(unsigned int seconds) { - struct itimerval it_new, it_old; - unsigned int oldalarm; - - it_new.it_interval.tv_sec = it_new.it_interval.tv_usec = 0; - it_new.it_value.tv_sec = seconds; - it_new.it_value.tv_usec = 0; - do_setitimer(ITIMER_REAL, &it_new, &it_old); - oldalarm = it_old.it_value.tv_sec; - /* ehhh.. We can't return 0 if we have an alarm pending.. */ - /* And we'd better return too much than too little anyway */ - if (it_old.it_value.tv_usec) - oldalarm++; - return oldalarm; + return alarm_setitimer(seconds); } /* Translations due to time_t size differences. Which affects all diff --git a/include/linux/time.h b/include/linux/time.h index d9cdba54b78..bf0e785e2e0 100644 --- a/include/linux/time.h +++ b/include/linux/time.h @@ -101,6 +101,7 @@ extern long do_utimes(int dfd, char __user *filename, struct timeval *times); struct itimerval; extern int do_setitimer(int which, struct itimerval *value, struct itimerval *ovalue); +extern unsigned int alarm_setitimer(unsigned int seconds); extern int do_getitimer(int which, struct itimerval *value); extern void getnstimeofday(struct timespec *tv); diff --git a/kernel/itimer.c b/kernel/itimer.c index 379be2f8c84..a2dc375927d 100644 --- a/kernel/itimer.c +++ b/kernel/itimer.c @@ -226,6 +226,43 @@ again: return 0; } +/** + * alarm_setitimer - set alarm in seconds + * + * @seconds: number of seconds until alarm + * 0 disables the alarm + * + * Returns the remaining time in seconds of a pending timer or 0 when + * the timer is not active. + * + * On 32 bit machines the seconds value is limited to (INT_MAX/2) to avoid + * negative timeval settings which would cause immediate expiry. + */ +unsigned int alarm_setitimer(unsigned int seconds) +{ + struct itimerval it_new, it_old; + +#if BITS_PER_LONG < 64 + if (seconds > INT_MAX) + seconds = INT_MAX; +#endif + it_new.it_value.tv_sec = seconds; + it_new.it_value.tv_usec = 0; + it_new.it_interval.tv_sec = it_new.it_interval.tv_usec = 0; + + do_setitimer(ITIMER_REAL, &it_new, &it_old); + + /* + * We can't return 0 if we have an alarm pending ... And we'd + * better return too much than too little anyway + */ + if ((!it_old.it_value.tv_sec && it_old.it_value.tv_usec) || + it_old.it_value.tv_usec >= 500000) + it_old.it_value.tv_sec++; + + return it_old.it_value.tv_sec; +} + asmlinkage long sys_setitimer(int which, struct itimerval __user *value, struct itimerval __user *ovalue) diff --git a/kernel/timer.c b/kernel/timer.c index 17d956cebcb..13fa72cac7d 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -956,19 +956,7 @@ void do_timer(struct pt_regs *regs) */ asmlinkage unsigned long sys_alarm(unsigned int seconds) { - struct itimerval it_new, it_old; - unsigned int oldalarm; - - it_new.it_interval.tv_sec = it_new.it_interval.tv_usec = 0; - it_new.it_value.tv_sec = seconds; - it_new.it_value.tv_usec = 0; - do_setitimer(ITIMER_REAL, &it_new, &it_old); - oldalarm = it_old.it_value.tv_sec; - /* ehhh.. We can't return 0 if we have an alarm pending.. */ - /* And we'd better return too much than too little anyway */ - if ((!oldalarm && it_old.it_value.tv_usec) || it_old.it_value.tv_usec >= 500000) - oldalarm++; - return oldalarm; + return alarm_setitimer(seconds); } #endif -- cgit v1.2.3-70-g09d2 From 3ed3bce846abc7ef460104b461cac793e41afe5e Mon Sep 17 00:00:00 2001 From: Matt Domsch Date: Sun, 26 Mar 2006 01:37:03 -0800 Subject: [PATCH] ia64: use i386 dmi_scan.c Enable DMI table parsing on ia64. Andi Kleen has a patch in his x86_64 tree which enables the use of i386 dmi_scan.c on x86_64. dmi_scan.c functions are being used by the drivers/char/ipmi/ipmi_si_intf.c driver for autodetecting the ports or memory spaces where the IPMI controllers may be found. This patch adds equivalent changes for ia64 as to what is in the x86_64 tree. In addition, I reworked the DMI detection, such that on EFI-capable systems, it uses the efi.smbios pointer to find the table, rather than brute-force searching from 0xF0000. On non-EFI systems, it continues the brute-force search. My test system, an Intel S870BN4 'Tiger4', aka Dell PowerEdge 7250, with latest BIOS, does not list the IPMI controller in the ACPI namespace, nor does it have an ACPI SPMI table. Also note, currently shipping Dell x8xx EM64T servers don't have these either, so DMI is the only method for obtaining the address of the IPMI controller. Signed-off-by: Matt Domsch Acked-by: "Luck, Tony" Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/dmi_scan.c | 90 +++++++++++++++++++++++++++++---------------- arch/ia64/Kconfig | 4 ++ arch/ia64/kernel/Makefile | 3 +- arch/ia64/kernel/setup.c | 8 ++++ include/asm-ia64/dmi.h | 6 +++ include/asm-ia64/io.h | 5 +++ 6 files changed, 83 insertions(+), 33 deletions(-) create mode 100644 include/asm-ia64/dmi.h (limited to 'arch/ia64') diff --git a/arch/i386/kernel/dmi_scan.c b/arch/i386/kernel/dmi_scan.c index ebc8dc116c4..bfecbd46f22 100644 --- a/arch/i386/kernel/dmi_scan.c +++ b/arch/i386/kernel/dmi_scan.c @@ -3,6 +3,7 @@ #include #include #include +#include #include #include #include @@ -185,47 +186,72 @@ static void __init dmi_decode(struct dmi_header *dm) } } -void __init dmi_scan_machine(void) +static int __init dmi_present(char __iomem *p) { u8 buf[15]; - char __iomem *p, *q; + memcpy_fromio(buf, p, 15); + if ((memcmp(buf, "_DMI_", 5) == 0) && dmi_checksum(buf)) { + u16 num = (buf[13] << 8) | buf[12]; + u16 len = (buf[7] << 8) | buf[6]; + u32 base = (buf[11] << 24) | (buf[10] << 16) | + (buf[9] << 8) | buf[8]; - /* - * no iounmap() for that ioremap(); it would be a no-op, but it's - * so early in setup that sucker gets confused into doing what - * it shouldn't if we actually call it. - */ - p = ioremap(0xF0000, 0x10000); - if (p == NULL) - goto out; - - for (q = p; q < p + 0x10000; q += 16) { - memcpy_fromio(buf, q, 15); - if ((memcmp(buf, "_DMI_", 5) == 0) && dmi_checksum(buf)) { - u16 num = (buf[13] << 8) | buf[12]; - u16 len = (buf[7] << 8) | buf[6]; - u32 base = (buf[11] << 24) | (buf[10] << 16) | - (buf[9] << 8) | buf[8]; - - /* - * DMI version 0.0 means that the real version is taken from - * the SMBIOS version, which we don't know at this point. - */ - if (buf[14] != 0) - printk(KERN_INFO "DMI %d.%d present.\n", - buf[14] >> 4, buf[14] & 0xF); - else - printk(KERN_INFO "DMI present.\n"); + /* + * DMI version 0.0 means that the real version is taken from + * the SMBIOS version, which we don't know at this point. + */ + if (buf[14] != 0) + printk(KERN_INFO "DMI %d.%d present.\n", + buf[14] >> 4, buf[14] & 0xF); + else + printk(KERN_INFO "DMI present.\n"); + if (dmi_table(base,len, num, dmi_decode) == 0) + return 0; + } + return 1; +} - if (dmi_table(base,len, num, dmi_decode) == 0) +void __init dmi_scan_machine(void) +{ + char __iomem *p, *q; + int rc; + + if (efi_enabled) { + if (!efi.smbios) + goto out; + + /* This is called as a core_initcall() because it isn't + * needed during early boot. This also means we can + * iounmap the space when we're done with it. + */ + p = dmi_ioremap((unsigned long)efi.smbios, 0x10000); + if (p == NULL) + goto out; + + rc = dmi_present(p + 0x10); /* offset of _DMI_ string */ + iounmap(p); + if (!rc) + return; + } + else { + /* + * no iounmap() for that ioremap(); it would be a no-op, but + * it's so early in setup that sucker gets confused into doing + * what it shouldn't if we actually call it. + */ + p = dmi_ioremap(0xF0000, 0x10000); + if (p == NULL) + goto out; + + for (q = p; q < p + 0x10000; q += 16) { + rc = dmi_present(q); + if (!rc) return; } } - -out: printk(KERN_INFO "DMI not present or invalid.\n"); + out: printk(KERN_INFO "DMI not present or invalid.\n"); } - /** * dmi_check_system - check system DMI data * @list: array of dmi_system_id structures to match against diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index 10b6b9e7716..d790a6d9026 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -42,6 +42,10 @@ config TIME_INTERPOLATION bool default y +config DMI + bool + default y + config EFI bool default y diff --git a/arch/ia64/kernel/Makefile b/arch/ia64/kernel/Makefile index 09a0dbc17fb..59e871dae74 100644 --- a/arch/ia64/kernel/Makefile +++ b/arch/ia64/kernel/Makefile @@ -7,7 +7,7 @@ extra-y := head.o init_task.o vmlinux.lds obj-y := acpi.o entry.o efi.o efi_stub.o gate-data.o fsys.o ia64_ksyms.o irq.o irq_ia64.o \ irq_lsapic.o ivt.o machvec.o pal.o patch.o process.o perfmon.o ptrace.o sal.o \ salinfo.o semaphore.o setup.o signal.o sys_ia64.o time.o traps.o unaligned.o \ - unwind.o mca.o mca_asm.o topology.o + unwind.o mca.o mca_asm.o topology.o dmi_scan.o obj-$(CONFIG_IA64_BRL_EMU) += brl_emu.o obj-$(CONFIG_IA64_GENERIC) += acpi-ext.o @@ -30,6 +30,7 @@ obj-$(CONFIG_IA64_MCA_RECOVERY) += mca_recovery.o obj-$(CONFIG_KPROBES) += kprobes.o jprobes.o obj-$(CONFIG_IA64_UNCACHED_ALLOCATOR) += uncached.o mca_recovery-y += mca_drv.o mca_drv_asm.o +dmi_scan-y += ../../i386/kernel/dmi_scan.o # The gate DSO image is built using a special linker script. targets += gate.so gate-syms.o diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c index eb388e271b2..a4421a66ea5 100644 --- a/arch/ia64/kernel/setup.c +++ b/arch/ia64/kernel/setup.c @@ -37,6 +37,7 @@ #include #include #include +#include #include #include #include @@ -887,3 +888,10 @@ check_bugs (void) ia64_patch_mckinley_e9((unsigned long) __start___mckinley_e9_bundles, (unsigned long) __end___mckinley_e9_bundles); } + +static int __init run_dmi_scan(void) +{ + dmi_scan_machine(); + return 0; +} +core_initcall(run_dmi_scan); diff --git a/include/asm-ia64/dmi.h b/include/asm-ia64/dmi.h new file mode 100644 index 00000000000..f3efaa22952 --- /dev/null +++ b/include/asm-ia64/dmi.h @@ -0,0 +1,6 @@ +#ifndef _ASM_DMI_H +#define _ASM_DMI_H 1 + +#include + +#endif diff --git a/include/asm-ia64/io.h b/include/asm-ia64/io.h index b64fdb98549..0d9bcc36f2a 100644 --- a/include/asm-ia64/io.h +++ b/include/asm-ia64/io.h @@ -435,6 +435,11 @@ iounmap (volatile void __iomem *addr) #define ioremap_nocache(o,s) ioremap(o,s) +/* Use normal IO mappings for DMI */ +#define dmi_ioremap ioremap +#define dmi_iounmap(x,l) iounmap(x) +#define dmi_alloc(l) kmalloc(l, GFP_ATOMIC) + # ifdef __KERNEL__ /* -- cgit v1.2.3-70-g09d2 From 136939a2b5aa4302281215745ccd567e1df2e8d4 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Sun, 26 Mar 2006 01:37:05 -0800 Subject: [PATCH] EFI, /dev/mem: simplify efi_mem_attribute_range() Pass the size, not a pointer to the size, to efi_mem_attribute_range(). This function validates memory regions for the /dev/mem read/write/mmap paths. The pointer allows arches to reduce the size of the range, but I think that's unnecessary complexity. Simplifying it will let me use efi_mem_attribute_range() to improve the ia64 ioremap() implementation. Signed-off-by: Bjorn Helgaas Cc: Matt Domsch Cc: "Tolentino, Matthew E" Cc: "Brown, Len" Cc: Andi Kleen Acked-by: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/kernel/efi.c | 41 ++++++++++++++++++++--------------------- drivers/char/mem.c | 18 ++++++------------ include/asm-ia64/io.h | 4 ++-- include/linux/efi.h | 2 ++ 4 files changed, 30 insertions(+), 35 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/efi.c b/arch/ia64/kernel/efi.c index 9990320b6f9..2993748c13d 100644 --- a/arch/ia64/kernel/efi.c +++ b/arch/ia64/kernel/efi.c @@ -677,27 +677,34 @@ EXPORT_SYMBOL(efi_mem_attributes); /* * Determines whether the memory at phys_addr supports the desired * attribute (WB, UC, etc). If this returns 1, the caller can safely - * access *size bytes at phys_addr with the specified attribute. + * access size bytes at phys_addr with the specified attribute. */ -static int -efi_mem_attribute_range (unsigned long phys_addr, unsigned long *size, u64 attr) +int +efi_mem_attribute_range (unsigned long phys_addr, unsigned long size, u64 attr) { + unsigned long end = phys_addr + size; efi_memory_desc_t *md = efi_memory_descriptor(phys_addr); - unsigned long md_end; - if (!md || (md->attribute & attr) != attr) + /* + * Some firmware doesn't report MMIO regions in the EFI memory + * map. The Intel BigSur (a.k.a. HP i2000) has this problem. + * On those platforms, we have to assume UC is valid everywhere. + */ + if (!md || (md->attribute & attr) != attr) { + if (attr == EFI_MEMORY_UC && !efi_memmap_has_mmio()) + return 1; return 0; + } do { - md_end = efi_md_end(md); - if (phys_addr + *size <= md_end) + unsigned long md_end = efi_md_end(md); + + if (end <= md_end) return 1; md = efi_memory_descriptor(md_end); - if (!md || (md->attribute & attr) != attr) { - *size = md_end - phys_addr; - return 1; - } + if (!md || (md->attribute & attr) != attr) + return 0; } while (md); return 0; } @@ -708,7 +715,7 @@ efi_mem_attribute_range (unsigned long phys_addr, unsigned long *size, u64 attr) * control access size. */ int -valid_phys_addr_range (unsigned long phys_addr, unsigned long *size) +valid_phys_addr_range (unsigned long phys_addr, unsigned long size) { return efi_mem_attribute_range(phys_addr, size, EFI_MEMORY_WB); } @@ -723,7 +730,7 @@ valid_phys_addr_range (unsigned long phys_addr, unsigned long *size) * because that doesn't appear in the boot-time EFI memory map. */ int -valid_mmap_phys_addr_range (unsigned long phys_addr, unsigned long *size) +valid_mmap_phys_addr_range (unsigned long phys_addr, unsigned long size) { if (efi_mem_attribute_range(phys_addr, size, EFI_MEMORY_WB)) return 1; @@ -731,14 +738,6 @@ valid_mmap_phys_addr_range (unsigned long phys_addr, unsigned long *size) if (efi_mem_attribute_range(phys_addr, size, EFI_MEMORY_UC)) return 1; - /* - * Some firmware doesn't report MMIO regions in the EFI memory map. - * The Intel BigSur (a.k.a. HP i2000) has this problem. In this - * case, we can't use the EFI memory map to validate mmap requests. - */ - if (!efi_memmap_has_mmio()) - return 1; - return 0; } diff --git a/drivers/char/mem.c b/drivers/char/mem.c index 26d0116b48d..5245ba1649e 100644 --- a/drivers/char/mem.c +++ b/drivers/char/mem.c @@ -88,21 +88,15 @@ static inline int uncached_access(struct file *file, unsigned long addr) } #ifndef ARCH_HAS_VALID_PHYS_ADDR_RANGE -static inline int valid_phys_addr_range(unsigned long addr, size_t *count) +static inline int valid_phys_addr_range(unsigned long addr, size_t count) { - unsigned long end_mem; - - end_mem = __pa(high_memory); - if (addr >= end_mem) + if (addr + count > __pa(high_memory)) return 0; - if (*count > end_mem - addr) - *count = end_mem - addr; - return 1; } -static inline int valid_mmap_phys_addr_range(unsigned long addr, size_t *size) +static inline int valid_mmap_phys_addr_range(unsigned long addr, size_t size) { return 1; } @@ -119,7 +113,7 @@ static ssize_t read_mem(struct file * file, char __user * buf, ssize_t read, sz; char *ptr; - if (!valid_phys_addr_range(p, &count)) + if (!valid_phys_addr_range(p, count)) return -EFAULT; read = 0; #ifdef __ARCH_HAS_NO_PAGE_ZERO_MAPPED @@ -177,7 +171,7 @@ static ssize_t write_mem(struct file * file, const char __user * buf, unsigned long copied; void *ptr; - if (!valid_phys_addr_range(p, &count)) + if (!valid_phys_addr_range(p, count)) return -EFAULT; written = 0; @@ -249,7 +243,7 @@ static int mmap_mem(struct file * file, struct vm_area_struct * vma) { size_t size = vma->vm_end - vma->vm_start; - if (!valid_mmap_phys_addr_range(vma->vm_pgoff << PAGE_SHIFT, &size)) + if (!valid_mmap_phys_addr_range(vma->vm_pgoff << PAGE_SHIFT, size)) return -EINVAL; vma->vm_page_prot = phys_mem_access_prot(file, vma->vm_pgoff, diff --git a/include/asm-ia64/io.h b/include/asm-ia64/io.h index 0d9bcc36f2a..acba019b08e 100644 --- a/include/asm-ia64/io.h +++ b/include/asm-ia64/io.h @@ -88,8 +88,8 @@ phys_to_virt (unsigned long address) } #define ARCH_HAS_VALID_PHYS_ADDR_RANGE -extern int valid_phys_addr_range (unsigned long addr, size_t *count); /* efi.c */ -extern int valid_mmap_phys_addr_range (unsigned long addr, size_t *count); +extern int valid_phys_addr_range (unsigned long addr, size_t count); /* efi.c */ +extern int valid_mmap_phys_addr_range (unsigned long addr, size_t count); /* * The following two macros are deprecated and scheduled for removal. diff --git a/include/linux/efi.h b/include/linux/efi.h index c7c5dd31618..d15725470aa 100644 --- a/include/linux/efi.h +++ b/include/linux/efi.h @@ -292,6 +292,8 @@ extern void efi_enter_virtual_mode (void); /* switch EFI to virtual mode, if pos extern u64 efi_get_iobase (void); extern u32 efi_mem_type (unsigned long phys_addr); extern u64 efi_mem_attributes (unsigned long phys_addr); +extern int efi_mem_attribute_range (unsigned long phys_addr, unsigned long size, + u64 attr); extern int __init efi_uart_console_only (void); extern void efi_initialize_iomem_resources(struct resource *code_resource, struct resource *data_resource); -- cgit v1.2.3-70-g09d2 From e9b0a0712148abe96ff717a2b9f8dab1d433e0d5 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Sun, 26 Mar 2006 01:37:06 -0800 Subject: [PATCH] ia64: ioremap: check EFI for valid memory attributes Check the EFI memory map so we can use the correct memory attributes for ioremap(). Previously, we always used uncacheable access, which blows up on some machines for regular system memory. Signed-off-by: Bjorn Helgaas Cc: Matt Domsch Cc: "Tolentino, Matthew E" Cc: "Brown, Len" Cc: Andi Kleen Acked-by: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/mm/Makefile | 2 +- arch/ia64/mm/ioremap.c | 43 +++++++++++++++++++++++++++++++++++++++++++ include/asm-ia64/io.h | 15 ++------------- 3 files changed, 46 insertions(+), 14 deletions(-) create mode 100644 arch/ia64/mm/ioremap.c (limited to 'arch/ia64') diff --git a/arch/ia64/mm/Makefile b/arch/ia64/mm/Makefile index d78d20f0a0f..bb0a01a8187 100644 --- a/arch/ia64/mm/Makefile +++ b/arch/ia64/mm/Makefile @@ -2,7 +2,7 @@ # Makefile for the ia64-specific parts of the memory manager. # -obj-y := init.o fault.o tlb.o extable.o +obj-y := init.o fault.o tlb.o extable.o ioremap.o obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o obj-$(CONFIG_NUMA) += numa.o diff --git a/arch/ia64/mm/ioremap.c b/arch/ia64/mm/ioremap.c new file mode 100644 index 00000000000..62328621f99 --- /dev/null +++ b/arch/ia64/mm/ioremap.c @@ -0,0 +1,43 @@ +/* + * (c) Copyright 2006 Hewlett-Packard Development Company, L.P. + * Bjorn Helgaas + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include +#include +#include +#include + +static inline void __iomem * +__ioremap (unsigned long offset, unsigned long size) +{ + return (void __iomem *) (__IA64_UNCACHED_OFFSET | offset); +} + +void __iomem * +ioremap (unsigned long offset, unsigned long size) +{ + if (efi_mem_attribute_range(offset, size, EFI_MEMORY_UC)) + return __ioremap(offset, size); + + if (efi_mem_attribute_range(offset, size, EFI_MEMORY_WB)) + return phys_to_virt(offset); + + /* + * Someday this should check ACPI resources so we + * can do the right thing for hot-plugged regions. + */ + return __ioremap(offset, size); +} +EXPORT_SYMBOL(ioremap); + +void __iomem * +ioremap_nocache (unsigned long offset, unsigned long size) +{ + return __ioremap(offset, size); +} +EXPORT_SYMBOL(ioremap_nocache); diff --git a/include/asm-ia64/io.h b/include/asm-ia64/io.h index acba019b08e..c2e3742108b 100644 --- a/include/asm-ia64/io.h +++ b/include/asm-ia64/io.h @@ -416,25 +416,14 @@ __writeq (unsigned long val, volatile void __iomem *addr) # define outl_p outl #endif -/* - * An "address" in IO memory space is not clearly either an integer or a pointer. We will - * accept both, thus the casts. - * - * On ia-64, we access the physical I/O memory space through the uncached kernel region. - */ -static inline void __iomem * -ioremap (unsigned long offset, unsigned long size) -{ - return (void __iomem *) (__IA64_UNCACHED_OFFSET | (offset)); -} +extern void __iomem * ioremap(unsigned long offset, unsigned long size); +extern void __iomem * ioremap_nocache (unsigned long offset, unsigned long size); static inline void iounmap (volatile void __iomem *addr) { } -#define ioremap_nocache(o,s) ioremap(o,s) - /* Use normal IO mappings for DMI */ #define dmi_ioremap ioremap #define dmi_iounmap(x,l) iounmap(x) -- cgit v1.2.3-70-g09d2 From b2c99e3c70d77fb194df5aa1642030080d28ea48 Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Sun, 26 Mar 2006 01:37:08 -0800 Subject: [PATCH] EFI: keep physical table addresses in efi structure Almost all users of the table addresses from the EFI system table want physical addresses. So rather than doing the pa->va->pa conversion, just keep physical addresses in struct efi. This fixes a DMI bug: the efi structure contained the physical SMBIOS address on x86 but the virtual address on ia64, so dmi_scan_machine() used ioremap() on a virtual address on ia64. This is essentially the same as an earlier patch by Matt Tolentino: http://marc.theaimsgroup.com/?l=linux-kernel&m=112130292316281&w=2 except that this changes all table addresses, not just ACPI addresses. Matt's original patch was backed out because it caused MCAs on HP sx1000 systems. That problem is resolved by the ioremap() attribute checking added for ia64. Signed-off-by: Bjorn Helgaas Cc: Matt Domsch Cc: "Tolentino, Matthew E" Cc: "Brown, Len" Cc: Andi Kleen Acked-by: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/acpi/boot.c | 8 ++++---- arch/i386/kernel/dmi_scan.c | 4 ++-- arch/i386/kernel/efi.c | 21 +++++++++++++++------ arch/ia64/kernel/acpi.c | 6 +++--- arch/ia64/kernel/efi.c | 21 +++++++++++++++------ arch/ia64/kernel/setup.c | 2 +- arch/ia64/sn/kernel/setup.c | 5 +++-- drivers/acpi/osl.c | 10 ++++------ drivers/firmware/efivars.c | 28 ++++++++++++++-------------- drivers/firmware/pcdp.c | 19 ++++++++++++------- include/asm-ia64/sn/sn_sal.h | 2 +- include/linux/efi.h | 18 ++++++++++-------- 12 files changed, 84 insertions(+), 60 deletions(-) (limited to 'arch/ia64') diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c index f1a21945963..033066176b3 100644 --- a/arch/i386/kernel/acpi/boot.c +++ b/arch/i386/kernel/acpi/boot.c @@ -668,10 +668,10 @@ unsigned long __init acpi_find_rsdp(void) unsigned long rsdp_phys = 0; if (efi_enabled) { - if (efi.acpi20) - return __pa(efi.acpi20); - else if (efi.acpi) - return __pa(efi.acpi); + if (efi.acpi20 != EFI_INVALID_TABLE_ADDR) + return efi.acpi20; + else if (efi.acpi != EFI_INVALID_TABLE_ADDR) + return efi.acpi; } /* * Scan memory looking for the RSDP signature. First search EBDA (low diff --git a/arch/i386/kernel/dmi_scan.c b/arch/i386/kernel/dmi_scan.c index c032f9e06bb..170d4c9f9bc 100644 --- a/arch/i386/kernel/dmi_scan.c +++ b/arch/i386/kernel/dmi_scan.c @@ -217,14 +217,14 @@ void __init dmi_scan_machine(void) int rc; if (efi_enabled) { - if (!efi.smbios) + if (efi.smbios == EFI_INVALID_TABLE_ADDR) goto out; /* This is called as a core_initcall() because it isn't * needed during early boot. This also means we can * iounmap the space when we're done with it. */ - p = dmi_ioremap((unsigned long)efi.smbios, 32); + p = dmi_ioremap(efi.smbios, 32); if (p == NULL) goto out; diff --git a/arch/i386/kernel/efi.c b/arch/i386/kernel/efi.c index 7ec6cfa01fb..c224c2aebba 100644 --- a/arch/i386/kernel/efi.c +++ b/arch/i386/kernel/efi.c @@ -381,29 +381,38 @@ void __init efi_init(void) if (config_tables == NULL) printk(KERN_ERR PFX "Could not map EFI Configuration Table!\n"); + efi.mps = EFI_INVALID_TABLE_ADDR; + efi.acpi = EFI_INVALID_TABLE_ADDR; + efi.acpi20 = EFI_INVALID_TABLE_ADDR; + efi.smbios = EFI_INVALID_TABLE_ADDR; + efi.sal_systab = EFI_INVALID_TABLE_ADDR; + efi.boot_info = EFI_INVALID_TABLE_ADDR; + efi.hcdp = EFI_INVALID_TABLE_ADDR; + efi.uga = EFI_INVALID_TABLE_ADDR; + for (i = 0; i < num_config_tables; i++) { if (efi_guidcmp(config_tables[i].guid, MPS_TABLE_GUID) == 0) { - efi.mps = (void *)config_tables[i].table; + efi.mps = config_tables[i].table; printk(KERN_INFO " MPS=0x%lx ", config_tables[i].table); } else if (efi_guidcmp(config_tables[i].guid, ACPI_20_TABLE_GUID) == 0) { - efi.acpi20 = __va(config_tables[i].table); + efi.acpi20 = config_tables[i].table; printk(KERN_INFO " ACPI 2.0=0x%lx ", config_tables[i].table); } else if (efi_guidcmp(config_tables[i].guid, ACPI_TABLE_GUID) == 0) { - efi.acpi = __va(config_tables[i].table); + efi.acpi = config_tables[i].table; printk(KERN_INFO " ACPI=0x%lx ", config_tables[i].table); } else if (efi_guidcmp(config_tables[i].guid, SMBIOS_TABLE_GUID) == 0) { - efi.smbios = (void *) config_tables[i].table; + efi.smbios = config_tables[i].table; printk(KERN_INFO " SMBIOS=0x%lx ", config_tables[i].table); } else if (efi_guidcmp(config_tables[i].guid, HCDP_TABLE_GUID) == 0) { - efi.hcdp = (void *)config_tables[i].table; + efi.hcdp = config_tables[i].table; printk(KERN_INFO " HCDP=0x%lx ", config_tables[i].table); } else if (efi_guidcmp(config_tables[i].guid, UGA_IO_PROTOCOL_GUID) == 0) { - efi.uga = (void *)config_tables[i].table; + efi.uga = config_tables[i].table; printk(KERN_INFO " UGA=0x%lx ", config_tables[i].table); } } diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c index a4e218ce2ed..58c93a30348 100644 --- a/arch/ia64/kernel/acpi.c +++ b/arch/ia64/kernel/acpi.c @@ -651,9 +651,9 @@ unsigned long __init acpi_find_rsdp(void) { unsigned long rsdp_phys = 0; - if (efi.acpi20) - rsdp_phys = __pa(efi.acpi20); - else if (efi.acpi) + if (efi.acpi20 != EFI_INVALID_TABLE_ADDR) + rsdp_phys = efi.acpi20; + else if (efi.acpi != EFI_INVALID_TABLE_ADDR) printk(KERN_WARNING PREFIX "v1.0/r0.71 tables no longer supported\n"); return rsdp_phys; diff --git a/arch/ia64/kernel/efi.c b/arch/ia64/kernel/efi.c index 2993748c13d..12cfedce73b 100644 --- a/arch/ia64/kernel/efi.c +++ b/arch/ia64/kernel/efi.c @@ -458,24 +458,33 @@ efi_init (void) printk(KERN_INFO "EFI v%u.%.02u by %s:", efi.systab->hdr.revision >> 16, efi.systab->hdr.revision & 0xffff, vendor); + efi.mps = EFI_INVALID_TABLE_ADDR; + efi.acpi = EFI_INVALID_TABLE_ADDR; + efi.acpi20 = EFI_INVALID_TABLE_ADDR; + efi.smbios = EFI_INVALID_TABLE_ADDR; + efi.sal_systab = EFI_INVALID_TABLE_ADDR; + efi.boot_info = EFI_INVALID_TABLE_ADDR; + efi.hcdp = EFI_INVALID_TABLE_ADDR; + efi.uga = EFI_INVALID_TABLE_ADDR; + for (i = 0; i < (int) efi.systab->nr_tables; i++) { if (efi_guidcmp(config_tables[i].guid, MPS_TABLE_GUID) == 0) { - efi.mps = __va(config_tables[i].table); + efi.mps = config_tables[i].table; printk(" MPS=0x%lx", config_tables[i].table); } else if (efi_guidcmp(config_tables[i].guid, ACPI_20_TABLE_GUID) == 0) { - efi.acpi20 = __va(config_tables[i].table); + efi.acpi20 = config_tables[i].table; printk(" ACPI 2.0=0x%lx", config_tables[i].table); } else if (efi_guidcmp(config_tables[i].guid, ACPI_TABLE_GUID) == 0) { - efi.acpi = __va(config_tables[i].table); + efi.acpi = config_tables[i].table; printk(" ACPI=0x%lx", config_tables[i].table); } else if (efi_guidcmp(config_tables[i].guid, SMBIOS_TABLE_GUID) == 0) { - efi.smbios = __va(config_tables[i].table); + efi.smbios = config_tables[i].table; printk(" SMBIOS=0x%lx", config_tables[i].table); } else if (efi_guidcmp(config_tables[i].guid, SAL_SYSTEM_TABLE_GUID) == 0) { - efi.sal_systab = __va(config_tables[i].table); + efi.sal_systab = config_tables[i].table; printk(" SALsystab=0x%lx", config_tables[i].table); } else if (efi_guidcmp(config_tables[i].guid, HCDP_TABLE_GUID) == 0) { - efi.hcdp = __va(config_tables[i].table); + efi.hcdp = config_tables[i].table; printk(" HCDP=0x%lx", config_tables[i].table); } } diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c index a4421a66ea5..e4dfda1eb7d 100644 --- a/arch/ia64/kernel/setup.c +++ b/arch/ia64/kernel/setup.c @@ -434,7 +434,7 @@ setup_arch (char **cmdline_p) find_memory(); /* process SAL system table: */ - ia64_sal_init(efi.sal_systab); + ia64_sal_init(__va(efi.sal_systab)); ia64_setup_printk_clock(); diff --git a/arch/ia64/sn/kernel/setup.c b/arch/ia64/sn/kernel/setup.c index 8b6d5c84470..30988dfbddf 100644 --- a/arch/ia64/sn/kernel/setup.c +++ b/arch/ia64/sn/kernel/setup.c @@ -327,10 +327,11 @@ sn_scan_pcdp(void) struct pcdp_interface_pci if_pci; extern struct efi efi; - pcdp = efi.hcdp; - if (! pcdp) + if (efi.hcdp == EFI_INVALID_TABLE_ADDR) return; /* no hcdp/pcdp table */ + pcdp = __va(efi.hcdp); + if (pcdp->rev < 3) return; /* only support PCDP (rev >= 3) */ diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c index ac5bbaedac1..fc8a3bce6cb 100644 --- a/drivers/acpi/osl.c +++ b/drivers/acpi/osl.c @@ -156,12 +156,10 @@ acpi_status acpi_os_get_root_pointer(u32 flags, struct acpi_pointer *addr) { if (efi_enabled) { addr->pointer_type = ACPI_PHYSICAL_POINTER; - if (efi.acpi20) - addr->pointer.physical = - (acpi_physical_address) virt_to_phys(efi.acpi20); - else if (efi.acpi) - addr->pointer.physical = - (acpi_physical_address) virt_to_phys(efi.acpi); + if (efi.acpi20 != EFI_INVALID_TABLE_ADDR) + addr->pointer.physical = efi.acpi20; + else if (efi.acpi != EFI_INVALID_TABLE_ADDR) + addr->pointer.physical = efi.acpi; else { printk(KERN_ERR PREFIX "System description tables not found\n"); diff --git a/drivers/firmware/efivars.c b/drivers/firmware/efivars.c index 343379f23a5..9b7e4d52ffd 100644 --- a/drivers/firmware/efivars.c +++ b/drivers/firmware/efivars.c @@ -568,20 +568,20 @@ systab_read(struct subsystem *entry, char *buf) if (!entry || !buf) return -EINVAL; - if (efi.mps) - str += sprintf(str, "MPS=0x%lx\n", __pa(efi.mps)); - if (efi.acpi20) - str += sprintf(str, "ACPI20=0x%lx\n", __pa(efi.acpi20)); - if (efi.acpi) - str += sprintf(str, "ACPI=0x%lx\n", __pa(efi.acpi)); - if (efi.smbios) - str += sprintf(str, "SMBIOS=0x%lx\n", __pa(efi.smbios)); - if (efi.hcdp) - str += sprintf(str, "HCDP=0x%lx\n", __pa(efi.hcdp)); - if (efi.boot_info) - str += sprintf(str, "BOOTINFO=0x%lx\n", __pa(efi.boot_info)); - if (efi.uga) - str += sprintf(str, "UGA=0x%lx\n", __pa(efi.uga)); + if (efi.mps != EFI_INVALID_TABLE_ADDR) + str += sprintf(str, "MPS=0x%lx\n", efi.mps); + if (efi.acpi20 != EFI_INVALID_TABLE_ADDR) + str += sprintf(str, "ACPI20=0x%lx\n", efi.acpi20); + if (efi.acpi != EFI_INVALID_TABLE_ADDR) + str += sprintf(str, "ACPI=0x%lx\n", efi.acpi); + if (efi.smbios != EFI_INVALID_TABLE_ADDR) + str += sprintf(str, "SMBIOS=0x%lx\n", efi.smbios); + if (efi.hcdp != EFI_INVALID_TABLE_ADDR) + str += sprintf(str, "HCDP=0x%lx\n", efi.hcdp); + if (efi.boot_info != EFI_INVALID_TABLE_ADDR) + str += sprintf(str, "BOOTINFO=0x%lx\n", efi.boot_info); + if (efi.uga != EFI_INVALID_TABLE_ADDR) + str += sprintf(str, "UGA=0x%lx\n", efi.uga); return str - buf; } diff --git a/drivers/firmware/pcdp.c b/drivers/firmware/pcdp.c index ae1fb45dbb4..c37baf9448b 100644 --- a/drivers/firmware/pcdp.c +++ b/drivers/firmware/pcdp.c @@ -89,19 +89,20 @@ efi_setup_pcdp_console(char *cmdline) struct pcdp_uart *uart; struct pcdp_device *dev, *end; int i, serial = 0; + int rc = -ENODEV; - pcdp = efi.hcdp; - if (!pcdp) + if (efi.hcdp == EFI_INVALID_TABLE_ADDR) return -ENODEV; - printk(KERN_INFO "PCDP: v%d at 0x%lx\n", pcdp->rev, __pa(pcdp)); + pcdp = ioremap(efi.hcdp, 4096); + printk(KERN_INFO "PCDP: v%d at 0x%lx\n", pcdp->rev, efi.hcdp); if (strstr(cmdline, "console=hcdp")) { if (pcdp->rev < 3) serial = 1; } else if (strstr(cmdline, "console=")) { printk(KERN_INFO "Explicit \"console=\"; ignoring PCDP\n"); - return -ENODEV; + goto out; } if (pcdp->rev < 3 && efi_uart_console_only()) @@ -110,7 +111,8 @@ efi_setup_pcdp_console(char *cmdline) for (i = 0, uart = pcdp->uart; i < pcdp->num_uarts; i++, uart++) { if (uart->flags & PCDP_UART_PRIMARY_CONSOLE || serial) { if (uart->type == PCDP_CONSOLE_UART) { - return setup_serial_console(uart); + rc = setup_serial_console(uart); + goto out; } } } @@ -121,10 +123,13 @@ efi_setup_pcdp_console(char *cmdline) dev = (struct pcdp_device *) ((u8 *) dev + dev->length)) { if (dev->flags & PCDP_PRIMARY_CONSOLE) { if (dev->type == PCDP_CONSOLE_VGA) { - return setup_vga_console(dev); + rc = setup_vga_console(dev); + goto out; } } } - return -ENODEV; +out: + iounmap(pcdp); + return rc; } diff --git a/include/asm-ia64/sn/sn_sal.h b/include/asm-ia64/sn/sn_sal.h index 244449df741..bf4cc867a69 100644 --- a/include/asm-ia64/sn/sn_sal.h +++ b/include/asm-ia64/sn/sn_sal.h @@ -159,7 +159,7 @@ static inline u32 sn_sal_rev(void) { - struct ia64_sal_systab *systab = efi.sal_systab; + struct ia64_sal_systab *systab = __va(efi.sal_systab); return (u32)(systab->sal_b_rev_major << 8 | systab->sal_b_rev_minor); } diff --git a/include/linux/efi.h b/include/linux/efi.h index d15725470aa..e203613d3ae 100644 --- a/include/linux/efi.h +++ b/include/linux/efi.h @@ -240,19 +240,21 @@ struct efi_memory_map { unsigned long desc_size; }; +#define EFI_INVALID_TABLE_ADDR (~0UL) + /* * All runtime access to EFI goes through this structure: */ extern struct efi { efi_system_table_t *systab; /* EFI system table */ - void *mps; /* MPS table */ - void *acpi; /* ACPI table (IA64 ext 0.71) */ - void *acpi20; /* ACPI table (ACPI 2.0) */ - void *smbios; /* SM BIOS table */ - void *sal_systab; /* SAL system table */ - void *boot_info; /* boot info table */ - void *hcdp; /* HCDP table */ - void *uga; /* UGA table */ + unsigned long mps; /* MPS table */ + unsigned long acpi; /* ACPI table (IA64 ext 0.71) */ + unsigned long acpi20; /* ACPI table (ACPI 2.0) */ + unsigned long smbios; /* SM BIOS table */ + unsigned long sal_systab; /* SAL system table */ + unsigned long boot_info; /* boot info table */ + unsigned long hcdp; /* HCDP table */ + unsigned long uga; /* UGA table */ efi_get_time_t *get_time; efi_set_time_t *set_time; efi_get_wakeup_time_t *get_wakeup_time; -- cgit v1.2.3-70-g09d2 From 88959ea968709c35e8b979ac9f5a398fa748091a Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Sun, 26 Mar 2006 01:37:27 -0800 Subject: [PATCH] create struct compat_timex and use it everywhere We had a copy of the compatibility version of struct timex in each 64 bit architecture. This patch just creates a global one and replaces all the usages of the old ones. Signed-off-by: Stephen Rothwell Cc: Arnd Bergmann Acked-by: Kyle McMartin Acked-by: Tony Luck Acked-by: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/ia32/sys_ia32.c | 15 +-------------- arch/mips/kernel/linux32.c | 15 +-------------- arch/parisc/kernel/sys_parisc32.c | 37 ++++--------------------------------- arch/powerpc/kernel/sys_ppc32.c | 15 +-------------- arch/s390/kernel/compat_linux.c | 15 +-------------- arch/sparc64/kernel/sys_sparc32.c | 15 +-------------- arch/x86_64/ia32/sys_ia32.c | 19 +++---------------- include/linux/compat.h | 26 ++++++++++++++++++++++++++ 8 files changed, 38 insertions(+), 119 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/ia32/sys_ia32.c b/arch/ia64/ia32/sys_ia32.c index 13e739e4c84..1e1a2353aef 100644 --- a/arch/ia64/ia32/sys_ia32.c +++ b/arch/ia64/ia32/sys_ia32.c @@ -2594,23 +2594,10 @@ sys32_setresgid(compat_gid_t rgid, compat_gid_t egid, /* Handle adjtimex compatibility. */ -struct timex32 { - u32 modes; - s32 offset, freq, maxerror, esterror; - s32 status, constant, precision, tolerance; - struct compat_timeval time; - s32 tick; - s32 ppsfreq, jitter, shift, stabil; - s32 jitcnt, calcnt, errcnt, stbcnt; - s32 :32; s32 :32; s32 :32; s32 :32; - s32 :32; s32 :32; s32 :32; s32 :32; - s32 :32; s32 :32; s32 :32; s32 :32; -}; - extern int do_adjtimex(struct timex *); asmlinkage long -sys32_adjtimex(struct timex32 *utp) +sys32_adjtimex(struct compat_timex *utp) { struct timex txc; int ret; diff --git a/arch/mips/kernel/linux32.c b/arch/mips/kernel/linux32.c index 013bc93688e..f33e779796f 100644 --- a/arch/mips/kernel/linux32.c +++ b/arch/mips/kernel/linux32.c @@ -1159,22 +1159,9 @@ out: /* Handle adjtimex compatibility. */ -struct timex32 { - u32 modes; - s32 offset, freq, maxerror, esterror; - s32 status, constant, precision, tolerance; - struct compat_timeval time; - s32 tick; - s32 ppsfreq, jitter, shift, stabil; - s32 jitcnt, calcnt, errcnt, stbcnt; - s32 :32; s32 :32; s32 :32; s32 :32; - s32 :32; s32 :32; s32 :32; s32 :32; - s32 :32; s32 :32; s32 :32; s32 :32; -}; - extern int do_adjtimex(struct timex *); -asmlinkage int sys32_adjtimex(struct timex32 __user *utp) +asmlinkage int sys32_adjtimex(struct compat_timex __user *utp) { struct timex txc; int ret; diff --git a/arch/parisc/kernel/sys_parisc32.c b/arch/parisc/kernel/sys_parisc32.c index 61356901841..ca1d8dbbadf 100644 --- a/arch/parisc/kernel/sys_parisc32.c +++ b/arch/parisc/kernel/sys_parisc32.c @@ -567,43 +567,14 @@ asmlinkage int sys32_sendfile64(int out_fd, int in_fd, compat_loff_t __user *off } -struct timex32 { - unsigned int modes; /* mode selector */ - int offset; /* time offset (usec) */ - int freq; /* frequency offset (scaled ppm) */ - int maxerror; /* maximum error (usec) */ - int esterror; /* estimated error (usec) */ - int status; /* clock command/status */ - int constant; /* pll time constant */ - int precision; /* clock precision (usec) (read only) */ - int tolerance; /* clock frequency tolerance (ppm) - * (read only) - */ - struct compat_timeval time; /* (read only) */ - int tick; /* (modified) usecs between clock ticks */ - - int ppsfreq; /* pps frequency (scaled ppm) (ro) */ - int jitter; /* pps jitter (us) (ro) */ - int shift; /* interval duration (s) (shift) (ro) */ - int stabil; /* pps stability (scaled ppm) (ro) */ - int jitcnt; /* jitter limit exceeded (ro) */ - int calcnt; /* calibration intervals (ro) */ - int errcnt; /* calibration errors (ro) */ - int stbcnt; /* stability limit exceeded (ro) */ - - int :32; int :32; int :32; int :32; - int :32; int :32; int :32; int :32; - int :32; int :32; int :32; int :32; -}; - -asmlinkage long sys32_adjtimex(struct timex32 __user *txc_p32) +asmlinkage long sys32_adjtimex(struct compat_timex __user *txc_p32) { struct timex txc; - struct timex32 t32; + struct compat_timex t32; int ret; extern int do_adjtimex(struct timex *txc); - if(copy_from_user(&t32, txc_p32, sizeof(struct timex32))) + if(copy_from_user(&t32, txc_p32, sizeof(struct compat_timex))) return -EFAULT; #undef CP #define CP(x) txc.x = t32.x @@ -620,7 +591,7 @@ asmlinkage long sys32_adjtimex(struct timex32 __user *txc_p32) CP(time.tv_sec); CP(time.tv_usec); CP(tick); CP(ppsfreq); CP(jitter); CP(shift); CP(stabil); CP(jitcnt); CP(calcnt); CP(errcnt); CP(stbcnt); - return copy_to_user(txc_p32, &t32, sizeof(struct timex32)) ? -EFAULT : ret; + return copy_to_user(txc_p32, &t32, sizeof(struct compat_timex)) ? -EFAULT : ret; } diff --git a/arch/powerpc/kernel/sys_ppc32.c b/arch/powerpc/kernel/sys_ppc32.c index cd75ab2908f..d9867f583c6 100644 --- a/arch/powerpc/kernel/sys_ppc32.c +++ b/arch/powerpc/kernel/sys_ppc32.c @@ -162,22 +162,9 @@ asmlinkage long compat_sys_sysfs(u32 option, u32 arg1, u32 arg2) } /* Handle adjtimex compatibility. */ -struct timex32 { - u32 modes; - s32 offset, freq, maxerror, esterror; - s32 status, constant, precision, tolerance; - struct compat_timeval time; - s32 tick; - s32 ppsfreq, jitter, shift, stabil; - s32 jitcnt, calcnt, errcnt, stbcnt; - s32 :32; s32 :32; s32 :32; s32 :32; - s32 :32; s32 :32; s32 :32; s32 :32; - s32 :32; s32 :32; s32 :32; s32 :32; -}; - extern int do_adjtimex(struct timex *); -asmlinkage long compat_sys_adjtimex(struct timex32 __user *utp) +asmlinkage long compat_sys_adjtimex(struct compat_timex __user *utp) { struct timex txc; int ret; diff --git a/arch/s390/kernel/compat_linux.c b/arch/s390/kernel/compat_linux.c index cc058dc3bc8..9809264f2f4 100644 --- a/arch/s390/kernel/compat_linux.c +++ b/arch/s390/kernel/compat_linux.c @@ -707,22 +707,9 @@ asmlinkage long sys32_sendfile64(int out_fd, int in_fd, /* Handle adjtimex compatibility. */ -struct timex32 { - u32 modes; - s32 offset, freq, maxerror, esterror; - s32 status, constant, precision, tolerance; - struct compat_timeval time; - s32 tick; - s32 ppsfreq, jitter, shift, stabil; - s32 jitcnt, calcnt, errcnt, stbcnt; - s32 :32; s32 :32; s32 :32; s32 :32; - s32 :32; s32 :32; s32 :32; s32 :32; - s32 :32; s32 :32; s32 :32; s32 :32; -}; - extern int do_adjtimex(struct timex *); -asmlinkage long sys32_adjtimex(struct timex32 __user *utp) +asmlinkage long sys32_adjtimex(struct compat_timex __user *utp) { struct timex txc; int ret; diff --git a/arch/sparc64/kernel/sys_sparc32.c b/arch/sparc64/kernel/sys_sparc32.c index 0e41df02448..48f02f20191 100644 --- a/arch/sparc64/kernel/sys_sparc32.c +++ b/arch/sparc64/kernel/sys_sparc32.c @@ -947,22 +947,9 @@ asmlinkage long compat_sys_sendfile64(int out_fd, int in_fd, /* Handle adjtimex compatibility. */ -struct timex32 { - u32 modes; - s32 offset, freq, maxerror, esterror; - s32 status, constant, precision, tolerance; - struct compat_timeval time; - s32 tick; - s32 ppsfreq, jitter, shift, stabil; - s32 jitcnt, calcnt, errcnt, stbcnt; - s32 :32; s32 :32; s32 :32; s32 :32; - s32 :32; s32 :32; s32 :32; s32 :32; - s32 :32; s32 :32; s32 :32; s32 :32; -}; - extern int do_adjtimex(struct timex *); -asmlinkage long sys32_adjtimex(struct timex32 __user *utp) +asmlinkage long sys32_adjtimex(struct compat_timex __user *utp) { struct timex txc; int ret; diff --git a/arch/x86_64/ia32/sys_ia32.c b/arch/x86_64/ia32/sys_ia32.c index 2b2d029f477..b13121e451a 100644 --- a/arch/x86_64/ia32/sys_ia32.c +++ b/arch/x86_64/ia32/sys_ia32.c @@ -769,30 +769,17 @@ sys32_sendfile(int out_fd, int in_fd, compat_off_t __user *offset, s32 count) /* Handle adjtimex compatibility. */ -struct timex32 { - u32 modes; - s32 offset, freq, maxerror, esterror; - s32 status, constant, precision, tolerance; - struct compat_timeval time; - s32 tick; - s32 ppsfreq, jitter, shift, stabil; - s32 jitcnt, calcnt, errcnt, stbcnt; - s32 :32; s32 :32; s32 :32; s32 :32; - s32 :32; s32 :32; s32 :32; s32 :32; - s32 :32; s32 :32; s32 :32; s32 :32; -}; - extern int do_adjtimex(struct timex *); asmlinkage long -sys32_adjtimex(struct timex32 __user *utp) +sys32_adjtimex(struct compat_timex __user *utp) { struct timex txc; int ret; memset(&txc, 0, sizeof(struct timex)); - if (!access_ok(VERIFY_READ, utp, sizeof(struct timex32)) || + if (!access_ok(VERIFY_READ, utp, sizeof(struct compat_timex)) || __get_user(txc.modes, &utp->modes) || __get_user(txc.offset, &utp->offset) || __get_user(txc.freq, &utp->freq) || @@ -817,7 +804,7 @@ sys32_adjtimex(struct timex32 __user *utp) ret = do_adjtimex(&txc); - if (!access_ok(VERIFY_WRITE, utp, sizeof(struct timex32)) || + if (!access_ok(VERIFY_WRITE, utp, sizeof(struct compat_timex)) || __put_user(txc.modes, &utp->modes) || __put_user(txc.offset, &utp->offset) || __put_user(txc.freq, &utp->freq) || diff --git a/include/linux/compat.h b/include/linux/compat.h index c9ab2a26348..859f95700d3 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -45,6 +45,32 @@ struct compat_tms { compat_clock_t tms_cstime; }; +struct compat_timex { + compat_uint_t modes; + compat_long_t offset; + compat_long_t freq; + compat_long_t maxerror; + compat_long_t esterror; + compat_int_t status; + compat_long_t constant; + compat_long_t precision; + compat_long_t tolerance; + struct compat_timeval time; + compat_long_t tick; + compat_long_t ppsfreq; + compat_long_t jitter; + compat_int_t shift; + compat_long_t stabil; + compat_long_t jitcnt; + compat_long_t calcnt; + compat_long_t errcnt; + compat_long_t stbcnt; + + compat_int_t :32; compat_int_t :32; compat_int_t :32; compat_int_t :32; + compat_int_t :32; compat_int_t :32; compat_int_t :32; compat_int_t :32; + compat_int_t :32; compat_int_t :32; compat_int_t :32; compat_int_t :32; +}; + #define _COMPAT_NSIG_WORDS (_COMPAT_NSIG / _COMPAT_NSIG_BPW) typedef struct { -- cgit v1.2.3-70-g09d2 From 3158e9411a66fb98d495ac441c242264f31aaf3e Mon Sep 17 00:00:00 2001 From: Stephen Rothwell Date: Sun, 26 Mar 2006 01:37:29 -0800 Subject: [PATCH] consolidate sys32/compat_adjtimex Create compat_sys_adjtimex and use it an all appropriate places. Signed-off-by: Stephen Rothwell Cc: Arnd Bergmann Acked-by: Paul Mackerras Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/alpha/kernel/osf_sys.c | 1 - arch/ia64/ia32/sys_ia32.c | 62 ------------------------------------ arch/mips/kernel/linux32.c | 61 ------------------------------------ arch/mips/kernel/scall64-n32.S | 2 +- arch/mips/kernel/scall64-o32.S | 2 +- arch/parisc/kernel/sys_parisc32.c | 29 ----------------- arch/parisc/kernel/syscall_table.S | 2 +- arch/powerpc/kernel/sys_ppc32.c | 60 ----------------------------------- arch/s390/kernel/compat_linux.c | 61 ------------------------------------ arch/s390/kernel/compat_wrapper.S | 8 ++--- arch/s390/kernel/syscalls.S | 2 +- arch/sparc64/kernel/sys_sparc32.c | 61 ------------------------------------ arch/sparc64/kernel/systbls.S | 2 +- arch/x86_64/ia32/ia32entry.S | 2 +- arch/x86_64/ia32/sys_ia32.c | 64 -------------------------------------- include/linux/compat.h | 2 ++ include/linux/timex.h | 2 ++ kernel/compat.c | 59 +++++++++++++++++++++++++++++++++++ 18 files changed, 73 insertions(+), 409 deletions(-) (limited to 'arch/ia64') diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c index 7fb14f42a12..31afe3d91ac 100644 --- a/arch/alpha/kernel/osf_sys.c +++ b/arch/alpha/kernel/osf_sys.c @@ -821,7 +821,6 @@ osf_setsysinfo(unsigned long op, void __user *buffer, unsigned long nbytes, affects all sorts of things, like timeval and itimerval. */ extern struct timezone sys_tz; -extern int do_adjtimex(struct timex *); struct timeval32 { diff --git a/arch/ia64/ia32/sys_ia32.c b/arch/ia64/ia32/sys_ia32.c index 1e1a2353aef..5366b3b23d0 100644 --- a/arch/ia64/ia32/sys_ia32.c +++ b/arch/ia64/ia32/sys_ia32.c @@ -25,7 +25,6 @@ #include #include #include -#include #include #include #include @@ -2591,65 +2590,4 @@ sys32_setresgid(compat_gid_t rgid, compat_gid_t egid, ssgid = (sgid == (compat_gid_t)-1) ? ((gid_t)-1) : ((gid_t)sgid); return sys_setresgid(srgid, segid, ssgid); } - -/* Handle adjtimex compatibility. */ - -extern int do_adjtimex(struct timex *); - -asmlinkage long -sys32_adjtimex(struct compat_timex *utp) -{ - struct timex txc; - int ret; - - memset(&txc, 0, sizeof(struct timex)); - - if(get_user(txc.modes, &utp->modes) || - __get_user(txc.offset, &utp->offset) || - __get_user(txc.freq, &utp->freq) || - __get_user(txc.maxerror, &utp->maxerror) || - __get_user(txc.esterror, &utp->esterror) || - __get_user(txc.status, &utp->status) || - __get_user(txc.constant, &utp->constant) || - __get_user(txc.precision, &utp->precision) || - __get_user(txc.tolerance, &utp->tolerance) || - __get_user(txc.time.tv_sec, &utp->time.tv_sec) || - __get_user(txc.time.tv_usec, &utp->time.tv_usec) || - __get_user(txc.tick, &utp->tick) || - __get_user(txc.ppsfreq, &utp->ppsfreq) || - __get_user(txc.jitter, &utp->jitter) || - __get_user(txc.shift, &utp->shift) || - __get_user(txc.stabil, &utp->stabil) || - __get_user(txc.jitcnt, &utp->jitcnt) || - __get_user(txc.calcnt, &utp->calcnt) || - __get_user(txc.errcnt, &utp->errcnt) || - __get_user(txc.stbcnt, &utp->stbcnt)) - return -EFAULT; - - ret = do_adjtimex(&txc); - - if(put_user(txc.modes, &utp->modes) || - __put_user(txc.offset, &utp->offset) || - __put_user(txc.freq, &utp->freq) || - __put_user(txc.maxerror, &utp->maxerror) || - __put_user(txc.esterror, &utp->esterror) || - __put_user(txc.status, &utp->status) || - __put_user(txc.constant, &utp->constant) || - __put_user(txc.precision, &utp->precision) || - __put_user(txc.tolerance, &utp->tolerance) || - __put_user(txc.time.tv_sec, &utp->time.tv_sec) || - __put_user(txc.time.tv_usec, &utp->time.tv_usec) || - __put_user(txc.tick, &utp->tick) || - __put_user(txc.ppsfreq, &utp->ppsfreq) || - __put_user(txc.jitter, &utp->jitter) || - __put_user(txc.shift, &utp->shift) || - __put_user(txc.stabil, &utp->stabil) || - __put_user(txc.jitcnt, &utp->jitcnt) || - __put_user(txc.calcnt, &utp->calcnt) || - __put_user(txc.errcnt, &utp->errcnt) || - __put_user(txc.stbcnt, &utp->stbcnt)) - ret = -EFAULT; - - return ret; -} #endif /* NOTYET */ diff --git a/arch/mips/kernel/linux32.c b/arch/mips/kernel/linux32.c index f33e779796f..3f40c37a9ee 100644 --- a/arch/mips/kernel/linux32.c +++ b/arch/mips/kernel/linux32.c @@ -30,7 +30,6 @@ #include #include #include -#include #include #include #include @@ -1157,66 +1156,6 @@ out: return err; } -/* Handle adjtimex compatibility. */ - -extern int do_adjtimex(struct timex *); - -asmlinkage int sys32_adjtimex(struct compat_timex __user *utp) -{ - struct timex txc; - int ret; - - memset(&txc, 0, sizeof(struct timex)); - - if (get_user(txc.modes, &utp->modes) || - __get_user(txc.offset, &utp->offset) || - __get_user(txc.freq, &utp->freq) || - __get_user(txc.maxerror, &utp->maxerror) || - __get_user(txc.esterror, &utp->esterror) || - __get_user(txc.status, &utp->status) || - __get_user(txc.constant, &utp->constant) || - __get_user(txc.precision, &utp->precision) || - __get_user(txc.tolerance, &utp->tolerance) || - __get_user(txc.time.tv_sec, &utp->time.tv_sec) || - __get_user(txc.time.tv_usec, &utp->time.tv_usec) || - __get_user(txc.tick, &utp->tick) || - __get_user(txc.ppsfreq, &utp->ppsfreq) || - __get_user(txc.jitter, &utp->jitter) || - __get_user(txc.shift, &utp->shift) || - __get_user(txc.stabil, &utp->stabil) || - __get_user(txc.jitcnt, &utp->jitcnt) || - __get_user(txc.calcnt, &utp->calcnt) || - __get_user(txc.errcnt, &utp->errcnt) || - __get_user(txc.stbcnt, &utp->stbcnt)) - return -EFAULT; - - ret = do_adjtimex(&txc); - - if (put_user(txc.modes, &utp->modes) || - __put_user(txc.offset, &utp->offset) || - __put_user(txc.freq, &utp->freq) || - __put_user(txc.maxerror, &utp->maxerror) || - __put_user(txc.esterror, &utp->esterror) || - __put_user(txc.status, &utp->status) || - __put_user(txc.constant, &utp->constant) || - __put_user(txc.precision, &utp->precision) || - __put_user(txc.tolerance, &utp->tolerance) || - __put_user(txc.time.tv_sec, &utp->time.tv_sec) || - __put_user(txc.time.tv_usec, &utp->time.tv_usec) || - __put_user(txc.tick, &utp->tick) || - __put_user(txc.ppsfreq, &utp->ppsfreq) || - __put_user(txc.jitter, &utp->jitter) || - __put_user(txc.shift, &utp->shift) || - __put_user(txc.stabil, &utp->stabil) || - __put_user(txc.jitcnt, &utp->jitcnt) || - __put_user(txc.calcnt, &utp->calcnt) || - __put_user(txc.errcnt, &utp->errcnt) || - __put_user(txc.stbcnt, &utp->stbcnt)) - ret = -EFAULT; - - return ret; -} - asmlinkage int sys32_sendfile(int out_fd, int in_fd, compat_off_t __user *offset, s32 count) { diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S index 02c8267e45e..05a2c0567da 100644 --- a/arch/mips/kernel/scall64-n32.S +++ b/arch/mips/kernel/scall64-n32.S @@ -273,7 +273,7 @@ EXPORT(sysn32_call_table) PTR sys_pivot_root PTR sys32_sysctl PTR sys_prctl - PTR sys32_adjtimex + PTR compat_sys_adjtimex PTR compat_sys_setrlimit /* 6155 */ PTR sys_chroot PTR sys_sync diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S index 797e0d87488..19c4ca481b0 100644 --- a/arch/mips/kernel/scall64-o32.S +++ b/arch/mips/kernel/scall64-o32.S @@ -328,7 +328,7 @@ sys_call_table: PTR sys_setdomainname PTR sys32_newuname PTR sys_ni_syscall /* sys_modify_ldt */ - PTR sys32_adjtimex + PTR compat_sys_adjtimex PTR sys_mprotect /* 4125 */ PTR compat_sys_sigprocmask PTR sys_ni_syscall /* was creat_module */ diff --git a/arch/parisc/kernel/sys_parisc32.c b/arch/parisc/kernel/sys_parisc32.c index ca1d8dbbadf..d286f68a3d3 100644 --- a/arch/parisc/kernel/sys_parisc32.c +++ b/arch/parisc/kernel/sys_parisc32.c @@ -21,7 +21,6 @@ #include #include #include -#include #include #include #include @@ -567,34 +566,6 @@ asmlinkage int sys32_sendfile64(int out_fd, int in_fd, compat_loff_t __user *off } -asmlinkage long sys32_adjtimex(struct compat_timex __user *txc_p32) -{ - struct timex txc; - struct compat_timex t32; - int ret; - extern int do_adjtimex(struct timex *txc); - - if(copy_from_user(&t32, txc_p32, sizeof(struct compat_timex))) - return -EFAULT; -#undef CP -#define CP(x) txc.x = t32.x - CP(modes); CP(offset); CP(freq); CP(maxerror); CP(esterror); - CP(status); CP(constant); CP(precision); CP(tolerance); - CP(time.tv_sec); CP(time.tv_usec); CP(tick); CP(ppsfreq); CP(jitter); - CP(shift); CP(stabil); CP(jitcnt); CP(calcnt); CP(errcnt); - CP(stbcnt); - ret = do_adjtimex(&txc); -#undef CP -#define CP(x) t32.x = txc.x - CP(modes); CP(offset); CP(freq); CP(maxerror); CP(esterror); - CP(status); CP(constant); CP(precision); CP(tolerance); - CP(time.tv_sec); CP(time.tv_usec); CP(tick); CP(ppsfreq); CP(jitter); - CP(shift); CP(stabil); CP(jitcnt); CP(calcnt); CP(errcnt); - CP(stbcnt); - return copy_to_user(txc_p32, &t32, sizeof(struct compat_timex)) ? -EFAULT : ret; -} - - struct sysinfo32 { s32 uptime; u32 loads[3]; diff --git a/arch/parisc/kernel/syscall_table.S b/arch/parisc/kernel/syscall_table.S index 71011eadb87..89b6c56ea0a 100644 --- a/arch/parisc/kernel/syscall_table.S +++ b/arch/parisc/kernel/syscall_table.S @@ -207,7 +207,7 @@ /* struct sockaddr... */ ENTRY_SAME(recvfrom) /* struct timex contains longs */ - ENTRY_DIFF(adjtimex) + ENTRY_COMP(adjtimex) ENTRY_SAME(mprotect) /* 125 */ /* old_sigset_t forced to 32 bits. Beware glibc sigset_t */ ENTRY_COMP(sigprocmask) diff --git a/arch/powerpc/kernel/sys_ppc32.c b/arch/powerpc/kernel/sys_ppc32.c index d9867f583c6..ec274e68881 100644 --- a/arch/powerpc/kernel/sys_ppc32.c +++ b/arch/powerpc/kernel/sys_ppc32.c @@ -24,7 +24,6 @@ #include #include #include -#include #include #include #include @@ -161,65 +160,6 @@ asmlinkage long compat_sys_sysfs(u32 option, u32 arg1, u32 arg2) return sys_sysfs((int)option, arg1, arg2); } -/* Handle adjtimex compatibility. */ -extern int do_adjtimex(struct timex *); - -asmlinkage long compat_sys_adjtimex(struct compat_timex __user *utp) -{ - struct timex txc; - int ret; - - memset(&txc, 0, sizeof(struct timex)); - - if(get_user(txc.modes, &utp->modes) || - __get_user(txc.offset, &utp->offset) || - __get_user(txc.freq, &utp->freq) || - __get_user(txc.maxerror, &utp->maxerror) || - __get_user(txc.esterror, &utp->esterror) || - __get_user(txc.status, &utp->status) || - __get_user(txc.constant, &utp->constant) || - __get_user(txc.precision, &utp->precision) || - __get_user(txc.tolerance, &utp->tolerance) || - __get_user(txc.time.tv_sec, &utp->time.tv_sec) || - __get_user(txc.time.tv_usec, &utp->time.tv_usec) || - __get_user(txc.tick, &utp->tick) || - __get_user(txc.ppsfreq, &utp->ppsfreq) || - __get_user(txc.jitter, &utp->jitter) || - __get_user(txc.shift, &utp->shift) || - __get_user(txc.stabil, &utp->stabil) || - __get_user(txc.jitcnt, &utp->jitcnt) || - __get_user(txc.calcnt, &utp->calcnt) || - __get_user(txc.errcnt, &utp->errcnt) || - __get_user(txc.stbcnt, &utp->stbcnt)) - return -EFAULT; - - ret = do_adjtimex(&txc); - - if(put_user(txc.modes, &utp->modes) || - __put_user(txc.offset, &utp->offset) || - __put_user(txc.freq, &utp->freq) || - __put_user(txc.maxerror, &utp->maxerror) || - __put_user(txc.esterror, &utp->esterror) || - __put_user(txc.status, &utp->status) || - __put_user(txc.constant, &utp->constant) || - __put_user(txc.precision, &utp->precision) || - __put_user(txc.tolerance, &utp->tolerance) || - __put_user(txc.time.tv_sec, &utp->time.tv_sec) || - __put_user(txc.time.tv_usec, &utp->time.tv_usec) || - __put_user(txc.tick, &utp->tick) || - __put_user(txc.ppsfreq, &utp->ppsfreq) || - __put_user(txc.jitter, &utp->jitter) || - __put_user(txc.shift, &utp->shift) || - __put_user(txc.stabil, &utp->stabil) || - __put_user(txc.jitcnt, &utp->jitcnt) || - __put_user(txc.calcnt, &utp->calcnt) || - __put_user(txc.errcnt, &utp->errcnt) || - __put_user(txc.stbcnt, &utp->stbcnt)) - ret = -EFAULT; - - return ret; -} - asmlinkage long compat_sys_pause(void) { current->state = TASK_INTERRUPTIBLE; diff --git a/arch/s390/kernel/compat_linux.c b/arch/s390/kernel/compat_linux.c index 9809264f2f4..5e14de37c17 100644 --- a/arch/s390/kernel/compat_linux.c +++ b/arch/s390/kernel/compat_linux.c @@ -26,7 +26,6 @@ #include #include #include -#include #include #include #include @@ -705,66 +704,6 @@ asmlinkage long sys32_sendfile64(int out_fd, int in_fd, return ret; } -/* Handle adjtimex compatibility. */ - -extern int do_adjtimex(struct timex *); - -asmlinkage long sys32_adjtimex(struct compat_timex __user *utp) -{ - struct timex txc; - int ret; - - memset(&txc, 0, sizeof(struct timex)); - - if(get_user(txc.modes, &utp->modes) || - __get_user(txc.offset, &utp->offset) || - __get_user(txc.freq, &utp->freq) || - __get_user(txc.maxerror, &utp->maxerror) || - __get_user(txc.esterror, &utp->esterror) || - __get_user(txc.status, &utp->status) || - __get_user(txc.constant, &utp->constant) || - __get_user(txc.precision, &utp->precision) || - __get_user(txc.tolerance, &utp->tolerance) || - __get_user(txc.time.tv_sec, &utp->time.tv_sec) || - __get_user(txc.time.tv_usec, &utp->time.tv_usec) || - __get_user(txc.tick, &utp->tick) || - __get_user(txc.ppsfreq, &utp->ppsfreq) || - __get_user(txc.jitter, &utp->jitter) || - __get_user(txc.shift, &utp->shift) || - __get_user(txc.stabil, &utp->stabil) || - __get_user(txc.jitcnt, &utp->jitcnt) || - __get_user(txc.calcnt, &utp->calcnt) || - __get_user(txc.errcnt, &utp->errcnt) || - __get_user(txc.stbcnt, &utp->stbcnt)) - return -EFAULT; - - ret = do_adjtimex(&txc); - - if(put_user(txc.modes, &utp->modes) || - __put_user(txc.offset, &utp->offset) || - __put_user(txc.freq, &utp->freq) || - __put_user(txc.maxerror, &utp->maxerror) || - __put_user(txc.esterror, &utp->esterror) || - __put_user(txc.status, &utp->status) || - __put_user(txc.constant, &utp->constant) || - __put_user(txc.precision, &utp->precision) || - __put_user(txc.tolerance, &utp->tolerance) || - __put_user(txc.time.tv_sec, &utp->time.tv_sec) || - __put_user(txc.time.tv_usec, &utp->time.tv_usec) || - __put_user(txc.tick, &utp->tick) || - __put_user(txc.ppsfreq, &utp->ppsfreq) || - __put_user(txc.jitter, &utp->jitter) || - __put_user(txc.shift, &utp->shift) || - __put_user(txc.stabil, &utp->stabil) || - __put_user(txc.jitcnt, &utp->jitcnt) || - __put_user(txc.calcnt, &utp->calcnt) || - __put_user(txc.errcnt, &utp->errcnt) || - __put_user(txc.stbcnt, &utp->stbcnt)) - ret = -EFAULT; - - return ret; -} - #ifdef CONFIG_SYSCTL struct __sysctl_args32 { u32 name; diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S index 50e80138e7a..199da68bd7b 100644 --- a/arch/s390/kernel/compat_wrapper.S +++ b/arch/s390/kernel/compat_wrapper.S @@ -551,10 +551,10 @@ sys32_newuname_wrapper: llgtr %r2,%r2 # struct new_utsname * jg s390x_newuname # branch to system call - .globl sys32_adjtimex_wrapper -sys32_adjtimex_wrapper: - llgtr %r2,%r2 # struct timex_emu31 * - jg sys32_adjtimex # branch to system call + .globl compat_sys_adjtimex_wrapper +compat_sys_adjtimex_wrapper: + llgtr %r2,%r2 # struct compat_timex * + jg compat_sys_adjtimex # branch to system call .globl sys32_mprotect_wrapper sys32_mprotect_wrapper: diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S index 7c88d85c359..2f56654da82 100644 --- a/arch/s390/kernel/syscalls.S +++ b/arch/s390/kernel/syscalls.S @@ -132,7 +132,7 @@ SYSCALL(sys_clone_glue,sys_clone_glue,sys32_clone_glue) /* 120 */ SYSCALL(sys_setdomainname,sys_setdomainname,sys32_setdomainname_wrapper) SYSCALL(sys_newuname,s390x_newuname,sys32_newuname_wrapper) NI_SYSCALL /* modify_ldt for i386 */ -SYSCALL(sys_adjtimex,sys_adjtimex,sys32_adjtimex_wrapper) +SYSCALL(sys_adjtimex,sys_adjtimex,compat_sys_adjtimex_wrapper) SYSCALL(sys_mprotect,sys_mprotect,sys32_mprotect_wrapper) /* 125 */ SYSCALL(sys_sigprocmask,sys_sigprocmask,compat_sys_sigprocmask_wrapper) NI_SYSCALL /* old "create module" */ diff --git a/arch/sparc64/kernel/sys_sparc32.c b/arch/sparc64/kernel/sys_sparc32.c index 48f02f20191..2e906bad56f 100644 --- a/arch/sparc64/kernel/sys_sparc32.c +++ b/arch/sparc64/kernel/sys_sparc32.c @@ -19,7 +19,6 @@ #include #include #include -#include #include #include #include @@ -945,66 +944,6 @@ asmlinkage long compat_sys_sendfile64(int out_fd, int in_fd, return ret; } -/* Handle adjtimex compatibility. */ - -extern int do_adjtimex(struct timex *); - -asmlinkage long sys32_adjtimex(struct compat_timex __user *utp) -{ - struct timex txc; - int ret; - - memset(&txc, 0, sizeof(struct timex)); - - if (get_user(txc.modes, &utp->modes) || - __get_user(txc.offset, &utp->offset) || - __get_user(txc.freq, &utp->freq) || - __get_user(txc.maxerror, &utp->maxerror) || - __get_user(txc.esterror, &utp->esterror) || - __get_user(txc.status, &utp->status) || - __get_user(txc.constant, &utp->constant) || - __get_user(txc.precision, &utp->precision) || - __get_user(txc.tolerance, &utp->tolerance) || - __get_user(txc.time.tv_sec, &utp->time.tv_sec) || - __get_user(txc.time.tv_usec, &utp->time.tv_usec) || - __get_user(txc.tick, &utp->tick) || - __get_user(txc.ppsfreq, &utp->ppsfreq) || - __get_user(txc.jitter, &utp->jitter) || - __get_user(txc.shift, &utp->shift) || - __get_user(txc.stabil, &utp->stabil) || - __get_user(txc.jitcnt, &utp->jitcnt) || - __get_user(txc.calcnt, &utp->calcnt) || - __get_user(txc.errcnt, &utp->errcnt) || - __get_user(txc.stbcnt, &utp->stbcnt)) - return -EFAULT; - - ret = do_adjtimex(&txc); - - if (put_user(txc.modes, &utp->modes) || - __put_user(txc.offset, &utp->offset) || - __put_user(txc.freq, &utp->freq) || - __put_user(txc.maxerror, &utp->maxerror) || - __put_user(txc.esterror, &utp->esterror) || - __put_user(txc.status, &utp->status) || - __put_user(txc.constant, &utp->constant) || - __put_user(txc.precision, &utp->precision) || - __put_user(txc.tolerance, &utp->tolerance) || - __put_user(txc.time.tv_sec, &utp->time.tv_sec) || - __put_user(txc.time.tv_usec, &utp->time.tv_usec) || - __put_user(txc.tick, &utp->tick) || - __put_user(txc.ppsfreq, &utp->ppsfreq) || - __put_user(txc.jitter, &utp->jitter) || - __put_user(txc.shift, &utp->shift) || - __put_user(txc.stabil, &utp->stabil) || - __put_user(txc.jitcnt, &utp->jitcnt) || - __put_user(txc.calcnt, &utp->calcnt) || - __put_user(txc.errcnt, &utp->errcnt) || - __put_user(txc.stbcnt, &utp->stbcnt)) - ret = -EFAULT; - - return ret; -} - /* This is just a version for 32-bit applications which does * not force O_LARGEFILE on. */ diff --git a/arch/sparc64/kernel/systbls.S b/arch/sparc64/kernel/systbls.S index c3adb7ac167..3b250f2318f 100644 --- a/arch/sparc64/kernel/systbls.S +++ b/arch/sparc64/kernel/systbls.S @@ -63,7 +63,7 @@ sys_call_table32: /*200*/ .word sys32_ssetmask, sys_sigsuspend, compat_sys_newlstat, sys_uselib, compat_sys_old_readdir .word sys32_readahead, sys32_socketcall, sys32_syslog, sys32_lookup_dcookie, sys32_fadvise64 /*210*/ .word sys32_fadvise64_64, sys32_tgkill, sys32_waitpid, sys_swapoff, sys32_sysinfo - .word sys32_ipc, sys32_sigreturn, sys_clone, sys32_ioprio_get, sys32_adjtimex + .word sys32_ipc, sys32_sigreturn, sys_clone, sys32_ioprio_get, compat_sys_adjtimex /*220*/ .word sys32_sigprocmask, sys_ni_syscall, sys32_delete_module, sys_ni_syscall, sys32_getpgid .word sys32_bdflush, sys32_sysfs, sys_nis_syscall, sys32_setfsuid16, sys32_setfsgid16 /*230*/ .word sys32_select, compat_sys_time, sys_nis_syscall, compat_sys_stime, compat_sys_statfs64 diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S index 00dee176c08..7549a4389fb 100644 --- a/arch/x86_64/ia32/ia32entry.S +++ b/arch/x86_64/ia32/ia32entry.S @@ -501,7 +501,7 @@ ia32_sys_call_table: .quad sys_setdomainname .quad sys_uname .quad sys_modify_ldt - .quad sys32_adjtimex + .quad compat_sys_adjtimex .quad sys32_mprotect /* 125 */ .quad compat_sys_sigprocmask .quad quiet_ni_syscall /* create_module */ diff --git a/arch/x86_64/ia32/sys_ia32.c b/arch/x86_64/ia32/sys_ia32.c index b13121e451a..f182b20858e 100644 --- a/arch/x86_64/ia32/sys_ia32.c +++ b/arch/x86_64/ia32/sys_ia32.c @@ -30,7 +30,6 @@ #include #include #include -#include #include #include #include @@ -767,69 +766,6 @@ sys32_sendfile(int out_fd, int in_fd, compat_off_t __user *offset, s32 count) return ret; } -/* Handle adjtimex compatibility. */ - -extern int do_adjtimex(struct timex *); - -asmlinkage long -sys32_adjtimex(struct compat_timex __user *utp) -{ - struct timex txc; - int ret; - - memset(&txc, 0, sizeof(struct timex)); - - if (!access_ok(VERIFY_READ, utp, sizeof(struct compat_timex)) || - __get_user(txc.modes, &utp->modes) || - __get_user(txc.offset, &utp->offset) || - __get_user(txc.freq, &utp->freq) || - __get_user(txc.maxerror, &utp->maxerror) || - __get_user(txc.esterror, &utp->esterror) || - __get_user(txc.status, &utp->status) || - __get_user(txc.constant, &utp->constant) || - __get_user(txc.precision, &utp->precision) || - __get_user(txc.tolerance, &utp->tolerance) || - __get_user(txc.time.tv_sec, &utp->time.tv_sec) || - __get_user(txc.time.tv_usec, &utp->time.tv_usec) || - __get_user(txc.tick, &utp->tick) || - __get_user(txc.ppsfreq, &utp->ppsfreq) || - __get_user(txc.jitter, &utp->jitter) || - __get_user(txc.shift, &utp->shift) || - __get_user(txc.stabil, &utp->stabil) || - __get_user(txc.jitcnt, &utp->jitcnt) || - __get_user(txc.calcnt, &utp->calcnt) || - __get_user(txc.errcnt, &utp->errcnt) || - __get_user(txc.stbcnt, &utp->stbcnt)) - return -EFAULT; - - ret = do_adjtimex(&txc); - - if (!access_ok(VERIFY_WRITE, utp, sizeof(struct compat_timex)) || - __put_user(txc.modes, &utp->modes) || - __put_user(txc.offset, &utp->offset) || - __put_user(txc.freq, &utp->freq) || - __put_user(txc.maxerror, &utp->maxerror) || - __put_user(txc.esterror, &utp->esterror) || - __put_user(txc.status, &utp->status) || - __put_user(txc.constant, &utp->constant) || - __put_user(txc.precision, &utp->precision) || - __put_user(txc.tolerance, &utp->tolerance) || - __put_user(txc.time.tv_sec, &utp->time.tv_sec) || - __put_user(txc.time.tv_usec, &utp->time.tv_usec) || - __put_user(txc.tick, &utp->tick) || - __put_user(txc.ppsfreq, &utp->ppsfreq) || - __put_user(txc.jitter, &utp->jitter) || - __put_user(txc.shift, &utp->shift) || - __put_user(txc.stabil, &utp->stabil) || - __put_user(txc.jitcnt, &utp->jitcnt) || - __put_user(txc.calcnt, &utp->calcnt) || - __put_user(txc.errcnt, &utp->errcnt) || - __put_user(txc.stbcnt, &utp->stbcnt)) - ret = -EFAULT; - - return ret; -} - asmlinkage long sys32_mmap2(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, unsigned long pgoff) diff --git a/include/linux/compat.h b/include/linux/compat.h index 859f95700d3..24d659cdbaf 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -207,5 +207,7 @@ static inline int compat_timespec_compare(struct compat_timespec *lhs, return lhs->tv_nsec - rhs->tv_nsec; } +asmlinkage long compat_sys_adjtimex(struct compat_timex __user *utp); + #endif /* CONFIG_COMPAT */ #endif /* _LINUX_COMPAT_H */ diff --git a/include/linux/timex.h b/include/linux/timex.h index 82dc9ae79d3..03914b7e41b 100644 --- a/include/linux/timex.h +++ b/include/linux/timex.h @@ -307,6 +307,8 @@ time_interpolator_reset(void) /* Returns how long ticks are at present, in ns / 2^(SHIFT_SCALE-10). */ extern u64 current_tick_length(void); +extern int do_adjtimex(struct timex *); + #endif /* KERNEL */ #endif /* LINUX_TIMEX_H */ diff --git a/kernel/compat.c b/kernel/compat.c index 8c9cd88b678..b9bdd1271f4 100644 --- a/kernel/compat.c +++ b/kernel/compat.c @@ -21,6 +21,7 @@ #include #include #include +#include #include @@ -898,3 +899,61 @@ asmlinkage long compat_sys_rt_sigsuspend(compat_sigset_t __user *unewset, compat return -ERESTARTNOHAND; } #endif /* __ARCH_WANT_COMPAT_SYS_RT_SIGSUSPEND */ + +asmlinkage long compat_sys_adjtimex(struct compat_timex __user *utp) +{ + struct timex txc; + int ret; + + memset(&txc, 0, sizeof(struct timex)); + + if (!access_ok(VERIFY_READ, utp, sizeof(struct compat_timex)) || + __get_user(txc.modes, &utp->modes) || + __get_user(txc.offset, &utp->offset) || + __get_user(txc.freq, &utp->freq) || + __get_user(txc.maxerror, &utp->maxerror) || + __get_user(txc.esterror, &utp->esterror) || + __get_user(txc.status, &utp->status) || + __get_user(txc.constant, &utp->constant) || + __get_user(txc.precision, &utp->precision) || + __get_user(txc.tolerance, &utp->tolerance) || + __get_user(txc.time.tv_sec, &utp->time.tv_sec) || + __get_user(txc.time.tv_usec, &utp->time.tv_usec) || + __get_user(txc.tick, &utp->tick) || + __get_user(txc.ppsfreq, &utp->ppsfreq) || + __get_user(txc.jitter, &utp->jitter) || + __get_user(txc.shift, &utp->shift) || + __get_user(txc.stabil, &utp->stabil) || + __get_user(txc.jitcnt, &utp->jitcnt) || + __get_user(txc.calcnt, &utp->calcnt) || + __get_user(txc.errcnt, &utp->errcnt) || + __get_user(txc.stbcnt, &utp->stbcnt)) + return -EFAULT; + + ret = do_adjtimex(&txc); + + if (!access_ok(VERIFY_WRITE, utp, sizeof(struct compat_timex)) || + __put_user(txc.modes, &utp->modes) || + __put_user(txc.offset, &utp->offset) || + __put_user(txc.freq, &utp->freq) || + __put_user(txc.maxerror, &utp->maxerror) || + __put_user(txc.esterror, &utp->esterror) || + __put_user(txc.status, &utp->status) || + __put_user(txc.constant, &utp->constant) || + __put_user(txc.precision, &utp->precision) || + __put_user(txc.tolerance, &utp->tolerance) || + __put_user(txc.time.tv_sec, &utp->time.tv_sec) || + __put_user(txc.time.tv_usec, &utp->time.tv_usec) || + __put_user(txc.tick, &utp->tick) || + __put_user(txc.ppsfreq, &utp->ppsfreq) || + __put_user(txc.jitter, &utp->jitter) || + __put_user(txc.shift, &utp->shift) || + __put_user(txc.stabil, &utp->stabil) || + __put_user(txc.jitcnt, &utp->jitcnt) || + __put_user(txc.calcnt, &utp->calcnt) || + __put_user(txc.errcnt, &utp->errcnt) || + __put_user(txc.stbcnt, &utp->stbcnt)) + ret = -EFAULT; + + return ret; +} -- cgit v1.2.3-70-g09d2 From c6fd91f0bdcd294a0ae0ba2b2a7f7456ef4b7144 Mon Sep 17 00:00:00 2001 From: bibo mao Date: Sun, 26 Mar 2006 01:38:20 -0800 Subject: [PATCH] kretprobe instance recycled by parent process When kretprobe probes the schedule() function, if the probed process exits then schedule() will never return, so some kretprobe instances will never be recycled. In this patch the parent process will recycle retprobe instances of the probed function and there will be no memory leak of kretprobe instances. Signed-off-by: bibo mao Cc: Masami Hiramatsu Cc: Prasanna S Panchamukhi Cc: Ananth N Mavinakayanahalli Cc: Anil S Keshavamurthy Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/process.c | 8 -------- arch/ia64/kernel/process.c | 8 -------- arch/powerpc/kernel/process.c | 2 -- arch/x86_64/kernel/process.c | 9 +-------- kernel/kprobes.c | 10 +++++----- kernel/sched.c | 9 ++++++++- 6 files changed, 14 insertions(+), 32 deletions(-) (limited to 'arch/ia64') diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c index 299e6167408..24b3e745478 100644 --- a/arch/i386/kernel/process.c +++ b/arch/i386/kernel/process.c @@ -38,7 +38,6 @@ #include #include #include -#include #include #include @@ -364,13 +363,6 @@ void exit_thread(void) struct task_struct *tsk = current; struct thread_struct *t = &tsk->thread; - /* - * Remove function-return probe instances associated with this task - * and put them back on the free list. Do not insert an exit probe for - * this function, it will be disabled by kprobe_flush_task if you do. - */ - kprobe_flush_task(tsk); - /* The process may have allocated an io port bitmap... nuke it. */ if (unlikely(NULL != t->io_bitmap_ptr)) { int cpu = get_cpu(); diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c index 309d59658e5..355d57970ba 100644 --- a/arch/ia64/kernel/process.c +++ b/arch/ia64/kernel/process.c @@ -30,7 +30,6 @@ #include #include #include -#include #include #include @@ -738,13 +737,6 @@ void exit_thread (void) { - /* - * Remove function-return probe instances associated with this task - * and put them back on the free list. Do not insert an exit probe for - * this function, it will be disabled by kprobe_flush_task if you do. - */ - kprobe_flush_task(current); - ia64_drop_fpu(current); #ifdef CONFIG_PERFMON /* if needed, stop monitoring and flush state to perfmon context */ diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c index 1770a066c21..f698aa77127 100644 --- a/arch/powerpc/kernel/process.c +++ b/arch/powerpc/kernel/process.c @@ -35,7 +35,6 @@ #include #include #include -#include #include #include @@ -460,7 +459,6 @@ void show_regs(struct pt_regs * regs) void exit_thread(void) { - kprobe_flush_task(current); discard_lazy_cpu_state(); } diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c index 81111835722..0370720515f 100644 --- a/arch/x86_64/kernel/process.c +++ b/arch/x86_64/kernel/process.c @@ -35,8 +35,8 @@ #include #include #include -#include #include +#include #include #include @@ -353,13 +353,6 @@ void exit_thread(void) struct task_struct *me = current; struct thread_struct *t = &me->thread; - /* - * Remove function-return probe instances associated with this task - * and put them back on the free list. Do not insert an exit probe for - * this function, it will be disabled by kprobe_flush_task if you do. - */ - kprobe_flush_task(me); - if (me->thread.io_bitmap_ptr) { struct tss_struct *tss = &per_cpu(init_tss, get_cpu()); diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 1fb9f753ef6..1156eb0977d 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -323,10 +323,10 @@ struct hlist_head __kprobes *kretprobe_inst_table_head(struct task_struct *tsk) } /* - * This function is called from exit_thread or flush_thread when task tk's - * stack is being recycled so that we can recycle any function-return probe - * instances associated with this task. These left over instances represent - * probed functions that have been called but will never return. + * This function is called from finish_task_switch when task tk becomes dead, + * so that we can recycle any function-return probe instances associated + * with this task. These left over instances represent probed functions + * that have been called but will never return. */ void __kprobes kprobe_flush_task(struct task_struct *tk) { @@ -336,7 +336,7 @@ void __kprobes kprobe_flush_task(struct task_struct *tk) unsigned long flags = 0; spin_lock_irqsave(&kretprobe_lock, flags); - head = kretprobe_inst_table_head(current); + head = kretprobe_inst_table_head(tk); hlist_for_each_entry_safe(ri, node, tmp, head, hlist) { if (ri->task == tk) recycle_rp_inst(ri); diff --git a/kernel/sched.c b/kernel/sched.c index 7ffaabd64f8..78acdefeccc 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -49,6 +49,7 @@ #include #include #include +#include #include #include @@ -1546,8 +1547,14 @@ static inline void finish_task_switch(runqueue_t *rq, task_t *prev) finish_lock_switch(rq, prev); if (mm) mmdrop(mm); - if (unlikely(prev_task_flags & PF_DEAD)) + if (unlikely(prev_task_flags & PF_DEAD)) { + /* + * Remove function-return probe instances associated with this + * task and put them back on the free list. + */ + kprobe_flush_task(prev); put_task_struct(prev); + } } /** -- cgit v1.2.3-70-g09d2 From 2326c77017c79fd6d55c69d8a49a57a252921bcd Mon Sep 17 00:00:00 2001 From: "bibo,mao" Date: Sun, 26 Mar 2006 01:38:21 -0800 Subject: [PATCH] kprobe handler: discard user space trap Currently kprobe handler traps only happen in kernel space, so function kprobe_exceptions_notify should skip traps which happen in user space. This patch modifies this, and it is based on 2.6.16-rc4. Signed-off-by: bibo mao Cc: Ananth N Mavinakayanahalli Cc: "Keshavamurthy, Anil S" Cc: Signed-off-by: Prasanna S Panchamukhi Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/kernel/kprobes.c | 19 ++++++------------- arch/ia64/kernel/kprobes.c | 3 +++ arch/powerpc/kernel/kprobes.c | 3 +++ arch/sparc64/kernel/kprobes.c | 3 +++ arch/x86_64/kernel/kprobes.c | 3 +++ 5 files changed, 18 insertions(+), 13 deletions(-) (limited to 'arch/ia64') diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c index acdcc640a72..df1b346d36f 100644 --- a/arch/i386/kernel/kprobes.c +++ b/arch/i386/kernel/kprobes.c @@ -203,13 +203,14 @@ static int __kprobes kprobe_handler(struct pt_regs *regs) { struct kprobe *p; int ret = 0; - kprobe_opcode_t *addr = NULL; - unsigned long *lp; + kprobe_opcode_t *addr; struct kprobe_ctlblk *kcb; #ifdef CONFIG_PREEMPT unsigned pre_preempt_count = preempt_count(); #endif /* CONFIG_PREEMPT */ + addr = (kprobe_opcode_t *)(regs->eip - sizeof(kprobe_opcode_t)); + /* * We don't want to be preempted for the entire * duration of kprobe processing @@ -217,17 +218,6 @@ static int __kprobes kprobe_handler(struct pt_regs *regs) preempt_disable(); kcb = get_kprobe_ctlblk(); - /* Check if the application is using LDT entry for its code segment and - * calculate the address by reading the base address from the LDT entry. - */ - if ((regs->xcs & 4) && (current->mm)) { - lp = (unsigned long *) ((unsigned long)((regs->xcs >> 3) * 8) - + (char *) current->mm->context.ldt); - addr = (kprobe_opcode_t *) (get_desc_base(lp) + regs->eip - - sizeof(kprobe_opcode_t)); - } else { - addr = (kprobe_opcode_t *)(regs->eip - sizeof(kprobe_opcode_t)); - } /* Check we're not actually recursing */ if (kprobe_running()) { p = get_kprobe(addr); @@ -579,6 +569,9 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self, struct die_args *args = (struct die_args *)data; int ret = NOTIFY_DONE; + if (args->regs && user_mode(args->regs)) + return ret; + switch (val) { case DIE_INT3: if (kprobe_handler(args->regs)) diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c index 50ae8c7d453..45b8479c986 100644 --- a/arch/ia64/kernel/kprobes.c +++ b/arch/ia64/kernel/kprobes.c @@ -740,6 +740,9 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self, struct die_args *args = (struct die_args *)data; int ret = NOTIFY_DONE; + if (args->regs && user_mode(args->regs)) + return ret; + switch(val) { case DIE_BREAK: /* err is break number from ia64_bad_break() */ diff --git a/arch/powerpc/kernel/kprobes.c b/arch/powerpc/kernel/kprobes.c index cb1fe5878e8..aea25dd18da 100644 --- a/arch/powerpc/kernel/kprobes.c +++ b/arch/powerpc/kernel/kprobes.c @@ -396,6 +396,9 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self, struct die_args *args = (struct die_args *)data; int ret = NOTIFY_DONE; + if (args->regs && user_mode(args->regs)) + return ret; + switch (val) { case DIE_BPT: if (kprobe_handler(args->regs)) diff --git a/arch/sparc64/kernel/kprobes.c b/arch/sparc64/kernel/kprobes.c index b9a9ce70e55..d91c31870ac 100644 --- a/arch/sparc64/kernel/kprobes.c +++ b/arch/sparc64/kernel/kprobes.c @@ -324,6 +324,9 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self, struct die_args *args = (struct die_args *)data; int ret = NOTIFY_DONE; + if (args->regs && user_mode(args->regs)) + return ret; + switch (val) { case DIE_DEBUG: if (kprobe_handler(args->regs)) diff --git a/arch/x86_64/kernel/kprobes.c b/arch/x86_64/kernel/kprobes.c index 14f0ced613b..218e015c319 100644 --- a/arch/x86_64/kernel/kprobes.c +++ b/arch/x86_64/kernel/kprobes.c @@ -601,6 +601,9 @@ int __kprobes kprobe_exceptions_notify(struct notifier_block *self, struct die_args *args = (struct die_args *)data; int ret = NOTIFY_DONE; + if (args->regs && user_mode(args->regs)) + return ret; + switch (val) { case DIE_INT3: if (kprobe_handler(args->regs)) -- cgit v1.2.3-70-g09d2 From c04c1c81e211c2c95b548add7923e1d4ba4847ab Mon Sep 17 00:00:00 2001 From: Prasanna S Panchamukhi Date: Sun, 26 Mar 2006 01:38:25 -0800 Subject: [PATCH] kprobes: fix broken fault handling for ia64 Provide proper kprobes fault handling, if a user-specified pre/post handlers tries to access user address space, through copy_from_user(), get_user() etc. The user-specified fault handler gets called only if the fault occurs while executing user-specified handlers. In such a case user-specified handler is allowed to fix it first, later if the user-specifed fault handler does not fix it, we try to fix it by calling fix_exception(). The user-specified handler will not be called if the fault happens when single stepping the original instruction, instead we reset the current probe and allow the system page fault handler to fix it up. Signed-off-by: Prasanna S Panchamukhi Acked-by: Anil S Keshavamurthy Cc: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/kernel/kprobes.c | 48 +++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 43 insertions(+), 5 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c index 45b8479c986..789881ca83d 100644 --- a/arch/ia64/kernel/kprobes.c +++ b/arch/ia64/kernel/kprobes.c @@ -34,6 +34,7 @@ #include #include #include +#include extern void jprobe_inst_return(void); @@ -722,13 +723,50 @@ static int __kprobes kprobes_fault_handler(struct pt_regs *regs, int trapnr) struct kprobe *cur = kprobe_running(); struct kprobe_ctlblk *kcb = get_kprobe_ctlblk(); - if (cur->fault_handler && cur->fault_handler(cur, regs, trapnr)) - return 1; - if (kcb->kprobe_status & KPROBE_HIT_SS) { - resume_execution(cur, regs); - reset_current_kprobe(); + switch(kcb->kprobe_status) { + case KPROBE_HIT_SS: + case KPROBE_REENTER: + /* + * We are here because the instruction being single + * stepped caused a page fault. We reset the current + * kprobe and the instruction pointer points back to + * the probe address and allow the page fault handler + * to continue as a normal page fault. + */ + regs->cr_iip = ((unsigned long)cur->addr) & ~0xFULL; + ia64_psr(regs)->ri = ((unsigned long)cur->addr) & 0xf; + if (kcb->kprobe_status == KPROBE_REENTER) + restore_previous_kprobe(kcb); + else + reset_current_kprobe(); preempt_enable_no_resched(); + break; + case KPROBE_HIT_ACTIVE: + case KPROBE_HIT_SSDONE: + /* + * We increment the nmissed count for accounting, + * we can also use npre/npostfault count for accouting + * these specific fault cases. + */ + kprobes_inc_nmissed_count(cur); + + /* + * We come here because instructions in the pre/post + * handler caused the page_fault, this could happen + * if handler tries to access user space by + * copy_from_user(), get_user() etc. Let the + * user-specified handler try to fix it first. + */ + if (cur->fault_handler && cur->fault_handler(cur, regs, trapnr)) + return 1; + + /* + * Let ia64_do_page_fault() fix it. + */ + break; + default: + break; } return 0; -- cgit v1.2.3-70-g09d2 From 4668f0cd0abd41a908ec7aa292e91fd158e0c8f8 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Sun, 26 Mar 2006 01:39:03 -0800 Subject: [PATCH] bitops: ia64: use cpu_set() instead of __set_bit() __set_bit() --> cpu_set() cleanup Signed-off-by: Akinobu Mita Cc: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/kernel/mca.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c index 87ff7fe33cf..8963171788d 100644 --- a/arch/ia64/kernel/mca.c +++ b/arch/ia64/kernel/mca.c @@ -69,6 +69,7 @@ #include #include #include +#include #include #include @@ -1505,7 +1506,7 @@ format_mca_init_stack(void *mca_data, unsigned long offset, ti->cpu = cpu; p->thread_info = ti; p->state = TASK_UNINTERRUPTIBLE; - __set_bit(cpu, &p->cpus_allowed); + cpu_set(cpu, p->cpus_allowed); INIT_LIST_HEAD(&p->tasks); p->parent = p->real_parent = p->group_leader = p; INIT_LIST_HEAD(&p->children); -- cgit v1.2.3-70-g09d2 From 2875aef8bd0e42367a66a78ef7abe10f3bba27b5 Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Sun, 26 Mar 2006 01:39:25 -0800 Subject: [PATCH] bitops: ia64: use generic bitops - remove generic_fls64() - remove find_{next,first}{,_zero}_bit() - remove ext2_{set,clear,test,find_first_zero,find_next_zero}_bit() - remove minix_{test,set,test_and_clear,test,find_first_zero}_bit() - remove sched_find_first_bit() Signed-off-by: Akinobu Mita Cc: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/Kconfig | 4 +++ arch/ia64/lib/Makefile | 2 +- arch/ia64/lib/bitop.c | 88 ----------------------------------------------- include/asm-ia64/bitops.h | 67 +++++++++--------------------------- 4 files changed, 22 insertions(+), 139 deletions(-) delete mode 100644 arch/ia64/lib/bitop.c (limited to 'arch/ia64') diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index d790a6d9026..edffe25a477 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -34,6 +34,10 @@ config RWSEM_XCHGADD_ALGORITHM bool default y +config GENERIC_FIND_NEXT_BIT + bool + default y + config GENERIC_CALIBRATE_DELAY bool default y diff --git a/arch/ia64/lib/Makefile b/arch/ia64/lib/Makefile index ac64664a180..d8536a2c22a 100644 --- a/arch/ia64/lib/Makefile +++ b/arch/ia64/lib/Makefile @@ -6,7 +6,7 @@ obj-y := io.o lib-y := __divsi3.o __udivsi3.o __modsi3.o __umodsi3.o \ __divdi3.o __udivdi3.o __moddi3.o __umoddi3.o \ - bitop.o checksum.o clear_page.o csum_partial_copy.o \ + checksum.o clear_page.o csum_partial_copy.o \ clear_user.o strncpy_from_user.o strlen_user.o strnlen_user.o \ flush.o ip_fast_csum.o do_csum.o \ memset.o strlen.o diff --git a/arch/ia64/lib/bitop.c b/arch/ia64/lib/bitop.c deleted file mode 100644 index 82e299c8464..00000000000 --- a/arch/ia64/lib/bitop.c +++ /dev/null @@ -1,88 +0,0 @@ -#include -#include -#include -#include -#include - -/* - * Find next zero bit in a bitmap reasonably efficiently.. - */ - -int __find_next_zero_bit (const void *addr, unsigned long size, unsigned long offset) -{ - unsigned long *p = ((unsigned long *) addr) + (offset >> 6); - unsigned long result = offset & ~63UL; - unsigned long tmp; - - if (offset >= size) - return size; - size -= result; - offset &= 63UL; - if (offset) { - tmp = *(p++); - tmp |= ~0UL >> (64-offset); - if (size < 64) - goto found_first; - if (~tmp) - goto found_middle; - size -= 64; - result += 64; - } - while (size & ~63UL) { - if (~(tmp = *(p++))) - goto found_middle; - result += 64; - size -= 64; - } - if (!size) - return result; - tmp = *p; -found_first: - tmp |= ~0UL << size; - if (tmp == ~0UL) /* any bits zero? */ - return result + size; /* nope */ -found_middle: - return result + ffz(tmp); -} -EXPORT_SYMBOL(__find_next_zero_bit); - -/* - * Find next bit in a bitmap reasonably efficiently.. - */ -int __find_next_bit(const void *addr, unsigned long size, unsigned long offset) -{ - unsigned long *p = ((unsigned long *) addr) + (offset >> 6); - unsigned long result = offset & ~63UL; - unsigned long tmp; - - if (offset >= size) - return size; - size -= result; - offset &= 63UL; - if (offset) { - tmp = *(p++); - tmp &= ~0UL << offset; - if (size < 64) - goto found_first; - if (tmp) - goto found_middle; - size -= 64; - result += 64; - } - while (size & ~63UL) { - if ((tmp = *(p++))) - goto found_middle; - result += 64; - size -= 64; - } - if (!size) - return result; - tmp = *p; - found_first: - tmp &= ~0UL >> (64-size); - if (tmp == 0UL) /* Are any bits set? */ - return result + size; /* Nope. */ - found_middle: - return result + __ffs(tmp); -} -EXPORT_SYMBOL(__find_next_bit); diff --git a/include/asm-ia64/bitops.h b/include/asm-ia64/bitops.h index eccb01c79c1..90921e16279 100644 --- a/include/asm-ia64/bitops.h +++ b/include/asm-ia64/bitops.h @@ -5,8 +5,8 @@ * Copyright (C) 1998-2003 Hewlett-Packard Co * David Mosberger-Tang * - * 02/06/02 find_next_bit() and find_first_bit() added from Erich Focht's ia64 O(1) - * scheduler patch + * 02/06/02 find_next_bit() and find_first_bit() added from Erich Focht's ia64 + * O(1) scheduler patch */ #include @@ -25,9 +25,9 @@ * restricted to acting on a single-word quantity. * * The address must be (at least) "long" aligned. - * Note that there are driver (e.g., eepro100) which use these operations to operate on - * hw-defined data-structures, so we can't easily change these operations to force a - * bigger alignment. + * Note that there are driver (e.g., eepro100) which use these operations to + * operate on hw-defined data-structures, so we can't easily change these + * operations to force a bigger alignment. * * bit 0 is the LSB of addr; bit 32 is the LSB of (addr+1). */ @@ -284,8 +284,8 @@ test_bit (int nr, const volatile void *addr) * ffz - find the first zero bit in a long word * @x: The long word to find the bit in * - * Returns the bit-number (0..63) of the first (least significant) zero bit. Undefined if - * no zero exists, so code should check against ~0UL first... + * Returns the bit-number (0..63) of the first (least significant) zero bit. + * Undefined if no zero exists, so code should check against ~0UL first... */ static inline unsigned long ffz (unsigned long x) @@ -345,13 +345,14 @@ fls (int t) x |= x >> 16; return ia64_popcnt(x); } -#define fls64(x) generic_fls64(x) + +#include /* - * ffs: find first bit set. This is defined the same way as the libc and compiler builtin - * ffs routines, therefore differs in spirit from the above ffz (man ffs): it operates on - * "int" values only and the result value is the bit number + 1. ffs(0) is defined to - * return zero. + * ffs: find first bit set. This is defined the same way as the libc and + * compiler builtin ffs routines, therefore differs in spirit from the above + * ffz (man ffs): it operates on "int" values only and the result value is the + * bit number + 1. ffs(0) is defined to return zero. */ #define ffs(x) __builtin_ffs(x) @@ -373,51 +374,17 @@ hweight64 (unsigned long x) #endif /* __KERNEL__ */ -extern int __find_next_zero_bit (const void *addr, unsigned long size, - unsigned long offset); -extern int __find_next_bit(const void *addr, unsigned long size, - unsigned long offset); - -#define find_next_zero_bit(addr, size, offset) \ - __find_next_zero_bit((addr), (size), (offset)) -#define find_next_bit(addr, size, offset) \ - __find_next_bit((addr), (size), (offset)) - -/* - * The optimizer actually does good code for this case.. - */ -#define find_first_zero_bit(addr, size) find_next_zero_bit((addr), (size), 0) - -#define find_first_bit(addr, size) find_next_bit((addr), (size), 0) +#include #ifdef __KERNEL__ -#define __clear_bit(nr, addr) clear_bit(nr, addr) +#include -#define ext2_set_bit __test_and_set_bit #define ext2_set_bit_atomic(l,n,a) test_and_set_bit(n,a) -#define ext2_clear_bit __test_and_clear_bit #define ext2_clear_bit_atomic(l,n,a) test_and_clear_bit(n,a) -#define ext2_test_bit test_bit -#define ext2_find_first_zero_bit find_first_zero_bit -#define ext2_find_next_zero_bit find_next_zero_bit - -/* Bitmap functions for the minix filesystem. */ -#define minix_test_and_set_bit(nr,addr) __test_and_set_bit(nr,addr) -#define minix_set_bit(nr,addr) __set_bit(nr,addr) -#define minix_test_and_clear_bit(nr,addr) __test_and_clear_bit(nr,addr) -#define minix_test_bit(nr,addr) test_bit(nr,addr) -#define minix_find_first_zero_bit(addr,size) find_first_zero_bit(addr,size) -static inline int -sched_find_first_bit (unsigned long *b) -{ - if (unlikely(b[0])) - return __ffs(b[0]); - if (unlikely(b[1])) - return 64 + __ffs(b[1]); - return __ffs(b[2]) + 128; -} +#include +#include #endif /* __KERNEL__ */ -- cgit v1.2.3-70-g09d2 From 55b0f8a68ab475cc68f88d1e7a873490869d022e Mon Sep 17 00:00:00 2001 From: Akinobu Mita Date: Sun, 26 Mar 2006 01:39:53 -0800 Subject: [PATCH] bitops: ia64: make partial_page.bitmap an unsigned long The find_*_bit() routines are defined to work on a pointer to unsigned long. But partial_page.bitmap is unsigned int and it is passed to find_*_bit() in arch/ia64/ia32/sys_ia32.c. So the compiler will print warnings. This patch changes to unsigned long instead. Signed-off-by: Akinobu Mita Cc: "Luck, Tony" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/ia32/ia32priv.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/ia32/ia32priv.h b/arch/ia64/ia32/ia32priv.h index 68ceb4e690c..ccb98ed48e5 100644 --- a/arch/ia64/ia32/ia32priv.h +++ b/arch/ia64/ia32/ia32priv.h @@ -29,9 +29,9 @@ struct partial_page { struct partial_page *next; /* linked list, sorted by address */ struct rb_node pp_rb; - /* 64K is the largest "normal" page supported by ia64 ABI. So 4K*32 + /* 64K is the largest "normal" page supported by ia64 ABI. So 4K*64 * should suffice.*/ - unsigned int bitmap; + unsigned long bitmap; unsigned int base; }; -- cgit v1.2.3-70-g09d2 From ec936fc563715a9e2b2e363eb060655b49529325 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Mon, 27 Mar 2006 01:15:59 -0800 Subject: [PATCH] for_each_online_pgdat: renaming for_each_pgdat Replace for_each_pgdat() with for_each_online_pgdat(). Signed-off-by: KAMEZAWA Hiroyuki Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/mm/pgtable.c | 2 +- arch/ia64/mm/discontig.c | 4 ++-- arch/ia64/mm/init.c | 2 +- arch/m32r/mm/init.c | 2 +- arch/powerpc/mm/mem.c | 4 ++-- arch/x86_64/mm/init.c | 2 +- fs/buffer.c | 2 +- mm/page_alloc.c | 6 +++--- mm/vmscan.c | 6 +++--- 9 files changed, 15 insertions(+), 15 deletions(-) (limited to 'arch/ia64') diff --git a/arch/i386/mm/pgtable.c b/arch/i386/mm/pgtable.c index 9db3242103b..2889567e21a 100644 --- a/arch/i386/mm/pgtable.c +++ b/arch/i386/mm/pgtable.c @@ -36,7 +36,7 @@ void show_mem(void) printk(KERN_INFO "Mem-info:\n"); show_free_areas(); printk(KERN_INFO "Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); - for_each_pgdat(pgdat) { + for_each_online_pgdat(pgdat) { pgdat_resize_lock(pgdat, &flags); for (i = 0; i < pgdat->node_spanned_pages; ++i) { page = pgdat_page_nr(pgdat, i); diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c index 2f5e44862e9..384f1d7dce9 100644 --- a/arch/ia64/mm/discontig.c +++ b/arch/ia64/mm/discontig.c @@ -386,7 +386,7 @@ static void __init pgdat_insert(pg_data_t *pgdat) { pg_data_t *prev = NULL, *next; - for_each_pgdat(next) + for_each_online_pgdat(next) if (pgdat->node_id < next->node_id) break; else @@ -560,7 +560,7 @@ void show_mem(void) printk("Mem-info:\n"); show_free_areas(); printk("Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); - for_each_pgdat(pgdat) { + for_each_online_pgdat(pgdat) { unsigned long present; unsigned long flags; int shared = 0, cached = 0, reserved = 0; diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c index ff4f31fcd33..2ef1151cde9 100644 --- a/arch/ia64/mm/init.c +++ b/arch/ia64/mm/init.c @@ -600,7 +600,7 @@ mem_init (void) kclist_add(&kcore_vmem, (void *)VMALLOC_START, VMALLOC_END-VMALLOC_START); kclist_add(&kcore_kernel, _stext, _end - _stext); - for_each_pgdat(pgdat) + for_each_online_pgdat(pgdat) if (pgdat->bdata->node_bootmem_map) totalram_pages += free_all_bootmem_node(pgdat); diff --git a/arch/m32r/mm/init.c b/arch/m32r/mm/init.c index c9e7dad860b..2e0fe199ce3 100644 --- a/arch/m32r/mm/init.c +++ b/arch/m32r/mm/init.c @@ -47,7 +47,7 @@ void show_mem(void) printk("Mem-info:\n"); show_free_areas(); printk("Free swap: %6ldkB\n",nr_swap_pages<<(PAGE_SHIFT-10)); - for_each_pgdat(pgdat) { + for_each_online_pgdat(pgdat) { unsigned long flags; pgdat_resize_lock(pgdat, &flags); for (i = 0; i < pgdat->node_spanned_pages; ++i) { diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c index badac10d700..5e435a9c343 100644 --- a/arch/powerpc/mm/mem.c +++ b/arch/powerpc/mm/mem.c @@ -195,7 +195,7 @@ void show_mem(void) printk("Mem-info:\n"); show_free_areas(); printk("Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); - for_each_pgdat(pgdat) { + for_each_online_pgdat(pgdat) { unsigned long flags; pgdat_resize_lock(pgdat, &flags); for (i = 0; i < pgdat->node_spanned_pages; i++) { @@ -351,7 +351,7 @@ void __init mem_init(void) max_mapnr = max_pfn; totalram_pages += free_all_bootmem(); #endif - for_each_pgdat(pgdat) { + for_each_online_pgdat(pgdat) { for (i = 0; i < pgdat->node_spanned_pages; i++) { if (!pfn_valid(pgdat->node_start_pfn + i)) continue; diff --git a/arch/x86_64/mm/init.c b/arch/x86_64/mm/init.c index b0441562544..e5f7f1c3446 100644 --- a/arch/x86_64/mm/init.c +++ b/arch/x86_64/mm/init.c @@ -72,7 +72,7 @@ void show_mem(void) show_free_areas(); printk(KERN_INFO "Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); - for_each_pgdat(pgdat) { + for_each_online_pgdat(pgdat) { for (i = 0; i < pgdat->node_spanned_pages; ++i) { page = pfn_to_page(pgdat->node_start_pfn + i); total++; diff --git a/fs/buffer.c b/fs/buffer.c index d597758dd12..23f1f3a6807 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -493,7 +493,7 @@ static void free_more_memory(void) wakeup_pdflush(1024); yield(); - for_each_pgdat(pgdat) { + for_each_online_pgdat(pgdat) { zones = pgdat->node_zonelists[gfp_zone(GFP_NOFS)].zones; if (*zones) try_to_free_pages(zones, GFP_NOFS); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 8dc8f2735d2..ccc3713dd40 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1201,7 +1201,7 @@ unsigned int nr_free_highpages (void) pg_data_t *pgdat; unsigned int pages = 0; - for_each_pgdat(pgdat) + for_each_online_pgdat(pgdat) pages += pgdat->node_zones[ZONE_HIGHMEM].free_pages; return pages; @@ -1343,7 +1343,7 @@ void get_zone_counts(unsigned long *active, *active = 0; *inactive = 0; *free = 0; - for_each_pgdat(pgdat) { + for_each_online_pgdat(pgdat) { unsigned long l, m, n; __get_zone_counts(&l, &m, &n, pgdat); *active += l; @@ -2482,7 +2482,7 @@ static void setup_per_zone_lowmem_reserve(void) struct pglist_data *pgdat; int j, idx; - for_each_pgdat(pgdat) { + for_each_online_pgdat(pgdat) { for (j = 0; j < MAX_NR_ZONES; j++) { struct zone *zone = pgdat->node_zones + j; unsigned long present_pages = zone->present_pages; diff --git a/mm/vmscan.c b/mm/vmscan.c index 78865c849f8..acdf001d694 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1305,7 +1305,7 @@ unsigned long shrink_all_memory(unsigned long nr_pages) current->reclaim_state = &reclaim_state; repeat: - for_each_pgdat(pgdat) { + for_each_online_pgdat(pgdat) { unsigned long freed; freed = balance_pgdat(pgdat, nr_to_free, 0); @@ -1335,7 +1335,7 @@ static int __devinit cpu_callback(struct notifier_block *nfb, cpumask_t mask; if (action == CPU_ONLINE) { - for_each_pgdat(pgdat) { + for_each_online_pgdat(pgdat) { mask = node_to_cpumask(pgdat->node_id); if (any_online_cpu(mask) != NR_CPUS) /* One of our CPUs online: restore mask */ @@ -1351,7 +1351,7 @@ static int __init kswapd_init(void) pg_data_t *pgdat; swap_setup(); - for_each_pgdat(pgdat) { + for_each_online_pgdat(pgdat) { pid_t pid; pid = kernel_thread(kswapd, pgdat, CLONE_KERNEL); -- cgit v1.2.3-70-g09d2 From 3571761fe49d960bb720c2308ffb9401f0a5e161 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Mon, 27 Mar 2006 01:16:00 -0800 Subject: [PATCH] for_each_online_pgdat: remove sorting pgdat Because pgdat_list was linked to pgdat_list in *reverse* order, (By default) some of arch has to sort it by themselves. for_each_pgdat has gone..for_each_online_pgdat() uses node_online_map, which doesn't need to be sorted. This patch removes codes for sorting pgdat. Signed-off-by: KAMEZAWA Hiroyuki Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/mm/discontig.c | 11 ----------- arch/ia64/mm/discontig.c | 31 ------------------------------- arch/m32r/mm/discontig.c | 6 ------ 3 files changed, 48 deletions(-) (limited to 'arch/ia64') diff --git a/arch/i386/mm/discontig.c b/arch/i386/mm/discontig.c index c4af9638dbf..c3f3ae95e22 100644 --- a/arch/i386/mm/discontig.c +++ b/arch/i386/mm/discontig.c @@ -352,17 +352,6 @@ void __init zone_sizes_init(void) { int nid; - /* - * Insert nodes into pgdat_list backward so they appear in order. - * Clobber node 0's links and NULL out pgdat_list before starting. - */ - pgdat_list = NULL; - for (nid = MAX_NUMNODES - 1; nid >= 0; nid--) { - if (!node_online(nid)) - continue; - NODE_DATA(nid)->pgdat_next = pgdat_list; - pgdat_list = NODE_DATA(nid); - } for_each_online_node(nid) { unsigned long zones_size[MAX_NR_ZONES] = {0, 0, 0}; diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c index 384f1d7dce9..ec9eeb89975 100644 --- a/arch/ia64/mm/discontig.c +++ b/arch/ia64/mm/discontig.c @@ -378,31 +378,6 @@ static void __init *memory_less_node_alloc(int nid, unsigned long pernodesize) return ptr; } -/** - * pgdat_insert - insert the pgdat into global pgdat_list - * @pgdat: the pgdat for a node. - */ -static void __init pgdat_insert(pg_data_t *pgdat) -{ - pg_data_t *prev = NULL, *next; - - for_each_online_pgdat(next) - if (pgdat->node_id < next->node_id) - break; - else - prev = next; - - if (prev) { - prev->pgdat_next = pgdat; - pgdat->pgdat_next = next; - } else { - pgdat->pgdat_next = pgdat_list; - pgdat_list = pgdat; - } - - return; -} - /** * memory_less_nodes - allocate and initialize CPU only nodes pernode * information. @@ -745,11 +720,5 @@ void __init paging_init(void) pfn_offset, zholes_size); } - /* - * Make memory less nodes become a member of the known nodes. - */ - for_each_node_mask(node, memory_less_mask) - pgdat_insert(mem_data[node].pgdat); - zero_page_memmap_ptr = virt_to_page(ia64_imva(empty_zero_page)); } diff --git a/arch/m32r/mm/discontig.c b/arch/m32r/mm/discontig.c index 08e72795555..70c8528a0ad 100644 --- a/arch/m32r/mm/discontig.c +++ b/arch/m32r/mm/discontig.c @@ -137,12 +137,6 @@ unsigned long __init zone_sizes_init(void) int nid, i; mem_prof_t *mp; - pgdat_list = NULL; - for (nid = num_online_nodes() - 1 ; nid >= 0 ; nid--) { - NODE_DATA(nid)->pgdat_next = pgdat_list; - pgdat_list = NODE_DATA(nid); - } - for_each_online_node(nid) { mp = &mem_prof[nid]; for (i = 0 ; i < MAX_NR_ZONES ; i++) { -- cgit v1.2.3-70-g09d2 From e041c683412d5bf44dc2b109053e3b837b71742d Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Mon, 27 Mar 2006 01:16:30 -0800 Subject: [PATCH] Notifier chain update: API changes The kernel's implementation of notifier chains is unsafe. There is no protection against entries being added to or removed from a chain while the chain is in use. The issues were discussed in this thread: http://marc.theaimsgroup.com/?l=linux-kernel&m=113018709002036&w=2 We noticed that notifier chains in the kernel fall into two basic usage classes: "Blocking" chains are always called from a process context and the callout routines are allowed to sleep; "Atomic" chains can be called from an atomic context and the callout routines are not allowed to sleep. We decided to codify this distinction and make it part of the API. Therefore this set of patches introduces three new, parallel APIs: one for blocking notifiers, one for atomic notifiers, and one for "raw" notifiers (which is really just the old API under a new name). New kinds of data structures are used for the heads of the chains, and new routines are defined for registration, unregistration, and calling a chain. The three APIs are explained in include/linux/notifier.h and their implementation is in kernel/sys.c. With atomic and blocking chains, the implementation guarantees that the chain links will not be corrupted and that chain callers will not get messed up by entries being added or removed. For raw chains the implementation provides no guarantees at all; users of this API must provide their own protections. (The idea was that situations may come up where the assumptions of the atomic and blocking APIs are not appropriate, so it should be possible for users to handle these things in their own way.) There are some limitations, which should not be too hard to live with. For atomic/blocking chains, registration and unregistration must always be done in a process context since the chain is protected by a mutex/rwsem. Also, a callout routine for a non-raw chain must not try to register or unregister entries on its own chain. (This did happen in a couple of places and the code had to be changed to avoid it.) Since atomic chains may be called from within an NMI handler, they cannot use spinlocks for synchronization. Instead we use RCU. The overhead falls almost entirely in the unregister routine, which is okay since unregistration is much less frequent that calling a chain. Here is the list of chains that we adjusted and their classifications. None of them use the raw API, so for the moment it is only a placeholder. ATOMIC CHAINS ------------- arch/i386/kernel/traps.c: i386die_chain arch/ia64/kernel/traps.c: ia64die_chain arch/powerpc/kernel/traps.c: powerpc_die_chain arch/sparc64/kernel/traps.c: sparc64die_chain arch/x86_64/kernel/traps.c: die_chain drivers/char/ipmi/ipmi_si_intf.c: xaction_notifier_list kernel/panic.c: panic_notifier_list kernel/profile.c: task_free_notifier net/bluetooth/hci_core.c: hci_notifier net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_chain net/ipv4/netfilter/ip_conntrack_core.c: ip_conntrack_expect_chain net/ipv6/addrconf.c: inet6addr_chain net/netfilter/nf_conntrack_core.c: nf_conntrack_chain net/netfilter/nf_conntrack_core.c: nf_conntrack_expect_chain net/netlink/af_netlink.c: netlink_chain BLOCKING CHAINS --------------- arch/powerpc/platforms/pseries/reconfig.c: pSeries_reconfig_chain arch/s390/kernel/process.c: idle_chain arch/x86_64/kernel/process.c idle_notifier drivers/base/memory.c: memory_chain drivers/cpufreq/cpufreq.c cpufreq_policy_notifier_list drivers/cpufreq/cpufreq.c cpufreq_transition_notifier_list drivers/macintosh/adb.c: adb_client_list drivers/macintosh/via-pmu.c sleep_notifier_list drivers/macintosh/via-pmu68k.c sleep_notifier_list drivers/macintosh/windfarm_core.c wf_client_list drivers/usb/core/notify.c usb_notifier_list drivers/video/fbmem.c fb_notifier_list kernel/cpu.c cpu_chain kernel/module.c module_notify_list kernel/profile.c munmap_notifier kernel/profile.c task_exit_notifier kernel/sys.c reboot_notifier_list net/core/dev.c netdev_chain net/decnet/dn_dev.c: dnaddr_chain net/ipv4/devinet.c: inetaddr_chain It's possible that some of these classifications are wrong. If they are, please let us know or submit a patch to fix them. Note that any chain that gets called very frequently should be atomic, because the rwsem read-locking used for blocking chains is very likely to incur cache misses on SMP systems. (However, if the chain's callout routines may sleep then the chain cannot be atomic.) The patch set was written by Alan Stern and Chandra Seetharaman, incorporating material written by Keith Owens and suggestions from Paul McKenney and Andrew Morton. [jes@sgi.com: restructure the notifier chain initialization macros] Signed-off-by: Alan Stern Signed-off-by: Chandra Seetharaman Signed-off-by: Jes Sorensen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/alpha/kernel/setup.c | 5 +- arch/arm/mach-omap1/board-netstar.c | 2 +- arch/arm/mach-omap1/board-voiceblue.c | 2 +- arch/i386/kernel/traps.c | 17 +- arch/ia64/kernel/traps.c | 6 +- arch/mips/lasat/setup.c | 3 +- arch/mips/sgi-ip22/ip22-reset.c | 2 +- arch/mips/sgi-ip32/ip32-reset.c | 2 +- arch/parisc/kernel/pdc_chassis.c | 3 +- arch/powerpc/kernel/setup_64.c | 3 +- arch/powerpc/kernel/traps.c | 16 +- arch/powerpc/platforms/pseries/reconfig.c | 10 +- arch/ppc/platforms/prep_setup.c | 2 +- arch/s390/kernel/process.c | 11 +- arch/sparc64/kernel/traps.c | 17 +- arch/um/drivers/mconsole_kern.c | 3 +- arch/um/kernel/um_arch.c | 3 +- arch/x86_64/kernel/process.c | 17 +- arch/x86_64/kernel/traps.c | 18 +- arch/xtensa/platform-iss/setup.c | 2 +- drivers/base/memory.c | 8 +- drivers/char/ipmi/ipmi_msghandler.c | 4 +- drivers/char/ipmi/ipmi_si_intf.c | 7 +- drivers/char/ipmi/ipmi_watchdog.c | 6 +- drivers/cpufreq/cpufreq.c | 61 +++--- drivers/firmware/dcdbas.c | 19 +- drivers/macintosh/adb.c | 11 +- drivers/macintosh/adbhid.c | 3 +- drivers/macintosh/via-pmu.c | 2 +- drivers/macintosh/via-pmu68k.c | 7 +- drivers/macintosh/windfarm_core.c | 8 +- drivers/misc/ibmasm/heartbeat.c | 5 +- drivers/net/bonding/bond_main.c | 2 +- drivers/parisc/led.c | 14 +- drivers/parisc/power.c | 6 +- drivers/scsi/gdth.c | 9 +- drivers/usb/core/notify.c | 65 +----- drivers/video/fbmem.c | 31 +-- include/asm-i386/kdebug.h | 10 +- include/asm-ia64/kdebug.h | 4 +- include/asm-powerpc/kdebug.h | 12 +- include/asm-sparc64/kdebug.h | 11 +- include/asm-x86_64/kdebug.h | 23 +- include/linux/adb.h | 2 +- include/linux/kernel.h | 2 +- include/linux/memory.h | 1 - include/linux/netfilter_ipv4/ip_conntrack.h | 17 +- include/linux/notifier.h | 96 +++++++- include/net/netfilter/nf_conntrack.h | 17 +- kernel/cpu.c | 29 +-- kernel/module.c | 20 +- kernel/panic.c | 4 +- kernel/profile.c | 53 ++--- kernel/softlockup.c | 2 +- kernel/sys.c | 327 ++++++++++++++++++++++------ net/bluetooth/hci_core.c | 8 +- net/core/dev.c | 42 ++-- net/decnet/dn_dev.c | 10 +- net/ipv4/devinet.c | 16 +- net/ipv4/netfilter/ip_conntrack_core.c | 6 +- net/ipv6/addrconf.c | 10 +- net/netfilter/nf_conntrack_core.c | 6 +- net/netlink/af_netlink.c | 9 +- 63 files changed, 677 insertions(+), 472 deletions(-) (limited to 'arch/ia64') diff --git a/arch/alpha/kernel/setup.c b/arch/alpha/kernel/setup.c index 9402624453c..dd876967059 100644 --- a/arch/alpha/kernel/setup.c +++ b/arch/alpha/kernel/setup.c @@ -43,7 +43,7 @@ #include #include -extern struct notifier_block *panic_notifier_list; +extern struct atomic_notifier_head panic_notifier_list; static int alpha_panic_event(struct notifier_block *, unsigned long, void *); static struct notifier_block alpha_panic_block = { alpha_panic_event, @@ -500,7 +500,8 @@ setup_arch(char **cmdline_p) } /* Register a call for panic conditions. */ - notifier_chain_register(&panic_notifier_list, &alpha_panic_block); + atomic_notifier_chain_register(&panic_notifier_list, + &alpha_panic_block); #ifdef CONFIG_ALPHA_GENERIC /* Assume that we've booted from SRM if we haven't booted from MILO. diff --git a/arch/arm/mach-omap1/board-netstar.c b/arch/arm/mach-omap1/board-netstar.c index 60d5f8a3339..7520e602d7a 100644 --- a/arch/arm/mach-omap1/board-netstar.c +++ b/arch/arm/mach-omap1/board-netstar.c @@ -141,7 +141,7 @@ static int __init netstar_late_init(void) /* TODO: Setup front panel switch here */ /* Setup panic notifier */ - notifier_chain_register(&panic_notifier_list, &panic_block); + atomic_notifier_chain_register(&panic_notifier_list, &panic_block); return 0; } diff --git a/arch/arm/mach-omap1/board-voiceblue.c b/arch/arm/mach-omap1/board-voiceblue.c index bfd5fdd1a87..52e4a9d6964 100644 --- a/arch/arm/mach-omap1/board-voiceblue.c +++ b/arch/arm/mach-omap1/board-voiceblue.c @@ -235,7 +235,7 @@ static struct notifier_block panic_block = { static int __init voiceblue_setup(void) { /* Setup panic notifier */ - notifier_chain_register(&panic_notifier_list, &panic_block); + atomic_notifier_chain_register(&panic_notifier_list, &panic_block); return 0; } diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 4624f8ca245..6b63a5aa1e4 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -92,22 +92,21 @@ asmlinkage void spurious_interrupt_bug(void); asmlinkage void machine_check(void); static int kstack_depth_to_print = 24; -struct notifier_block *i386die_chain; -static DEFINE_SPINLOCK(die_notifier_lock); +ATOMIC_NOTIFIER_HEAD(i386die_chain); int register_die_notifier(struct notifier_block *nb) { - int err = 0; - unsigned long flags; - vmalloc_sync_all(); - spin_lock_irqsave(&die_notifier_lock, flags); - err = notifier_chain_register(&i386die_chain, nb); - spin_unlock_irqrestore(&die_notifier_lock, flags); - return err; + return atomic_notifier_chain_register(&i386die_chain, nb); } EXPORT_SYMBOL(register_die_notifier); +int unregister_die_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_unregister(&i386die_chain, nb); +} +EXPORT_SYMBOL(unregister_die_notifier); + static inline int valid_stack_ptr(struct thread_info *tinfo, void *p) { return p > (void *)tinfo && diff --git a/arch/ia64/kernel/traps.c b/arch/ia64/kernel/traps.c index dabd6c32641..7c1ddc8ac44 100644 --- a/arch/ia64/kernel/traps.c +++ b/arch/ia64/kernel/traps.c @@ -30,19 +30,19 @@ extern spinlock_t timerlist_lock; fpswa_interface_t *fpswa_interface; EXPORT_SYMBOL(fpswa_interface); -struct notifier_block *ia64die_chain; +ATOMIC_NOTIFIER_HEAD(ia64die_chain); int register_die_notifier(struct notifier_block *nb) { - return notifier_chain_register(&ia64die_chain, nb); + return atomic_notifier_chain_register(&ia64die_chain, nb); } EXPORT_SYMBOL_GPL(register_die_notifier); int unregister_die_notifier(struct notifier_block *nb) { - return notifier_chain_unregister(&ia64die_chain, nb); + return atomic_notifier_chain_unregister(&ia64die_chain, nb); } EXPORT_SYMBOL_GPL(unregister_die_notifier); diff --git a/arch/mips/lasat/setup.c b/arch/mips/lasat/setup.c index 83eb08b7a07..e9e9a89c674 100644 --- a/arch/mips/lasat/setup.c +++ b/arch/mips/lasat/setup.c @@ -165,7 +165,8 @@ void __init plat_setup(void) /* Set up panic notifier */ for (i = 0; i < sizeof(lasat_panic_block) / sizeof(struct notifier_block); i++) - notifier_chain_register(&panic_notifier_list, &lasat_panic_block[i]); + atomic_notifier_chain_register(&panic_notifier_list, + &lasat_panic_block[i]); lasat_reboot_setup(); diff --git a/arch/mips/sgi-ip22/ip22-reset.c b/arch/mips/sgi-ip22/ip22-reset.c index 92a3b3c15ed..a9c58e067b5 100644 --- a/arch/mips/sgi-ip22/ip22-reset.c +++ b/arch/mips/sgi-ip22/ip22-reset.c @@ -238,7 +238,7 @@ static int __init reboot_setup(void) request_irq(SGI_PANEL_IRQ, panel_int, 0, "Front Panel", NULL); init_timer(&blink_timer); blink_timer.function = blink_timeout; - notifier_chain_register(&panic_notifier_list, &panic_block); + atomic_notifier_chain_register(&panic_notifier_list, &panic_block); return 0; } diff --git a/arch/mips/sgi-ip32/ip32-reset.c b/arch/mips/sgi-ip32/ip32-reset.c index 0c948008b02..ab9d9cef089 100644 --- a/arch/mips/sgi-ip32/ip32-reset.c +++ b/arch/mips/sgi-ip32/ip32-reset.c @@ -193,7 +193,7 @@ static __init int ip32_reboot_setup(void) init_timer(&blink_timer); blink_timer.function = blink_timeout; - notifier_chain_register(&panic_notifier_list, &panic_block); + atomic_notifier_chain_register(&panic_notifier_list, &panic_block); request_irq(MACEISA_RTC_IRQ, ip32_rtc_int, 0, "rtc", NULL); diff --git a/arch/parisc/kernel/pdc_chassis.c b/arch/parisc/kernel/pdc_chassis.c index 2a01fe1bdc9..0cea6958f42 100644 --- a/arch/parisc/kernel/pdc_chassis.c +++ b/arch/parisc/kernel/pdc_chassis.c @@ -150,7 +150,8 @@ void __init parisc_pdc_chassis_init(void) if (handle) { /* initialize panic notifier chain */ - notifier_chain_register(&panic_notifier_list, &pdc_chassis_panic_block); + atomic_notifier_chain_register(&panic_notifier_list, + &pdc_chassis_panic_block); /* initialize reboot notifier chain */ register_reboot_notifier(&pdc_chassis_reboot_block); diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c index 2f3fdad3559..e20c1fae342 100644 --- a/arch/powerpc/kernel/setup_64.c +++ b/arch/powerpc/kernel/setup_64.c @@ -579,7 +579,8 @@ void __init setup_arch(char **cmdline_p) panic_timeout = 180; if (ppc_md.panic) - notifier_chain_register(&panic_notifier_list, &ppc64_panic_block); + atomic_notifier_chain_register(&panic_notifier_list, + &ppc64_panic_block); init_mm.start_code = PAGE_OFFSET; init_mm.end_code = (unsigned long) _etext; diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c index 98660aedeeb..9763faab673 100644 --- a/arch/powerpc/kernel/traps.c +++ b/arch/powerpc/kernel/traps.c @@ -74,19 +74,19 @@ EXPORT_SYMBOL(__debugger_dabr_match); EXPORT_SYMBOL(__debugger_fault_handler); #endif -struct notifier_block *powerpc_die_chain; -static DEFINE_SPINLOCK(die_notifier_lock); +ATOMIC_NOTIFIER_HEAD(powerpc_die_chain); int register_die_notifier(struct notifier_block *nb) { - int err = 0; - unsigned long flags; + return atomic_notifier_chain_register(&powerpc_die_chain, nb); +} +EXPORT_SYMBOL(register_die_notifier); - spin_lock_irqsave(&die_notifier_lock, flags); - err = notifier_chain_register(&powerpc_die_chain, nb); - spin_unlock_irqrestore(&die_notifier_lock, flags); - return err; +int unregister_die_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_unregister(&powerpc_die_chain, nb); } +EXPORT_SYMBOL(unregister_die_notifier); /* * Trap & Exception support diff --git a/arch/powerpc/platforms/pseries/reconfig.c b/arch/powerpc/platforms/pseries/reconfig.c index 86cfa6ecdcf..5ad90676567 100644 --- a/arch/powerpc/platforms/pseries/reconfig.c +++ b/arch/powerpc/platforms/pseries/reconfig.c @@ -94,16 +94,16 @@ static struct device_node *derive_parent(const char *path) return parent; } -static struct notifier_block *pSeries_reconfig_chain; +static BLOCKING_NOTIFIER_HEAD(pSeries_reconfig_chain); int pSeries_reconfig_notifier_register(struct notifier_block *nb) { - return notifier_chain_register(&pSeries_reconfig_chain, nb); + return blocking_notifier_chain_register(&pSeries_reconfig_chain, nb); } void pSeries_reconfig_notifier_unregister(struct notifier_block *nb) { - notifier_chain_unregister(&pSeries_reconfig_chain, nb); + blocking_notifier_chain_unregister(&pSeries_reconfig_chain, nb); } static int pSeries_reconfig_add_node(const char *path, struct property *proplist) @@ -131,7 +131,7 @@ static int pSeries_reconfig_add_node(const char *path, struct property *proplist goto out_err; } - err = notifier_call_chain(&pSeries_reconfig_chain, + err = blocking_notifier_call_chain(&pSeries_reconfig_chain, PSERIES_RECONFIG_ADD, np); if (err == NOTIFY_BAD) { printk(KERN_ERR "Failed to add device node %s\n", path); @@ -171,7 +171,7 @@ static int pSeries_reconfig_remove_node(struct device_node *np) remove_node_proc_entries(np); - notifier_call_chain(&pSeries_reconfig_chain, + blocking_notifier_call_chain(&pSeries_reconfig_chain, PSERIES_RECONFIG_REMOVE, np); of_detach_node(np); diff --git a/arch/ppc/platforms/prep_setup.c b/arch/ppc/platforms/prep_setup.c index a0fc628ffb1..d95c05d9824 100644 --- a/arch/ppc/platforms/prep_setup.c +++ b/arch/ppc/platforms/prep_setup.c @@ -736,7 +736,7 @@ ibm_statusled_progress(char *s, unsigned short hex) hex = 0xfff; if (!notifier_installed) { ++notifier_installed; - notifier_chain_register(&panic_notifier_list, + atomic_notifier_chain_register(&panic_notifier_list, &ibm_statusled_block); } } diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c index 99182a415fe..4a0f5a1551e 100644 --- a/arch/s390/kernel/process.c +++ b/arch/s390/kernel/process.c @@ -76,17 +76,17 @@ unsigned long thread_saved_pc(struct task_struct *tsk) /* * Need to know about CPUs going idle? */ -static struct notifier_block *idle_chain; +static ATOMIC_NOTIFIER_HEAD(idle_chain); int register_idle_notifier(struct notifier_block *nb) { - return notifier_chain_register(&idle_chain, nb); + return atomic_notifier_chain_register(&idle_chain, nb); } EXPORT_SYMBOL(register_idle_notifier); int unregister_idle_notifier(struct notifier_block *nb) { - return notifier_chain_unregister(&idle_chain, nb); + return atomic_notifier_chain_unregister(&idle_chain, nb); } EXPORT_SYMBOL(unregister_idle_notifier); @@ -95,7 +95,7 @@ void do_monitor_call(struct pt_regs *regs, long interruption_code) /* disable monitor call class 0 */ __ctl_clear_bit(8, 15); - notifier_call_chain(&idle_chain, CPU_NOT_IDLE, + atomic_notifier_call_chain(&idle_chain, CPU_NOT_IDLE, (void *)(long) smp_processor_id()); } @@ -116,7 +116,8 @@ static void default_idle(void) return; } - rc = notifier_call_chain(&idle_chain, CPU_IDLE, (void *)(long) cpu); + rc = atomic_notifier_call_chain(&idle_chain, + CPU_IDLE, (void *)(long) cpu); if (rc != NOTIFY_OK && rc != NOTIFY_DONE) BUG(); if (rc != NOTIFY_OK) { diff --git a/arch/sparc64/kernel/traps.c b/arch/sparc64/kernel/traps.c index df612e4f75f..ff090bb9734 100644 --- a/arch/sparc64/kernel/traps.c +++ b/arch/sparc64/kernel/traps.c @@ -43,18 +43,19 @@ #include #endif -struct notifier_block *sparc64die_chain; -static DEFINE_SPINLOCK(die_notifier_lock); +ATOMIC_NOTIFIER_HEAD(sparc64die_chain); int register_die_notifier(struct notifier_block *nb) { - int err = 0; - unsigned long flags; - spin_lock_irqsave(&die_notifier_lock, flags); - err = notifier_chain_register(&sparc64die_chain, nb); - spin_unlock_irqrestore(&die_notifier_lock, flags); - return err; + return atomic_notifier_chain_register(&sparc64die_chain, nb); } +EXPORT_SYMBOL(register_die_notifier); + +int unregister_die_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_unregister(&sparc64die_chain, nb); +} +EXPORT_SYMBOL(unregister_die_notifier); /* When an irrecoverable trap occurs at tl > 0, the trap entry * code logs the trap state registers at every level in the trap diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c index 54388d10bcf..1488816588e 100644 --- a/arch/um/drivers/mconsole_kern.c +++ b/arch/um/drivers/mconsole_kern.c @@ -762,7 +762,8 @@ static struct notifier_block panic_exit_notifier = { static int add_notifier(void) { - notifier_chain_register(&panic_notifier_list, &panic_exit_notifier); + atomic_notifier_chain_register(&panic_notifier_list, + &panic_exit_notifier); return(0); } diff --git a/arch/um/kernel/um_arch.c b/arch/um/kernel/um_arch.c index bb1c87211ac..7d51dd7201c 100644 --- a/arch/um/kernel/um_arch.c +++ b/arch/um/kernel/um_arch.c @@ -477,7 +477,8 @@ static struct notifier_block panic_exit_notifier = { void __init setup_arch(char **cmdline_p) { - notifier_chain_register(&panic_notifier_list, &panic_exit_notifier); + atomic_notifier_chain_register(&panic_notifier_list, + &panic_exit_notifier); paging_init(); strlcpy(saved_command_line, command_line, COMMAND_LINE_SIZE); *cmdline_p = command_line; diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c index 0370720515f..70dd8e5c688 100644 --- a/arch/x86_64/kernel/process.c +++ b/arch/x86_64/kernel/process.c @@ -66,24 +66,17 @@ EXPORT_SYMBOL(boot_option_idle_override); void (*pm_idle)(void); static DEFINE_PER_CPU(unsigned int, cpu_idle_state); -static struct notifier_block *idle_notifier; -static DEFINE_SPINLOCK(idle_notifier_lock); +static ATOMIC_NOTIFIER_HEAD(idle_notifier); void idle_notifier_register(struct notifier_block *n) { - unsigned long flags; - spin_lock_irqsave(&idle_notifier_lock, flags); - notifier_chain_register(&idle_notifier, n); - spin_unlock_irqrestore(&idle_notifier_lock, flags); + atomic_notifier_chain_register(&idle_notifier, n); } EXPORT_SYMBOL_GPL(idle_notifier_register); void idle_notifier_unregister(struct notifier_block *n) { - unsigned long flags; - spin_lock_irqsave(&idle_notifier_lock, flags); - notifier_chain_unregister(&idle_notifier, n); - spin_unlock_irqrestore(&idle_notifier_lock, flags); + atomic_notifier_chain_unregister(&idle_notifier, n); } EXPORT_SYMBOL(idle_notifier_unregister); @@ -93,13 +86,13 @@ static DEFINE_PER_CPU(enum idle_state, idle_state) = CPU_NOT_IDLE; void enter_idle(void) { __get_cpu_var(idle_state) = CPU_IDLE; - notifier_call_chain(&idle_notifier, IDLE_START, NULL); + atomic_notifier_call_chain(&idle_notifier, IDLE_START, NULL); } static void __exit_idle(void) { __get_cpu_var(idle_state) = CPU_NOT_IDLE; - notifier_call_chain(&idle_notifier, IDLE_END, NULL); + atomic_notifier_call_chain(&idle_notifier, IDLE_END, NULL); } /* Called from interrupts to signify idle end */ diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index 7b148309c52..edaa9fe654d 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -69,20 +69,20 @@ asmlinkage void alignment_check(void); asmlinkage void machine_check(void); asmlinkage void spurious_interrupt_bug(void); -struct notifier_block *die_chain; -static DEFINE_SPINLOCK(die_notifier_lock); +ATOMIC_NOTIFIER_HEAD(die_chain); int register_die_notifier(struct notifier_block *nb) { - int err = 0; - unsigned long flags; - vmalloc_sync_all(); - spin_lock_irqsave(&die_notifier_lock, flags); - err = notifier_chain_register(&die_chain, nb); - spin_unlock_irqrestore(&die_notifier_lock, flags); - return err; + return atomic_notifier_chain_register(&die_chain, nb); +} +EXPORT_SYMBOL(register_die_notifier); + +int unregister_die_notifier(struct notifier_block *nb) +{ + return atomic_notifier_chain_unregister(&die_chain, nb); } +EXPORT_SYMBOL(unregister_die_notifier); static inline void conditional_sti(struct pt_regs *regs) { diff --git a/arch/xtensa/platform-iss/setup.c b/arch/xtensa/platform-iss/setup.c index 2e6dcbf0cc0..23790a5610e 100644 --- a/arch/xtensa/platform-iss/setup.c +++ b/arch/xtensa/platform-iss/setup.c @@ -108,5 +108,5 @@ static struct notifier_block iss_panic_block = { void __init platform_setup(char **p_cmdline) { - notifier_chain_register(&panic_notifier_list, &iss_panic_block); + atomic_notifier_chain_register(&panic_notifier_list, &iss_panic_block); } diff --git a/drivers/base/memory.c b/drivers/base/memory.c index 105a0d61eb1..dd547af4681 100644 --- a/drivers/base/memory.c +++ b/drivers/base/memory.c @@ -47,16 +47,16 @@ static struct kset_uevent_ops memory_uevent_ops = { .uevent = memory_uevent, }; -static struct notifier_block *memory_chain; +static BLOCKING_NOTIFIER_HEAD(memory_chain); int register_memory_notifier(struct notifier_block *nb) { - return notifier_chain_register(&memory_chain, nb); + return blocking_notifier_chain_register(&memory_chain, nb); } void unregister_memory_notifier(struct notifier_block *nb) { - notifier_chain_unregister(&memory_chain, nb); + blocking_notifier_chain_unregister(&memory_chain, nb); } /* @@ -140,7 +140,7 @@ static ssize_t show_mem_state(struct sys_device *dev, char *buf) static inline int memory_notify(unsigned long val, void *v) { - return notifier_call_chain(&memory_chain, val, v); + return blocking_notifier_call_chain(&memory_chain, val, v); } /* diff --git a/drivers/char/ipmi/ipmi_msghandler.c b/drivers/char/ipmi/ipmi_msghandler.c index b8fb87c6c29..40eb005b9d7 100644 --- a/drivers/char/ipmi/ipmi_msghandler.c +++ b/drivers/char/ipmi/ipmi_msghandler.c @@ -3744,7 +3744,7 @@ static int ipmi_init_msghandler(void) ipmi_timer.expires = jiffies + IPMI_TIMEOUT_JIFFIES; add_timer(&ipmi_timer); - notifier_chain_register(&panic_notifier_list, &panic_block); + atomic_notifier_chain_register(&panic_notifier_list, &panic_block); initialized = 1; @@ -3764,7 +3764,7 @@ static __exit void cleanup_ipmi(void) if (!initialized) return; - notifier_chain_unregister(&panic_notifier_list, &panic_block); + atomic_notifier_chain_unregister(&panic_notifier_list, &panic_block); /* This can't be called if any interfaces exist, so no worry about shutting down the interfaces. */ diff --git a/drivers/char/ipmi/ipmi_si_intf.c b/drivers/char/ipmi/ipmi_si_intf.c index 12f858dc999..35fbd4d8ed4 100644 --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -237,10 +237,10 @@ struct smi_info static int try_smi_init(struct smi_info *smi); -static struct notifier_block *xaction_notifier_list; +static ATOMIC_NOTIFIER_HEAD(xaction_notifier_list); static int register_xaction_notifier(struct notifier_block * nb) { - return notifier_chain_register(&xaction_notifier_list, nb); + return atomic_notifier_chain_register(&xaction_notifier_list, nb); } static void si_restart_short_timer(struct smi_info *smi_info); @@ -302,7 +302,8 @@ static enum si_sm_result start_next_msg(struct smi_info *smi_info) do_gettimeofday(&t); printk("**Start2: %d.%9.9d\n", t.tv_sec, t.tv_usec); #endif - err = notifier_call_chain(&xaction_notifier_list, 0, smi_info); + err = atomic_notifier_call_chain(&xaction_notifier_list, + 0, smi_info); if (err & NOTIFY_STOP_MASK) { rv = SI_SM_CALL_WITHOUT_DELAY; goto out; diff --git a/drivers/char/ipmi/ipmi_watchdog.c b/drivers/char/ipmi/ipmi_watchdog.c index 616539310d9..7ece9f3c8f7 100644 --- a/drivers/char/ipmi/ipmi_watchdog.c +++ b/drivers/char/ipmi/ipmi_watchdog.c @@ -1158,7 +1158,8 @@ static int __init ipmi_wdog_init(void) } register_reboot_notifier(&wdog_reboot_notifier); - notifier_chain_register(&panic_notifier_list, &wdog_panic_notifier); + atomic_notifier_chain_register(&panic_notifier_list, + &wdog_panic_notifier); printk(KERN_INFO PFX "driver initialized\n"); @@ -1176,7 +1177,8 @@ static __exit void ipmi_unregister_watchdog(void) release_nmi(&ipmi_nmi_handler); #endif - notifier_chain_unregister(&panic_notifier_list, &wdog_panic_notifier); + atomic_notifier_chain_unregister(&panic_notifier_list, + &wdog_panic_notifier); unregister_reboot_notifier(&wdog_reboot_notifier); if (! watchdog_user) diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index aed80e6aec6..9b6ae7dc8b8 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -52,9 +52,8 @@ static void handle_update(void *data); * changes to devices when the CPU clock speed changes. * The mutex locks both lists. */ -static struct notifier_block *cpufreq_policy_notifier_list; -static struct notifier_block *cpufreq_transition_notifier_list; -static DECLARE_RWSEM (cpufreq_notifier_rwsem); +static BLOCKING_NOTIFIER_HEAD(cpufreq_policy_notifier_list); +static BLOCKING_NOTIFIER_HEAD(cpufreq_transition_notifier_list); static LIST_HEAD(cpufreq_governor_list); @@ -247,8 +246,6 @@ void cpufreq_notify_transition(struct cpufreq_freqs *freqs, unsigned int state) dprintk("notification %u of frequency transition to %u kHz\n", state, freqs->new); - down_read(&cpufreq_notifier_rwsem); - policy = cpufreq_cpu_data[freqs->cpu]; switch (state) { @@ -266,20 +263,19 @@ void cpufreq_notify_transition(struct cpufreq_freqs *freqs, unsigned int state) freqs->old = policy->cur; } } - notifier_call_chain(&cpufreq_transition_notifier_list, - CPUFREQ_PRECHANGE, freqs); + blocking_notifier_call_chain(&cpufreq_transition_notifier_list, + CPUFREQ_PRECHANGE, freqs); adjust_jiffies(CPUFREQ_PRECHANGE, freqs); break; case CPUFREQ_POSTCHANGE: adjust_jiffies(CPUFREQ_POSTCHANGE, freqs); - notifier_call_chain(&cpufreq_transition_notifier_list, - CPUFREQ_POSTCHANGE, freqs); + blocking_notifier_call_chain(&cpufreq_transition_notifier_list, + CPUFREQ_POSTCHANGE, freqs); if (likely(policy) && likely(policy->cpu == freqs->cpu)) policy->cur = freqs->new; break; } - up_read(&cpufreq_notifier_rwsem); } EXPORT_SYMBOL_GPL(cpufreq_notify_transition); @@ -1007,7 +1003,7 @@ static int cpufreq_suspend(struct sys_device * sysdev, pm_message_t pmsg) freqs.old = cpu_policy->cur; freqs.new = cur_freq; - notifier_call_chain(&cpufreq_transition_notifier_list, + blocking_notifier_call_chain(&cpufreq_transition_notifier_list, CPUFREQ_SUSPENDCHANGE, &freqs); adjust_jiffies(CPUFREQ_SUSPENDCHANGE, &freqs); @@ -1088,7 +1084,8 @@ static int cpufreq_resume(struct sys_device * sysdev) freqs.old = cpu_policy->cur; freqs.new = cur_freq; - notifier_call_chain(&cpufreq_transition_notifier_list, + blocking_notifier_call_chain( + &cpufreq_transition_notifier_list, CPUFREQ_RESUMECHANGE, &freqs); adjust_jiffies(CPUFREQ_RESUMECHANGE, &freqs); @@ -1125,24 +1122,24 @@ static struct sysdev_driver cpufreq_sysdev_driver = { * changes in cpufreq policy. * * This function may sleep, and has the same return conditions as - * notifier_chain_register. + * blocking_notifier_chain_register. */ int cpufreq_register_notifier(struct notifier_block *nb, unsigned int list) { int ret; - down_write(&cpufreq_notifier_rwsem); switch (list) { case CPUFREQ_TRANSITION_NOTIFIER: - ret = notifier_chain_register(&cpufreq_transition_notifier_list, nb); + ret = blocking_notifier_chain_register( + &cpufreq_transition_notifier_list, nb); break; case CPUFREQ_POLICY_NOTIFIER: - ret = notifier_chain_register(&cpufreq_policy_notifier_list, nb); + ret = blocking_notifier_chain_register( + &cpufreq_policy_notifier_list, nb); break; default: ret = -EINVAL; } - up_write(&cpufreq_notifier_rwsem); return ret; } @@ -1157,24 +1154,24 @@ EXPORT_SYMBOL(cpufreq_register_notifier); * Remove a driver from the CPU frequency notifier list. * * This function may sleep, and has the same return conditions as - * notifier_chain_unregister. + * blocking_notifier_chain_unregister. */ int cpufreq_unregister_notifier(struct notifier_block *nb, unsigned int list) { int ret; - down_write(&cpufreq_notifier_rwsem); switch (list) { case CPUFREQ_TRANSITION_NOTIFIER: - ret = notifier_chain_unregister(&cpufreq_transition_notifier_list, nb); + ret = blocking_notifier_chain_unregister( + &cpufreq_transition_notifier_list, nb); break; case CPUFREQ_POLICY_NOTIFIER: - ret = notifier_chain_unregister(&cpufreq_policy_notifier_list, nb); + ret = blocking_notifier_chain_unregister( + &cpufreq_policy_notifier_list, nb); break; default: ret = -EINVAL; } - up_write(&cpufreq_notifier_rwsem); return ret; } @@ -1346,29 +1343,23 @@ static int __cpufreq_set_policy(struct cpufreq_policy *data, struct cpufreq_poli if (ret) goto error_out; - down_read(&cpufreq_notifier_rwsem); - /* adjust if necessary - all reasons */ - notifier_call_chain(&cpufreq_policy_notifier_list, CPUFREQ_ADJUST, - policy); + blocking_notifier_call_chain(&cpufreq_policy_notifier_list, + CPUFREQ_ADJUST, policy); /* adjust if necessary - hardware incompatibility*/ - notifier_call_chain(&cpufreq_policy_notifier_list, CPUFREQ_INCOMPATIBLE, - policy); + blocking_notifier_call_chain(&cpufreq_policy_notifier_list, + CPUFREQ_INCOMPATIBLE, policy); /* verify the cpu speed can be set within this limit, which might be different to the first one */ ret = cpufreq_driver->verify(policy); - if (ret) { - up_read(&cpufreq_notifier_rwsem); + if (ret) goto error_out; - } /* notification of the new policy */ - notifier_call_chain(&cpufreq_policy_notifier_list, CPUFREQ_NOTIFY, - policy); - - up_read(&cpufreq_notifier_rwsem); + blocking_notifier_call_chain(&cpufreq_policy_notifier_list, + CPUFREQ_NOTIFY, policy); data->min = policy->min; data->max = policy->max; diff --git a/drivers/firmware/dcdbas.c b/drivers/firmware/dcdbas.c index d6543fc4a92..339f405ff70 100644 --- a/drivers/firmware/dcdbas.c +++ b/drivers/firmware/dcdbas.c @@ -484,26 +484,15 @@ static void dcdbas_host_control(void) static int dcdbas_reboot_notify(struct notifier_block *nb, unsigned long code, void *unused) { - static unsigned int notify_cnt = 0; - switch (code) { case SYS_DOWN: case SYS_HALT: case SYS_POWER_OFF: if (host_control_on_shutdown) { /* firmware is going to perform host control action */ - if (++notify_cnt == 2) { - printk(KERN_WARNING - "Please wait for shutdown " - "action to complete...\n"); - dcdbas_host_control(); - } - /* - * register again and initiate the host control - * action on the second notification to allow - * everyone that registered to be notified - */ - register_reboot_notifier(nb); + printk(KERN_WARNING "Please wait for shutdown " + "action to complete...\n"); + dcdbas_host_control(); } break; } @@ -514,7 +503,7 @@ static int dcdbas_reboot_notify(struct notifier_block *nb, unsigned long code, static struct notifier_block dcdbas_reboot_nb = { .notifier_call = dcdbas_reboot_notify, .next = NULL, - .priority = 0 + .priority = INT_MIN }; static DCDBAS_BIN_ATTR_RW(smi_data); diff --git a/drivers/macintosh/adb.c b/drivers/macintosh/adb.c index d2ead1776c1..34fcabac5fd 100644 --- a/drivers/macintosh/adb.c +++ b/drivers/macintosh/adb.c @@ -80,7 +80,7 @@ static struct adb_driver *adb_driver_list[] = { static struct class *adb_dev_class; struct adb_driver *adb_controller; -struct notifier_block *adb_client_list = NULL; +BLOCKING_NOTIFIER_HEAD(adb_client_list); static int adb_got_sleep; static int adb_inited; static pid_t adb_probe_task_pid; @@ -354,7 +354,8 @@ adb_notify_sleep(struct pmu_sleep_notifier *self, int when) /* Stop autopoll */ if (adb_controller->autopoll) adb_controller->autopoll(0); - ret = notifier_call_chain(&adb_client_list, ADB_MSG_POWERDOWN, NULL); + ret = blocking_notifier_call_chain(&adb_client_list, + ADB_MSG_POWERDOWN, NULL); if (ret & NOTIFY_STOP_MASK) { up(&adb_probe_mutex); return PBOOK_SLEEP_REFUSE; @@ -391,7 +392,8 @@ do_adb_reset_bus(void) if (adb_controller->autopoll) adb_controller->autopoll(0); - nret = notifier_call_chain(&adb_client_list, ADB_MSG_PRE_RESET, NULL); + nret = blocking_notifier_call_chain(&adb_client_list, + ADB_MSG_PRE_RESET, NULL); if (nret & NOTIFY_STOP_MASK) { if (adb_controller->autopoll) adb_controller->autopoll(autopoll_devs); @@ -426,7 +428,8 @@ do_adb_reset_bus(void) } up(&adb_handler_sem); - nret = notifier_call_chain(&adb_client_list, ADB_MSG_POST_RESET, NULL); + nret = blocking_notifier_call_chain(&adb_client_list, + ADB_MSG_POST_RESET, NULL); if (nret & NOTIFY_STOP_MASK) return -EBUSY; diff --git a/drivers/macintosh/adbhid.c b/drivers/macintosh/adbhid.c index c0b46bceb5d..f5779a73184 100644 --- a/drivers/macintosh/adbhid.c +++ b/drivers/macintosh/adbhid.c @@ -1214,7 +1214,8 @@ static int __init adbhid_init(void) adbhid_probe(); - notifier_chain_register(&adb_client_list, &adbhid_adb_notifier); + blocking_notifier_chain_register(&adb_client_list, + &adbhid_adb_notifier); return 0; } diff --git a/drivers/macintosh/via-pmu.c b/drivers/macintosh/via-pmu.c index 4f5f3abc9cb..0b5ff553e39 100644 --- a/drivers/macintosh/via-pmu.c +++ b/drivers/macintosh/via-pmu.c @@ -187,7 +187,7 @@ extern int disable_kernel_backlight; int __fake_sleep; int asleep; -struct notifier_block *sleep_notifier_list; +BLOCKING_NOTIFIER_HEAD(sleep_notifier_list); #ifdef CONFIG_ADB static int adb_dev_map = 0; diff --git a/drivers/macintosh/via-pmu68k.c b/drivers/macintosh/via-pmu68k.c index f08e52f2107..35b70323e7e 100644 --- a/drivers/macintosh/via-pmu68k.c +++ b/drivers/macintosh/via-pmu68k.c @@ -102,7 +102,7 @@ static int pmu_kind = PMU_UNKNOWN; static int pmu_fully_inited = 0; int asleep; -struct notifier_block *sleep_notifier_list; +BLOCKING_NOTIFIER_HEAD(sleep_notifier_list); static int pmu_probe(void); static int pmu_init(void); @@ -913,7 +913,8 @@ int powerbook_sleep(void) struct adb_request sleep_req; /* Notify device drivers */ - ret = notifier_call_chain(&sleep_notifier_list, PBOOK_SLEEP, NULL); + ret = blocking_notifier_call_chain(&sleep_notifier_list, + PBOOK_SLEEP, NULL); if (ret & NOTIFY_STOP_MASK) return -EBUSY; @@ -984,7 +985,7 @@ int powerbook_sleep(void) enable_irq(i); /* Notify drivers */ - notifier_call_chain(&sleep_notifier_list, PBOOK_WAKE, NULL); + blocking_notifier_call_chain(&sleep_notifier_list, PBOOK_WAKE, NULL); /* reenable ADB autopoll */ pmu_adb_autopoll(adb_dev_map); diff --git a/drivers/macintosh/windfarm_core.c b/drivers/macintosh/windfarm_core.c index 6c0ba04bc57..ab3faa702d5 100644 --- a/drivers/macintosh/windfarm_core.c +++ b/drivers/macintosh/windfarm_core.c @@ -52,7 +52,7 @@ static LIST_HEAD(wf_controls); static LIST_HEAD(wf_sensors); static DEFINE_MUTEX(wf_lock); -static struct notifier_block *wf_client_list; +static BLOCKING_NOTIFIER_HEAD(wf_client_list); static int wf_client_count; static unsigned int wf_overtemp; static unsigned int wf_overtemp_counter; @@ -68,7 +68,7 @@ static struct platform_device wf_platform_device = { static inline void wf_notify(int event, void *param) { - notifier_call_chain(&wf_client_list, event, param); + blocking_notifier_call_chain(&wf_client_list, event, param); } int wf_critical_overtemp(void) @@ -398,7 +398,7 @@ int wf_register_client(struct notifier_block *nb) struct wf_sensor *sr; mutex_lock(&wf_lock); - rc = notifier_chain_register(&wf_client_list, nb); + rc = blocking_notifier_chain_register(&wf_client_list, nb); if (rc != 0) goto bail; wf_client_count++; @@ -417,7 +417,7 @@ EXPORT_SYMBOL_GPL(wf_register_client); int wf_unregister_client(struct notifier_block *nb) { mutex_lock(&wf_lock); - notifier_chain_unregister(&wf_client_list, nb); + blocking_notifier_chain_unregister(&wf_client_list, nb); wf_client_count++; if (wf_client_count == 0) wf_stop_thread(); diff --git a/drivers/misc/ibmasm/heartbeat.c b/drivers/misc/ibmasm/heartbeat.c index f295401fac2..7fd7a43e38d 100644 --- a/drivers/misc/ibmasm/heartbeat.c +++ b/drivers/misc/ibmasm/heartbeat.c @@ -52,12 +52,13 @@ static struct notifier_block panic_notifier = { panic_happened, NULL, 1 }; void ibmasm_register_panic_notifier(void) { - notifier_chain_register(&panic_notifier_list, &panic_notifier); + atomic_notifier_chain_register(&panic_notifier_list, &panic_notifier); } void ibmasm_unregister_panic_notifier(void) { - notifier_chain_unregister(&panic_notifier_list, &panic_notifier); + atomic_notifier_chain_unregister(&panic_notifier_list, + &panic_notifier); } diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 2d0ac169a86..f13a539dc16 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -3159,7 +3159,7 @@ static int bond_slave_netdev_event(unsigned long event, struct net_device *slave * bond_netdev_event: handle netdev notifier chain events. * * This function receives events for the netdev chain. The caller (an - * ioctl handler calling notifier_call_chain) holds the necessary + * ioctl handler calling blocking_notifier_call_chain) holds the necessary * locks for us to safely manipulate the slave devices (RTNL lock, * dev_probe_lock). */ diff --git a/drivers/parisc/led.c b/drivers/parisc/led.c index 3627a2d7f79..298f2ddb2c1 100644 --- a/drivers/parisc/led.c +++ b/drivers/parisc/led.c @@ -499,11 +499,16 @@ static int led_halt(struct notifier_block *, unsigned long, void *); static struct notifier_block led_notifier = { .notifier_call = led_halt, }; +static int notifier_disabled = 0; static int led_halt(struct notifier_block *nb, unsigned long event, void *buf) { char *txt; - + + if (notifier_disabled) + return NOTIFY_OK; + + notifier_disabled = 1; switch (event) { case SYS_RESTART: txt = "SYSTEM RESTART"; break; @@ -527,7 +532,6 @@ static int led_halt(struct notifier_block *nb, unsigned long event, void *buf) if (led_func_ptr) led_func_ptr(0xff); /* turn all LEDs ON */ - unregister_reboot_notifier(&led_notifier); return NOTIFY_OK; } @@ -758,6 +762,12 @@ not_found: return 1; } +static void __exit led_exit(void) +{ + unregister_reboot_notifier(&led_notifier); + return; +} + #ifdef CONFIG_PROC_FS module_init(led_create_procfs) #endif diff --git a/drivers/parisc/power.c b/drivers/parisc/power.c index 54b2b7f20b9..0bcab83b408 100644 --- a/drivers/parisc/power.c +++ b/drivers/parisc/power.c @@ -251,7 +251,8 @@ static int __init power_init(void) } /* Register a call for panic conditions. */ - notifier_chain_register(&panic_notifier_list, &parisc_panic_block); + atomic_notifier_chain_register(&panic_notifier_list, + &parisc_panic_block); tasklet_enable(&power_tasklet); @@ -264,7 +265,8 @@ static void __exit power_exit(void) return; tasklet_disable(&power_tasklet); - notifier_chain_unregister(&panic_notifier_list, &parisc_panic_block); + atomic_notifier_chain_unregister(&panic_notifier_list, + &parisc_panic_block); power_tasklet.func = NULL; pdc_soft_power_button(0); } diff --git a/drivers/scsi/gdth.c b/drivers/scsi/gdth.c index 62e3cda859a..7f7013e80a8 100644 --- a/drivers/scsi/gdth.c +++ b/drivers/scsi/gdth.c @@ -671,7 +671,7 @@ static struct file_operations gdth_fops = { static struct notifier_block gdth_notifier = { gdth_halt, NULL, 0 }; - +static int notifier_disabled = 0; static void gdth_delay(int milliseconds) { @@ -4595,13 +4595,13 @@ static int __init gdth_detect(struct scsi_host_template *shtp) add_timer(&gdth_timer); #endif major = register_chrdev(0,"gdth",&gdth_fops); + notifier_disabled = 0; register_reboot_notifier(&gdth_notifier); } gdth_polling = FALSE; return gdth_ctr_vcount; } - static int gdth_release(struct Scsi_Host *shp) { int hanum; @@ -5632,10 +5632,14 @@ static int gdth_halt(struct notifier_block *nb, ulong event, void *buf) char cmnd[MAX_COMMAND_SIZE]; #endif + if (notifier_disabled) + return NOTIFY_OK; + TRACE2(("gdth_halt() event %d\n",(int)event)); if (event != SYS_RESTART && event != SYS_HALT && event != SYS_POWER_OFF) return NOTIFY_DONE; + notifier_disabled = 1; printk("GDT-HA: Flushing all host drives .. "); for (hanum = 0; hanum < gdth_ctr_count; ++hanum) { gdth_flush(hanum); @@ -5679,7 +5683,6 @@ static int gdth_halt(struct notifier_block *nb, ulong event, void *buf) #ifdef GDTH_STATISTICS del_timer(&gdth_timer); #endif - unregister_reboot_notifier(&gdth_notifier); return NOTIFY_OK; } diff --git a/drivers/usb/core/notify.c b/drivers/usb/core/notify.c index 4b55285de9a..fe0ed54fa0a 100644 --- a/drivers/usb/core/notify.c +++ b/drivers/usb/core/notify.c @@ -16,57 +16,7 @@ #include #include "usb.h" - -static struct notifier_block *usb_notifier_list; -static DEFINE_MUTEX(usb_notifier_lock); - -static void usb_notifier_chain_register(struct notifier_block **list, - struct notifier_block *n) -{ - mutex_lock(&usb_notifier_lock); - while (*list) { - if (n->priority > (*list)->priority) - break; - list = &((*list)->next); - } - n->next = *list; - *list = n; - mutex_unlock(&usb_notifier_lock); -} - -static void usb_notifier_chain_unregister(struct notifier_block **nl, - struct notifier_block *n) -{ - mutex_lock(&usb_notifier_lock); - while ((*nl)!=NULL) { - if ((*nl)==n) { - *nl = n->next; - goto exit; - } - nl=&((*nl)->next); - } -exit: - mutex_unlock(&usb_notifier_lock); -} - -static int usb_notifier_call_chain(struct notifier_block **n, - unsigned long val, void *v) -{ - int ret=NOTIFY_DONE; - struct notifier_block *nb = *n; - - mutex_lock(&usb_notifier_lock); - while (nb) { - ret = nb->notifier_call(nb,val,v); - if (ret&NOTIFY_STOP_MASK) { - goto exit; - } - nb = nb->next; - } -exit: - mutex_unlock(&usb_notifier_lock); - return ret; -} +static BLOCKING_NOTIFIER_HEAD(usb_notifier_list); /** * usb_register_notify - register a notifier callback whenever a usb change happens @@ -76,7 +26,7 @@ exit: */ void usb_register_notify(struct notifier_block *nb) { - usb_notifier_chain_register(&usb_notifier_list, nb); + blocking_notifier_chain_register(&usb_notifier_list, nb); } EXPORT_SYMBOL_GPL(usb_register_notify); @@ -89,27 +39,28 @@ EXPORT_SYMBOL_GPL(usb_register_notify); */ void usb_unregister_notify(struct notifier_block *nb) { - usb_notifier_chain_unregister(&usb_notifier_list, nb); + blocking_notifier_chain_unregister(&usb_notifier_list, nb); } EXPORT_SYMBOL_GPL(usb_unregister_notify); void usb_notify_add_device(struct usb_device *udev) { - usb_notifier_call_chain(&usb_notifier_list, USB_DEVICE_ADD, udev); + blocking_notifier_call_chain(&usb_notifier_list, USB_DEVICE_ADD, udev); } void usb_notify_remove_device(struct usb_device *udev) { - usb_notifier_call_chain(&usb_notifier_list, USB_DEVICE_REMOVE, udev); + blocking_notifier_call_chain(&usb_notifier_list, + USB_DEVICE_REMOVE, udev); } void usb_notify_add_bus(struct usb_bus *ubus) { - usb_notifier_call_chain(&usb_notifier_list, USB_BUS_ADD, ubus); + blocking_notifier_call_chain(&usb_notifier_list, USB_BUS_ADD, ubus); } void usb_notify_remove_bus(struct usb_bus *ubus) { - usb_notifier_call_chain(&usb_notifier_list, USB_BUS_REMOVE, ubus); + blocking_notifier_call_chain(&usb_notifier_list, USB_BUS_REMOVE, ubus); } diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c index 07d882b1439..b1a8dca7643 100644 --- a/drivers/video/fbmem.c +++ b/drivers/video/fbmem.c @@ -55,7 +55,7 @@ #define FBPIXMAPSIZE (1024 * 8) -static struct notifier_block *fb_notifier_list; +static BLOCKING_NOTIFIER_HEAD(fb_notifier_list); struct fb_info *registered_fb[FB_MAX]; int num_registered_fb; @@ -784,7 +784,7 @@ fb_set_var(struct fb_info *info, struct fb_var_screeninfo *var) event.info = info; event.data = &mode1; - ret = notifier_call_chain(&fb_notifier_list, + ret = blocking_notifier_call_chain(&fb_notifier_list, FB_EVENT_MODE_DELETE, &event); } @@ -830,8 +830,8 @@ fb_set_var(struct fb_info *info, struct fb_var_screeninfo *var) info->flags &= ~FBINFO_MISC_USEREVENT; event.info = info; - notifier_call_chain(&fb_notifier_list, evnt, - &event); + blocking_notifier_call_chain(&fb_notifier_list, + evnt, &event); } } } @@ -854,7 +854,8 @@ fb_blank(struct fb_info *info, int blank) event.info = info; event.data = ␣ - notifier_call_chain(&fb_notifier_list, FB_EVENT_BLANK, &event); + blocking_notifier_call_chain(&fb_notifier_list, + FB_EVENT_BLANK, &event); } return ret; @@ -925,7 +926,7 @@ fb_ioctl(struct inode *inode, struct file *file, unsigned int cmd, con2fb.framebuffer = -1; event.info = info; event.data = &con2fb; - notifier_call_chain(&fb_notifier_list, + blocking_notifier_call_chain(&fb_notifier_list, FB_EVENT_GET_CONSOLE_MAP, &event); return copy_to_user(argp, &con2fb, sizeof(con2fb)) ? -EFAULT : 0; @@ -944,7 +945,7 @@ fb_ioctl(struct inode *inode, struct file *file, unsigned int cmd, return -EINVAL; event.info = info; event.data = &con2fb; - return notifier_call_chain(&fb_notifier_list, + return blocking_notifier_call_chain(&fb_notifier_list, FB_EVENT_SET_CONSOLE_MAP, &event); case FBIOBLANK: @@ -1324,7 +1325,7 @@ register_framebuffer(struct fb_info *fb_info) devfs_mk_cdev(MKDEV(FB_MAJOR, i), S_IFCHR | S_IRUGO | S_IWUGO, "fb/%d", i); event.info = fb_info; - notifier_call_chain(&fb_notifier_list, + blocking_notifier_call_chain(&fb_notifier_list, FB_EVENT_FB_REGISTERED, &event); return 0; } @@ -1366,7 +1367,7 @@ unregister_framebuffer(struct fb_info *fb_info) */ int fb_register_client(struct notifier_block *nb) { - return notifier_chain_register(&fb_notifier_list, nb); + return blocking_notifier_chain_register(&fb_notifier_list, nb); } /** @@ -1375,7 +1376,7 @@ int fb_register_client(struct notifier_block *nb) */ int fb_unregister_client(struct notifier_block *nb) { - return notifier_chain_unregister(&fb_notifier_list, nb); + return blocking_notifier_chain_unregister(&fb_notifier_list, nb); } /** @@ -1393,11 +1394,13 @@ void fb_set_suspend(struct fb_info *info, int state) event.info = info; if (state) { - notifier_call_chain(&fb_notifier_list, FB_EVENT_SUSPEND, &event); + blocking_notifier_call_chain(&fb_notifier_list, + FB_EVENT_SUSPEND, &event); info->state = FBINFO_STATE_SUSPENDED; } else { info->state = FBINFO_STATE_RUNNING; - notifier_call_chain(&fb_notifier_list, FB_EVENT_RESUME, &event); + blocking_notifier_call_chain(&fb_notifier_list, + FB_EVENT_RESUME, &event); } } @@ -1469,7 +1472,7 @@ int fb_new_modelist(struct fb_info *info) if (!list_empty(&info->modelist)) { event.info = info; - err = notifier_call_chain(&fb_notifier_list, + err = blocking_notifier_call_chain(&fb_notifier_list, FB_EVENT_NEW_MODELIST, &event); } @@ -1495,7 +1498,7 @@ int fb_con_duit(struct fb_info *info, int event, void *data) evnt.info = info; evnt.data = data; - return notifier_call_chain(&fb_notifier_list, event, &evnt); + return blocking_notifier_call_chain(&fb_notifier_list, event, &evnt); } EXPORT_SYMBOL(fb_con_duit); diff --git a/include/asm-i386/kdebug.h b/include/asm-i386/kdebug.h index 316138e8991..96d0828ce09 100644 --- a/include/asm-i386/kdebug.h +++ b/include/asm-i386/kdebug.h @@ -17,11 +17,9 @@ struct die_args { int signr; }; -/* Note - you should never unregister because that can race with NMIs. - If you really want to do it first unregister - then synchronize_sched - then free. - */ -int register_die_notifier(struct notifier_block *nb); -extern struct notifier_block *i386die_chain; +extern int register_die_notifier(struct notifier_block *); +extern int unregister_die_notifier(struct notifier_block *); +extern struct atomic_notifier_head i386die_chain; /* Grossly misnamed. */ @@ -51,7 +49,7 @@ static inline int notify_die(enum die_val val, const char *str, .trapnr = trap, .signr = sig }; - return notifier_call_chain(&i386die_chain, val, &args); + return atomic_notifier_call_chain(&i386die_chain, val, &args); } #endif diff --git a/include/asm-ia64/kdebug.h b/include/asm-ia64/kdebug.h index 8b01a083dde..218c458ab60 100644 --- a/include/asm-ia64/kdebug.h +++ b/include/asm-ia64/kdebug.h @@ -40,7 +40,7 @@ struct die_args { extern int register_die_notifier(struct notifier_block *); extern int unregister_die_notifier(struct notifier_block *); -extern struct notifier_block *ia64die_chain; +extern struct atomic_notifier_head ia64die_chain; enum die_val { DIE_BREAK = 1, @@ -81,7 +81,7 @@ static inline int notify_die(enum die_val val, char *str, struct pt_regs *regs, .signr = sig }; - return notifier_call_chain(&ia64die_chain, val, &args); + return atomic_notifier_call_chain(&ia64die_chain, val, &args); } #endif diff --git a/include/asm-powerpc/kdebug.h b/include/asm-powerpc/kdebug.h index 7c16265568e..c01786ab5fa 100644 --- a/include/asm-powerpc/kdebug.h +++ b/include/asm-powerpc/kdebug.h @@ -16,13 +16,9 @@ struct die_args { int signr; }; -/* - Note - you should never unregister because that can race with NMIs. - If you really want to do it first unregister - then synchronize_sched - - then free. - */ -int register_die_notifier(struct notifier_block *nb); -extern struct notifier_block *powerpc_die_chain; +extern int register_die_notifier(struct notifier_block *); +extern int unregister_die_notifier(struct notifier_block *); +extern struct atomic_notifier_head powerpc_die_chain; /* Grossly misnamed. */ enum die_val { @@ -37,7 +33,7 @@ enum die_val { static inline int notify_die(enum die_val val,char *str,struct pt_regs *regs,long err,int trap, int sig) { struct die_args args = { .regs=regs, .str=str, .err=err, .trapnr=trap,.signr=sig }; - return notifier_call_chain(&powerpc_die_chain, val, &args); + return atomic_notifier_call_chain(&powerpc_die_chain, val, &args); } #endif /* __KERNEL__ */ diff --git a/include/asm-sparc64/kdebug.h b/include/asm-sparc64/kdebug.h index 6321f5a0198..4040d127ac3 100644 --- a/include/asm-sparc64/kdebug.h +++ b/include/asm-sparc64/kdebug.h @@ -15,12 +15,9 @@ struct die_args { int signr; }; -/* Note - you should never unregister because that can race with NMIs. - * If you really want to do it first unregister - then synchronize_sched - * - then free. - */ -int register_die_notifier(struct notifier_block *nb); -extern struct notifier_block *sparc64die_chain; +extern int register_die_notifier(struct notifier_block *); +extern int unregister_die_notifier(struct notifier_block *); +extern struct atomic_notifier_head sparc64die_chain; extern void bad_trap(struct pt_regs *, long); @@ -46,7 +43,7 @@ static inline int notify_die(enum die_val val,char *str, struct pt_regs *regs, .trapnr = trap, .signr = sig }; - return notifier_call_chain(&sparc64die_chain, val, &args); + return atomic_notifier_call_chain(&sparc64die_chain, val, &args); } #endif diff --git a/include/asm-x86_64/kdebug.h b/include/asm-x86_64/kdebug.h index b9ed4c0c878..cf795631d9b 100644 --- a/include/asm-x86_64/kdebug.h +++ b/include/asm-x86_64/kdebug.h @@ -5,21 +5,20 @@ struct pt_regs; -struct die_args { +struct die_args { struct pt_regs *regs; const char *str; - long err; + long err; int trapnr; int signr; -}; +}; + +extern int register_die_notifier(struct notifier_block *); +extern int unregister_die_notifier(struct notifier_block *); +extern struct atomic_notifier_head die_chain; -/* Note - you should never unregister because that can race with NMIs. - If you really want to do it first unregister - then synchronize_sched - then free. - */ -int register_die_notifier(struct notifier_block *nb); -extern struct notifier_block *die_chain; /* Grossly misnamed. */ -enum die_val { +enum die_val { DIE_OOPS = 1, DIE_INT3, DIE_DEBUG, @@ -33,8 +32,8 @@ enum die_val { DIE_CALL, DIE_NMI_IPI, DIE_PAGE_FAULT, -}; - +}; + static inline int notify_die(enum die_val val, const char *str, struct pt_regs *regs, long err, int trap, int sig) { @@ -45,7 +44,7 @@ static inline int notify_die(enum die_val val, const char *str, .trapnr = trap, .signr = sig }; - return notifier_call_chain(&die_chain, val, &args); + return atomic_notifier_call_chain(&die_chain, val, &args); } extern int printk_address(unsigned long address); diff --git a/include/linux/adb.h b/include/linux/adb.h index e9fdc63483c..b7305b17827 100644 --- a/include/linux/adb.h +++ b/include/linux/adb.h @@ -85,7 +85,7 @@ enum adb_message { ADB_MSG_POST_RESET /* Called after resetting the bus (re-do init & register) */ }; extern struct adb_driver *adb_controller; -extern struct notifier_block *adb_client_list; +extern struct blocking_notifier_head adb_client_list; int adb_request(struct adb_request *req, void (*done)(struct adb_request *), int flags, int nbytes, ...); diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 03d6cfaa5b8..a3720f973ea 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -87,7 +87,7 @@ extern int cond_resched(void); (__x < 0) ? -__x : __x; \ }) -extern struct notifier_block *panic_notifier_list; +extern struct atomic_notifier_head panic_notifier_list; extern long (*panic_blink)(long time); NORET_TYPE void panic(const char * fmt, ...) __attribute__ ((NORET_AND format (printf, 1, 2))); diff --git a/include/linux/memory.h b/include/linux/memory.h index e251dc43d0f..8f04143ca36 100644 --- a/include/linux/memory.h +++ b/include/linux/memory.h @@ -77,7 +77,6 @@ extern int remove_memory_block(unsigned long, struct mem_section *, int); #define CONFIG_MEM_BLOCK_SIZE (PAGES_PER_SECTION< +#include +#include -struct notifier_block -{ - int (*notifier_call)(struct notifier_block *self, unsigned long, void *); +/* + * Notifier chains are of three types: + * + * Atomic notifier chains: Chain callbacks run in interrupt/atomic + * context. Callouts are not allowed to block. + * Blocking notifier chains: Chain callbacks run in process context. + * Callouts are allowed to block. + * Raw notifier chains: There are no restrictions on callbacks, + * registration, or unregistration. All locking and protection + * must be provided by the caller. + * + * atomic_notifier_chain_register() may be called from an atomic context, + * but blocking_notifier_chain_register() must be called from a process + * context. Ditto for the corresponding _unregister() routines. + * + * atomic_notifier_chain_unregister() and blocking_notifier_chain_unregister() + * _must not_ be called from within the call chain. + */ + +struct notifier_block { + int (*notifier_call)(struct notifier_block *, unsigned long, void *); struct notifier_block *next; int priority; }; +struct atomic_notifier_head { + spinlock_t lock; + struct notifier_block *head; +}; + +struct blocking_notifier_head { + struct rw_semaphore rwsem; + struct notifier_block *head; +}; + +struct raw_notifier_head { + struct notifier_block *head; +}; + +#define ATOMIC_INIT_NOTIFIER_HEAD(name) do { \ + spin_lock_init(&(name)->lock); \ + (name)->head = NULL; \ + } while (0) +#define BLOCKING_INIT_NOTIFIER_HEAD(name) do { \ + init_rwsem(&(name)->rwsem); \ + (name)->head = NULL; \ + } while (0) +#define RAW_INIT_NOTIFIER_HEAD(name) do { \ + (name)->head = NULL; \ + } while (0) + +#define ATOMIC_NOTIFIER_INIT(name) { \ + .lock = SPIN_LOCK_UNLOCKED, \ + .head = NULL } +#define BLOCKING_NOTIFIER_INIT(name) { \ + .rwsem = __RWSEM_INITIALIZER((name).rwsem), \ + .head = NULL } +#define RAW_NOTIFIER_INIT(name) { \ + .head = NULL } + +#define ATOMIC_NOTIFIER_HEAD(name) \ + struct atomic_notifier_head name = \ + ATOMIC_NOTIFIER_INIT(name) +#define BLOCKING_NOTIFIER_HEAD(name) \ + struct blocking_notifier_head name = \ + BLOCKING_NOTIFIER_INIT(name) +#define RAW_NOTIFIER_HEAD(name) \ + struct raw_notifier_head name = \ + RAW_NOTIFIER_INIT(name) #ifdef __KERNEL__ -extern int notifier_chain_register(struct notifier_block **list, struct notifier_block *n); -extern int notifier_chain_unregister(struct notifier_block **nl, struct notifier_block *n); -extern int notifier_call_chain(struct notifier_block **n, unsigned long val, void *v); +extern int atomic_notifier_chain_register(struct atomic_notifier_head *, + struct notifier_block *); +extern int blocking_notifier_chain_register(struct blocking_notifier_head *, + struct notifier_block *); +extern int raw_notifier_chain_register(struct raw_notifier_head *, + struct notifier_block *); + +extern int atomic_notifier_chain_unregister(struct atomic_notifier_head *, + struct notifier_block *); +extern int blocking_notifier_chain_unregister(struct blocking_notifier_head *, + struct notifier_block *); +extern int raw_notifier_chain_unregister(struct raw_notifier_head *, + struct notifier_block *); + +extern int atomic_notifier_call_chain(struct atomic_notifier_head *, + unsigned long val, void *v); +extern int blocking_notifier_call_chain(struct blocking_notifier_head *, + unsigned long val, void *v); +extern int raw_notifier_call_chain(struct raw_notifier_head *, + unsigned long val, void *v); #define NOTIFY_DONE 0x0000 /* Don't care */ #define NOTIFY_OK 0x0001 /* Suits me */ #define NOTIFY_STOP_MASK 0x8000 /* Don't call further */ -#define NOTIFY_BAD (NOTIFY_STOP_MASK|0x0002) /* Bad/Veto action */ +#define NOTIFY_BAD (NOTIFY_STOP_MASK|0x0002) + /* Bad/Veto action */ /* * Clean way to return from the notifier and stop further calls. */ diff --git a/include/net/netfilter/nf_conntrack.h b/include/net/netfilter/nf_conntrack.h index b6f0905a4ee..916013ca4a5 100644 --- a/include/net/netfilter/nf_conntrack.h +++ b/include/net/netfilter/nf_conntrack.h @@ -300,29 +300,30 @@ DECLARE_PER_CPU(struct nf_conntrack_ecache, nf_conntrack_ecache); #define CONNTRACK_ECACHE(x) (__get_cpu_var(nf_conntrack_ecache).x) -extern struct notifier_block *nf_conntrack_chain; -extern struct notifier_block *nf_conntrack_expect_chain; +extern struct atomic_notifier_head nf_conntrack_chain; +extern struct atomic_notifier_head nf_conntrack_expect_chain; static inline int nf_conntrack_register_notifier(struct notifier_block *nb) { - return notifier_chain_register(&nf_conntrack_chain, nb); + return atomic_notifier_chain_register(&nf_conntrack_chain, nb); } static inline int nf_conntrack_unregister_notifier(struct notifier_block *nb) { - return notifier_chain_unregister(&nf_conntrack_chain, nb); + return atomic_notifier_chain_unregister(&nf_conntrack_chain, nb); } static inline int nf_conntrack_expect_register_notifier(struct notifier_block *nb) { - return notifier_chain_register(&nf_conntrack_expect_chain, nb); + return atomic_notifier_chain_register(&nf_conntrack_expect_chain, nb); } static inline int nf_conntrack_expect_unregister_notifier(struct notifier_block *nb) { - return notifier_chain_unregister(&nf_conntrack_expect_chain, nb); + return atomic_notifier_chain_unregister(&nf_conntrack_expect_chain, + nb); } extern void nf_ct_deliver_cached_events(const struct nf_conn *ct); @@ -347,14 +348,14 @@ static inline void nf_conntrack_event(enum ip_conntrack_events event, struct nf_conn *ct) { if (nf_ct_is_confirmed(ct) && !nf_ct_is_dying(ct)) - notifier_call_chain(&nf_conntrack_chain, event, ct); + atomic_notifier_call_chain(&nf_conntrack_chain, event, ct); } static inline void nf_conntrack_expect_event(enum ip_conntrack_expect_events event, struct nf_conntrack_expect *exp) { - notifier_call_chain(&nf_conntrack_expect_chain, event, exp); + atomic_notifier_call_chain(&nf_conntrack_expect_chain, event, exp); } #else /* CONFIG_NF_CONNTRACK_EVENTS */ static inline void nf_conntrack_event_cache(enum ip_conntrack_events event, diff --git a/kernel/cpu.c b/kernel/cpu.c index 8be22bd8093..fe2b8d0bfe4 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -18,7 +18,7 @@ /* This protects CPUs going up and down... */ static DECLARE_MUTEX(cpucontrol); -static struct notifier_block *cpu_chain; +static BLOCKING_NOTIFIER_HEAD(cpu_chain); #ifdef CONFIG_HOTPLUG_CPU static struct task_struct *lock_cpu_hotplug_owner; @@ -71,21 +71,13 @@ EXPORT_SYMBOL_GPL(lock_cpu_hotplug_interruptible); /* Need to know about CPUs going up/down? */ int register_cpu_notifier(struct notifier_block *nb) { - int ret; - - if ((ret = lock_cpu_hotplug_interruptible()) != 0) - return ret; - ret = notifier_chain_register(&cpu_chain, nb); - unlock_cpu_hotplug(); - return ret; + return blocking_notifier_chain_register(&cpu_chain, nb); } EXPORT_SYMBOL(register_cpu_notifier); void unregister_cpu_notifier(struct notifier_block *nb) { - lock_cpu_hotplug(); - notifier_chain_unregister(&cpu_chain, nb); - unlock_cpu_hotplug(); + blocking_notifier_chain_unregister(&cpu_chain, nb); } EXPORT_SYMBOL(unregister_cpu_notifier); @@ -141,7 +133,7 @@ int cpu_down(unsigned int cpu) goto out; } - err = notifier_call_chain(&cpu_chain, CPU_DOWN_PREPARE, + err = blocking_notifier_call_chain(&cpu_chain, CPU_DOWN_PREPARE, (void *)(long)cpu); if (err == NOTIFY_BAD) { printk("%s: attempt to take down CPU %u failed\n", @@ -159,7 +151,7 @@ int cpu_down(unsigned int cpu) p = __stop_machine_run(take_cpu_down, NULL, cpu); if (IS_ERR(p)) { /* CPU didn't die: tell everyone. Can't complain. */ - if (notifier_call_chain(&cpu_chain, CPU_DOWN_FAILED, + if (blocking_notifier_call_chain(&cpu_chain, CPU_DOWN_FAILED, (void *)(long)cpu) == NOTIFY_BAD) BUG(); @@ -182,8 +174,8 @@ int cpu_down(unsigned int cpu) put_cpu(); /* CPU is completely dead: tell everyone. Too late to complain. */ - if (notifier_call_chain(&cpu_chain, CPU_DEAD, (void *)(long)cpu) - == NOTIFY_BAD) + if (blocking_notifier_call_chain(&cpu_chain, CPU_DEAD, + (void *)(long)cpu) == NOTIFY_BAD) BUG(); check_for_tasks(cpu); @@ -211,7 +203,7 @@ int __devinit cpu_up(unsigned int cpu) goto out; } - ret = notifier_call_chain(&cpu_chain, CPU_UP_PREPARE, hcpu); + ret = blocking_notifier_call_chain(&cpu_chain, CPU_UP_PREPARE, hcpu); if (ret == NOTIFY_BAD) { printk("%s: attempt to bring up CPU %u failed\n", __FUNCTION__, cpu); @@ -226,11 +218,12 @@ int __devinit cpu_up(unsigned int cpu) BUG_ON(!cpu_online(cpu)); /* Now call notifier in preparation. */ - notifier_call_chain(&cpu_chain, CPU_ONLINE, hcpu); + blocking_notifier_call_chain(&cpu_chain, CPU_ONLINE, hcpu); out_notify: if (ret != 0) - notifier_call_chain(&cpu_chain, CPU_UP_CANCELED, hcpu); + blocking_notifier_call_chain(&cpu_chain, + CPU_UP_CANCELED, hcpu); out: unlock_cpu_hotplug(); return ret; diff --git a/kernel/module.c b/kernel/module.c index ddfe45ac2fd..4fafd58038a 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -64,26 +64,17 @@ static DEFINE_SPINLOCK(modlist_lock); static DEFINE_MUTEX(module_mutex); static LIST_HEAD(modules); -static DEFINE_MUTEX(notify_mutex); -static struct notifier_block * module_notify_list; +static BLOCKING_NOTIFIER_HEAD(module_notify_list); int register_module_notifier(struct notifier_block * nb) { - int err; - mutex_lock(¬ify_mutex); - err = notifier_chain_register(&module_notify_list, nb); - mutex_unlock(¬ify_mutex); - return err; + return blocking_notifier_chain_register(&module_notify_list, nb); } EXPORT_SYMBOL(register_module_notifier); int unregister_module_notifier(struct notifier_block * nb) { - int err; - mutex_lock(¬ify_mutex); - err = notifier_chain_unregister(&module_notify_list, nb); - mutex_unlock(¬ify_mutex); - return err; + return blocking_notifier_chain_unregister(&module_notify_list, nb); } EXPORT_SYMBOL(unregister_module_notifier); @@ -1816,9 +1807,8 @@ sys_init_module(void __user *umod, /* Drop lock so they can recurse */ mutex_unlock(&module_mutex); - mutex_lock(¬ify_mutex); - notifier_call_chain(&module_notify_list, MODULE_STATE_COMING, mod); - mutex_unlock(¬ify_mutex); + blocking_notifier_call_chain(&module_notify_list, + MODULE_STATE_COMING, mod); /* Start the module */ if (mod->init != NULL) diff --git a/kernel/panic.c b/kernel/panic.c index acd95adddb9..f895c7c01d5 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -29,7 +29,7 @@ static DEFINE_SPINLOCK(pause_on_oops_lock); int panic_timeout; EXPORT_SYMBOL(panic_timeout); -struct notifier_block *panic_notifier_list; +ATOMIC_NOTIFIER_HEAD(panic_notifier_list); EXPORT_SYMBOL(panic_notifier_list); @@ -97,7 +97,7 @@ NORET_TYPE void panic(const char * fmt, ...) smp_send_stop(); #endif - notifier_call_chain(&panic_notifier_list, 0, buf); + atomic_notifier_call_chain(&panic_notifier_list, 0, buf); if (!panic_blink) panic_blink = no_blink; diff --git a/kernel/profile.c b/kernel/profile.c index ad81f799a9b..5a730fdb1a2 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -87,72 +87,52 @@ void __init profile_init(void) #ifdef CONFIG_PROFILING -static DECLARE_RWSEM(profile_rwsem); -static DEFINE_RWLOCK(handoff_lock); -static struct notifier_block * task_exit_notifier; -static struct notifier_block * task_free_notifier; -static struct notifier_block * munmap_notifier; +static BLOCKING_NOTIFIER_HEAD(task_exit_notifier); +static ATOMIC_NOTIFIER_HEAD(task_free_notifier); +static BLOCKING_NOTIFIER_HEAD(munmap_notifier); void profile_task_exit(struct task_struct * task) { - down_read(&profile_rwsem); - notifier_call_chain(&task_exit_notifier, 0, task); - up_read(&profile_rwsem); + blocking_notifier_call_chain(&task_exit_notifier, 0, task); } int profile_handoff_task(struct task_struct * task) { int ret; - read_lock(&handoff_lock); - ret = notifier_call_chain(&task_free_notifier, 0, task); - read_unlock(&handoff_lock); + ret = atomic_notifier_call_chain(&task_free_notifier, 0, task); return (ret == NOTIFY_OK) ? 1 : 0; } void profile_munmap(unsigned long addr) { - down_read(&profile_rwsem); - notifier_call_chain(&munmap_notifier, 0, (void *)addr); - up_read(&profile_rwsem); + blocking_notifier_call_chain(&munmap_notifier, 0, (void *)addr); } int task_handoff_register(struct notifier_block * n) { - int err = -EINVAL; - - write_lock(&handoff_lock); - err = notifier_chain_register(&task_free_notifier, n); - write_unlock(&handoff_lock); - return err; + return atomic_notifier_chain_register(&task_free_notifier, n); } int task_handoff_unregister(struct notifier_block * n) { - int err = -EINVAL; - - write_lock(&handoff_lock); - err = notifier_chain_unregister(&task_free_notifier, n); - write_unlock(&handoff_lock); - return err; + return atomic_notifier_chain_unregister(&task_free_notifier, n); } int profile_event_register(enum profile_type type, struct notifier_block * n) { int err = -EINVAL; - down_write(&profile_rwsem); - switch (type) { case PROFILE_TASK_EXIT: - err = notifier_chain_register(&task_exit_notifier, n); + err = blocking_notifier_chain_register( + &task_exit_notifier, n); break; case PROFILE_MUNMAP: - err = notifier_chain_register(&munmap_notifier, n); + err = blocking_notifier_chain_register( + &munmap_notifier, n); break; } - up_write(&profile_rwsem); - return err; } @@ -161,18 +141,17 @@ int profile_event_unregister(enum profile_type type, struct notifier_block * n) { int err = -EINVAL; - down_write(&profile_rwsem); - switch (type) { case PROFILE_TASK_EXIT: - err = notifier_chain_unregister(&task_exit_notifier, n); + err = blocking_notifier_chain_unregister( + &task_exit_notifier, n); break; case PROFILE_MUNMAP: - err = notifier_chain_unregister(&munmap_notifier, n); + err = blocking_notifier_chain_unregister( + &munmap_notifier, n); break; } - up_write(&profile_rwsem); return err; } diff --git a/kernel/softlockup.c b/kernel/softlockup.c index d9b3d5847ed..ced91e1ff56 100644 --- a/kernel/softlockup.c +++ b/kernel/softlockup.c @@ -152,5 +152,5 @@ __init void spawn_softlockup_task(void) cpu_callback(&cpu_nfb, CPU_ONLINE, cpu); register_cpu_notifier(&cpu_nfb); - notifier_chain_register(&panic_notifier_list, &panic_block); + atomic_notifier_chain_register(&panic_notifier_list, &panic_block); } diff --git a/kernel/sys.c b/kernel/sys.c index 38bc73ede2b..c93d37f71ae 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -95,99 +95,304 @@ int cad_pid = 1; * and the like. */ -static struct notifier_block *reboot_notifier_list; -static DEFINE_RWLOCK(notifier_lock); +static BLOCKING_NOTIFIER_HEAD(reboot_notifier_list); + +/* + * Notifier chain core routines. The exported routines below + * are layered on top of these, with appropriate locking added. + */ + +static int notifier_chain_register(struct notifier_block **nl, + struct notifier_block *n) +{ + while ((*nl) != NULL) { + if (n->priority > (*nl)->priority) + break; + nl = &((*nl)->next); + } + n->next = *nl; + rcu_assign_pointer(*nl, n); + return 0; +} + +static int notifier_chain_unregister(struct notifier_block **nl, + struct notifier_block *n) +{ + while ((*nl) != NULL) { + if ((*nl) == n) { + rcu_assign_pointer(*nl, n->next); + return 0; + } + nl = &((*nl)->next); + } + return -ENOENT; +} + +static int __kprobes notifier_call_chain(struct notifier_block **nl, + unsigned long val, void *v) +{ + int ret = NOTIFY_DONE; + struct notifier_block *nb; + + nb = rcu_dereference(*nl); + while (nb) { + ret = nb->notifier_call(nb, val, v); + if ((ret & NOTIFY_STOP_MASK) == NOTIFY_STOP_MASK) + break; + nb = rcu_dereference(nb->next); + } + return ret; +} + +/* + * Atomic notifier chain routines. Registration and unregistration + * use a mutex, and call_chain is synchronized by RCU (no locks). + */ /** - * notifier_chain_register - Add notifier to a notifier chain - * @list: Pointer to root list pointer + * atomic_notifier_chain_register - Add notifier to an atomic notifier chain + * @nh: Pointer to head of the atomic notifier chain * @n: New entry in notifier chain * - * Adds a notifier to a notifier chain. + * Adds a notifier to an atomic notifier chain. * * Currently always returns zero. */ + +int atomic_notifier_chain_register(struct atomic_notifier_head *nh, + struct notifier_block *n) +{ + unsigned long flags; + int ret; + + spin_lock_irqsave(&nh->lock, flags); + ret = notifier_chain_register(&nh->head, n); + spin_unlock_irqrestore(&nh->lock, flags); + return ret; +} + +EXPORT_SYMBOL_GPL(atomic_notifier_chain_register); + +/** + * atomic_notifier_chain_unregister - Remove notifier from an atomic notifier chain + * @nh: Pointer to head of the atomic notifier chain + * @n: Entry to remove from notifier chain + * + * Removes a notifier from an atomic notifier chain. + * + * Returns zero on success or %-ENOENT on failure. + */ +int atomic_notifier_chain_unregister(struct atomic_notifier_head *nh, + struct notifier_block *n) +{ + unsigned long flags; + int ret; + + spin_lock_irqsave(&nh->lock, flags); + ret = notifier_chain_unregister(&nh->head, n); + spin_unlock_irqrestore(&nh->lock, flags); + synchronize_rcu(); + return ret; +} + +EXPORT_SYMBOL_GPL(atomic_notifier_chain_unregister); + +/** + * atomic_notifier_call_chain - Call functions in an atomic notifier chain + * @nh: Pointer to head of the atomic notifier chain + * @val: Value passed unmodified to notifier function + * @v: Pointer passed unmodified to notifier function + * + * Calls each function in a notifier chain in turn. The functions + * run in an atomic context, so they must not block. + * This routine uses RCU to synchronize with changes to the chain. + * + * If the return value of the notifier can be and'ed + * with %NOTIFY_STOP_MASK then atomic_notifier_call_chain + * will return immediately, with the return value of + * the notifier function which halted execution. + * Otherwise the return value is the return value + * of the last notifier function called. + */ -int notifier_chain_register(struct notifier_block **list, struct notifier_block *n) +int atomic_notifier_call_chain(struct atomic_notifier_head *nh, + unsigned long val, void *v) { - write_lock(¬ifier_lock); - while(*list) - { - if(n->priority > (*list)->priority) - break; - list= &((*list)->next); - } - n->next = *list; - *list=n; - write_unlock(¬ifier_lock); - return 0; + int ret; + + rcu_read_lock(); + ret = notifier_call_chain(&nh->head, val, v); + rcu_read_unlock(); + return ret; } -EXPORT_SYMBOL(notifier_chain_register); +EXPORT_SYMBOL_GPL(atomic_notifier_call_chain); + +/* + * Blocking notifier chain routines. All access to the chain is + * synchronized by an rwsem. + */ /** - * notifier_chain_unregister - Remove notifier from a notifier chain - * @nl: Pointer to root list pointer + * blocking_notifier_chain_register - Add notifier to a blocking notifier chain + * @nh: Pointer to head of the blocking notifier chain * @n: New entry in notifier chain * - * Removes a notifier from a notifier chain. + * Adds a notifier to a blocking notifier chain. + * Must be called in process context. * - * Returns zero on success, or %-ENOENT on failure. + * Currently always returns zero. */ -int notifier_chain_unregister(struct notifier_block **nl, struct notifier_block *n) +int blocking_notifier_chain_register(struct blocking_notifier_head *nh, + struct notifier_block *n) { - write_lock(¬ifier_lock); - while((*nl)!=NULL) - { - if((*nl)==n) - { - *nl=n->next; - write_unlock(¬ifier_lock); - return 0; - } - nl=&((*nl)->next); - } - write_unlock(¬ifier_lock); - return -ENOENT; + int ret; + + /* + * This code gets used during boot-up, when task switching is + * not yet working and interrupts must remain disabled. At + * such times we must not call down_write(). + */ + if (unlikely(system_state == SYSTEM_BOOTING)) + return notifier_chain_register(&nh->head, n); + + down_write(&nh->rwsem); + ret = notifier_chain_register(&nh->head, n); + up_write(&nh->rwsem); + return ret; } -EXPORT_SYMBOL(notifier_chain_unregister); +EXPORT_SYMBOL_GPL(blocking_notifier_chain_register); /** - * notifier_call_chain - Call functions in a notifier chain - * @n: Pointer to root pointer of notifier chain + * blocking_notifier_chain_unregister - Remove notifier from a blocking notifier chain + * @nh: Pointer to head of the blocking notifier chain + * @n: Entry to remove from notifier chain + * + * Removes a notifier from a blocking notifier chain. + * Must be called from process context. + * + * Returns zero on success or %-ENOENT on failure. + */ +int blocking_notifier_chain_unregister(struct blocking_notifier_head *nh, + struct notifier_block *n) +{ + int ret; + + /* + * This code gets used during boot-up, when task switching is + * not yet working and interrupts must remain disabled. At + * such times we must not call down_write(). + */ + if (unlikely(system_state == SYSTEM_BOOTING)) + return notifier_chain_unregister(&nh->head, n); + + down_write(&nh->rwsem); + ret = notifier_chain_unregister(&nh->head, n); + up_write(&nh->rwsem); + return ret; +} + +EXPORT_SYMBOL_GPL(blocking_notifier_chain_unregister); + +/** + * blocking_notifier_call_chain - Call functions in a blocking notifier chain + * @nh: Pointer to head of the blocking notifier chain * @val: Value passed unmodified to notifier function * @v: Pointer passed unmodified to notifier function * - * Calls each function in a notifier chain in turn. + * Calls each function in a notifier chain in turn. The functions + * run in a process context, so they are allowed to block. * - * If the return value of the notifier can be and'd - * with %NOTIFY_STOP_MASK, then notifier_call_chain + * If the return value of the notifier can be and'ed + * with %NOTIFY_STOP_MASK then blocking_notifier_call_chain * will return immediately, with the return value of * the notifier function which halted execution. - * Otherwise, the return value is the return value + * Otherwise the return value is the return value * of the last notifier function called. */ -int __kprobes notifier_call_chain(struct notifier_block **n, unsigned long val, void *v) +int blocking_notifier_call_chain(struct blocking_notifier_head *nh, + unsigned long val, void *v) { - int ret=NOTIFY_DONE; - struct notifier_block *nb = *n; + int ret; - while(nb) - { - ret=nb->notifier_call(nb,val,v); - if(ret&NOTIFY_STOP_MASK) - { - return ret; - } - nb=nb->next; - } + down_read(&nh->rwsem); + ret = notifier_call_chain(&nh->head, val, v); + up_read(&nh->rwsem); return ret; } -EXPORT_SYMBOL(notifier_call_chain); +EXPORT_SYMBOL_GPL(blocking_notifier_call_chain); + +/* + * Raw notifier chain routines. There is no protection; + * the caller must provide it. Use at your own risk! + */ + +/** + * raw_notifier_chain_register - Add notifier to a raw notifier chain + * @nh: Pointer to head of the raw notifier chain + * @n: New entry in notifier chain + * + * Adds a notifier to a raw notifier chain. + * All locking must be provided by the caller. + * + * Currently always returns zero. + */ + +int raw_notifier_chain_register(struct raw_notifier_head *nh, + struct notifier_block *n) +{ + return notifier_chain_register(&nh->head, n); +} + +EXPORT_SYMBOL_GPL(raw_notifier_chain_register); + +/** + * raw_notifier_chain_unregister - Remove notifier from a raw notifier chain + * @nh: Pointer to head of the raw notifier chain + * @n: Entry to remove from notifier chain + * + * Removes a notifier from a raw notifier chain. + * All locking must be provided by the caller. + * + * Returns zero on success or %-ENOENT on failure. + */ +int raw_notifier_chain_unregister(struct raw_notifier_head *nh, + struct notifier_block *n) +{ + return notifier_chain_unregister(&nh->head, n); +} + +EXPORT_SYMBOL_GPL(raw_notifier_chain_unregister); + +/** + * raw_notifier_call_chain - Call functions in a raw notifier chain + * @nh: Pointer to head of the raw notifier chain + * @val: Value passed unmodified to notifier function + * @v: Pointer passed unmodified to notifier function + * + * Calls each function in a notifier chain in turn. The functions + * run in an undefined context. + * All locking must be provided by the caller. + * + * If the return value of the notifier can be and'ed + * with %NOTIFY_STOP_MASK then raw_notifier_call_chain + * will return immediately, with the return value of + * the notifier function which halted execution. + * Otherwise the return value is the return value + * of the last notifier function called. + */ + +int raw_notifier_call_chain(struct raw_notifier_head *nh, + unsigned long val, void *v) +{ + return notifier_call_chain(&nh->head, val, v); +} + +EXPORT_SYMBOL_GPL(raw_notifier_call_chain); /** * register_reboot_notifier - Register function to be called at reboot time @@ -196,13 +401,13 @@ EXPORT_SYMBOL(notifier_call_chain); * Registers a function with the list of functions * to be called at reboot time. * - * Currently always returns zero, as notifier_chain_register + * Currently always returns zero, as blocking_notifier_chain_register * always returns zero. */ int register_reboot_notifier(struct notifier_block * nb) { - return notifier_chain_register(&reboot_notifier_list, nb); + return blocking_notifier_chain_register(&reboot_notifier_list, nb); } EXPORT_SYMBOL(register_reboot_notifier); @@ -219,7 +424,7 @@ EXPORT_SYMBOL(register_reboot_notifier); int unregister_reboot_notifier(struct notifier_block * nb) { - return notifier_chain_unregister(&reboot_notifier_list, nb); + return blocking_notifier_chain_unregister(&reboot_notifier_list, nb); } EXPORT_SYMBOL(unregister_reboot_notifier); @@ -380,7 +585,7 @@ EXPORT_SYMBOL_GPL(emergency_restart); void kernel_restart_prepare(char *cmd) { - notifier_call_chain(&reboot_notifier_list, SYS_RESTART, cmd); + blocking_notifier_call_chain(&reboot_notifier_list, SYS_RESTART, cmd); system_state = SYSTEM_RESTART; device_shutdown(); } @@ -430,7 +635,7 @@ EXPORT_SYMBOL_GPL(kernel_kexec); void kernel_shutdown_prepare(enum system_states state) { - notifier_call_chain(&reboot_notifier_list, + blocking_notifier_call_chain(&reboot_notifier_list, (state == SYSTEM_HALT)?SYS_HALT:SYS_POWER_OFF, NULL); system_state = state; device_shutdown(); diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 9106354c781..a49a6975092 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -73,23 +73,23 @@ DEFINE_RWLOCK(hci_cb_list_lock); struct hci_proto *hci_proto[HCI_MAX_PROTO]; /* HCI notifiers list */ -static struct notifier_block *hci_notifier; +static ATOMIC_NOTIFIER_HEAD(hci_notifier); /* ---- HCI notifications ---- */ int hci_register_notifier(struct notifier_block *nb) { - return notifier_chain_register(&hci_notifier, nb); + return atomic_notifier_chain_register(&hci_notifier, nb); } int hci_unregister_notifier(struct notifier_block *nb) { - return notifier_chain_unregister(&hci_notifier, nb); + return atomic_notifier_chain_unregister(&hci_notifier, nb); } static void hci_notify(struct hci_dev *hdev, int event) { - notifier_call_chain(&hci_notifier, event, hdev); + atomic_notifier_call_chain(&hci_notifier, event, hdev); } /* ---- HCI requests ---- */ diff --git a/net/core/dev.c b/net/core/dev.c index 8e1dc305122..a3ab11f3415 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -193,7 +193,7 @@ static inline struct hlist_head *dev_index_hash(int ifindex) * Our notifier list */ -static struct notifier_block *netdev_chain; +static BLOCKING_NOTIFIER_HEAD(netdev_chain); /* * Device drivers call our routines to queue packets here. We empty the @@ -736,7 +736,8 @@ int dev_change_name(struct net_device *dev, char *newname) if (!err) { hlist_del(&dev->name_hlist); hlist_add_head(&dev->name_hlist, dev_name_hash(dev->name)); - notifier_call_chain(&netdev_chain, NETDEV_CHANGENAME, dev); + blocking_notifier_call_chain(&netdev_chain, + NETDEV_CHANGENAME, dev); } return err; @@ -750,7 +751,7 @@ int dev_change_name(struct net_device *dev, char *newname) */ void netdev_features_change(struct net_device *dev) { - notifier_call_chain(&netdev_chain, NETDEV_FEAT_CHANGE, dev); + blocking_notifier_call_chain(&netdev_chain, NETDEV_FEAT_CHANGE, dev); } EXPORT_SYMBOL(netdev_features_change); @@ -765,7 +766,8 @@ EXPORT_SYMBOL(netdev_features_change); void netdev_state_change(struct net_device *dev) { if (dev->flags & IFF_UP) { - notifier_call_chain(&netdev_chain, NETDEV_CHANGE, dev); + blocking_notifier_call_chain(&netdev_chain, + NETDEV_CHANGE, dev); rtmsg_ifinfo(RTM_NEWLINK, dev, 0); } } @@ -862,7 +864,7 @@ int dev_open(struct net_device *dev) /* * ... and announce new interface. */ - notifier_call_chain(&netdev_chain, NETDEV_UP, dev); + blocking_notifier_call_chain(&netdev_chain, NETDEV_UP, dev); } return ret; } @@ -885,7 +887,7 @@ int dev_close(struct net_device *dev) * Tell people we are going down, so that they can * prepare to death, when device is still operating. */ - notifier_call_chain(&netdev_chain, NETDEV_GOING_DOWN, dev); + blocking_notifier_call_chain(&netdev_chain, NETDEV_GOING_DOWN, dev); dev_deactivate(dev); @@ -922,7 +924,7 @@ int dev_close(struct net_device *dev) /* * Tell people we are down */ - notifier_call_chain(&netdev_chain, NETDEV_DOWN, dev); + blocking_notifier_call_chain(&netdev_chain, NETDEV_DOWN, dev); return 0; } @@ -953,7 +955,7 @@ int register_netdevice_notifier(struct notifier_block *nb) int err; rtnl_lock(); - err = notifier_chain_register(&netdev_chain, nb); + err = blocking_notifier_chain_register(&netdev_chain, nb); if (!err) { for (dev = dev_base; dev; dev = dev->next) { nb->notifier_call(nb, NETDEV_REGISTER, dev); @@ -981,7 +983,7 @@ int unregister_netdevice_notifier(struct notifier_block *nb) int err; rtnl_lock(); - err = notifier_chain_unregister(&netdev_chain, nb); + err = blocking_notifier_chain_unregister(&netdev_chain, nb); rtnl_unlock(); return err; } @@ -992,12 +994,12 @@ int unregister_netdevice_notifier(struct notifier_block *nb) * @v: pointer passed unmodified to notifier function * * Call all network notifier blocks. Parameters and return value - * are as for notifier_call_chain(). + * are as for blocking_notifier_call_chain(). */ int call_netdevice_notifiers(unsigned long val, void *v) { - return notifier_call_chain(&netdev_chain, val, v); + return blocking_notifier_call_chain(&netdev_chain, val, v); } /* When > 0 there are consumers of rx skb time stamps */ @@ -2242,7 +2244,8 @@ int dev_change_flags(struct net_device *dev, unsigned flags) if (dev->flags & IFF_UP && ((old_flags ^ dev->flags) &~ (IFF_UP | IFF_PROMISC | IFF_ALLMULTI | IFF_VOLATILE))) - notifier_call_chain(&netdev_chain, NETDEV_CHANGE, dev); + blocking_notifier_call_chain(&netdev_chain, + NETDEV_CHANGE, dev); if ((flags ^ dev->gflags) & IFF_PROMISC) { int inc = (flags & IFF_PROMISC) ? +1 : -1; @@ -2286,8 +2289,8 @@ int dev_set_mtu(struct net_device *dev, int new_mtu) else dev->mtu = new_mtu; if (!err && dev->flags & IFF_UP) - notifier_call_chain(&netdev_chain, - NETDEV_CHANGEMTU, dev); + blocking_notifier_call_chain(&netdev_chain, + NETDEV_CHANGEMTU, dev); return err; } @@ -2303,7 +2306,8 @@ int dev_set_mac_address(struct net_device *dev, struct sockaddr *sa) return -ENODEV; err = dev->set_mac_address(dev, sa); if (!err) - notifier_call_chain(&netdev_chain, NETDEV_CHANGEADDR, dev); + blocking_notifier_call_chain(&netdev_chain, + NETDEV_CHANGEADDR, dev); return err; } @@ -2359,7 +2363,7 @@ static int dev_ifsioc(struct ifreq *ifr, unsigned int cmd) return -EINVAL; memcpy(dev->broadcast, ifr->ifr_hwaddr.sa_data, min(sizeof ifr->ifr_hwaddr.sa_data, (size_t) dev->addr_len)); - notifier_call_chain(&netdev_chain, + blocking_notifier_call_chain(&netdev_chain, NETDEV_CHANGEADDR, dev); return 0; @@ -2813,7 +2817,7 @@ int register_netdevice(struct net_device *dev) write_unlock_bh(&dev_base_lock); /* Notify protocols, that a new device appeared. */ - notifier_call_chain(&netdev_chain, NETDEV_REGISTER, dev); + blocking_notifier_call_chain(&netdev_chain, NETDEV_REGISTER, dev); /* Finish registration after unlock */ net_set_todo(dev); @@ -2892,7 +2896,7 @@ static void netdev_wait_allrefs(struct net_device *dev) rtnl_lock(); /* Rebroadcast unregister notification */ - notifier_call_chain(&netdev_chain, + blocking_notifier_call_chain(&netdev_chain, NETDEV_UNREGISTER, dev); if (test_bit(__LINK_STATE_LINKWATCH_PENDING, @@ -3148,7 +3152,7 @@ int unregister_netdevice(struct net_device *dev) /* Notify protocols, that we are about to destroy this device. They should clean all the things. */ - notifier_call_chain(&netdev_chain, NETDEV_UNREGISTER, dev); + blocking_notifier_call_chain(&netdev_chain, NETDEV_UNREGISTER, dev); /* * Flush the multicast chain diff --git a/net/decnet/dn_dev.c b/net/decnet/dn_dev.c index cc7b9d9255e..d2ae9893ca1 100644 --- a/net/decnet/dn_dev.c +++ b/net/decnet/dn_dev.c @@ -68,7 +68,7 @@ __le16 decnet_address = 0; static DEFINE_RWLOCK(dndev_lock); static struct net_device *decnet_default_device; -static struct notifier_block *dnaddr_chain; +static BLOCKING_NOTIFIER_HEAD(dnaddr_chain); static struct dn_dev *dn_dev_create(struct net_device *dev, int *err); static void dn_dev_delete(struct net_device *dev); @@ -446,7 +446,7 @@ static void dn_dev_del_ifa(struct dn_dev *dn_db, struct dn_ifaddr **ifap, int de } rtmsg_ifa(RTM_DELADDR, ifa1); - notifier_call_chain(&dnaddr_chain, NETDEV_DOWN, ifa1); + blocking_notifier_call_chain(&dnaddr_chain, NETDEV_DOWN, ifa1); if (destroy) { dn_dev_free_ifa(ifa1); @@ -481,7 +481,7 @@ static int dn_dev_insert_ifa(struct dn_dev *dn_db, struct dn_ifaddr *ifa) dn_db->ifa_list = ifa; rtmsg_ifa(RTM_NEWADDR, ifa); - notifier_call_chain(&dnaddr_chain, NETDEV_UP, ifa); + blocking_notifier_call_chain(&dnaddr_chain, NETDEV_UP, ifa); return 0; } @@ -1285,12 +1285,12 @@ void dn_dev_devices_on(void) int register_dnaddr_notifier(struct notifier_block *nb) { - return notifier_chain_register(&dnaddr_chain, nb); + return blocking_notifier_chain_register(&dnaddr_chain, nb); } int unregister_dnaddr_notifier(struct notifier_block *nb) { - return notifier_chain_unregister(&dnaddr_chain, nb); + return blocking_notifier_chain_unregister(&dnaddr_chain, nb); } #ifdef CONFIG_PROC_FS diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index 44fdf1413e2..81c2f788529 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -81,7 +81,7 @@ static struct ipv4_devconf ipv4_devconf_dflt = { static void rtmsg_ifa(int event, struct in_ifaddr *); -static struct notifier_block *inetaddr_chain; +static BLOCKING_NOTIFIER_HEAD(inetaddr_chain); static void inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, int destroy); #ifdef CONFIG_SYSCTL @@ -267,7 +267,8 @@ static void inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, *ifap1 = ifa->ifa_next; rtmsg_ifa(RTM_DELADDR, ifa); - notifier_call_chain(&inetaddr_chain, NETDEV_DOWN, ifa); + blocking_notifier_call_chain(&inetaddr_chain, + NETDEV_DOWN, ifa); inet_free_ifa(ifa); } else { promote = ifa; @@ -291,7 +292,7 @@ static void inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, So that, this order is correct. */ rtmsg_ifa(RTM_DELADDR, ifa1); - notifier_call_chain(&inetaddr_chain, NETDEV_DOWN, ifa1); + blocking_notifier_call_chain(&inetaddr_chain, NETDEV_DOWN, ifa1); if (promote) { @@ -303,7 +304,8 @@ static void inet_del_ifa(struct in_device *in_dev, struct in_ifaddr **ifap, promote->ifa_flags &= ~IFA_F_SECONDARY; rtmsg_ifa(RTM_NEWADDR, promote); - notifier_call_chain(&inetaddr_chain, NETDEV_UP, promote); + blocking_notifier_call_chain(&inetaddr_chain, + NETDEV_UP, promote); for (ifa = promote->ifa_next; ifa; ifa = ifa->ifa_next) { if (ifa1->ifa_mask != ifa->ifa_mask || !inet_ifa_match(ifa1->ifa_address, ifa)) @@ -366,7 +368,7 @@ static int inet_insert_ifa(struct in_ifaddr *ifa) Notifier will trigger FIB update, so that listeners of netlink will know about new ifaddr */ rtmsg_ifa(RTM_NEWADDR, ifa); - notifier_call_chain(&inetaddr_chain, NETDEV_UP, ifa); + blocking_notifier_call_chain(&inetaddr_chain, NETDEV_UP, ifa); return 0; } @@ -938,12 +940,12 @@ u32 inet_confirm_addr(const struct net_device *dev, u32 dst, u32 local, int scop int register_inetaddr_notifier(struct notifier_block *nb) { - return notifier_chain_register(&inetaddr_chain, nb); + return blocking_notifier_chain_register(&inetaddr_chain, nb); } int unregister_inetaddr_notifier(struct notifier_block *nb) { - return notifier_chain_unregister(&inetaddr_chain, nb); + return blocking_notifier_chain_unregister(&inetaddr_chain, nb); } /* Rename ifa_labels for a device name change. Make some effort to preserve existing diff --git a/net/ipv4/netfilter/ip_conntrack_core.c b/net/ipv4/netfilter/ip_conntrack_core.c index 9e34034729a..ceaabc18202 100644 --- a/net/ipv4/netfilter/ip_conntrack_core.c +++ b/net/ipv4/netfilter/ip_conntrack_core.c @@ -80,8 +80,8 @@ static int ip_conntrack_vmalloc; static unsigned int ip_conntrack_next_id; static unsigned int ip_conntrack_expect_next_id; #ifdef CONFIG_IP_NF_CONNTRACK_EVENTS -struct notifier_block *ip_conntrack_chain; -struct notifier_block *ip_conntrack_expect_chain; +ATOMIC_NOTIFIER_HEAD(ip_conntrack_chain); +ATOMIC_NOTIFIER_HEAD(ip_conntrack_expect_chain); DEFINE_PER_CPU(struct ip_conntrack_ecache, ip_conntrack_ecache); @@ -92,7 +92,7 @@ __ip_ct_deliver_cached_events(struct ip_conntrack_ecache *ecache) { DEBUGP("ecache: delivering events for %p\n", ecache->ct); if (is_confirmed(ecache->ct) && !is_dying(ecache->ct) && ecache->events) - notifier_call_chain(&ip_conntrack_chain, ecache->events, + atomic_notifier_call_chain(&ip_conntrack_chain, ecache->events, ecache->ct); ecache->events = 0; ip_conntrack_put(ecache->ct); diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 01c62a0d374..445006ee452 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -143,7 +143,7 @@ static void inet6_prefix_notify(int event, struct inet6_dev *idev, struct prefix_info *pinfo); static int ipv6_chk_same_addr(const struct in6_addr *addr, struct net_device *dev); -static struct notifier_block *inet6addr_chain; +static ATOMIC_NOTIFIER_HEAD(inet6addr_chain); struct ipv6_devconf ipv6_devconf = { .forwarding = 0, @@ -593,7 +593,7 @@ out2: read_unlock_bh(&addrconf_lock); if (likely(err == 0)) - notifier_call_chain(&inet6addr_chain, NETDEV_UP, ifa); + atomic_notifier_call_chain(&inet6addr_chain, NETDEV_UP, ifa); else { kfree(ifa); ifa = ERR_PTR(err); @@ -688,7 +688,7 @@ static void ipv6_del_addr(struct inet6_ifaddr *ifp) ipv6_ifa_notify(RTM_DELADDR, ifp); - notifier_call_chain(&inet6addr_chain,NETDEV_DOWN,ifp); + atomic_notifier_call_chain(&inet6addr_chain, NETDEV_DOWN, ifp); addrconf_del_timer(ifp); @@ -3767,12 +3767,12 @@ static void addrconf_sysctl_unregister(struct ipv6_devconf *p) int register_inet6addr_notifier(struct notifier_block *nb) { - return notifier_chain_register(&inet6addr_chain, nb); + return atomic_notifier_chain_register(&inet6addr_chain, nb); } int unregister_inet6addr_notifier(struct notifier_block *nb) { - return notifier_chain_unregister(&inet6addr_chain,nb); + return atomic_notifier_chain_unregister(&inet6addr_chain,nb); } /* diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 0ae281d9bfc..56389c83557 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -90,8 +90,8 @@ static int nf_conntrack_vmalloc; static unsigned int nf_conntrack_next_id; static unsigned int nf_conntrack_expect_next_id; #ifdef CONFIG_NF_CONNTRACK_EVENTS -struct notifier_block *nf_conntrack_chain; -struct notifier_block *nf_conntrack_expect_chain; +ATOMIC_NOTIFIER_HEAD(nf_conntrack_chain); +ATOMIC_NOTIFIER_HEAD(nf_conntrack_expect_chain); DEFINE_PER_CPU(struct nf_conntrack_ecache, nf_conntrack_ecache); @@ -103,7 +103,7 @@ __nf_ct_deliver_cached_events(struct nf_conntrack_ecache *ecache) DEBUGP("ecache: delivering events for %p\n", ecache->ct); if (nf_ct_is_confirmed(ecache->ct) && !nf_ct_is_dying(ecache->ct) && ecache->events) - notifier_call_chain(&nf_conntrack_chain, ecache->events, + atomic_notifier_call_chain(&nf_conntrack_chain, ecache->events, ecache->ct); ecache->events = 0; diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index d00a9034cb5..2a233ffcf61 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -123,7 +123,7 @@ static void netlink_destroy_callback(struct netlink_callback *cb); static DEFINE_RWLOCK(nl_table_lock); static atomic_t nl_table_users = ATOMIC_INIT(0); -static struct notifier_block *netlink_chain; +static ATOMIC_NOTIFIER_HEAD(netlink_chain); static u32 netlink_group_mask(u32 group) { @@ -469,7 +469,8 @@ static int netlink_release(struct socket *sock) .protocol = sk->sk_protocol, .pid = nlk->pid, }; - notifier_call_chain(&netlink_chain, NETLINK_URELEASE, &n); + atomic_notifier_call_chain(&netlink_chain, + NETLINK_URELEASE, &n); } if (nlk->module) @@ -1695,12 +1696,12 @@ static struct file_operations netlink_seq_fops = { int netlink_register_notifier(struct notifier_block *nb) { - return notifier_chain_register(&netlink_chain, nb); + return atomic_notifier_chain_register(&netlink_chain, nb); } int netlink_unregister_notifier(struct notifier_block *nb) { - return notifier_chain_unregister(&netlink_chain, nb); + return atomic_notifier_chain_unregister(&netlink_chain, nb); } static const struct proto_ops netlink_ops = { -- cgit v1.2.3-70-g09d2 From 5e48521e869de7d904bb5ffe9739258ffa026927 Mon Sep 17 00:00:00 2001 From: "Zhang, Yanmin" Date: Thu, 23 Feb 2006 11:07:20 +0800 Subject: [IA64] lazy_mmu_prot_update needs to be aware of huge pages Function lazy_mmu_prot_update is also used on huge pages when it is called by set_huge_ptep_writable, but it isn't aware of huge pages. Signed-off-by: Zhang Yanmin Acked-by: Ken Chen Signed-off-by: Tony Luck --- arch/ia64/mm/init.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) (limited to 'arch/ia64') diff --git a/arch/ia64/mm/init.c b/arch/ia64/mm/init.c index ff4f31fcd33..674736129b8 100644 --- a/arch/ia64/mm/init.c +++ b/arch/ia64/mm/init.c @@ -109,6 +109,7 @@ lazy_mmu_prot_update (pte_t pte) { unsigned long addr; struct page *page; + unsigned long order; if (!pte_exec(pte)) return; /* not an executable page... */ @@ -119,7 +120,12 @@ lazy_mmu_prot_update (pte_t pte) if (test_bit(PG_arch_1, &page->flags)) return; /* i-cache is already coherent with d-cache */ - flush_icache_range(addr, addr + PAGE_SIZE); + if (PageCompound(page)) { + order = (unsigned long) (page[1].lru.prev); + flush_icache_range(addr, addr + (1UL << order << PAGE_SHIFT)); + } + else + flush_icache_range(addr, addr + PAGE_SIZE); set_bit(PG_arch_1, &page->flags); /* mark page as clean */ } -- cgit v1.2.3-70-g09d2 From ce9eed5a98efacb896551d3470d9d46826caaee5 Mon Sep 17 00:00:00 2001 From: "Chen, Kenneth W" Date: Mon, 6 Mar 2006 14:12:54 -0800 Subject: [IA64] optimize flush_tlb_range on large numa box It was reported from a field customer that global spin lock ptcg_lock is giving a lot of grief on munmap performance running on a large numa machine. What appears to be a problem coming from flush_tlb_range(), which currently unconditionally calls platform_global_tlb_purge(). For some of the numa machines in existence today, this function is mapped into ia64_global_tlb_purge(), which holds ptcg_lock spin lock while executing ptc.ga instruction. Here is a patch that attempt to avoid global tlb purge whenever possible. It will use local tlb purge as much as possible. Though the conditions to use local tlb purge is pretty restrictive. One of the side effect of having flush tlb range instruction on ia64 is that kernel don't get a chance to clear out cpu_vm_mask. On ia64, this mask is sticky and it will accumulate if process bounces around. Thus diminishing the possible use of ptc.l. Thoughts? Signed-off-by: Ken Chen Acked-by: Jack Steiner Acked-by: KAMEZAWA Hiroyuki Signed-off-by: Tony Luck --- arch/ia64/mm/tlb.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/mm/tlb.c b/arch/ia64/mm/tlb.c index 6a4eec9113e..4dbbca0b5e9 100644 --- a/arch/ia64/mm/tlb.c +++ b/arch/ia64/mm/tlb.c @@ -156,17 +156,19 @@ flush_tlb_range (struct vm_area_struct *vma, unsigned long start, nbits = purge.max_bits; start &= ~((1UL << nbits) - 1); -# ifdef CONFIG_SMP - platform_global_tlb_purge(mm, start, end, nbits); -# else preempt_disable(); +#ifdef CONFIG_SMP + if (mm != current->active_mm || cpus_weight(mm->cpu_vm_mask) != 1) { + platform_global_tlb_purge(mm, start, end, nbits); + preempt_enable(); + return; + } +#endif do { ia64_ptcl(start, (nbits<<2)); start += (1UL << nbits); } while (start < end); preempt_enable(); -# endif - ia64_srlz_i(); /* srlz.i implies srlz.d */ } EXPORT_SYMBOL(flush_tlb_range); -- cgit v1.2.3-70-g09d2 From e6ef0fca2c61f7e773178be8c3833a709a47e022 Mon Sep 17 00:00:00 2001 From: "hawkes@sgi.com" Date: Mon, 27 Mar 2006 13:18:17 -0800 Subject: [IA64-SGI] sn_hwperf use of num_online_cpus() Eliminate an unnecessary -- and flawed -- use of the expensive num_online_cpus(). Signed-off-by: John Hawkes Signed-off-by: Tony Luck --- arch/ia64/sn/kernel/sn2/sn_hwperf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/ia64') diff --git a/arch/ia64/sn/kernel/sn2/sn_hwperf.c b/arch/ia64/sn/kernel/sn2/sn_hwperf.c index 70db21f3df2..6ff34fc05c8 100644 --- a/arch/ia64/sn/kernel/sn2/sn_hwperf.c +++ b/arch/ia64/sn/kernel/sn2/sn_hwperf.c @@ -605,7 +605,7 @@ static int sn_hwperf_op_cpu(struct sn_hwperf_op_info *op_info) op_info->a->arg &= SN_HWPERF_ARG_OBJID_MASK; if (cpu != SN_HWPERF_ARG_ANY_CPU) { - if (cpu >= num_online_cpus() || !cpu_online(cpu)) { + if (cpu >= NR_CPUS || !cpu_online(cpu)) { r = -EINVAL; goto out; } -- cgit v1.2.3-70-g09d2 From 769ebc66dec24770ea1e99019099d2bc636a61d3 Mon Sep 17 00:00:00 2001 From: Dean Roe Date: Mon, 6 Mar 2006 10:32:05 -0600 Subject: [IA64-SGI] fix for-loop in sn_hwperf_geoid_to_cnode() Fix a for-loop in sn_hwperf_geoid_to_cnode(). It needs to loop over num_cnodes to ensure it can still process TIO nodes in addition to compute nodes on systems with many nodes. Interim fix until better support for many (>265) nodes is complete. Signed-off-by: Dean Roe Signed-off-by: Tony Luck --- arch/ia64/sn/kernel/sn2/sn_hwperf.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'arch/ia64') diff --git a/arch/ia64/sn/kernel/sn2/sn_hwperf.c b/arch/ia64/sn/kernel/sn2/sn_hwperf.c index 6ff34fc05c8..d917afa30b2 100644 --- a/arch/ia64/sn/kernel/sn2/sn_hwperf.c +++ b/arch/ia64/sn/kernel/sn2/sn_hwperf.c @@ -110,7 +110,11 @@ static int sn_hwperf_geoid_to_cnode(char *location) if (sn_hwperf_location_to_bpos(location, &rack, &bay, &slot, &slab)) return -1; - for_each_node(cnode) { + /* + * FIXME: replace with cleaner for_each_XXX macro which addresses + * both compute and IO nodes once ACPI3.0 is available. + */ + for (cnode = 0; cnode < num_cnodes; cnode++) { geoid = cnodeid_get_geoid(cnode); module_id = geo_module(geoid); this_rack = MODULE_GET_RACK(module_id); -- cgit v1.2.3-70-g09d2 From ec1b9466cb4f6ae6d950bd67055d9410d1056d2a Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 28 Mar 2006 01:56:40 -0800 Subject: [PATCH] ia64: const f_ops fix Tweak the proc setup code so things work OK with const proc_dir_entry.proc_fops. Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/sn/kernel/sn2/sn_proc_fs.c | 39 ++++++++++++++++++------------------ 1 file changed, 20 insertions(+), 19 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/sn/kernel/sn2/sn_proc_fs.c b/arch/ia64/sn/kernel/sn2/sn_proc_fs.c index c686d9c12f7..5100261310f 100644 --- a/arch/ia64/sn/kernel/sn2/sn_proc_fs.c +++ b/arch/ia64/sn/kernel/sn2/sn_proc_fs.c @@ -93,19 +93,22 @@ static int coherence_id_open(struct inode *inode, struct file *file) static struct proc_dir_entry *sn_procfs_create_entry(const char *name, struct proc_dir_entry *parent, int (*openfunc)(struct inode *, struct file *), - int (*releasefunc)(struct inode *, struct file *)) + int (*releasefunc)(struct inode *, struct file *), + ssize_t (*write) (struct file *, const char __user *, size_t, loff_t *)) { struct proc_dir_entry *e = create_proc_entry(name, 0444, parent); if (e) { - e->proc_fops = (struct file_operations *)kmalloc( - sizeof(struct file_operations), GFP_KERNEL); - if (e->proc_fops) { - memset(e->proc_fops, 0, sizeof(struct file_operations)); - e->proc_fops->open = openfunc; - e->proc_fops->read = seq_read; - e->proc_fops->llseek = seq_lseek; - e->proc_fops->release = releasefunc; + struct file_operations *f; + + f = kzalloc(sizeof(*f), GFP_KERNEL); + if (f) { + f->open = openfunc; + f->read = seq_read; + f->llseek = seq_lseek; + f->release = releasefunc; + f->write = write; + e->proc_fops = f; } } @@ -119,31 +122,29 @@ extern int sn_topology_release(struct inode *, struct file *); void register_sn_procfs(void) { static struct proc_dir_entry *sgi_proc_dir = NULL; - struct proc_dir_entry *e; BUG_ON(sgi_proc_dir != NULL); if (!(sgi_proc_dir = proc_mkdir("sgi_sn", NULL))) return; sn_procfs_create_entry("partition_id", sgi_proc_dir, - partition_id_open, single_release); + partition_id_open, single_release, NULL); sn_procfs_create_entry("system_serial_number", sgi_proc_dir, - system_serial_number_open, single_release); + system_serial_number_open, single_release, NULL); sn_procfs_create_entry("licenseID", sgi_proc_dir, - licenseID_open, single_release); + licenseID_open, single_release, NULL); - e = sn_procfs_create_entry("sn_force_interrupt", sgi_proc_dir, - sn_force_interrupt_open, single_release); - if (e) - e->proc_fops->write = sn_force_interrupt_write_proc; + sn_procfs_create_entry("sn_force_interrupt", sgi_proc_dir, + sn_force_interrupt_open, single_release, + sn_force_interrupt_write_proc); sn_procfs_create_entry("coherence_id", sgi_proc_dir, - coherence_id_open, single_release); + coherence_id_open, single_release, NULL); sn_procfs_create_entry("sn_topology", sgi_proc_dir, - sn_topology_open, sn_topology_release); + sn_topology_open, sn_topology_release, NULL); } #endif /* CONFIG_PROC_FS */ -- cgit v1.2.3-70-g09d2 From 46cba3dcaea4eb51b28fac78d15542a90619a6ef Mon Sep 17 00:00:00 2001 From: Satoru Takeuchi Date: Mon, 27 Mar 2006 17:12:19 +0900 Subject: [IA64] correct some messages and fixes some minor things This patch corrects some wrong comments and a printk message. It also fixes some minor things, and makes all lines fit in 80 columns. Signed-off-by: Satoru Takeuchi Signed-off-by: Tony Luck --- arch/ia64/kernel/iosapic.c | 260 ++++++++++++++++++++++++++++----------------- 1 file changed, 161 insertions(+), 99 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c index 8832c553230..22d845f8536 100644 --- a/arch/ia64/kernel/iosapic.c +++ b/arch/ia64/kernel/iosapic.c @@ -9,54 +9,65 @@ * Copyright (C) 1999 VA Linux Systems * Copyright (C) 1999,2000 Walt Drummond * - * 00/04/19 D. Mosberger Rewritten to mirror more closely the x86 I/O APIC code. - * In particular, we now have separate handlers for edge - * and level triggered interrupts. - * 00/10/27 Asit Mallick, Goutham Rao IRQ vector allocation - * PCI to vector mapping, shared PCI interrupts. - * 00/10/27 D. Mosberger Document things a bit more to make them more understandable. - * Clean up much of the old IOSAPIC cruft. - * 01/07/27 J.I. Lee PCI irq routing, Platform/Legacy interrupts and fixes for - * ACPI S5(SoftOff) support. + * 00/04/19 D. Mosberger Rewritten to mirror more closely the x86 I/O + * APIC code. In particular, we now have separate + * handlers for edge and level triggered + * interrupts. + * 00/10/27 Asit Mallick, Goutham Rao IRQ vector + * allocation PCI to vector mapping, shared PCI + * interrupts. + * 00/10/27 D. Mosberger Document things a bit more to make them more + * understandable. Clean up much of the old + * IOSAPIC cruft. + * 01/07/27 J.I. Lee PCI irq routing, Platform/Legacy interrupts + * and fixes for ACPI S5(SoftOff) support. * 02/01/23 J.I. Lee iosapic pgm fixes for PCI irq routing from _PRT - * 02/01/07 E. Focht Redirectable interrupt vectors in - * iosapic_set_affinity(), initializations for - * /proc/irq/#/smp_affinity + * 02/01/07 E. Focht Redirectable interrupt + * vectors in iosapic_set_affinity(), + * initializations for /proc/irq/#/smp_affinity * 02/04/02 P. Diefenbaugh Cleaned up ACPI PCI IRQ routing. * 02/04/18 J.I. Lee bug fix in iosapic_init_pci_irq - * 02/04/30 J.I. Lee bug fix in find_iosapic to fix ACPI PCI IRQ to IOSAPIC mapping - * error + * 02/04/30 J.I. Lee bug fix in find_iosapic to fix ACPI PCI IRQ to + * IOSAPIC mapping error * 02/07/29 T. Kochi Allocate interrupt vectors dynamically - * 02/08/04 T. Kochi Cleaned up terminology (irq, global system interrupt, vector, etc.) - * 02/09/20 D. Mosberger Simplified by taking advantage of ACPI's pci_irq code. + * 02/08/04 T. Kochi Cleaned up terminology (irq, global system + * interrupt, vector, etc.) + * 02/09/20 D. Mosberger Simplified by taking advantage of ACPI's + * pci_irq code. * 03/02/19 B. Helgaas Make pcat_compat system-wide, not per-IOSAPIC. - * Remove iosapic_address & gsi_base from external interfaces. - * Rationalize __init/__devinit attributes. + * Remove iosapic_address & gsi_base from + * external interfaces. Rationalize + * __init/__devinit attributes. * 04/12/04 Ashok Raj Intel Corporation 2004 - * Updated to work with irq migration necessary for CPU Hotplug + * Updated to work with irq migration necessary + * for CPU Hotplug */ /* - * Here is what the interrupt logic between a PCI device and the kernel looks like: + * Here is what the interrupt logic between a PCI device and the kernel looks + * like: * - * (1) A PCI device raises one of the four interrupt pins (INTA, INTB, INTC, INTD). The - * device is uniquely identified by its bus--, and slot-number (the function - * number does not matter here because all functions share the same interrupt - * lines). + * (1) A PCI device raises one of the four interrupt pins (INTA, INTB, INTC, + * INTD). The device is uniquely identified by its bus-, and slot-number + * (the function number does not matter here because all functions share + * the same interrupt lines). * - * (2) The motherboard routes the interrupt line to a pin on a IOSAPIC controller. - * Multiple interrupt lines may have to share the same IOSAPIC pin (if they're level - * triggered and use the same polarity). Each interrupt line has a unique Global - * System Interrupt (GSI) number which can be calculated as the sum of the controller's - * base GSI number and the IOSAPIC pin number to which the line connects. + * (2) The motherboard routes the interrupt line to a pin on a IOSAPIC + * controller. Multiple interrupt lines may have to share the same + * IOSAPIC pin (if they're level triggered and use the same polarity). + * Each interrupt line has a unique Global System Interrupt (GSI) number + * which can be calculated as the sum of the controller's base GSI number + * and the IOSAPIC pin number to which the line connects. * - * (3) The IOSAPIC uses an internal routing table entries (RTEs) to map the IOSAPIC pin - * into the IA-64 interrupt vector. This interrupt vector is then sent to the CPU. + * (3) The IOSAPIC uses an internal routing table entries (RTEs) to map the + * IOSAPIC pin into the IA-64 interrupt vector. This interrupt vector is then + * sent to the CPU. * - * (4) The kernel recognizes an interrupt as an IRQ. The IRQ interface is used as - * architecture-independent interrupt handling mechanism in Linux. As an - * IRQ is a number, we have to have IA-64 interrupt vector number <-> IRQ number - * mapping. On smaller systems, we use one-to-one mapping between IA-64 vector and - * IRQ. A platform can implement platform_irq_to_vector(irq) and + * (4) The kernel recognizes an interrupt as an IRQ. The IRQ interface is + * used as architecture-independent interrupt handling mechanism in Linux. + * As an IRQ is a number, we have to have + * IA-64 interrupt vector number <-> IRQ number mapping. On smaller + * systems, we use one-to-one mapping between IA-64 vector and IRQ. A + * platform can implement platform_irq_to_vector(irq) and * platform_local_vector_to_irq(vector) APIs to differentiate the mapping. * Please see also include/asm-ia64/hw_irq.h for those APIs. * @@ -64,9 +75,9 @@ * * PCI pin -> global system interrupt (GSI) -> IA-64 vector <-> IRQ * - * Note: The term "IRQ" is loosely used everywhere in Linux kernel to describe interrupts. - * Now we use "IRQ" only for Linux IRQ's. ISA IRQ (isa_irq) is the only exception in this - * source code. + * Note: The term "IRQ" is loosely used everywhere in Linux kernel to + * describeinterrupts. Now we use "IRQ" only for Linux IRQ's. ISA IRQ + * (isa_irq) is the only exception in this source code. */ #include @@ -90,7 +101,6 @@ #include #include - #undef DEBUG_INTERRUPT_ROUTING #ifdef DEBUG_INTERRUPT_ROUTING @@ -99,36 +109,46 @@ #define DBG(fmt...) #endif -#define NR_PREALLOCATE_RTE_ENTRIES (PAGE_SIZE / sizeof(struct iosapic_rte_info)) +#define NR_PREALLOCATE_RTE_ENTRIES \ + (PAGE_SIZE / sizeof(struct iosapic_rte_info)) #define RTE_PREALLOCATED (1) static DEFINE_SPINLOCK(iosapic_lock); -/* These tables map IA-64 vectors to the IOSAPIC pin that generates this vector. */ +/* + * These tables map IA-64 vectors to the IOSAPIC pin that generates this + * vector. + */ struct iosapic_rte_info { - struct list_head rte_list; /* node in list of RTEs sharing the same vector */ + struct list_head rte_list; /* node in list of RTEs sharing the + * same vector */ char __iomem *addr; /* base address of IOSAPIC */ - unsigned int gsi_base; /* first GSI assigned to this IOSAPIC */ + unsigned int gsi_base; /* first GSI assigned to this + * IOSAPIC */ char rte_index; /* IOSAPIC RTE index */ int refcnt; /* reference counter */ unsigned int flags; /* flags */ } ____cacheline_aligned; static struct iosapic_intr_info { - struct list_head rtes; /* RTEs using this vector (empty => not an IOSAPIC interrupt) */ + struct list_head rtes; /* RTEs using this vector (empty => + * not an IOSAPIC interrupt) */ int count; /* # of RTEs that shares this vector */ - u32 low32; /* current value of low word of Redirection table entry */ + u32 low32; /* current value of low word of + * Redirection table entry */ unsigned int dest; /* destination CPU physical ID */ unsigned char dmode : 3; /* delivery mode (see iosapic.h) */ - unsigned char polarity: 1; /* interrupt polarity (see iosapic.h) */ + unsigned char polarity: 1; /* interrupt polarity + * (see iosapic.h) */ unsigned char trigger : 1; /* trigger mode (see iosapic.h) */ } iosapic_intr_info[IA64_NUM_VECTORS]; static struct iosapic { char __iomem *addr; /* base address of IOSAPIC */ - unsigned int gsi_base; /* first GSI assigned to this IOSAPIC */ - unsigned short num_rte; /* number of RTE in this IOSAPIC */ + unsigned int gsi_base; /* first GSI assigned to this + * IOSAPIC */ + unsigned short num_rte; /* # of RTEs on this IOSAPIC */ int rtes_inuse; /* # of RTEs in use on this IOSAPIC */ #ifdef CONFIG_NUMA unsigned short node; /* numa node association via pxm */ @@ -149,7 +169,8 @@ find_iosapic (unsigned int gsi) int i; for (i = 0; i < NR_IOSAPICS; i++) { - if ((unsigned) (gsi - iosapic_lists[i].gsi_base) < iosapic_lists[i].num_rte) + if ((unsigned) (gsi - iosapic_lists[i].gsi_base) < + iosapic_lists[i].num_rte) return i; } @@ -162,7 +183,8 @@ _gsi_to_vector (unsigned int gsi) struct iosapic_intr_info *info; struct iosapic_rte_info *rte; - for (info = iosapic_intr_info; info < iosapic_intr_info + IA64_NUM_VECTORS; ++info) + for (info = iosapic_intr_info; info < + iosapic_intr_info + IA64_NUM_VECTORS; ++info) list_for_each_entry(rte, &info->rtes, rte_list) if (rte->gsi_base + rte->rte_index == gsi) return info - iosapic_intr_info; @@ -185,8 +207,8 @@ gsi_to_irq (unsigned int gsi) unsigned long flags; int irq; /* - * XXX fix me: this assumes an identity mapping vetween IA-64 vector and Linux irq - * numbers... + * XXX fix me: this assumes an identity mapping between IA-64 vector + * and Linux irq numbers... */ spin_lock_irqsave(&iosapic_lock, flags); { @@ -197,7 +219,8 @@ gsi_to_irq (unsigned int gsi) return irq; } -static struct iosapic_rte_info *gsi_vector_to_rte(unsigned int gsi, unsigned int vec) +static struct iosapic_rte_info *gsi_vector_to_rte(unsigned int gsi, + unsigned int vec) { struct iosapic_rte_info *rte; @@ -237,7 +260,9 @@ set_rte (unsigned int gsi, unsigned int vector, unsigned int dest, int mask) for (irq = 0; irq < NR_IRQS; ++irq) if (irq_to_vector(irq) == vector) { - set_irq_affinity_info(irq, (int)(dest & 0xffff), redir); + set_irq_affinity_info(irq, + (int)(dest & 0xffff), + redir); break; } } @@ -259,7 +284,7 @@ set_rte (unsigned int gsi, unsigned int vector, unsigned int dest, int mask) } static void -nop (unsigned int vector) +nop (unsigned int irq) { /* do nothing... */ } @@ -281,7 +306,8 @@ mask_irq (unsigned int irq) { /* set only the mask bit */ low32 = iosapic_intr_info[vec].low32 |= IOSAPIC_MASK; - list_for_each_entry(rte, &iosapic_intr_info[vec].rtes, rte_list) { + list_for_each_entry(rte, &iosapic_intr_info[vec].rtes, + rte_list) { addr = rte->addr; rte_index = rte->rte_index; iosapic_write(addr, IOSAPIC_RTE_LOW(rte_index), low32); @@ -306,7 +332,8 @@ unmask_irq (unsigned int irq) spin_lock_irqsave(&iosapic_lock, flags); { low32 = iosapic_intr_info[vec].low32 &= ~IOSAPIC_MASK; - list_for_each_entry(rte, &iosapic_intr_info[vec].rtes, rte_list) { + list_for_each_entry(rte, &iosapic_intr_info[vec].rtes, + rte_list) { addr = rte->addr; rte_index = rte->rte_index; iosapic_write(addr, IOSAPIC_RTE_LOW(rte_index), low32); @@ -346,21 +373,25 @@ iosapic_set_affinity (unsigned int irq, cpumask_t mask) spin_lock_irqsave(&iosapic_lock, flags); { - low32 = iosapic_intr_info[vec].low32 & ~(7 << IOSAPIC_DELIVERY_SHIFT); + low32 = iosapic_intr_info[vec].low32 & + ~(7 << IOSAPIC_DELIVERY_SHIFT); if (redir) /* change delivery mode to lowest priority */ - low32 |= (IOSAPIC_LOWEST_PRIORITY << IOSAPIC_DELIVERY_SHIFT); + low32 |= (IOSAPIC_LOWEST_PRIORITY << + IOSAPIC_DELIVERY_SHIFT); else /* change delivery mode to fixed */ low32 |= (IOSAPIC_FIXED << IOSAPIC_DELIVERY_SHIFT); iosapic_intr_info[vec].low32 = low32; iosapic_intr_info[vec].dest = dest; - list_for_each_entry(rte, &iosapic_intr_info[vec].rtes, rte_list) { + list_for_each_entry(rte, &iosapic_intr_info[vec].rtes, + rte_list) { addr = rte->addr; rte_index = rte->rte_index; - iosapic_write(addr, IOSAPIC_RTE_HIGH(rte_index), high32); + iosapic_write(addr, IOSAPIC_RTE_HIGH(rte_index), + high32); iosapic_write(addr, IOSAPIC_RTE_LOW(rte_index), low32); } } @@ -433,7 +464,8 @@ iosapic_ack_edge_irq (unsigned int irq) * interrupt for real. This prevents IRQ storms from unhandled * devices. */ - if ((idesc->status & (IRQ_PENDING|IRQ_DISABLED)) == (IRQ_PENDING|IRQ_DISABLED)) + if ((idesc->status & (IRQ_PENDING|IRQ_DISABLED)) == + (IRQ_PENDING|IRQ_DISABLED)) mask_irq(irq); } @@ -467,7 +499,8 @@ iosapic_version (char __iomem *addr) return iosapic_read(addr, IOSAPIC_VERSION); } -static int iosapic_find_sharable_vector (unsigned long trigger, unsigned long pol) +static int iosapic_find_sharable_vector (unsigned long trigger, + unsigned long pol) { int i, vector = -1, min_count = -1; struct iosapic_intr_info *info; @@ -482,7 +515,8 @@ static int iosapic_find_sharable_vector (unsigned long trigger, unsigned long po for (i = IA64_FIRST_DEVICE_VECTOR; i <= IA64_LAST_DEVICE_VECTOR; i++) { info = &iosapic_intr_info[i]; if (info->trigger == trigger && info->polarity == pol && - (info->dmode == IOSAPIC_FIXED || info->dmode == IOSAPIC_LOWEST_PRIORITY)) { + (info->dmode == IOSAPIC_FIXED || info->dmode == + IOSAPIC_LOWEST_PRIORITY)) { if (min_count == -1 || info->count < min_count) { vector = i; min_count = info->count; @@ -506,12 +540,15 @@ iosapic_reassign_vector (int vector) new_vector = assign_irq_vector(AUTO_ASSIGN); if (new_vector < 0) panic("%s: out of interrupt vectors!\n", __FUNCTION__); - printk(KERN_INFO "Reassigning vector %d to %d\n", vector, new_vector); + printk(KERN_INFO "Reassigning vector %d to %d\n", + vector, new_vector); memcpy(&iosapic_intr_info[new_vector], &iosapic_intr_info[vector], sizeof(struct iosapic_intr_info)); INIT_LIST_HEAD(&iosapic_intr_info[new_vector].rtes); - list_move(iosapic_intr_info[vector].rtes.next, &iosapic_intr_info[new_vector].rtes); - memset(&iosapic_intr_info[vector], 0, sizeof(struct iosapic_intr_info)); + list_move(iosapic_intr_info[vector].rtes.next, + &iosapic_intr_info[new_vector].rtes); + memset(&iosapic_intr_info[vector], 0, + sizeof(struct iosapic_intr_info)); iosapic_intr_info[vector].low32 = IOSAPIC_MASK; INIT_LIST_HEAD(&iosapic_intr_info[vector].rtes); } @@ -524,7 +561,8 @@ static struct iosapic_rte_info *iosapic_alloc_rte (void) int preallocated = 0; if (!iosapic_kmalloc_ok && list_empty(&free_rte_list)) { - rte = alloc_bootmem(sizeof(struct iosapic_rte_info) * NR_PREALLOCATE_RTE_ENTRIES); + rte = alloc_bootmem(sizeof(struct iosapic_rte_info) * + NR_PREALLOCATE_RTE_ENTRIES); if (!rte) return NULL; for (i = 0; i < NR_PREALLOCATE_RTE_ENTRIES; i++, rte++) @@ -532,7 +570,8 @@ static struct iosapic_rte_info *iosapic_alloc_rte (void) } if (!list_empty(&free_rte_list)) { - rte = list_entry(free_rte_list.next, struct iosapic_rte_info, rte_list); + rte = list_entry(free_rte_list.next, struct iosapic_rte_info, + rte_list); list_del(&rte->rte_list); preallocated++; } else { @@ -575,7 +614,8 @@ register_intr (unsigned int gsi, int vector, unsigned char delivery, index = find_iosapic(gsi); if (index < 0) { - printk(KERN_WARNING "%s: No IOSAPIC for GSI %u\n", __FUNCTION__, gsi); + printk(KERN_WARNING "%s: No IOSAPIC for GSI %u\n", + __FUNCTION__, gsi); return -ENODEV; } @@ -586,7 +626,8 @@ register_intr (unsigned int gsi, int vector, unsigned char delivery, if (!rte) { rte = iosapic_alloc_rte(); if (!rte) { - printk(KERN_WARNING "%s: cannot allocate memory\n", __FUNCTION__); + printk(KERN_WARNING "%s: cannot allocate memory\n", + __FUNCTION__); return -ENOMEM; } @@ -602,7 +643,9 @@ register_intr (unsigned int gsi, int vector, unsigned char delivery, else if (vector_is_shared(vector)) { struct iosapic_intr_info *info = &iosapic_intr_info[vector]; if (info->trigger != trigger || info->polarity != polarity) { - printk (KERN_WARNING "%s: cannot override the interrupt\n", __FUNCTION__); + printk (KERN_WARNING + "%s: cannot override the interrupt\n", + __FUNCTION__); return -EINVAL; } } @@ -619,8 +662,10 @@ register_intr (unsigned int gsi, int vector, unsigned char delivery, idesc = irq_descp(vector); if (idesc->handler != irq_type) { if (idesc->handler != &no_irq_type) - printk(KERN_WARNING "%s: changing vector %d from %s to %s\n", - __FUNCTION__, vector, idesc->handler->typename, irq_type->typename); + printk(KERN_WARNING + "%s: changing vector %d from %s to %s\n", + __FUNCTION__, vector, + idesc->handler->typename, irq_type->typename); idesc->handler = irq_type; } return 0; @@ -681,7 +726,7 @@ get_target_cpu (unsigned int gsi, int vector) if (!num_cpus) goto skip_numa_setup; - /* Use vector assigment to distribute across cpus in node */ + /* Use vector assignment to distribute across cpus in node */ cpu_index = vector % num_cpus; for (numa_cpu = first_cpu(cpu_mask) ; i < cpu_index ; i++) @@ -703,7 +748,7 @@ skip_numa_setup: } while (!cpu_online(cpu)); return cpu_physical_id(cpu); -#else +#else /* CONFIG_SMP */ return cpu_physical_id(smp_processor_id()); #endif } @@ -755,7 +800,8 @@ again: if (list_empty(&iosapic_intr_info[vector].rtes)) free_irq_vector(vector); spin_unlock(&iosapic_lock); - spin_unlock_irqrestore(&irq_descp(vector)->lock, flags); + spin_unlock_irqrestore(&irq_descp(vector)->lock, + flags); goto again; } @@ -764,7 +810,8 @@ again: polarity, trigger); if (err < 0) { spin_unlock(&iosapic_lock); - spin_unlock_irqrestore(&irq_descp(vector)->lock, flags); + spin_unlock_irqrestore(&irq_descp(vector)->lock, + flags); return err; } @@ -806,7 +853,8 @@ iosapic_unregister_intr (unsigned int gsi) */ irq = gsi_to_irq(gsi); if (irq < 0) { - printk(KERN_ERR "iosapic_unregister_intr(%u) unbalanced\n", gsi); + printk(KERN_ERR "iosapic_unregister_intr(%u) unbalanced\n", + gsi); WARN_ON(1); return; } @@ -817,7 +865,9 @@ iosapic_unregister_intr (unsigned int gsi) spin_lock(&iosapic_lock); { if ((rte = gsi_vector_to_rte(gsi, vector)) == NULL) { - printk(KERN_ERR "iosapic_unregister_intr(%u) unbalanced\n", gsi); + printk(KERN_ERR + "iosapic_unregister_intr(%u) unbalanced\n", + gsi); WARN_ON(1); goto out; } @@ -827,7 +877,8 @@ iosapic_unregister_intr (unsigned int gsi) /* Mask the interrupt */ low32 = iosapic_intr_info[vector].low32 | IOSAPIC_MASK; - iosapic_write(rte->addr, IOSAPIC_RTE_LOW(rte->rte_index), low32); + iosapic_write(rte->addr, IOSAPIC_RTE_LOW(rte->rte_index), + low32); /* Remove the rte entry from the list */ list_del(&rte->rte_list); @@ -840,7 +891,9 @@ iosapic_unregister_intr (unsigned int gsi) trigger = iosapic_intr_info[vector].trigger; polarity = iosapic_intr_info[vector].polarity; dest = iosapic_intr_info[vector].dest; - printk(KERN_INFO "GSI %u (%s, %s) -> CPU %d (0x%04x) vector %d unregistered\n", + printk(KERN_INFO + "GSI %u (%s, %s) -> CPU %d (0x%04x)" + " vector %d unregistered\n", gsi, (trigger == IOSAPIC_EDGE ? "edge" : "level"), (polarity == IOSAPIC_POL_HIGH ? "high" : "low"), cpu_logical_id(dest), dest, vector); @@ -853,12 +906,15 @@ iosapic_unregister_intr (unsigned int gsi) idesc->handler = &no_irq_type; /* Clear the interrupt information */ - memset(&iosapic_intr_info[vector], 0, sizeof(struct iosapic_intr_info)); + memset(&iosapic_intr_info[vector], 0, + sizeof(struct iosapic_intr_info)); iosapic_intr_info[vector].low32 |= IOSAPIC_MASK; INIT_LIST_HEAD(&iosapic_intr_info[vector].rtes); if (idesc->action) { - printk(KERN_ERR "interrupt handlers still exist on IRQ %u\n", irq); + printk(KERN_ERR + "interrupt handlers still exist on" + "IRQ %u\n", irq); WARN_ON(1); } @@ -873,7 +929,6 @@ iosapic_unregister_intr (unsigned int gsi) /* * ACPI calls this when it finds an entry for a platform interrupt. - * Note that the irq_base and IOSAPIC address must be set in iosapic_init(). */ int __init iosapic_register_platform_intr (u32 int_type, unsigned int gsi, @@ -907,13 +962,16 @@ iosapic_register_platform_intr (u32 int_type, unsigned int gsi, mask = 1; break; default: - printk(KERN_ERR "iosapic_register_platform_irq(): invalid int type 0x%x\n", int_type); + printk(KERN_ERR "%s: invalid int type 0x%x\n", __FUNCTION__, + int_type); return -1; } register_intr(gsi, vector, delivery, polarity, trigger); - printk(KERN_INFO "PLATFORM int %s (0x%x): GSI %u (%s, %s) -> CPU %d (0x%04x) vector %d\n", + printk(KERN_INFO + "PLATFORM int %s (0x%x): GSI %u (%s, %s) -> CPU %d (0x%04x)" + " vector %d\n", int_type < ARRAY_SIZE(name) ? name[int_type] : "unknown", int_type, gsi, (trigger == IOSAPIC_EDGE ? "edge" : "level"), (polarity == IOSAPIC_POL_HIGH ? "high" : "low"), @@ -923,10 +981,8 @@ iosapic_register_platform_intr (u32 int_type, unsigned int gsi, return vector; } - /* * ACPI calls this when it finds an entry for a legacy ISA IRQ override. - * Note that the gsi_base and IOSAPIC address must be set in iosapic_init(). */ void __init iosapic_override_isa_irq (unsigned int isa_irq, unsigned int gsi, @@ -955,16 +1011,19 @@ iosapic_system_init (int system_pcat_compat) for (vector = 0; vector < IA64_NUM_VECTORS; ++vector) { iosapic_intr_info[vector].low32 = IOSAPIC_MASK; - INIT_LIST_HEAD(&iosapic_intr_info[vector].rtes); /* mark as unused */ + /* mark as unused */ + INIT_LIST_HEAD(&iosapic_intr_info[vector].rtes); } pcat_compat = system_pcat_compat; if (pcat_compat) { /* - * Disable the compatibility mode interrupts (8259 style), needs IN/OUT support - * enabled. + * Disable the compatibility mode interrupts (8259 style), + * needs IN/OUT support enabled. */ - printk(KERN_INFO "%s: Disabling PC-AT compatible 8259 interrupts\n", __FUNCTION__); + printk(KERN_INFO + "%s: Disabling PC-AT compatible 8259 interrupts\n", + __FUNCTION__); outb(0xff, 0xA1); outb(0xff, 0x21); } @@ -1053,12 +1112,14 @@ iosapic_init (unsigned long phys_addr, unsigned int gsi_base) if ((gsi_base == 0) && pcat_compat) { /* - * Map the legacy ISA devices into the IOSAPIC data. Some of these may - * get reprogrammed later on with data from the ACPI Interrupt Source - * Override table. + * Map the legacy ISA devices into the IOSAPIC data. Some of + * these may get reprogrammed later on with data from the ACPI + * Interrupt Source Override table. */ for (isa_irq = 0; isa_irq < 16; ++isa_irq) - iosapic_override_isa_irq(isa_irq, isa_irq, IOSAPIC_POL_HIGH, IOSAPIC_EDGE); + iosapic_override_isa_irq(isa_irq, isa_irq, + IOSAPIC_POL_HIGH, + IOSAPIC_EDGE); } return 0; } @@ -1081,7 +1142,8 @@ iosapic_remove (unsigned int gsi_base) if (iosapic_lists[index].rtes_inuse) { err = -EBUSY; - printk(KERN_WARNING "%s: IOSAPIC for GSI base %u is busy\n", + printk(KERN_WARNING + "%s: IOSAPIC for GSI base %u is busy\n", __FUNCTION__, gsi_base); goto out; } -- cgit v1.2.3-70-g09d2 From e6d1ba5cd9954e7c15538aaef7585b235a69e0e1 Mon Sep 17 00:00:00 2001 From: Satoru Takeuchi Date: Mon, 27 Mar 2006 17:13:46 +0900 Subject: [IA64] simplify some condition checks in iosapic_check_gsi_range Some condition checks on iosapic_check_gsi_range() can be omitted because always `base <= end' is assured. This patch simplifies those checks. Signed-off-by: Satoru Takeuchi Signed-off-by: Tony Luck --- arch/ia64/kernel/iosapic.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c index 22d845f8536..7956eb9058f 100644 --- a/arch/ia64/kernel/iosapic.c +++ b/arch/ia64/kernel/iosapic.c @@ -1063,10 +1063,7 @@ iosapic_check_gsi_range (unsigned int gsi_base, unsigned int ver) base = iosapic_lists[index].gsi_base; end = base + iosapic_lists[index].num_rte - 1; - if (gsi_base < base && gsi_end < base) - continue;/* OK */ - - if (gsi_base > end && gsi_end > end) + if (gsi_end < base || end < gsi_base) continue; /* OK */ return -EBUSY; -- cgit v1.2.3-70-g09d2 From d89cfe7f1e82d758a7983584c1593795d4e2c098 Mon Sep 17 00:00:00 2001 From: Russ Anderson Date: Wed, 29 Mar 2006 11:31:23 -0600 Subject: [IA64] Move __mca_table out of the __init section Move __mca_table out of the __init section. Signed-off-by: Tony Luck --- arch/ia64/kernel/vmlinux.lds.S | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/vmlinux.lds.S b/arch/ia64/kernel/vmlinux.lds.S index 0b9e56dd7f0..783600fe52b 100644 --- a/arch/ia64/kernel/vmlinux.lds.S +++ b/arch/ia64/kernel/vmlinux.lds.S @@ -70,6 +70,15 @@ SECTIONS __stop___ex_table = .; } + /* MCA table */ + . = ALIGN(16); + __mca_table : AT(ADDR(__mca_table) - LOAD_OFFSET) + { + __start___mca_table = .; + *(__mca_table) + __stop___mca_table = .; + } + /* Global data */ _data = .; @@ -130,15 +139,6 @@ SECTIONS __initcall_end = .; } - /* MCA table */ - . = ALIGN(16); - __mca_table : AT(ADDR(__mca_table) - LOAD_OFFSET) - { - __start___mca_table = .; - *(__mca_table) - __stop___mca_table = .; - } - .data.patch.vtop : AT(ADDR(.data.patch.vtop) - LOAD_OFFSET) { __start___vtop_patchlist = .; -- cgit v1.2.3-70-g09d2 From 3283a67d8618c9a292eced23e8753ab64adc6dba Mon Sep 17 00:00:00 2001 From: Jes Sorensen Date: Thu, 30 Mar 2006 10:13:22 -0500 Subject: [IA64] Add __mca_table to the DISCARD list in gate.lds Add __mca_table to the DISCARD list for the gate.lds linker script to avoid broken linker references when linking the final vmlinux file. Also add comment to include/asm-ia64/asmmacros.h to avoid anyone else hitting this problem in the future. Credits to James Bottomley for spotting the DISCARD list in gate.lds.S Signed-off-by: Jes Sorensen Signed-off-by: Tony Luck --- arch/ia64/kernel/gate.lds.S | 1 + include/asm-ia64/asmmacro.h | 4 ++++ 2 files changed, 5 insertions(+) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/gate.lds.S b/arch/ia64/kernel/gate.lds.S index e1e4aba9ecd..7c99e6ec3da 100644 --- a/arch/ia64/kernel/gate.lds.S +++ b/arch/ia64/kernel/gate.lds.S @@ -59,6 +59,7 @@ SECTIONS *(.dynbss) *(.bss .bss.* .gnu.linkonce.b.*) *(__ex_table) + *(__mca_table) } } diff --git a/include/asm-ia64/asmmacro.h b/include/asm-ia64/asmmacro.h index d4cec32083d..edf2cebb296 100644 --- a/include/asm-ia64/asmmacro.h +++ b/include/asm-ia64/asmmacro.h @@ -38,6 +38,10 @@ name: /* * Helper macros for accessing user memory. + * + * When adding any new .section/.previous entries here, make sure to + * also add it to the DISCARD section in arch/ia64/kernel/gate.lds.S or + * unpleasant things will happen. */ .section "__ex_table", "a" // declare section & section attributes -- cgit v1.2.3-70-g09d2 From c1c57d767100417f63c18da52d7e96f82b2b9e1a Mon Sep 17 00:00:00 2001 From: Bjorn Helgaas Date: Thu, 30 Mar 2006 09:53:39 -0700 Subject: [IA64] ioremap() should prefer WB over UC efi_memmap_init() collects full granules of WB memory, without regard for whether they also support UC. So in order for ioremap() to work for main memory, it must prefer WB mappings when possible. Signed-off-by: Bjorn Helgaas Signed-off-by: Tony Luck --- arch/ia64/mm/ioremap.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/mm/ioremap.c b/arch/ia64/mm/ioremap.c index 62328621f99..643ccc6960c 100644 --- a/arch/ia64/mm/ioremap.c +++ b/arch/ia64/mm/ioremap.c @@ -21,12 +21,12 @@ __ioremap (unsigned long offset, unsigned long size) void __iomem * ioremap (unsigned long offset, unsigned long size) { - if (efi_mem_attribute_range(offset, size, EFI_MEMORY_UC)) - return __ioremap(offset, size); - if (efi_mem_attribute_range(offset, size, EFI_MEMORY_WB)) return phys_to_virt(offset); + if (efi_mem_attribute_range(offset, size, EFI_MEMORY_UC)) + return __ioremap(offset, size); + /* * Someday this should check ACPI resources so we * can do the right thing for hot-plugged regions. -- cgit v1.2.3-70-g09d2 From 5274f052e7b3dbd81935772eb551dfd0325dfa9d Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 30 Mar 2006 15:15:30 +0200 Subject: [PATCH] Introduce sys_splice() system call This adds support for the sys_splice system call. Using a pipe as a transport, it can connect to files or sockets (latter as output only). From the splice.c comments: "splice": joining two ropes together by interweaving their strands. This is the "extended pipe" functionality, where a pipe is used as an arbitrary in-memory buffer. Think of a pipe as a small kernel buffer that you can use to transfer data from one end to the other. The traditional unix read/write is extended with a "splice()" operation that transfers data buffers to or from a pipe buffer. Named by Larry McVoy, original implementation from Linus, extended by Jens to support splicing to files and fixing the initial implementation bugs. Signed-off-by: Jens Axboe Signed-off-by: Linus Torvalds --- arch/i386/kernel/syscall_table.S | 1 + arch/ia64/kernel/entry.S | 1 + fs/Makefile | 2 +- fs/ext2/file.c | 2 + fs/ext3/file.c | 2 + fs/pipe.c | 33 ++- fs/reiserfs/file.c | 2 + fs/splice.c | 612 +++++++++++++++++++++++++++++++++++++++ include/asm-i386/unistd.h | 3 +- include/asm-ia64/unistd.h | 3 +- include/asm-powerpc/unistd.h | 3 +- include/asm-x86_64/unistd.h | 4 +- include/linux/fs.h | 4 + include/linux/syscalls.h | 2 + net/socket.c | 6 +- 15 files changed, 669 insertions(+), 11 deletions(-) create mode 100644 fs/splice.c (limited to 'arch/ia64') diff --git a/arch/i386/kernel/syscall_table.S b/arch/i386/kernel/syscall_table.S index 326595f3fa4..ce3ef4fa055 100644 --- a/arch/i386/kernel/syscall_table.S +++ b/arch/i386/kernel/syscall_table.S @@ -312,3 +312,4 @@ ENTRY(sys_call_table) .long sys_unshare /* 310 */ .long sys_set_robust_list .long sys_get_robust_list + .long sys_splice diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S index 0e3eda99e54..750e8e7fbdc 100644 --- a/arch/ia64/kernel/entry.S +++ b/arch/ia64/kernel/entry.S @@ -1605,5 +1605,6 @@ sys_call_table: data8 sys_ni_syscall // reserved for pselect data8 sys_ni_syscall // 1295 reserved for ppoll data8 sys_unshare + data8 sys_splice .org sys_call_table + 8*NR_syscalls // guard against failures to increase NR_syscalls diff --git a/fs/Makefile b/fs/Makefile index 080b3867be4..f3a4f707717 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -10,7 +10,7 @@ obj-y := open.o read_write.o file_table.o buffer.o bio.o super.o \ ioctl.o readdir.o select.o fifo.o locks.o dcache.o inode.o \ attr.o bad_inode.o file.o filesystems.o namespace.o aio.o \ seq_file.o xattr.o libfs.o fs-writeback.o mpage.o direct-io.o \ - ioprio.o pnode.o drop_caches.o + ioprio.o pnode.o drop_caches.o splice.o obj-$(CONFIG_INOTIFY) += inotify.o obj-$(CONFIG_EPOLL) += eventpoll.o diff --git a/fs/ext2/file.c b/fs/ext2/file.c index 509cceca04d..23e2c7ccec1 100644 --- a/fs/ext2/file.c +++ b/fs/ext2/file.c @@ -53,6 +53,8 @@ const struct file_operations ext2_file_operations = { .readv = generic_file_readv, .writev = generic_file_writev, .sendfile = generic_file_sendfile, + .splice_read = generic_file_splice_read, + .splice_write = generic_file_splice_write, }; #ifdef CONFIG_EXT2_FS_XIP diff --git a/fs/ext3/file.c b/fs/ext3/file.c index 783a796220b..1efefb630ea 100644 --- a/fs/ext3/file.c +++ b/fs/ext3/file.c @@ -119,6 +119,8 @@ const struct file_operations ext3_file_operations = { .release = ext3_release_file, .fsync = ext3_sync_file, .sendfile = generic_file_sendfile, + .splice_read = generic_file_splice_read, + .splice_write = generic_file_splice_write, }; struct inode_operations ext3_file_inode_operations = { diff --git a/fs/pipe.c b/fs/pipe.c index e2f4f1d9ffc..2414bf270db 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include @@ -94,11 +95,20 @@ static void anon_pipe_buf_release(struct pipe_inode_info *info, struct pipe_buff { struct page *page = buf->page; - if (info->tmp_page) { - __free_page(page); + /* + * If nobody else uses this page, and we don't already have a + * temporary page, let's keep track of it as a one-deep + * allocation cache + */ + if (page_count(page) == 1 && !info->tmp_page) { + info->tmp_page = page; return; } - info->tmp_page = page; + + /* + * Otherwise just release our reference to it + */ + page_cache_release(page); } static void *anon_pipe_buf_map(struct file *file, struct pipe_inode_info *info, struct pipe_buffer *buf) @@ -152,6 +162,11 @@ pipe_readv(struct file *filp, const struct iovec *_iov, chars = total_len; addr = ops->map(filp, info, buf); + if (IS_ERR(addr)) { + if (!ret) + ret = PTR_ERR(addr); + break; + } error = pipe_iov_copy_to_user(iov, addr + buf->offset, chars); ops->unmap(info, buf); if (unlikely(error)) { @@ -254,8 +269,16 @@ pipe_writev(struct file *filp, const struct iovec *_iov, struct pipe_buf_operations *ops = buf->ops; int offset = buf->offset + buf->len; if (ops->can_merge && offset + chars <= PAGE_SIZE) { - void *addr = ops->map(filp, info, buf); - int error = pipe_iov_copy_from_user(offset + addr, iov, chars); + void *addr; + int error; + + addr = ops->map(filp, info, buf); + if (IS_ERR(addr)) { + error = PTR_ERR(addr); + goto out; + } + error = pipe_iov_copy_from_user(offset + addr, iov, + chars); ops->unmap(info, buf); ret = error; do_wakeup = 1; diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c index 010094d14da..cf6e1cf4035 100644 --- a/fs/reiserfs/file.c +++ b/fs/reiserfs/file.c @@ -1576,6 +1576,8 @@ const struct file_operations reiserfs_file_operations = { .sendfile = generic_file_sendfile, .aio_read = generic_file_aio_read, .aio_write = reiserfs_aio_write, + .splice_read = generic_file_splice_read, + .splice_write = generic_file_splice_write, }; struct inode_operations reiserfs_file_inode_operations = { diff --git a/fs/splice.c b/fs/splice.c new file mode 100644 index 00000000000..efa47c1c4e1 --- /dev/null +++ b/fs/splice.c @@ -0,0 +1,612 @@ +/* + * "splice": joining two ropes together by interweaving their strands. + * + * This is the "extended pipe" functionality, where a pipe is used as + * an arbitrary in-memory buffer. Think of a pipe as a small kernel + * buffer that you can use to transfer data from one end to the other. + * + * The traditional unix read/write is extended with a "splice()" operation + * that transfers data buffers to or from a pipe buffer. + * + * Named by Larry McVoy, original implementation from Linus, extended by + * Jens to support splicing to files and fixing the initial implementation + * bugs. + * + * Copyright (C) 2005 Jens Axboe + * Copyright (C) 2005 Linus Torvalds + * + */ +#include +#include +#include +#include +#include + +/* + * Passed to the actors + */ +struct splice_desc { + unsigned int len, total_len; /* current and remaining length */ + unsigned int flags; /* splice flags */ + struct file *file; /* file to read/write */ + loff_t pos; /* file position */ +}; + +static void page_cache_pipe_buf_release(struct pipe_inode_info *info, + struct pipe_buffer *buf) +{ + page_cache_release(buf->page); + buf->page = NULL; +} + +static void *page_cache_pipe_buf_map(struct file *file, + struct pipe_inode_info *info, + struct pipe_buffer *buf) +{ + struct page *page = buf->page; + + lock_page(page); + + if (!PageUptodate(page)) { + unlock_page(page); + return ERR_PTR(-EIO); + } + + if (!page->mapping) { + unlock_page(page); + return ERR_PTR(-ENODATA); + } + + return kmap(buf->page); +} + +static void page_cache_pipe_buf_unmap(struct pipe_inode_info *info, + struct pipe_buffer *buf) +{ + unlock_page(buf->page); + kunmap(buf->page); +} + +static struct pipe_buf_operations page_cache_pipe_buf_ops = { + .can_merge = 0, + .map = page_cache_pipe_buf_map, + .unmap = page_cache_pipe_buf_unmap, + .release = page_cache_pipe_buf_release, +}; + +static ssize_t move_to_pipe(struct inode *inode, struct page **pages, + int nr_pages, unsigned long offset, + unsigned long len) +{ + struct pipe_inode_info *info; + int ret, do_wakeup, i; + + ret = 0; + do_wakeup = 0; + i = 0; + + mutex_lock(PIPE_MUTEX(*inode)); + + info = inode->i_pipe; + for (;;) { + int bufs; + + if (!PIPE_READERS(*inode)) { + send_sig(SIGPIPE, current, 0); + if (!ret) + ret = -EPIPE; + break; + } + + bufs = info->nrbufs; + if (bufs < PIPE_BUFFERS) { + int newbuf = (info->curbuf + bufs) & (PIPE_BUFFERS - 1); + struct pipe_buffer *buf = info->bufs + newbuf; + struct page *page = pages[i++]; + unsigned long this_len; + + this_len = PAGE_CACHE_SIZE - offset; + if (this_len > len) + this_len = len; + + buf->page = page; + buf->offset = offset; + buf->len = this_len; + buf->ops = &page_cache_pipe_buf_ops; + info->nrbufs = ++bufs; + do_wakeup = 1; + + ret += this_len; + len -= this_len; + offset = 0; + if (!--nr_pages) + break; + if (!len) + break; + if (bufs < PIPE_BUFFERS) + continue; + + break; + } + + if (signal_pending(current)) { + if (!ret) + ret = -ERESTARTSYS; + break; + } + + if (do_wakeup) { + wake_up_interruptible_sync(PIPE_WAIT(*inode)); + kill_fasync(PIPE_FASYNC_READERS(*inode), SIGIO, + POLL_IN); + do_wakeup = 0; + } + + PIPE_WAITING_WRITERS(*inode)++; + pipe_wait(inode); + PIPE_WAITING_WRITERS(*inode)--; + } + + mutex_unlock(PIPE_MUTEX(*inode)); + + if (do_wakeup) { + wake_up_interruptible(PIPE_WAIT(*inode)); + kill_fasync(PIPE_FASYNC_READERS(*inode), SIGIO, POLL_IN); + } + + while (i < nr_pages) + page_cache_release(pages[i++]); + + return ret; +} + +static int __generic_file_splice_read(struct file *in, struct inode *pipe, + size_t len) +{ + struct address_space *mapping = in->f_mapping; + unsigned int offset, nr_pages; + struct page *pages[PIPE_BUFFERS], *shadow[PIPE_BUFFERS]; + struct page *page; + pgoff_t index, pidx; + int i, j; + + index = in->f_pos >> PAGE_CACHE_SHIFT; + offset = in->f_pos & ~PAGE_CACHE_MASK; + nr_pages = (len + offset + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + + if (nr_pages > PIPE_BUFFERS) + nr_pages = PIPE_BUFFERS; + + /* + * initiate read-ahead on this page range + */ + do_page_cache_readahead(mapping, in, index, nr_pages); + + /* + * Get as many pages from the page cache as possible.. + * Start IO on the page cache entries we create (we + * can assume that any pre-existing ones we find have + * already had IO started on them). + */ + i = find_get_pages(mapping, index, nr_pages, pages); + + /* + * common case - we found all pages and they are contiguous, + * kick them off + */ + if (i && (pages[i - 1]->index == index + i - 1)) + goto splice_them; + + /* + * fill shadow[] with pages at the right locations, so we only + * have to fill holes + */ + memset(shadow, 0, i * sizeof(struct page *)); + for (j = 0, pidx = index; j < i; pidx++, j++) + shadow[pages[j]->index - pidx] = pages[j]; + + /* + * now fill in the holes + */ + for (i = 0, pidx = index; i < nr_pages; pidx++, i++) { + int error; + + if (shadow[i]) + continue; + + /* + * no page there, look one up / create it + */ + page = find_or_create_page(mapping, pidx, + mapping_gfp_mask(mapping)); + if (!page) + break; + + if (PageUptodate(page)) + unlock_page(page); + else { + error = mapping->a_ops->readpage(in, page); + + if (unlikely(error)) { + page_cache_release(page); + break; + } + } + shadow[i] = page; + } + + if (!i) { + for (i = 0; i < nr_pages; i++) { + if (shadow[i]) + page_cache_release(shadow[i]); + } + return 0; + } + + memcpy(pages, shadow, i * sizeof(struct page *)); + + /* + * Now we splice them into the pipe.. + */ +splice_them: + return move_to_pipe(pipe, pages, i, offset, len); +} + +ssize_t generic_file_splice_read(struct file *in, struct inode *pipe, + size_t len, unsigned int flags) +{ + ssize_t spliced; + int ret; + + ret = 0; + spliced = 0; + while (len) { + ret = __generic_file_splice_read(in, pipe, len); + + if (ret <= 0) + break; + + in->f_pos += ret; + len -= ret; + spliced += ret; + } + + if (spliced) + return spliced; + + return ret; +} + +/* + * Send 'len' bytes to socket from 'file' at position 'pos' using sendpage(). + */ +static int pipe_to_sendpage(struct pipe_inode_info *info, + struct pipe_buffer *buf, struct splice_desc *sd) +{ + struct file *file = sd->file; + loff_t pos = sd->pos; + unsigned int offset; + ssize_t ret; + void *ptr; + + /* + * sub-optimal, but we are limited by the pipe ->map. we don't + * need a kmap'ed buffer here, we just want to make sure we + * have the page pinned if the pipe page originates from the + * page cache + */ + ptr = buf->ops->map(file, info, buf); + if (IS_ERR(ptr)) + return PTR_ERR(ptr); + + offset = pos & ~PAGE_CACHE_MASK; + + ret = file->f_op->sendpage(file, buf->page, offset, sd->len, &pos, + sd->len < sd->total_len); + + buf->ops->unmap(info, buf); + if (ret == sd->len) + return 0; + + return -EIO; +} + +/* + * This is a little more tricky than the file -> pipe splicing. There are + * basically three cases: + * + * - Destination page already exists in the address space and there + * are users of it. For that case we have no other option that + * copying the data. Tough luck. + * - Destination page already exists in the address space, but there + * are no users of it. Make sure it's uptodate, then drop it. Fall + * through to last case. + * - Destination page does not exist, we can add the pipe page to + * the page cache and avoid the copy. + * + * For now we just do the slower thing and always copy pages over, it's + * easier than migrating pages from the pipe to the target file. For the + * case of doing file | file splicing, the migrate approach had some LRU + * nastiness... + */ +static int pipe_to_file(struct pipe_inode_info *info, struct pipe_buffer *buf, + struct splice_desc *sd) +{ + struct file *file = sd->file; + struct address_space *mapping = file->f_mapping; + unsigned int offset; + struct page *page; + char *src, *dst; + pgoff_t index; + int ret; + + /* + * after this, page will be locked and unmapped + */ + src = buf->ops->map(file, info, buf); + if (IS_ERR(src)) + return PTR_ERR(src); + + index = sd->pos >> PAGE_CACHE_SHIFT; + offset = sd->pos & ~PAGE_CACHE_MASK; + +find_page: + ret = -ENOMEM; + page = find_or_create_page(mapping, index, mapping_gfp_mask(mapping)); + if (!page) + goto out; + + /* + * If the page is uptodate, it is also locked. If it isn't + * uptodate, we can mark it uptodate if we are filling the + * full page. Otherwise we need to read it in first... + */ + if (!PageUptodate(page)) { + if (sd->len < PAGE_CACHE_SIZE) { + ret = mapping->a_ops->readpage(file, page); + if (unlikely(ret)) + goto out; + + lock_page(page); + + if (!PageUptodate(page)) { + /* + * page got invalidated, repeat + */ + if (!page->mapping) { + unlock_page(page); + page_cache_release(page); + goto find_page; + } + ret = -EIO; + goto out; + } + } else { + WARN_ON(!PageLocked(page)); + SetPageUptodate(page); + } + } + + ret = mapping->a_ops->prepare_write(file, page, 0, sd->len); + if (ret) + goto out; + + dst = kmap_atomic(page, KM_USER0); + memcpy(dst + offset, src + buf->offset, sd->len); + flush_dcache_page(page); + kunmap_atomic(dst, KM_USER0); + + ret = mapping->a_ops->commit_write(file, page, 0, sd->len); + if (ret < 0) + goto out; + + set_page_dirty(page); + ret = write_one_page(page, 0); +out: + if (ret < 0) + unlock_page(page); + page_cache_release(page); + buf->ops->unmap(info, buf); + return ret; +} + +typedef int (splice_actor)(struct pipe_inode_info *, struct pipe_buffer *, + struct splice_desc *); + +static ssize_t move_from_pipe(struct inode *inode, struct file *out, + size_t len, unsigned int flags, + splice_actor *actor) +{ + struct pipe_inode_info *info; + int ret, do_wakeup, err; + struct splice_desc sd; + + ret = 0; + do_wakeup = 0; + + sd.total_len = len; + sd.flags = flags; + sd.file = out; + sd.pos = out->f_pos; + + mutex_lock(PIPE_MUTEX(*inode)); + + info = inode->i_pipe; + for (;;) { + int bufs = info->nrbufs; + + if (bufs) { + int curbuf = info->curbuf; + struct pipe_buffer *buf = info->bufs + curbuf; + struct pipe_buf_operations *ops = buf->ops; + + sd.len = buf->len; + if (sd.len > sd.total_len) + sd.len = sd.total_len; + + err = actor(info, buf, &sd); + if (err) { + if (!ret && err != -ENODATA) + ret = err; + + break; + } + + ret += sd.len; + buf->offset += sd.len; + buf->len -= sd.len; + if (!buf->len) { + buf->ops = NULL; + ops->release(info, buf); + curbuf = (curbuf + 1) & (PIPE_BUFFERS - 1); + info->curbuf = curbuf; + info->nrbufs = --bufs; + do_wakeup = 1; + } + + sd.pos += sd.len; + sd.total_len -= sd.len; + if (!sd.total_len) + break; + } + + if (bufs) + continue; + if (!PIPE_WRITERS(*inode)) + break; + if (!PIPE_WAITING_WRITERS(*inode)) { + if (ret) + break; + } + + if (signal_pending(current)) { + if (!ret) + ret = -ERESTARTSYS; + break; + } + + if (do_wakeup) { + wake_up_interruptible_sync(PIPE_WAIT(*inode)); + kill_fasync(PIPE_FASYNC_WRITERS(*inode),SIGIO,POLL_OUT); + do_wakeup = 0; + } + + pipe_wait(inode); + } + + mutex_unlock(PIPE_MUTEX(*inode)); + + if (do_wakeup) { + wake_up_interruptible(PIPE_WAIT(*inode)); + kill_fasync(PIPE_FASYNC_WRITERS(*inode), SIGIO, POLL_OUT); + } + + mutex_lock(&out->f_mapping->host->i_mutex); + out->f_pos = sd.pos; + mutex_unlock(&out->f_mapping->host->i_mutex); + return ret; + +} + +ssize_t generic_file_splice_write(struct inode *inode, struct file *out, + size_t len, unsigned int flags) +{ + return move_from_pipe(inode, out, len, flags, pipe_to_file); +} + +ssize_t generic_splice_sendpage(struct inode *inode, struct file *out, + size_t len, unsigned int flags) +{ + return move_from_pipe(inode, out, len, flags, pipe_to_sendpage); +} + +static long do_splice_from(struct inode *pipe, struct file *out, size_t len, + unsigned int flags) +{ + loff_t pos; + int ret; + + if (!out->f_op || !out->f_op->splice_write) + return -EINVAL; + + if (!(out->f_mode & FMODE_WRITE)) + return -EBADF; + + pos = out->f_pos; + ret = rw_verify_area(WRITE, out, &pos, len); + if (unlikely(ret < 0)) + return ret; + + return out->f_op->splice_write(pipe, out, len, flags); +} + +static long do_splice_to(struct file *in, struct inode *pipe, size_t len, + unsigned int flags) +{ + loff_t pos, isize, left; + int ret; + + if (!in->f_op || !in->f_op->splice_read) + return -EINVAL; + + if (!(in->f_mode & FMODE_READ)) + return -EBADF; + + pos = in->f_pos; + ret = rw_verify_area(READ, in, &pos, len); + if (unlikely(ret < 0)) + return ret; + + isize = i_size_read(in->f_mapping->host); + if (unlikely(in->f_pos >= isize)) + return 0; + + left = isize - in->f_pos; + if (left < len) + len = left; + + return in->f_op->splice_read(in, pipe, len, flags); +} + +static long do_splice(struct file *in, struct file *out, size_t len, + unsigned int flags) +{ + struct inode *pipe; + + pipe = in->f_dentry->d_inode; + if (pipe->i_pipe) + return do_splice_from(pipe, out, len, flags); + + pipe = out->f_dentry->d_inode; + if (pipe->i_pipe) + return do_splice_to(in, pipe, len, flags); + + return -EINVAL; +} + +asmlinkage long sys_splice(int fdin, int fdout, size_t len, unsigned int flags) +{ + long error; + struct file *in, *out; + int fput_in, fput_out; + + if (unlikely(!len)) + return 0; + + error = -EBADF; + in = fget_light(fdin, &fput_in); + if (in) { + if (in->f_mode & FMODE_READ) { + out = fget_light(fdout, &fput_out); + if (out) { + if (out->f_mode & FMODE_WRITE) + error = do_splice(in, out, len, flags); + fput_light(out, fput_out); + } + } + + fput_light(in, fput_in); + } + + return error; +} diff --git a/include/asm-i386/unistd.h b/include/asm-i386/unistd.h index 014e3562895..789e9bdd0a4 100644 --- a/include/asm-i386/unistd.h +++ b/include/asm-i386/unistd.h @@ -318,8 +318,9 @@ #define __NR_unshare 310 #define __NR_set_robust_list 311 #define __NR_get_robust_list 312 +#define __NR_sys_splice 313 -#define NR_syscalls 313 +#define NR_syscalls 314 /* * user-visible error numbers are in the range -1 - -128: see diff --git a/include/asm-ia64/unistd.h b/include/asm-ia64/unistd.h index 019956c613e..36070c1014d 100644 --- a/include/asm-ia64/unistd.h +++ b/include/asm-ia64/unistd.h @@ -285,12 +285,13 @@ #define __NR_faccessat 1293 /* 1294, 1295 reserved for pselect/ppoll */ #define __NR_unshare 1296 +#define __NR_splice 1297 #ifdef __KERNEL__ #include -#define NR_syscalls 273 /* length of syscall table */ +#define NR_syscalls 274 /* length of syscall table */ #define __ARCH_WANT_SYS_RT_SIGACTION diff --git a/include/asm-powerpc/unistd.h b/include/asm-powerpc/unistd.h index 1e990747dce..536ba087305 100644 --- a/include/asm-powerpc/unistd.h +++ b/include/asm-powerpc/unistd.h @@ -301,8 +301,9 @@ #define __NR_pselect6 280 #define __NR_ppoll 281 #define __NR_unshare 282 +#define __NR_splice 283 -#define __NR_syscalls 283 +#define __NR_syscalls 284 #ifdef __KERNEL__ #define __NR__exit __NR_exit diff --git a/include/asm-x86_64/unistd.h b/include/asm-x86_64/unistd.h index fcc51635308..f21ff2c1e96 100644 --- a/include/asm-x86_64/unistd.h +++ b/include/asm-x86_64/unistd.h @@ -609,8 +609,10 @@ __SYSCALL(__NR_unshare, sys_unshare) __SYSCALL(__NR_set_robust_list, sys_set_robust_list) #define __NR_get_robust_list 274 __SYSCALL(__NR_get_robust_list, sys_get_robust_list) +#define __NR_splice 275 +__SYSCALL(__NR_splice, sys_splice) -#define __NR_syscall_max __NR_get_robust_list +#define __NR_syscall_max __NR_splice #ifndef __NO_STUBS diff --git a/include/linux/fs.h b/include/linux/fs.h index 408fe89498f..20fa5f6d726 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1032,6 +1032,8 @@ struct file_operations { int (*check_flags)(int); int (*dir_notify)(struct file *filp, unsigned long arg); int (*flock) (struct file *, int, struct file_lock *); + ssize_t (*splice_write)(struct inode *, struct file *, size_t, unsigned int); + ssize_t (*splice_read)(struct file *, struct inode *, size_t, unsigned int); }; struct inode_operations { @@ -1609,6 +1611,8 @@ extern ssize_t generic_file_sendfile(struct file *, loff_t *, size_t, read_actor extern void do_generic_mapping_read(struct address_space *mapping, struct file_ra_state *, struct file *, loff_t *, read_descriptor_t *, read_actor_t); +extern ssize_t generic_file_splice_read(struct file *, struct inode *, size_t, unsigned int); +extern ssize_t generic_file_splice_write(struct inode *, struct file *, size_t, unsigned int); extern void file_ra_state_init(struct file_ra_state *ra, struct address_space *mapping); extern ssize_t generic_file_readv(struct file *filp, const struct iovec *iov, diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index e487e3b60f6..e78ffc7d5b5 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -569,5 +569,7 @@ asmlinkage long compat_sys_newfstatat(unsigned int dfd, char __user * filename, asmlinkage long compat_sys_openat(unsigned int dfd, const char __user *filename, int flags, int mode); asmlinkage long sys_unshare(unsigned long unshare_flags); +asmlinkage long sys_splice(int fdin, int fdout, size_t len, + unsigned int flags); #endif diff --git a/net/socket.c b/net/socket.c index fcd77eac0cc..b13042f68c0 100644 --- a/net/socket.c +++ b/net/socket.c @@ -119,6 +119,9 @@ static ssize_t sock_writev(struct file *file, const struct iovec *vector, static ssize_t sock_sendpage(struct file *file, struct page *page, int offset, size_t size, loff_t *ppos, int more); +extern ssize_t generic_splice_sendpage(struct inode *inode, struct file *out, + size_t len, unsigned int flags); + /* * Socket files have a set of 'special' operations as well as the generic file ones. These don't appear @@ -141,7 +144,8 @@ static struct file_operations socket_file_ops = { .fasync = sock_fasync, .readv = sock_readv, .writev = sock_writev, - .sendpage = sock_sendpage + .sendpage = sock_sendpage, + .splice_write = generic_splice_sendpage, }; /* -- cgit v1.2.3-70-g09d2 From f19180056ea09ec6a5d32e741234451a1e6eba4d Mon Sep 17 00:00:00 2001 From: "Zhang, Yanmin" Date: Mon, 27 Feb 2006 11:37:45 +0800 Subject: [IA64] Export cpu cache info by sysfs The patch exports 8 attributes of cpu cache info under /sys/devices/system/cpu/cpuX/cache/indexX: 1) level 2) type 3) coherency_line_size 4) ways_of_associativity 5) size 6) shared_cpu_map 7) attributes 8) number_of_sets: number_of_sets=size/ways_of_associativity/coherency_line_size. Signed-off-by: Zhang Yanmin Signed-off-by: Tony Luck --- arch/ia64/kernel/topology.c | 367 ++++++++++++++++++++++++++++++++++++++++++++ include/asm-ia64/pal.h | 28 ++++ 2 files changed, 395 insertions(+) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/topology.c b/arch/ia64/kernel/topology.c index 3b6fd798c4d..b47476d655f 100644 --- a/arch/ia64/kernel/topology.c +++ b/arch/ia64/kernel/topology.c @@ -9,6 +9,8 @@ * 2002/08/07 Erich Focht * Populate cpu entries in sysfs for non-numa systems as well * Intel Corporation - Ashok Raj + * 02/27/2006 Zhang, Yanmin + * Populate cpu cache entries in sysfs for cpu cache info */ #include @@ -19,6 +21,7 @@ #include #include #include +#include #include #include #include @@ -101,3 +104,367 @@ out: } subsys_initcall(topology_init); + + +/* + * Export cpu cache information through sysfs + */ + +/* + * A bunch of string array to get pretty printing + */ +static const char *cache_types[] = { + "", /* not used */ + "Instruction", + "Data", + "Unified" /* unified */ +}; + +static const char *cache_mattrib[]={ + "WriteThrough", + "WriteBack", + "", /* reserved */ + "" /* reserved */ +}; + +struct cache_info { + pal_cache_config_info_t cci; + cpumask_t shared_cpu_map; + int level; + int type; + struct kobject kobj; +}; + +struct cpu_cache_info { + struct cache_info *cache_leaves; + int num_cache_leaves; + struct kobject kobj; +}; + +static struct cpu_cache_info all_cpu_cache_info[NR_CPUS]; +#define LEAF_KOBJECT_PTR(x,y) (&all_cpu_cache_info[x].cache_leaves[y]) + +#ifdef CONFIG_SMP +static void cache_shared_cpu_map_setup( unsigned int cpu, + struct cache_info * this_leaf) +{ + pal_cache_shared_info_t csi; + int num_shared, i = 0; + unsigned int j; + + if (cpu_data(cpu)->threads_per_core <= 1 && + cpu_data(cpu)->cores_per_socket <= 1) { + cpu_set(cpu, this_leaf->shared_cpu_map); + return; + } + + if (ia64_pal_cache_shared_info(this_leaf->level, + this_leaf->type, + 0, + &csi) != PAL_STATUS_SUCCESS) + return; + + num_shared = (int) csi.num_shared; + do { + for_each_cpu(j) + if (cpu_data(cpu)->socket_id == cpu_data(j)->socket_id + && cpu_data(j)->core_id == csi.log1_cid + && cpu_data(j)->thread_id == csi.log1_tid) + cpu_set(j, this_leaf->shared_cpu_map); + + i++; + } while (i < num_shared && + ia64_pal_cache_shared_info(this_leaf->level, + this_leaf->type, + i, + &csi) == PAL_STATUS_SUCCESS); +} +#else +static void cache_shared_cpu_map_setup(unsigned int cpu, + struct cache_info * this_leaf) +{ + cpu_set(cpu, this_leaf->shared_cpu_map); + return; +} +#endif + +static ssize_t show_coherency_line_size(struct cache_info *this_leaf, + char *buf) +{ + return sprintf(buf, "%u\n", 1 << this_leaf->cci.pcci_line_size); +} + +static ssize_t show_ways_of_associativity(struct cache_info *this_leaf, + char *buf) +{ + return sprintf(buf, "%u\n", this_leaf->cci.pcci_assoc); +} + +static ssize_t show_attributes(struct cache_info *this_leaf, char *buf) +{ + return sprintf(buf, + "%s\n", + cache_mattrib[this_leaf->cci.pcci_cache_attr]); +} + +static ssize_t show_size(struct cache_info *this_leaf, char *buf) +{ + return sprintf(buf, "%uK\n", this_leaf->cci.pcci_cache_size / 1024); +} + +static ssize_t show_number_of_sets(struct cache_info *this_leaf, char *buf) +{ + unsigned number_of_sets = this_leaf->cci.pcci_cache_size; + number_of_sets /= this_leaf->cci.pcci_assoc; + number_of_sets /= 1 << this_leaf->cci.pcci_line_size; + + return sprintf(buf, "%u\n", number_of_sets); +} + +static ssize_t show_shared_cpu_map(struct cache_info *this_leaf, char *buf) +{ + ssize_t len; + cpumask_t shared_cpu_map; + + cpus_and(shared_cpu_map, this_leaf->shared_cpu_map, cpu_online_map); + len = cpumask_scnprintf(buf, NR_CPUS+1, shared_cpu_map); + len += sprintf(buf+len, "\n"); + return len; +} + +static ssize_t show_type(struct cache_info *this_leaf, char *buf) +{ + int type = this_leaf->type + this_leaf->cci.pcci_unified; + return sprintf(buf, "%s\n", cache_types[type]); +} + +static ssize_t show_level(struct cache_info *this_leaf, char *buf) +{ + return sprintf(buf, "%u\n", this_leaf->level); +} + +struct cache_attr { + struct attribute attr; + ssize_t (*show)(struct cache_info *, char *); + ssize_t (*store)(struct cache_info *, const char *, size_t count); +}; + +#ifdef define_one_ro + #undef define_one_ro +#endif +#define define_one_ro(_name) \ + static struct cache_attr _name = \ +__ATTR(_name, 0444, show_##_name, NULL) + +define_one_ro(level); +define_one_ro(type); +define_one_ro(coherency_line_size); +define_one_ro(ways_of_associativity); +define_one_ro(size); +define_one_ro(number_of_sets); +define_one_ro(shared_cpu_map); +define_one_ro(attributes); + +static struct attribute * cache_default_attrs[] = { + &type.attr, + &level.attr, + &coherency_line_size.attr, + &ways_of_associativity.attr, + &attributes.attr, + &size.attr, + &number_of_sets.attr, + &shared_cpu_map.attr, + NULL +}; + +#define to_object(k) container_of(k, struct cache_info, kobj) +#define to_attr(a) container_of(a, struct cache_attr, attr) + +static ssize_t cache_show(struct kobject * kobj, struct attribute * attr, char * buf) +{ + struct cache_attr *fattr = to_attr(attr); + struct cache_info *this_leaf = to_object(kobj); + ssize_t ret; + + ret = fattr->show ? fattr->show(this_leaf, buf) : 0; + return ret; +} + +static struct sysfs_ops cache_sysfs_ops = { + .show = cache_show +}; + +static struct kobj_type cache_ktype = { + .sysfs_ops = &cache_sysfs_ops, + .default_attrs = cache_default_attrs, +}; + +static struct kobj_type cache_ktype_percpu_entry = { + .sysfs_ops = &cache_sysfs_ops, +}; + +static void __cpuinit cpu_cache_sysfs_exit(unsigned int cpu) +{ + if (all_cpu_cache_info[cpu].cache_leaves) { + kfree(all_cpu_cache_info[cpu].cache_leaves); + all_cpu_cache_info[cpu].cache_leaves = NULL; + } + all_cpu_cache_info[cpu].num_cache_leaves = 0; + memset(&all_cpu_cache_info[cpu].kobj, 0, sizeof(struct kobject)); + + return; +} + +static int __cpuinit cpu_cache_sysfs_init(unsigned int cpu) +{ + u64 i, levels, unique_caches; + pal_cache_config_info_t cci; + int j; + s64 status; + struct cache_info *this_cache; + int num_cache_leaves = 0; + + if ((status = ia64_pal_cache_summary(&levels, &unique_caches)) != 0) { + printk(KERN_ERR "ia64_pal_cache_summary=%ld\n", status); + return -1; + } + + this_cache=kzalloc(sizeof(struct cache_info)*unique_caches, + GFP_KERNEL); + if (this_cache == NULL) + return -ENOMEM; + + for (i=0; i < levels; i++) { + for (j=2; j >0 ; j--) { + if ((status=ia64_pal_cache_config_info(i,j, &cci)) != + PAL_STATUS_SUCCESS) + continue; + + this_cache[num_cache_leaves].cci = cci; + this_cache[num_cache_leaves].level = i + 1; + this_cache[num_cache_leaves].type = j; + + cache_shared_cpu_map_setup(cpu, + &this_cache[num_cache_leaves]); + num_cache_leaves ++; + } + } + + all_cpu_cache_info[cpu].cache_leaves = this_cache; + all_cpu_cache_info[cpu].num_cache_leaves = num_cache_leaves; + + memset(&all_cpu_cache_info[cpu].kobj, 0, sizeof(struct kobject)); + + return 0; +} + +/* Add cache interface for CPU device */ +static int __cpuinit cache_add_dev(struct sys_device * sys_dev) +{ + unsigned int cpu = sys_dev->id; + unsigned long i, j; + struct cache_info *this_object; + int retval = 0; + cpumask_t oldmask; + + if (all_cpu_cache_info[cpu].kobj.parent) + return 0; + + oldmask = current->cpus_allowed; + retval = set_cpus_allowed(current, cpumask_of_cpu(cpu)); + if (unlikely(retval)) + return retval; + + retval = cpu_cache_sysfs_init(cpu); + set_cpus_allowed(current, oldmask); + if (unlikely(retval < 0)) + return retval; + + all_cpu_cache_info[cpu].kobj.parent = &sys_dev->kobj; + kobject_set_name(&all_cpu_cache_info[cpu].kobj, "%s", "cache"); + all_cpu_cache_info[cpu].kobj.ktype = &cache_ktype_percpu_entry; + retval = kobject_register(&all_cpu_cache_info[cpu].kobj); + + for (i = 0; i < all_cpu_cache_info[cpu].num_cache_leaves; i++) { + this_object = LEAF_KOBJECT_PTR(cpu,i); + this_object->kobj.parent = &all_cpu_cache_info[cpu].kobj; + kobject_set_name(&(this_object->kobj), "index%1lu", i); + this_object->kobj.ktype = &cache_ktype; + retval = kobject_register(&(this_object->kobj)); + if (unlikely(retval)) { + for (j = 0; j < i; j++) { + kobject_unregister( + &(LEAF_KOBJECT_PTR(cpu,j)->kobj)); + } + kobject_unregister(&all_cpu_cache_info[cpu].kobj); + cpu_cache_sysfs_exit(cpu); + break; + } + } + return retval; +} + +/* Remove cache interface for CPU device */ +static int __cpuinit cache_remove_dev(struct sys_device * sys_dev) +{ + unsigned int cpu = sys_dev->id; + unsigned long i; + + for (i = 0; i < all_cpu_cache_info[cpu].num_cache_leaves; i++) + kobject_unregister(&(LEAF_KOBJECT_PTR(cpu,i)->kobj)); + + if (all_cpu_cache_info[cpu].kobj.parent) { + kobject_unregister(&all_cpu_cache_info[cpu].kobj); + memset(&all_cpu_cache_info[cpu].kobj, + 0, + sizeof(struct kobject)); + } + + cpu_cache_sysfs_exit(cpu); + + return 0; +} + +/* + * When a cpu is hot-plugged, do a check and initiate + * cache kobject if necessary + */ +static int __cpuinit cache_cpu_callback(struct notifier_block *nfb, + unsigned long action, void *hcpu) +{ + unsigned int cpu = (unsigned long)hcpu; + struct sys_device *sys_dev; + + sys_dev = get_cpu_sysdev(cpu); + switch (action) { + case CPU_ONLINE: + cache_add_dev(sys_dev); + break; + case CPU_DEAD: + cache_remove_dev(sys_dev); + break; + } + return NOTIFY_OK; +} + +static struct notifier_block cache_cpu_notifier = +{ + .notifier_call = cache_cpu_callback +}; + +static int __cpuinit cache_sysfs_init(void) +{ + int i; + + for_each_online_cpu(i) { + cache_cpu_callback(&cache_cpu_notifier, CPU_ONLINE, + (void *)(long)i); + } + + register_cpu_notifier(&cache_cpu_notifier); + + return 0; +} + +device_initcall(cache_sysfs_init); + diff --git a/include/asm-ia64/pal.h b/include/asm-ia64/pal.h index 4e7e6f23b08..5d229c5953e 100644 --- a/include/asm-ia64/pal.h +++ b/include/asm-ia64/pal.h @@ -68,6 +68,7 @@ #define PAL_SHUTDOWN 40 /* enter processor shutdown state */ #define PAL_PREFETCH_VISIBILITY 41 /* Make Processor Prefetches Visible */ #define PAL_LOGICAL_TO_PHYSICAL 42 /* returns information on logical to physical processor mapping */ +#define PAL_CACHE_SHARED_INFO 43 /* returns information on caches shared by logical processor */ #define PAL_COPY_PAL 256 /* relocate PAL procedures and PAL PMI */ #define PAL_HALT_INFO 257 /* return the low power capabilities of processor */ @@ -1647,6 +1648,33 @@ ia64_pal_logical_to_phys(u64 proc_number, pal_logical_to_physical_t *mapping) return iprv.status; } + +typedef struct pal_cache_shared_info_s +{ + u64 num_shared; + pal_proc_n_log_info1_t ppli1; + pal_proc_n_log_info2_t ppli2; +} pal_cache_shared_info_t; + +/* Get information on logical to physical processor mappings. */ +static inline s64 +ia64_pal_cache_shared_info(u64 level, + u64 type, + u64 proc_number, + pal_cache_shared_info_t *info) +{ + struct ia64_pal_retval iprv; + + PAL_CALL(iprv, PAL_CACHE_SHARED_INFO, level, type, proc_number); + + if (iprv.status == PAL_STATUS_SUCCESS) { + info->num_shared = iprv.v0; + info->ppli1.ppli1_data = iprv.v1; + info->ppli2.ppli2_data = iprv.v2; + } + + return iprv.status; +} #endif /* __ASSEMBLY__ */ #endif /* _ASM_IA64_PAL_H */ -- cgit v1.2.3-70-g09d2 From 2ab9391dea6e36fed13443c29bf97d3be05f5289 Mon Sep 17 00:00:00 2001 From: Tony Luck Date: Fri, 31 Mar 2006 10:28:29 -0800 Subject: [IA64] Avoid "u64 foo : 32;" for gcc3 vs. gcc4 compatibility gcc3 thinks that a 32-bit field of a u64 type is itself a u64, so should be printed with "%ld". gcc4 thinks it needs just "%d". Make both versions happy by avoiding this construct. Signed-off-by: Tony Luck --- arch/ia64/kernel/palinfo.c | 8 ++++---- arch/ia64/kernel/time.c | 2 +- include/asm-ia64/pal.h | 6 +++--- 3 files changed, 8 insertions(+), 8 deletions(-) (limited to 'arch/ia64') diff --git a/arch/ia64/kernel/palinfo.c b/arch/ia64/kernel/palinfo.c index 89faa603c6b..6386f63c413 100644 --- a/arch/ia64/kernel/palinfo.c +++ b/arch/ia64/kernel/palinfo.c @@ -240,7 +240,7 @@ cache_info(char *page) } p += sprintf(p, "%s Cache level %lu:\n" - "\tSize : %lu bytes\n" + "\tSize : %u bytes\n" "\tAttributes : ", cache_types[j+cci.pcci_unified], i+1, cci.pcci_cache_size); @@ -648,9 +648,9 @@ frequency_info(char *page) if (ia64_pal_freq_ratios(&proc, &bus, &itc) != 0) return 0; p += sprintf(p, - "Processor/Clock ratio : %ld/%ld\n" - "Bus/Clock ratio : %ld/%ld\n" - "ITC/Clock ratio : %ld/%ld\n", + "Processor/Clock ratio : %d/%d\n" + "Bus/Clock ratio : %d/%d\n" + "ITC/Clock ratio : %d/%d\n", proc.num, proc.den, bus.num, bus.den, itc.num, itc.den); return p - page; diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c index ac167436e93..49958904045 100644 --- a/arch/ia64/kernel/time.c +++ b/arch/ia64/kernel/time.c @@ -188,7 +188,7 @@ ia64_init_itm (void) itc_freq = (platform_base_freq*itc_ratio.num)/itc_ratio.den; local_cpu_data->itm_delta = (itc_freq + HZ/2) / HZ; - printk(KERN_DEBUG "CPU %d: base freq=%lu.%03luMHz, ITC ratio=%lu/%lu, " + printk(KERN_DEBUG "CPU %d: base freq=%lu.%03luMHz, ITC ratio=%u/%u, " "ITC freq=%lu.%03luMHz", smp_processor_id(), platform_base_freq / 1000000, (platform_base_freq / 1000) % 1000, itc_ratio.num, itc_ratio.den, itc_freq / 1000000, (itc_freq / 1000) % 1000); diff --git a/include/asm-ia64/pal.h b/include/asm-ia64/pal.h index 5d229c5953e..37e52a2836b 100644 --- a/include/asm-ia64/pal.h +++ b/include/asm-ia64/pal.h @@ -131,7 +131,7 @@ typedef u64 pal_cache_line_state_t; #define PAL_CACHE_LINE_STATE_MODIFIED 3 /* Modified */ typedef struct pal_freq_ratio { - u64 den : 32, num : 32; /* numerator & denominator */ + u32 den, num; /* numerator & denominator */ } itc_ratio, proc_ratio; typedef union pal_cache_config_info_1_s { @@ -152,10 +152,10 @@ typedef union pal_cache_config_info_1_s { typedef union pal_cache_config_info_2_s { struct { - u64 cache_size : 32, /*cache size in bytes*/ + u32 cache_size; /*cache size in bytes*/ - alias_boundary : 8, /* 39-32 aliased addr + u32 alias_boundary : 8, /* 39-32 aliased addr * separation for max * performance. */ -- cgit v1.2.3-70-g09d2