diff options
Diffstat (limited to 'arch/ia64/kernel')
34 files changed, 1986 insertions, 1068 deletions
diff --git a/arch/ia64/kernel/Makefile b/arch/ia64/kernel/Makefile index 33e5a598672d..13fd10e8699e 100644 --- a/arch/ia64/kernel/Makefile +++ b/arch/ia64/kernel/Makefile @@ -6,7 +6,7 @@ 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 \ + salinfo.o setup.o signal.o sys_ia64.o time.o traps.o unaligned.o \ unwind.o mca.o mca_asm.o topology.o obj-$(CONFIG_IA64_BRL_EMU) += brl_emu.o diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c index 78f28d825f30..c7467f863c7a 100644 --- a/arch/ia64/kernel/acpi.c +++ b/arch/ia64/kernel/acpi.c @@ -423,6 +423,7 @@ static u32 __devinitdata pxm_flag[PXM_FLAG_LEN]; #define pxm_bit_set(bit) (set_bit(bit,(void *)pxm_flag)) #define pxm_bit_test(bit) (test_bit(bit,(void *)pxm_flag)) static struct acpi_table_slit __initdata *slit_table; +cpumask_t early_cpu_possible_map = CPU_MASK_NONE; static int get_processor_proximity_domain(struct acpi_srat_cpu_affinity *pa) { @@ -482,6 +483,7 @@ acpi_numa_processor_affinity_init(struct acpi_srat_cpu_affinity *pa) (pa->apic_id << 8) | (pa->local_sapic_eid); /* nid should be overridden as logical node id later */ node_cpuid[srat_num_cpus].nid = pxm; + cpu_set(srat_num_cpus, early_cpu_possible_map); srat_num_cpus++; } @@ -559,7 +561,7 @@ void __init acpi_numa_arch_fixup(void) } /* set logical node id in cpu structure */ - for (i = 0; i < srat_num_cpus; i++) + for_each_possible_early_cpu(i) node_cpuid[i].nid = pxm_to_node(node_cpuid[i].nid); printk(KERN_INFO "Number of logical nodes in system = %d\n", diff --git a/arch/ia64/kernel/asm-offsets.c b/arch/ia64/kernel/asm-offsets.c index 0aebc6f79e95..230a6f92367f 100644 --- a/arch/ia64/kernel/asm-offsets.c +++ b/arch/ia64/kernel/asm-offsets.c @@ -7,6 +7,7 @@ #define ASM_OFFSETS_C 1 #include <linux/sched.h> +#include <linux/pid.h> #include <linux/clocksource.h> #include <asm-ia64/processor.h> @@ -34,17 +35,29 @@ void foo(void) DEFINE(SIGFRAME_SIZE, sizeof (struct sigframe)); DEFINE(UNW_FRAME_INFO_SIZE, sizeof (struct unw_frame_info)); + BUILD_BUG_ON(sizeof(struct upid) != 32); + DEFINE(IA64_UPID_SHIFT, 5); + BLANK(); DEFINE(TI_FLAGS, offsetof(struct thread_info, flags)); DEFINE(TI_CPU, offsetof(struct thread_info, cpu)); DEFINE(TI_PRE_COUNT, offsetof(struct thread_info, preempt_count)); +#ifdef CONFIG_VIRT_CPU_ACCOUNTING + DEFINE(TI_AC_STAMP, offsetof(struct thread_info, ac_stamp)); + DEFINE(TI_AC_LEAVE, offsetof(struct thread_info, ac_leave)); + DEFINE(TI_AC_STIME, offsetof(struct thread_info, ac_stime)); + DEFINE(TI_AC_UTIME, offsetof(struct thread_info, ac_utime)); +#endif BLANK(); DEFINE(IA64_TASK_BLOCKED_OFFSET,offsetof (struct task_struct, blocked)); DEFINE(IA64_TASK_CLEAR_CHILD_TID_OFFSET,offsetof (struct task_struct, clear_child_tid)); DEFINE(IA64_TASK_GROUP_LEADER_OFFSET, offsetof (struct task_struct, group_leader)); + DEFINE(IA64_TASK_TGIDLINK_OFFSET, offsetof (struct task_struct, pids[PIDTYPE_PID].pid)); + DEFINE(IA64_PID_LEVEL_OFFSET, offsetof (struct pid, level)); + DEFINE(IA64_PID_UPID_OFFSET, offsetof (struct pid, numbers[0])); DEFINE(IA64_TASK_PENDING_OFFSET,offsetof (struct task_struct, pending)); DEFINE(IA64_TASK_PID_OFFSET, offsetof (struct task_struct, pid)); DEFINE(IA64_TASK_REAL_PARENT_OFFSET, offsetof (struct task_struct, real_parent)); diff --git a/arch/ia64/kernel/crash.c b/arch/ia64/kernel/crash.c index f1cf2df97a2d..f065093f8e9b 100644 --- a/arch/ia64/kernel/crash.c +++ b/arch/ia64/kernel/crash.c @@ -24,6 +24,7 @@ int kdump_status[NR_CPUS]; static atomic_t kdump_cpu_frozen; atomic_t kdump_in_progress; static int kdump_on_init = 1; +static int kdump_on_fatal_mca = 1; static inline Elf64_Word *append_elf_note(Elf64_Word *buf, char *name, unsigned type, void *data, @@ -118,6 +119,7 @@ machine_crash_shutdown(struct pt_regs *pt) static void machine_kdump_on_init(void) { + crash_save_vmcoreinfo(); local_irq_disable(); kexec_disable_iosapic(); machine_kexec(ia64_kimage); @@ -148,14 +150,14 @@ kdump_init_notifier(struct notifier_block *self, unsigned long val, void *data) struct ia64_mca_notify_die *nd; struct die_args *args = data; - if (!kdump_on_init) + if (!kdump_on_init && !kdump_on_fatal_mca) return NOTIFY_DONE; if (!ia64_kimage) { if (val == DIE_INIT_MONARCH_LEAVE) ia64_mca_printk(KERN_NOTICE "%s: kdump not configured\n", - __FUNCTION__); + __func__); return NOTIFY_DONE; } @@ -173,32 +175,38 @@ kdump_init_notifier(struct notifier_block *self, unsigned long val, void *data) return NOTIFY_DONE; switch (val) { - case DIE_INIT_MONARCH_PROCESS: + case DIE_INIT_MONARCH_PROCESS: + if (kdump_on_init) { atomic_set(&kdump_in_progress, 1); *(nd->monarch_cpu) = -1; - break; - case DIE_INIT_MONARCH_LEAVE: + } + break; + case DIE_INIT_MONARCH_LEAVE: + if (kdump_on_init) machine_kdump_on_init(); - break; - case DIE_INIT_SLAVE_LEAVE: - if (atomic_read(&kdump_in_progress)) - unw_init_running(kdump_cpu_freeze, NULL); - break; - case DIE_MCA_RENDZVOUS_LEAVE: - if (atomic_read(&kdump_in_progress)) - unw_init_running(kdump_cpu_freeze, NULL); - break; - case DIE_MCA_MONARCH_LEAVE: - /* die_register->signr indicate if MCA is recoverable */ - if (!args->signr) - machine_kdump_on_init(); - break; + break; + case DIE_INIT_SLAVE_LEAVE: + if (atomic_read(&kdump_in_progress)) + unw_init_running(kdump_cpu_freeze, NULL); + break; + case DIE_MCA_RENDZVOUS_LEAVE: + if (atomic_read(&kdump_in_progress)) + unw_init_running(kdump_cpu_freeze, NULL); + break; + case DIE_MCA_MONARCH_LEAVE: + /* *(nd->data) indicate if MCA is recoverable */ + if (kdump_on_fatal_mca && !(*(nd->data))) { + atomic_set(&kdump_in_progress, 1); + *(nd->monarch_cpu) = -1; + machine_kdump_on_init(); + } + break; } return NOTIFY_DONE; } #ifdef CONFIG_SYSCTL -static ctl_table kdump_on_init_table[] = { +static ctl_table kdump_ctl_table[] = { { .ctl_name = CTL_UNNUMBERED, .procname = "kdump_on_init", @@ -207,6 +215,14 @@ static ctl_table kdump_on_init_table[] = { .mode = 0644, .proc_handler = &proc_dointvec, }, + { + .ctl_name = CTL_UNNUMBERED, + .procname = "kdump_on_fatal_mca", + .data = &kdump_on_fatal_mca, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, { .ctl_name = 0 } }; @@ -215,7 +231,7 @@ static ctl_table sys_table[] = { .ctl_name = CTL_KERN, .procname = "kernel", .mode = 0555, - .child = kdump_on_init_table, + .child = kdump_ctl_table, }, { .ctl_name = 0 } }; diff --git a/arch/ia64/kernel/efi.c b/arch/ia64/kernel/efi.c index 919070a9aed7..d45f215bc8fc 100644 --- a/arch/ia64/kernel/efi.c +++ b/arch/ia64/kernel/efi.c @@ -37,6 +37,7 @@ #include <asm/pgtable.h> #include <asm/processor.h> #include <asm/mca.h> +#include <asm/tlbflush.h> #define EFI_DEBUG 0 @@ -379,8 +380,8 @@ efi_get_pal_addr (void) * a dedicated ITR for the PAL code. */ if ((vaddr & mask) == (KERNEL_START & mask)) { - printk(KERN_INFO "%s: no need to install ITR for " - "PAL code\n", __FUNCTION__); + printk(KERN_INFO "%s: no need to install ITR for PAL code\n", + __func__); continue; } @@ -399,10 +400,45 @@ efi_get_pal_addr (void) return __va(md->phys_addr); } printk(KERN_WARNING "%s: no PAL-code memory-descriptor found\n", - __FUNCTION__); + __func__); return NULL; } + +static u8 __init palo_checksum(u8 *buffer, u32 length) +{ + u8 sum = 0; + u8 *end = buffer + length; + + while (buffer < end) + sum = (u8) (sum + *(buffer++)); + + return sum; +} + +/* + * Parse and handle PALO table which is published at: + * http://www.dig64.org/home/DIG64_PALO_R1_0.pdf + */ +static void __init handle_palo(unsigned long palo_phys) +{ + struct palo_table *palo = __va(palo_phys); + u8 checksum; + + if (strncmp(palo->signature, PALO_SIG, sizeof(PALO_SIG) - 1)) { + printk(KERN_INFO "PALO signature incorrect.\n"); + return; + } + + checksum = palo_checksum((u8 *)palo, palo->length); + if (checksum) { + printk(KERN_INFO "PALO checksum incorrect.\n"); + return; + } + + setup_ptcg_sem(palo->max_tlb_purges, NPTCG_FROM_PALO); +} + void efi_map_pal_code (void) { @@ -432,6 +468,7 @@ efi_init (void) u64 efi_desc_size; char *cp, vendor[100] = "unknown"; int i; + unsigned long palo_phys; /* * It's too early to be able to use the standard kernel command line @@ -496,6 +533,8 @@ efi_init (void) efi.hcdp = EFI_INVALID_TABLE_ADDR; efi.uga = EFI_INVALID_TABLE_ADDR; + palo_phys = 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 = config_tables[i].table; @@ -515,10 +554,17 @@ efi_init (void) } else if (efi_guidcmp(config_tables[i].guid, HCDP_TABLE_GUID) == 0) { efi.hcdp = config_tables[i].table; printk(" HCDP=0x%lx", config_tables[i].table); + } else if (efi_guidcmp(config_tables[i].guid, + PROCESSOR_ABSTRACTION_LAYER_OVERWRITE_GUID) == 0) { + palo_phys = config_tables[i].table; + printk(" PALO=0x%lx", config_tables[i].table); } } printk("\n"); + if (palo_phys != EFI_INVALID_TABLE_ADDR) + handle_palo(palo_phys); + runtime = __va(efi.systab->runtime); efi.get_time = phys_get_time; efi.set_time = phys_set_time; @@ -543,12 +589,30 @@ efi_init (void) for (i = 0, p = efi_map_start; p < efi_map_end; ++i, p += efi_desc_size) { + const char *unit; + unsigned long size; + md = p; - printk("mem%02u: type=%u, attr=0x%lx, " - "range=[0x%016lx-0x%016lx) (%luMB)\n", + size = md->num_pages << EFI_PAGE_SHIFT; + + if ((size >> 40) > 0) { + size >>= 40; + unit = "TB"; + } else if ((size >> 30) > 0) { + size >>= 30; + unit = "GB"; + } else if ((size >> 20) > 0) { + size >>= 20; + unit = "MB"; + } else { + size >>= 10; + unit = "KB"; + } + + printk("mem%02d: type=%2u, attr=0x%016lx, " + "range=[0x%016lx-0x%016lx) (%4lu%s)\n", i, md->type, md->attribute, md->phys_addr, - md->phys_addr + efi_md_size(md), - md->num_pages >> (20 - EFI_PAGE_SHIFT)); + md->phys_addr + efi_md_size(md), size, unit); } } #endif diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S index 3c331c464b40..e49ad8c5dc69 100644 --- a/arch/ia64/kernel/entry.S +++ b/arch/ia64/kernel/entry.S @@ -570,6 +570,7 @@ GLOBAL_ENTRY(ia64_trace_syscall) br.call.sptk.many rp=syscall_trace_leave // give parent a chance to catch return value .ret3: (pUStk) cmp.eq.unc p6,p0=r0,r0 // p6 <- pUStk +(pUStk) rsm psr.i // disable interrupts br.cond.sptk .work_pending_syscall_end strace_error: @@ -710,6 +711,16 @@ ENTRY(ia64_leave_syscall) (pUStk) cmp.eq.unc p6,p0=r0,r0 // p6 <- pUStk #endif .work_processed_syscall: +#ifdef CONFIG_VIRT_CPU_ACCOUNTING + adds r2=PT(LOADRS)+16,r12 +(pUStk) mov.m r22=ar.itc // fetch time at leave + adds r18=TI_FLAGS+IA64_TASK_SIZE,r13 + ;; +(p6) ld4 r31=[r18] // load current_thread_info()->flags + ld8 r19=[r2],PT(B6)-PT(LOADRS) // load ar.rsc value for "loadrs" + adds r3=PT(AR_BSPSTORE)+16,r12 // deferred + ;; +#else adds r2=PT(LOADRS)+16,r12 adds r3=PT(AR_BSPSTORE)+16,r12 adds r18=TI_FLAGS+IA64_TASK_SIZE,r13 @@ -718,6 +729,7 @@ ENTRY(ia64_leave_syscall) ld8 r19=[r2],PT(B6)-PT(LOADRS) // load ar.rsc value for "loadrs" nop.i 0 ;; +#endif mov r16=ar.bsp // M2 get existing backing store pointer ld8 r18=[r2],PT(R9)-PT(B6) // load b6 (p6) and r15=TIF_WORK_MASK,r31 // any work other than TIF_SYSCALL_TRACE? @@ -737,12 +749,21 @@ ENTRY(ia64_leave_syscall) ld8 r29=[r2],16 // M0|1 load cr.ipsr ld8 r28=[r3],16 // M0|1 load cr.iip +#ifdef CONFIG_VIRT_CPU_ACCOUNTING +(pUStk) add r14=TI_AC_LEAVE+IA64_TASK_SIZE,r13 + ;; + ld8 r30=[r2],16 // M0|1 load cr.ifs + ld8 r25=[r3],16 // M0|1 load ar.unat +(pUStk) add r15=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13 + ;; +#else mov r22=r0 // A clear r22 ;; ld8 r30=[r2],16 // M0|1 load cr.ifs ld8 r25=[r3],16 // M0|1 load ar.unat (pUStk) add r14=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13 ;; +#endif ld8 r26=[r2],PT(B0)-PT(AR_PFS) // M0|1 load ar.pfs (pKStk) mov r22=psr // M2 read PSR now that interrupts are disabled nop 0 @@ -759,7 +780,11 @@ ENTRY(ia64_leave_syscall) ld8.fill r1=[r3],16 // M0|1 load r1 (pUStk) mov r17=1 // A ;; +#ifdef CONFIG_VIRT_CPU_ACCOUNTING +(pUStk) st1 [r15]=r17 // M2|3 +#else (pUStk) st1 [r14]=r17 // M2|3 +#endif ld8.fill r13=[r3],16 // M0|1 mov f8=f0 // F clear f8 ;; @@ -775,12 +800,22 @@ ENTRY(ia64_leave_syscall) shr.u r18=r19,16 // I0|1 get byte size of existing "dirty" partition cover // B add current frame into dirty partition & set cr.ifs ;; +#ifdef CONFIG_VIRT_CPU_ACCOUNTING + mov r19=ar.bsp // M2 get new backing store pointer + st8 [r14]=r22 // M save time at leave + mov f10=f0 // F clear f10 + + mov r22=r0 // A clear r22 + movl r14=__kernel_syscall_via_epc // X + ;; +#else mov r19=ar.bsp // M2 get new backing store pointer mov f10=f0 // F clear f10 nop.m 0 movl r14=__kernel_syscall_via_epc // X ;; +#endif mov.m ar.csd=r0 // M2 clear ar.csd mov.m ar.ccv=r0 // M2 clear ar.ccv mov b7=r14 // I0 clear b7 (hint with __kernel_syscall_via_epc) @@ -913,10 +948,18 @@ GLOBAL_ENTRY(ia64_leave_kernel) adds r16=PT(CR_IPSR)+16,r12 adds r17=PT(CR_IIP)+16,r12 +#ifdef CONFIG_VIRT_CPU_ACCOUNTING + .pred.rel.mutex pUStk,pKStk +(pKStk) mov r22=psr // M2 read PSR now that interrupts are disabled +(pUStk) mov.m r22=ar.itc // M fetch time at leave + nop.i 0 + ;; +#else (pKStk) mov r22=psr // M2 read PSR now that interrupts are disabled nop.i 0 nop.i 0 ;; +#endif ld8 r29=[r16],16 // load cr.ipsr ld8 r28=[r17],16 // load cr.iip ;; @@ -938,15 +981,37 @@ GLOBAL_ENTRY(ia64_leave_kernel) ;; ld8.fill r12=[r16],16 ld8.fill r13=[r17],16 +#ifdef CONFIG_VIRT_CPU_ACCOUNTING +(pUStk) adds r3=TI_AC_LEAVE+IA64_TASK_SIZE,r18 +#else (pUStk) adds r18=IA64_TASK_THREAD_ON_USTACK_OFFSET,r18 +#endif ;; ld8 r20=[r16],16 // ar.fpsr ld8.fill r15=[r17],16 +#ifdef CONFIG_VIRT_CPU_ACCOUNTING +(pUStk) adds r18=IA64_TASK_THREAD_ON_USTACK_OFFSET,r18 // deferred +#endif ;; ld8.fill r14=[r16],16 ld8.fill r2=[r17] (pUStk) mov r17=1 ;; +#ifdef CONFIG_VIRT_CPU_ACCOUNTING + // mmi_ : ld8 st1 shr;; mmi_ : st8 st1 shr;; + // mib : mov add br -> mib : ld8 add br + // bbb_ : br nop cover;; mbb_ : mov br cover;; + // + // no one require bsp in r16 if (pKStk) branch is selected. +(pUStk) st8 [r3]=r22 // save time at leave +(pUStk) st1 [r18]=r17 // restore current->thread.on_ustack + shr.u r18=r19,16 // get byte size of existing "dirty" partition + ;; + ld8.fill r3=[r16] // deferred + LOAD_PHYS_STACK_REG_SIZE(r17) +(pKStk) br.cond.dpnt skip_rbs_switch + mov r16=ar.bsp // get existing backing store pointer +#else ld8.fill r3=[r16] (pUStk) st1 [r18]=r17 // restore current->thread.on_ustack shr.u r18=r19,16 // get byte size of existing "dirty" partition @@ -954,6 +1019,7 @@ GLOBAL_ENTRY(ia64_leave_kernel) mov r16=ar.bsp // get existing backing store pointer LOAD_PHYS_STACK_REG_SIZE(r17) (pKStk) br.cond.dpnt skip_rbs_switch +#endif /* * Restore user backing store. diff --git a/arch/ia64/kernel/fsys.S b/arch/ia64/kernel/fsys.S index 44841971f077..c1625c7e1779 100644 --- a/arch/ia64/kernel/fsys.S +++ b/arch/ia64/kernel/fsys.S @@ -61,13 +61,29 @@ ENTRY(fsys_getpid) .prologue .altrp b6 .body + add r17=IA64_TASK_GROUP_LEADER_OFFSET,r16 + ;; + ld8 r17=[r17] // r17 = current->group_leader add r9=TI_FLAGS+IA64_TASK_SIZE,r16 ;; ld4 r9=[r9] - add r8=IA64_TASK_TGID_OFFSET,r16 + add r17=IA64_TASK_TGIDLINK_OFFSET,r17 ;; and r9=TIF_ALLWORK_MASK,r9 - ld4 r8=[r8] // r8 = current->tgid + ld8 r17=[r17] // r17 = current->group_leader->pids[PIDTYPE_PID].pid + ;; + add r8=IA64_PID_LEVEL_OFFSET,r17 + ;; + ld4 r8=[r8] // r8 = pid->level + add r17=IA64_PID_UPID_OFFSET,r17 // r17 = &pid->numbers[0] + ;; + shl r8=r8,IA64_UPID_SHIFT + ;; + add r17=r17,r8 // r17 = &pid->numbers[pid->level] + ;; + ld4 r8=[r17] // r8 = pid->numbers[pid->level].nr + ;; + mov r17=0 ;; cmp.ne p8,p0=0,r9 (p8) br.spnt.many fsys_fallback_syscall @@ -126,15 +142,25 @@ ENTRY(fsys_set_tid_address) .altrp b6 .body add r9=TI_FLAGS+IA64_TASK_SIZE,r16 + add r17=IA64_TASK_TGIDLINK_OFFSET,r16 ;; ld4 r9=[r9] tnat.z p6,p7=r32 // check argument register for being NaT + ld8 r17=[r17] // r17 = current->pids[PIDTYPE_PID].pid ;; and r9=TIF_ALLWORK_MASK,r9 - add r8=IA64_TASK_PID_OFFSET,r16 + add r8=IA64_PID_LEVEL_OFFSET,r17 add r18=IA64_TASK_CLEAR_CHILD_TID_OFFSET,r16 ;; - ld4 r8=[r8] + ld4 r8=[r8] // r8 = pid->level + add r17=IA64_PID_UPID_OFFSET,r17 // r17 = &pid->numbers[0] + ;; + shl r8=r8,IA64_UPID_SHIFT + ;; + add r17=r17,r8 // r17 = &pid->numbers[pid->level] + ;; + ld4 r8=[r17] // r8 = pid->numbers[pid->level].nr + ;; cmp.ne p8,p0=0,r9 mov r17=-1 ;; @@ -210,27 +236,25 @@ ENTRY(fsys_gettimeofday) // Note that instructions are optimized for McKinley. McKinley can // process two bundles simultaneously and therefore we continuously // try to feed the CPU two bundles and then a stop. - // - // Additional note that code has changed a lot. Optimization is TBD. - // Comments begin with "?" are maybe outdated. - tnat.nz p6,p0 = r31 // ? branch deferred to fit later bundle - mov pr = r30,0xc000 // Set predicates according to function + add r2 = TI_FLAGS+IA64_TASK_SIZE,r16 + tnat.nz p6,p0 = r31 // guard against Nat argument +(p6) br.cond.spnt.few .fail_einval movl r20 = fsyscall_gtod_data // load fsyscall gettimeofday data address ;; + ld4 r2 = [r2] // process work pending flags movl r29 = itc_jitter_data // itc_jitter add r22 = IA64_GTOD_WALL_TIME_OFFSET,r20 // wall_time - ld4 r2 = [r2] // process work pending flags - ;; -(p15) add r22 = IA64_GTOD_MONO_TIME_OFFSET,r20 // monotonic_time add r21 = IA64_CLKSRC_MMIO_OFFSET,r20 - add r19 = IA64_ITC_LASTCYCLE_OFFSET,r29 + mov pr = r30,0xc000 // Set predicates according to function + ;; and r2 = TIF_ALLWORK_MASK,r2 -(p6) br.cond.spnt.few .fail_einval // ? deferred branch + add r19 = IA64_ITC_LASTCYCLE_OFFSET,r29 +(p15) add r22 = IA64_GTOD_MONO_TIME_OFFSET,r20 // monotonic_time ;; - add r26 = IA64_CLKSRC_CYCLE_LAST_OFFSET,r20 // clksrc_cycle_last + add r26 = IA64_CLKSRC_CYCLE_LAST_OFFSET,r20 // clksrc_cycle_last cmp.ne p6, p0 = 0, r2 // Fallback if work is scheduled -(p6) br.cond.spnt.many fsys_fallback_syscall +(p6) br.cond.spnt.many fsys_fallback_syscall ;; // Begin critical section .time_redo: @@ -258,7 +282,6 @@ ENTRY(fsys_gettimeofday) (p8) mov r2 = ar.itc // CPU_TIMER. 36 clocks latency!!! (p9) ld8 r2 = [r30] // MMIO_TIMER. Could also have latency issues.. (p13) ld8 r25 = [r19] // get itc_lastcycle value - ;; // ? could be removed by moving the last add upward ld8 r9 = [r22],IA64_TIMESPEC_TV_NSEC_OFFSET // tv_sec ;; ld8 r8 = [r22],-IA64_TIMESPEC_TV_NSEC_OFFSET // tv_nsec @@ -285,13 +308,12 @@ ENTRY(fsys_gettimeofday) EX(.fail_efault, probe.w.fault r31, 3) xmpy.l f8 = f8,f7 // nsec_per_cyc*(counter-last_counter) ;; - // ? simulate tbit.nz.or p7,p0 = r28,0 getf.sig r2 = f8 mf ;; ld4 r10 = [r20] // gtod_lock.sequence shr.u r2 = r2,r23 // shift by factor - ;; // ? overloaded 3 bundles! + ;; add r8 = r8,r2 // Add xtime.nsecs cmp4.ne p7,p0 = r28,r10 (p7) br.cond.dpnt.few .time_redo // sequence number changed, redo @@ -319,9 +341,9 @@ EX(.fail_efault, probe.w.fault r31, 3) EX(.fail_efault, probe.w.fault r23, 3) // This also costs 5 cycles (p14) xmpy.hu f8 = f8, f7 // xmpy has 5 cycles latency so use it ;; - mov r8 = r0 (p14) getf.sig r2 = f8 ;; + mov r8 = r0 (p14) shr.u r21 = r2, 4 ;; EX(.fail_efault, st8 [r31] = r9) @@ -660,7 +682,11 @@ GLOBAL_ENTRY(fsys_bubble_down) nop.i 0 ;; mov ar.rsc=0 // M2 set enforced lazy mode, pl 0, LE, loadrs=0 +#ifdef CONFIG_VIRT_CPU_ACCOUNTING + mov.m r30=ar.itc // M get cycle for accounting +#else nop.m 0 +#endif nop.i 0 ;; mov r23=ar.bspstore // M2 (12 cyc) save ar.bspstore @@ -682,6 +708,28 @@ GLOBAL_ENTRY(fsys_bubble_down) cmp.ne pKStk,pUStk=r0,r0 // A set pKStk <- 0, pUStk <- 1 br.call.sptk.many b7=ia64_syscall_setup // B ;; +#ifdef CONFIG_VIRT_CPU_ACCOUNTING + // mov.m r30=ar.itc is called in advance + add r16=TI_AC_STAMP+IA64_TASK_SIZE,r2 + add r17=TI_AC_LEAVE+IA64_TASK_SIZE,r2 + ;; + ld8 r18=[r16],TI_AC_STIME-TI_AC_STAMP // time at last check in kernel + ld8 r19=[r17],TI_AC_UTIME-TI_AC_LEAVE // time at leave kernel + ;; + ld8 r20=[r16],TI_AC_STAMP-TI_AC_STIME // cumulated stime + ld8 r21=[r17] // cumulated utime + sub r22=r19,r18 // stime before leave kernel + ;; + st8 [r16]=r30,TI_AC_STIME-TI_AC_STAMP // update stamp + sub r18=r30,r19 // elapsed time in user mode + ;; + add r20=r20,r22 // sum stime + add r21=r21,r18 // sum utime + ;; + st8 [r16]=r20 // update stime + st8 [r17]=r21 // update utime + ;; +#endif mov ar.rsc=0x3 // M2 set eager mode, pl 0, LE, loadrs=0 mov rp=r14 // I0 set the real return addr and r3=_TIF_SYSCALL_TRACEAUDIT,r3 // A diff --git a/arch/ia64/kernel/head.S b/arch/ia64/kernel/head.S index d3a41d5f8d12..ddeab4e36fd5 100644 --- a/arch/ia64/kernel/head.S +++ b/arch/ia64/kernel/head.S @@ -1002,6 +1002,26 @@ GLOBAL_ENTRY(sched_clock) br.ret.sptk.many rp END(sched_clock) +#ifdef CONFIG_VIRT_CPU_ACCOUNTING +GLOBAL_ENTRY(cycle_to_cputime) + alloc r16=ar.pfs,1,0,0,0 + addl r8=THIS_CPU(cpu_info) + IA64_CPUINFO_NSEC_PER_CYC_OFFSET,r0 + ;; + ldf8 f8=[r8] + ;; + setf.sig f9=r32 + ;; + xmpy.lu f10=f9,f8 // calculate low 64 bits of 128-bit product (4 cyc) + xmpy.hu f11=f9,f8 // calculate high 64 bits of 128-bit product + ;; + getf.sig r8=f10 // (5 cyc) + getf.sig r9=f11 + ;; + shrp r8=r9,r8,IA64_NSEC_PER_CYC_SHIFT + br.ret.sptk.many rp +END(cycle_to_cputime) +#endif /* CONFIG_VIRT_CPU_ACCOUNTING */ + GLOBAL_ENTRY(start_kernel_thread) .prologue .save rp, r0 // this is the end of the call-chain diff --git a/arch/ia64/kernel/ia64_ksyms.c b/arch/ia64/kernel/ia64_ksyms.c index 8e7193d55528..6da1f20d7372 100644 --- a/arch/ia64/kernel/ia64_ksyms.c +++ b/arch/ia64/kernel/ia64_ksyms.c @@ -19,12 +19,6 @@ EXPORT_SYMBOL_GPL(empty_zero_page); EXPORT_SYMBOL(ip_fast_csum); /* hand-coded assembly */ EXPORT_SYMBOL(csum_ipv6_magic); -#include <asm/semaphore.h> -EXPORT_SYMBOL(__down); -EXPORT_SYMBOL(__down_interruptible); -EXPORT_SYMBOL(__down_trylock); -EXPORT_SYMBOL(__up); - #include <asm/page.h> EXPORT_SYMBOL(clear_page); diff --git a/arch/ia64/kernel/iosapic.c b/arch/ia64/kernel/iosapic.c index 398e2fd1cd25..082c31dcfd99 100644 --- a/arch/ia64/kernel/iosapic.c +++ b/arch/ia64/kernel/iosapic.c @@ -345,7 +345,7 @@ iosapic_set_affinity (unsigned int irq, cpumask_t mask) if (cpus_empty(mask)) return; - if (reassign_irq_vector(irq, first_cpu(mask))) + if (irq_prepare_move(irq, first_cpu(mask))) return; dest = cpu_physical_id(first_cpu(mask)); @@ -397,6 +397,7 @@ iosapic_end_level_irq (unsigned int irq) struct iosapic_rte_info *rte; int do_unmask_irq = 0; + irq_complete_move(irq); if (unlikely(irq_desc[irq].status & IRQ_MOVE_PENDING)) { do_unmask_irq = 1; mask_irq(irq); @@ -450,6 +451,7 @@ iosapic_ack_edge_irq (unsigned int irq) { irq_desc_t *idesc = irq_desc + irq; + irq_complete_move(irq); move_native_irq(irq); /* * Once we have recorded IRQ_PENDING already, we can mask the @@ -532,7 +534,7 @@ iosapic_reassign_vector (int irq) if (iosapic_intr_info[irq].count) { new_irq = create_irq(); if (new_irq < 0) - panic("%s: out of interrupt vectors!\n", __FUNCTION__); + panic("%s: out of interrupt vectors!\n", __func__); printk(KERN_INFO "Reassigning vector %d to %d\n", irq_to_vector(irq), irq_to_vector(new_irq)); memcpy(&iosapic_intr_info[new_irq], &iosapic_intr_info[irq], @@ -597,7 +599,7 @@ register_intr (unsigned int gsi, int irq, unsigned char delivery, index = find_iosapic(gsi); if (index < 0) { printk(KERN_WARNING "%s: No IOSAPIC for GSI %u\n", - __FUNCTION__, gsi); + __func__, gsi); return -ENODEV; } @@ -606,7 +608,7 @@ register_intr (unsigned int gsi, int irq, unsigned char delivery, rte = iosapic_alloc_rte(); if (!rte) { printk(KERN_WARNING "%s: cannot allocate memory\n", - __FUNCTION__); + __func__); return -ENOMEM; } @@ -623,7 +625,7 @@ register_intr (unsigned int gsi, int irq, unsigned char delivery, (info->trigger != trigger || info->polarity != polarity)){ printk (KERN_WARNING "%s: cannot override the interrupt\n", - __FUNCTION__); + __func__); return -EINVAL; } rte->refcnt++; @@ -645,7 +647,7 @@ register_intr (unsigned int gsi, int irq, unsigned char delivery, if (idesc->chip != &no_irq_type) printk(KERN_WARNING "%s: changing vector %d from %s to %s\n", - __FUNCTION__, irq_to_vector(irq), + __func__, irq_to_vector(irq), idesc->chip->name, irq_type->name); idesc->chip = irq_type; } @@ -918,7 +920,7 @@ iosapic_register_platform_intr (u32 int_type, unsigned int gsi, case ACPI_INTERRUPT_INIT: irq = create_irq(); if (irq < 0) - panic("%s: out of interrupt vectors!\n", __FUNCTION__); + panic("%s: out of interrupt vectors!\n", __func__); vector = irq_to_vector(irq); delivery = IOSAPIC_INIT; break; @@ -929,7 +931,7 @@ iosapic_register_platform_intr (u32 int_type, unsigned int gsi, mask = 1; break; default: - printk(KERN_ERR "%s: invalid int type 0x%x\n", __FUNCTION__, + printk(KERN_ERR "%s: invalid int type 0x%x\n", __func__, int_type); return -1; } @@ -994,7 +996,7 @@ iosapic_system_init (int system_pcat_compat) */ printk(KERN_INFO "%s: Disabling PC-AT compatible 8259 interrupts\n", - __FUNCTION__); + __func__); outb(0xff, 0xA1); outb(0xff, 0x21); } @@ -1009,7 +1011,7 @@ iosapic_alloc (void) if (!iosapic_lists[index].addr) return index; - printk(KERN_WARNING "%s: failed to allocate iosapic\n", __FUNCTION__); + printk(KERN_WARNING "%s: failed to allocate iosapic\n", __func__); return -1; } @@ -1107,14 +1109,14 @@ iosapic_remove (unsigned int gsi_base) index = find_iosapic(gsi_base); if (index < 0) { printk(KERN_WARNING "%s: No IOSAPIC for GSI base %u\n", - __FUNCTION__, gsi_base); + __func__, gsi_base); goto out; } if (iosapic_lists[index].rtes_inuse) { err = -EBUSY; printk(KERN_WARNING "%s: IOSAPIC for GSI base %u is busy\n", - __FUNCTION__, gsi_base); + __func__, gsi_base); goto out; } @@ -1135,7 +1137,7 @@ map_iosapic_to_node(unsigned int gsi_base, int node) index = find_iosapic(gsi_base); if (index < 0) { printk(KERN_WARNING "%s: No IOSAPIC for GSI %u\n", - __FUNCTION__, gsi_base); + __func__, gsi_base); return; } iosapic_lists[index].node = node; diff --git a/arch/ia64/kernel/irq_ia64.c b/arch/ia64/kernel/irq_ia64.c index 0b52f19ed046..5538471e8d68 100644 --- a/arch/ia64/kernel/irq_ia64.c +++ b/arch/ia64/kernel/irq_ia64.c @@ -260,6 +260,8 @@ void __setup_vector_irq(int cpu) } #if defined(CONFIG_SMP) && (defined(CONFIG_IA64_GENERIC) || defined(CONFIG_IA64_DIG)) +#define IA64_IRQ_MOVE_VECTOR IA64_DEF_FIRST_DEVICE_VECTOR + static enum vector_domain_type { VECTOR_DOMAIN_NONE, VECTOR_DOMAIN_PERCPU @@ -272,6 +274,101 @@ static cpumask_t vector_allocation_domain(int cpu) return CPU_MASK_ALL; } +static int __irq_prepare_move(int irq, int cpu) +{ + struct irq_cfg *cfg = &irq_cfg[irq]; + int vector; + cpumask_t domain; + + if (cfg->move_in_progress || cfg->move_cleanup_count) + return -EBUSY; + if (cfg->vector == IRQ_VECTOR_UNASSIGNED || !cpu_online(cpu)) + return -EINVAL; + if (cpu_isset(cpu, cfg->domain)) + return 0; + domain = vector_allocation_domain(cpu); + vector = find_unassigned_vector(domain); + if (vector < 0) + return -ENOSPC; + cfg->move_in_progress = 1; + cfg->old_domain = cfg->domain; + cfg->vector = IRQ_VECTOR_UNASSIGNED; + cfg->domain = CPU_MASK_NONE; + BUG_ON(__bind_irq_vector(irq, vector, domain)); + return 0; +} + +int irq_prepare_move(int irq, int cpu) +{ + unsigned long flags; + int ret; + + spin_lock_irqsave(&vector_lock, flags); + ret = __irq_prepare_move(irq, cpu); + spin_unlock_irqrestore(&vector_lock, flags); + return ret; +} + +void irq_complete_move(unsigned irq) +{ + struct irq_cfg *cfg = &irq_cfg[irq]; + cpumask_t cleanup_mask; + int i; + + if (likely(!cfg->move_in_progress)) + return; + + if (unlikely(cpu_isset(smp_processor_id(), cfg->old_domain))) + return; + + cpus_and(cleanup_mask, cfg->old_domain, cpu_online_map); + cfg->move_cleanup_count = cpus_weight(cleanup_mask); + for_each_cpu_mask(i, cleanup_mask) + platform_send_ipi(i, IA64_IRQ_MOVE_VECTOR, IA64_IPI_DM_INT, 0); + cfg->move_in_progress = 0; +} + +static irqreturn_t smp_irq_move_cleanup_interrupt(int irq, void *dev_id) +{ + int me = smp_processor_id(); + ia64_vector vector; + unsigned long flags; + + for (vector = IA64_FIRST_DEVICE_VECTOR; + vector < IA64_LAST_DEVICE_VECTOR; vector++) { + int irq; + struct irq_desc *desc; + struct irq_cfg *cfg; + irq = __get_cpu_var(vector_irq)[vector]; + if (irq < 0) + continue; + + desc = irq_desc + irq; + cfg = irq_cfg + irq; + spin_lock(&desc->lock); + if (!cfg->move_cleanup_count) + goto unlock; + + if (!cpu_isset(me, cfg->old_domain)) + goto unlock; + + spin_lock_irqsave(&vector_lock, flags); + __get_cpu_var(vector_irq)[vector] = -1; + cpu_clear(me, vector_table[vector]); + spin_unlock_irqrestore(&vector_lock, flags); + cfg->move_cleanup_count--; + unlock: + spin_unlock(&desc->lock); + } + return IRQ_HANDLED; +} + +static struct irqaction irq_move_irqaction = { + .handler = smp_irq_move_cleanup_interrupt, + .flags = IRQF_DISABLED, + .name = "irq_move" +}; + static int __init parse_vector_domain(char *arg) { if (!arg) @@ -303,36 +400,6 @@ void destroy_and_reserve_irq(unsigned int irq) spin_unlock_irqrestore(&vector_lock, flags); } -static int __reassign_irq_vector(int irq, int cpu) -{ - struct irq_cfg *cfg = &irq_cfg[irq]; - int vector; - cpumask_t domain; - - if (cfg->vector == IRQ_VECTOR_UNASSIGNED || !cpu_online(cpu)) - return -EINVAL; - if (cpu_isset(cpu, cfg->domain)) - return 0; - domain = vector_allocation_domain(cpu); - vector = find_unassigned_vector(domain); - if (vector < 0) - return -ENOSPC; - __clear_irq_vector(irq); - BUG_ON(__bind_irq_vector(irq, vector, domain)); - return 0; -} - -int reassign_irq_vector(int irq, int cpu) -{ - unsigned long flags; - int ret; - - spin_lock_irqsave(&vector_lock, flags); - ret = __reassign_irq_vector(irq, cpu); - spin_unlock_irqrestore(&vector_lock, flags); - return ret; -} - /* * Dynamic irq allocate and deallocation for MSI */ @@ -405,7 +472,7 @@ ia64_handle_irq (ia64_vector vector, struct pt_regs *regs) static unsigned char count; static long last_time; - if (jiffies - last_time > 5*HZ) + if (time_after(jiffies, last_time + 5 * HZ)) count = 0; if (++count < 5) { last_time = jiffies; @@ -440,7 +507,7 @@ ia64_handle_irq (ia64_vector vector, struct pt_regs *regs) if (unlikely(irq < 0)) { printk(KERN_ERR "%s: Unexpected interrupt " "vector %d on CPU %d is not mapped " - "to any IRQ!\n", __FUNCTION__, vector, + "to any IRQ!\n", __func__, vector, smp_processor_id()); } else generic_handle_irq(irq); @@ -505,7 +572,7 @@ void ia64_process_pending_intr(void) if (unlikely(irq < 0)) { printk(KERN_ERR "%s: Unexpected interrupt " "vector %d on CPU %d not being mapped " - "to any IRQ!!\n", __FUNCTION__, vector, + "to any IRQ!!\n", __func__, vector, smp_processor_id()); } else { vectors_in_migration[irq]=0; @@ -578,6 +645,13 @@ init_IRQ (void) register_percpu_irq(IA64_IPI_VECTOR, &ipi_irqaction); register_percpu_irq(IA64_IPI_RESCHEDULE, &resched_irqaction); register_percpu_irq(IA64_IPI_LOCAL_TLB_FLUSH, &tlb_irqaction); +#if defined(CONFIG_IA64_GENERIC) || defined(CONFIG_IA64_DIG) + if (vector_domain_type != VECTOR_DOMAIN_NONE) { + BUG_ON(IA64_FIRST_DEVICE_VECTOR != IA64_IRQ_MOVE_VECTOR); + IA64_FIRST_DEVICE_VECTOR++; + register_percpu_irq(IA64_IRQ_MOVE_VECTOR, &irq_move_irqaction); + } +#endif #endif #ifdef CONFIG_PERFMON pfm_init_percpu(); @@ -592,11 +666,7 @@ ia64_send_ipi (int cpu, int vector, int delivery_mode, int redirect) unsigned long ipi_data; unsigned long phys_cpu_id; -#ifdef CONFIG_SMP phys_cpu_id = cpu_physical_id(cpu); -#else - phys_cpu_id = (ia64_getreg(_IA64_REG_CR_LID) >> 16) & 0xffff; -#endif /* * cpu number is in 8bit ID and 8bit EID diff --git a/arch/ia64/kernel/ivt.S b/arch/ia64/kernel/ivt.S index 34f44d8be00d..6678c49daba3 100644 --- a/arch/ia64/kernel/ivt.S +++ b/arch/ia64/kernel/ivt.S @@ -805,8 +805,13 @@ ENTRY(break_fault) (p8) adds r28=16,r28 // A switch cr.iip to next bundle (p9) adds r8=1,r8 // A increment ei to next slot +#ifdef CONFIG_VIRT_CPU_ACCOUNTING + ;; + mov b6=r30 // I0 setup syscall handler branch reg early +#else nop.i 0 ;; +#endif mov.m r25=ar.unat // M2 (5 cyc) dep r29=r8,r29,41,2 // I0 insert new ei into cr.ipsr @@ -817,7 +822,11 @@ ENTRY(break_fault) // /////////////////////////////////////////////////////////////////////// st1 [r16]=r0 // M2|3 clear current->thread.on_ustack flag +#ifdef CONFIG_VIRT_CPU_ACCOUNTING + mov.m r30=ar.itc // M get cycle for accounting +#else mov b6=r30 // I0 setup syscall handler branch reg early +#endif cmp.eq pKStk,pUStk=r0,r17 // A were we on kernel stacks already? and r9=_TIF_SYSCALL_TRACEAUDIT,r9 // A mask trace or audit @@ -829,6 +838,30 @@ ENTRY(break_fault) cmp.eq p14,p0=r9,r0 // A are syscalls being traced/audited? br.call.sptk.many b7=ia64_syscall_setup // B 1: +#ifdef CONFIG_VIRT_CPU_ACCOUNTING + // mov.m r30=ar.itc is called in advance, and r13 is current + add r16=TI_AC_STAMP+IA64_TASK_SIZE,r13 // A + add r17=TI_AC_LEAVE+IA64_TASK_SIZE,r13 // A +(pKStk) br.cond.spnt .skip_accounting // B unlikely skip + ;; + ld8 r18=[r16],TI_AC_STIME-TI_AC_STAMP // M get last stamp + ld8 r19=[r17],TI_AC_UTIME-TI_AC_LEAVE // M time at leave + ;; + ld8 r20=[r16],TI_AC_STAMP-TI_AC_STIME // M cumulated stime + ld8 r21=[r17] // M cumulated utime + sub r22=r19,r18 // A stime before leave + ;; + st8 [r16]=r30,TI_AC_STIME-TI_AC_STAMP // M update stamp + sub r18=r30,r19 // A elapsed time in user + ;; + add r20=r20,r22 // A sum stime + add r21=r21,r18 // A sum utime + ;; + st8 [r16]=r20 // M update stime + st8 [r17]=r21 // M update utime + ;; +.skip_accounting: +#endif mov ar.rsc=0x3 // M2 set eager mode, pl 0, LE, loadrs=0 nop 0 bsw.1 // B (6 cyc) regs are saved, switch to bank 1 @@ -928,6 +961,7 @@ END(interrupt) * - r27: saved ar.rsc * - r28: saved cr.iip * - r29: saved cr.ipsr + * - r30: ar.itc for accounting (don't touch) * - r31: saved pr * - b0: original contents (to be saved) * On exit: @@ -1090,6 +1124,41 @@ END(dispatch_illegal_op_fault) DBG_FAULT(16) FAULT(16) +#ifdef CONFIG_VIRT_CPU_ACCOUNTING + /* + * There is no particular reason for this code to be here, other than + * that there happens to be space here that would go unused otherwise. + * If this fault ever gets "unreserved", simply moved the following + * code to a more suitable spot... + * + * account_sys_enter is called from SAVE_MIN* macros if accounting is + * enabled and if the macro is entered from user mode. + */ +ENTRY(account_sys_enter) + // mov.m r20=ar.itc is called in advance, and r13 is current + add r16=TI_AC_STAMP+IA64_TASK_SIZE,r13 + add r17=TI_AC_LEAVE+IA64_TASK_SIZE,r13 + ;; + ld8 r18=[r16],TI_AC_STIME-TI_AC_STAMP // time at last check in kernel + ld8 r19=[r17],TI_AC_UTIME-TI_AC_LEAVE // time at left from kernel + ;; + ld8 r23=[r16],TI_AC_STAMP-TI_AC_STIME // cumulated stime + ld8 r21=[r17] // cumulated utime + sub r22=r19,r18 // stime before leave kernel + ;; + st8 [r16]=r20,TI_AC_STIME-TI_AC_STAMP // update stamp + sub r18=r20,r19 // elapsed time in user mode + ;; + add r23=r23,r22 // sum stime + add r21=r21,r18 // sum utime + ;; + st8 [r16]=r23 // update stime + st8 [r17]=r21 // update utime + ;; + br.ret.sptk.many rp +END(account_sys_enter) +#endif + .org ia64_ivt+0x4400 ///////////////////////////////////////////////////////////////////////////////////////// // 0x4400 Entry 17 (size 64 bundles) Reserved diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c index b618487cdc85..233434f4f88f 100644 --- a/arch/ia64/kernel/kprobes.c +++ b/arch/ia64/kernel/kprobes.c @@ -78,6 +78,20 @@ static enum instruction_type bundle_encoding[32][3] = { { u, u, u }, /* 1F */ }; +/* Insert a long branch code */ +static void __kprobes set_brl_inst(void *from, void *to) +{ + s64 rel = ((s64) to - (s64) from) >> 4; + bundle_t *brl; + brl = (bundle_t *) ((u64) from & ~0xf); + brl->quad0.template = 0x05; /* [MLX](stop) */ + brl->quad0.slot0 = NOP_M_INST; /* nop.m 0x0 */ + brl->quad0.slot1_p0 = ((rel >> 20) & 0x7fffffffff) << 2; + brl->quad1.slot1_p1 = (((rel >> 20) & 0x7fffffffff) << 2) >> (64 - 46); + /* brl.cond.sptk.many.clr rel<<4 (qp=0) */ + brl->quad1.slot2 = BRL_INST(rel >> 59, rel & 0xfffff); +} + /* * In this function we check to see if the instruction * is IP relative instruction and update the kprobe @@ -496,6 +510,77 @@ void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri, regs->b0 = ((struct fnptr *)kretprobe_trampoline)->ip; } +/* Check the instruction in the slot is break */ +static int __kprobes __is_ia64_break_inst(bundle_t *bundle, uint slot) +{ + unsigned int major_opcode; + unsigned int template = bundle->quad0.template; + unsigned long kprobe_inst; + + /* Move to slot 2, if bundle is MLX type and kprobe slot is 1 */ + if (slot == 1 && bundle_encoding[template][1] == L) + slot++; + + /* Get Kprobe probe instruction at given slot*/ + get_kprobe_inst(bundle, slot, &kprobe_inst, &major_opcode); + + /* For break instruction, + * Bits 37:40 Major opcode to be zero + * Bits 27:32 X6 to be zero + * Bits 32:35 X3 to be zero + */ + if (major_opcode || ((kprobe_inst >> 27) & 0x1FF)) { + /* Not a break instruction */ + return 0; + } + + /* Is a break instruction */ + return 1; +} + +/* + * In this function, we check whether the target bundle modifies IP or + * it triggers an exception. If so, it cannot be boostable. + */ +static int __kprobes can_boost(bundle_t *bundle, uint slot, + unsigned long bundle_addr) +{ + unsigned int template = bundle->quad0.template; + + do { + if (search_exception_tables(bundle_addr + slot) || + __is_ia64_break_inst(bundle, slot)) + return 0; /* exception may occur in this bundle*/ + } while ((++slot) < 3); + template &= 0x1e; + if (template >= 0x10 /* including B unit */ || + template == 0x04 /* including X unit */ || + template == 0x06) /* undefined */ + return 0; + + return 1; +} + +/* Prepare long jump bundle and disables other boosters if need */ +static void __kprobes prepare_booster(struct kprobe *p) +{ + unsigned long addr = (unsigned long)p->addr & ~0xFULL; + unsigned int slot = (unsigned long)p->addr & 0xf; + struct kprobe *other_kp; + + if (can_boost(&p->ainsn.insn[0].bundle, slot, addr)) { + set_brl_inst(&p->ainsn.insn[1].bundle, (bundle_t *)addr + 1); + p->ainsn.inst_flag |= INST_FLAG_BOOSTABLE; + } + + /* disables boosters in previous slots */ + for (; addr < (unsigned long)p->addr; addr++) { + other_kp = get_kprobe((void *)addr); + if (other_kp) + other_kp->ainsn.inst_flag &= ~INST_FLAG_BOOSTABLE; + } +} + int __kprobes arch_prepare_kprobe(struct kprobe *p) { unsigned long addr = (unsigned long) p->addr; @@ -530,6 +615,8 @@ int __kprobes arch_prepare_kprobe(struct kprobe *p) prepare_break_inst(template, slot, major_opcode, kprobe_inst, p, qp); + prepare_booster(p); + return 0; } @@ -543,7 +630,9 @@ void __kprobes arch_arm_kprobe(struct kprobe *p) src = &p->opcode.bundle; flush_icache_range((unsigned long)p->ainsn.insn, - (unsigned long)p->ainsn.insn + sizeof(kprobe_opcode_t)); + (unsigned long)p->ainsn.insn + + sizeof(kprobe_opcode_t) * MAX_INSN_SIZE); + switch (p->ainsn.slot) { case 0: dest->quad0.slot0 = src->quad0.slot0; @@ -584,13 +673,13 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p) void __kprobes arch_remove_kprobe(struct kprobe *p) { mutex_lock(&kprobe_mutex); - free_insn_slot(p->ainsn.insn, 0); + free_insn_slot(p->ainsn.insn, p->ainsn.inst_flag & INST_FLAG_BOOSTABLE); mutex_unlock(&kprobe_mutex); } /* * We are resuming execution after a single step fault, so the pt_regs * structure reflects the register state after we executed the instruction - * located in the kprobe (p->ainsn.insn.bundle). We still need to adjust + * located in the kprobe (p->ainsn.insn->bundle). We still need to adjust * the ip to point back to the original stack address. To set the IP address * to original stack address, handle the case where we need to fixup the * relative IP address and/or fixup branch register. @@ -607,7 +696,7 @@ static void __kprobes resume_execution(struct kprobe *p, struct pt_regs *regs) if (slot == 1 && bundle_encoding[template][1] == L) slot = 2; - if (p->ainsn.inst_flag) { + if (p->ainsn.inst_flag & ~INST_FLAG_BOOSTABLE) { if (p->ainsn.inst_flag & INST_FLAG_FIX_RELATIVE_IP_ADDR) { /* Fix relative IP address */ @@ -686,33 +775,12 @@ static void __kprobes prepare_ss(struct kprobe *p, struct pt_regs *regs) static int __kprobes is_ia64_break_inst(struct pt_regs *regs) { unsigned int slot = ia64_psr(regs)->ri; - unsigned int template, major_opcode; - unsigned long kprobe_inst; unsigned long *kprobe_addr = (unsigned long *)regs->cr_iip; bundle_t bundle; memcpy(&bundle, kprobe_addr, sizeof(bundle_t)); - template = bundle.quad0.template; - - /* Move to slot 2, if bundle is MLX type and kprobe slot is 1 */ - if (slot == 1 && bundle_encoding[template][1] == L) - slot++; - - /* Get Kprobe probe instruction at given slot*/ - get_kprobe_inst(&bundle, slot, &kprobe_inst, &major_opcode); - /* For break instruction, - * Bits 37:40 Major opcode to be zero - * Bits 27:32 X6 to be zero - * Bits 32:35 X3 to be zero - */ - if (major_opcode || ((kprobe_inst >> 27) & 0x1FF) ) { - /* Not a break instruction */ - return 0; - } - - /* Is a break instruction */ - return 1; + return __is_ia64_break_inst(&bundle, slot); } static int __kprobes pre_kprobes_handler(struct die_args *args) @@ -802,6 +870,19 @@ static int __kprobes pre_kprobes_handler(struct die_args *args) return 1; ss_probe: +#if !defined(CONFIG_PREEMPT) || defined(CONFIG_PM) + if (p->ainsn.inst_flag == INST_FLAG_BOOSTABLE && !p->post_handler) { + /* Boost up -- we can execute copied instructions directly */ + ia64_psr(regs)->ri = p->ainsn.slot; + regs->cr_iip = (unsigned long)&p->ainsn.insn->bundle & ~0xFULL; + /* turn single stepping off */ + ia64_psr(regs)->ss = 0; + + reset_current_kprobe(); + preempt_enable_no_resched(); + return 1; + } +#endif prepare_ss(p, regs); kcb->kprobe_status = KPROBE_HIT_SS; return 1; @@ -838,7 +919,7 @@ out: return 1; } -int __kprobes kprobes_fault_handler(struct pt_regs *regs, int trapnr) +int __kprobes kprobe_fault_handler(struct pt_regs *regs, int trapnr) { struct kprobe *cur = kprobe_running(); struct kprobe_ctlblk *kcb = get_kprobe_ctlblk(); @@ -1001,6 +1082,11 @@ int __kprobes setjmp_pre_handler(struct kprobe *p, struct pt_regs *regs) return 1; } +/* ia64 does not need this */ +void __kprobes jprobe_return(void) +{ +} + int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs) { struct kprobe_ctlblk *kcb = get_kprobe_ctlblk(); diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c index 6e17aed53135..705176b434b3 100644 --- a/arch/ia64/kernel/mca.c +++ b/arch/ia64/kernel/mca.c @@ -69,6 +69,7 @@ * 2007-04-27 Russ Anderson <rja@sgi.com> * Support multiple cpus going through OS_MCA in the same event. */ +#include <linux/jiffies.h> #include <linux/types.h> #include <linux/init.h> #include <linux/sched.h> @@ -97,6 +98,7 @@ #include <asm/irq.h> #include <asm/hw_irq.h> +#include <asm/tlb.h> #include "mca_drv.h" #include "entry.h" @@ -107,11 +109,26 @@ # define IA64_MCA_DEBUG(fmt...) #endif +#define NOTIFY_INIT(event, regs, arg, spin) \ +do { \ + if ((notify_die((event), "INIT", (regs), (arg), 0, 0) \ + == NOTIFY_STOP) && ((spin) == 1)) \ + ia64_mca_spin(__func__); \ +} while (0) + +#define NOTIFY_MCA(event, regs, arg, spin) \ +do { \ + if ((notify_die((event), "MCA", (regs), (arg), 0, 0) \ + == NOTIFY_STOP) && ((spin) == 1)) \ + ia64_mca_spin(__func__); \ +} while (0) + /* Used by mca_asm.S */ DEFINE_PER_CPU(u64, ia64_mca_data); /* == __per_cpu_mca[smp_processor_id()] */ DEFINE_PER_CPU(u64, ia64_mca_per_cpu_pte); /* PTE to map per-CPU area */ DEFINE_PER_CPU(u64, ia64_mca_pal_pte); /* PTE to map PAL code */ DEFINE_PER_CPU(u64, ia64_mca_pal_base); /* vaddr PAL code granule */ +DEFINE_PER_CPU(u64, ia64_mca_tr_reload); /* Flag for TR reload */ unsigned long __per_cpu_mca[NR_CPUS]; @@ -293,7 +310,8 @@ static void ia64_mlogbuf_dump_from_init(void) if (mlogbuf_finished) return; - if (mlogbuf_timestamp && (mlogbuf_timestamp + 30*HZ > jiffies)) { + if (mlogbuf_timestamp && + time_before(jiffies, mlogbuf_timestamp + 30 * HZ)) { printk(KERN_ERR "INIT: mlogbuf_dump is interrupted by INIT " " and the system seems to be messed up.\n"); ia64_mlogbuf_finish(0); @@ -413,8 +431,8 @@ ia64_log_get(int sal_info_type, u8 **buffer, int irq_safe) IA64_LOG_INDEX_INC(sal_info_type); IA64_LOG_UNLOCK(sal_info_type); if (irq_safe) { - IA64_MCA_DEBUG("%s: SAL error record type %d retrieved. " - "Record length = %ld\n", __FUNCTION__, sal_info_type, total_len); + IA64_MCA_DEBUG("%s: SAL error record type %d retrieved. Record length = %ld\n", + __func__, sal_info_type, total_len); } *buffer = (u8 *) log_buffer; return total_len; @@ -518,7 +536,7 @@ ia64_mca_cpe_int_handler (int cpe_irq, void *arg) static DEFINE_SPINLOCK(cpe_history_lock); IA64_MCA_DEBUG("%s: received interrupt vector = %#x on CPU %d\n", - __FUNCTION__, cpe_irq, smp_processor_id()); + __func__, cpe_irq, smp_processor_id()); /* SAL spec states this should run w/ interrupts enabled */ local_irq_enable(); @@ -594,7 +612,7 @@ ia64_mca_register_cpev (int cpev) } IA64_MCA_DEBUG("%s: corrected platform error " - "vector %#x registered\n", __FUNCTION__, cpev); + "vector %#x registered\n", __func__, cpev); } #endif /* CONFIG_ACPI */ @@ -621,12 +639,11 @@ ia64_mca_cmc_vector_setup (void) cmcv.cmcv_vector = IA64_CMC_VECTOR; ia64_setreg(_IA64_REG_CR_CMCV, cmcv.cmcv_regval); - IA64_MCA_DEBUG("%s: CPU %d corrected " - "machine check vector %#x registered.\n", - __FUNCTION__, smp_processor_id(), IA64_CMC_VECTOR); + IA64_MCA_DEBUG("%s: CPU %d corrected machine check vector %#x registered.\n", + __func__, smp_processor_id(), IA64_CMC_VECTOR); IA64_MCA_DEBUG("%s: CPU %d CMCV = %#016lx\n", - __FUNCTION__, smp_processor_id(), ia64_getreg(_IA64_REG_CR_CMCV)); + __func__, smp_processor_id(), ia64_getreg(_IA64_REG_CR_CMCV)); } /* @@ -651,9 +668,8 @@ ia64_mca_cmc_vector_disable (void *dummy) cmcv.cmcv_mask = 1; /* Mask/disable interrupt */ ia64_setreg(_IA64_REG_CR_CMCV, cmcv.cmcv_regval); - IA64_MCA_DEBUG("%s: CPU %d corrected " - "machine check vector %#x disabled.\n", - __FUNCTION__, smp_processor_id(), cmcv.cmcv_vector); + IA64_MCA_DEBUG("%s: CPU %d corrected machine check vector %#x disabled.\n", + __func__, smp_processor_id(), cmcv.cmcv_vector); } /* @@ -678,9 +694,8 @@ ia64_mca_cmc_vector_enable (void *dummy) cmcv.cmcv_mask = 0; /* Unmask/enable interrupt */ ia64_setreg(_IA64_REG_CR_CMCV, cmcv.cmcv_regval); - IA64_MCA_DEBUG("%s: CPU %d corrected " - "machine check vector %#x enabled.\n", - __FUNCTION__, smp_processor_id(), cmcv.cmcv_vector); + IA64_MCA_DEBUG("%s: CPU %d corrected machine check vector %#x enabled.\n", + __func__, smp_processor_id(), cmcv.cmcv_vector); } /* @@ -765,9 +780,8 @@ ia64_mca_rendez_int_handler(int rendez_irq, void *arg) /* Mask all interrupts */ local_irq_save(flags); - if (notify_die(DIE_MCA_RENDZVOUS_ENTER, "MCA", get_irq_regs(), - (long)&nd, 0, 0) == NOTIFY_STOP) - ia64_mca_spin(__FUNCTION__); + + NOTIFY_MCA(DIE_MCA_RENDZVOUS_ENTER, get_irq_regs(), (long)&nd, 1); ia64_mc_info.imi_rendez_checkin[cpu] = IA64_MCA_RENDEZ_CHECKIN_DONE; /* Register with the SAL monarch that the slave has @@ -775,17 +789,13 @@ ia64_mca_rendez_int_handler(int rendez_irq, void *arg) */ ia64_sal_mc_rendez(); - if (notify_die(DIE_MCA_RENDZVOUS_PROCESS, "MCA", get_irq_regs(), - (long)&nd, 0, 0) == NOTIFY_STOP) - ia64_mca_spin(__FUNCTION__); + NOTIFY_MCA(DIE_MCA_RENDZVOUS_PROCESS, get_irq_regs(), (long)&nd, 1); /* Wait for the monarch cpu to exit. */ while (monarch_cpu != -1) cpu_relax(); /* spin until monarch leaves */ - if (notify_die(DIE_MCA_RENDZVOUS_LEAVE, "MCA", get_irq_regs(), - (long)&nd, 0, 0) == NOTIFY_STOP) - ia64_mca_spin(__FUNCTION__); + NOTIFY_MCA(DIE_MCA_RENDZVOUS_LEAVE, get_irq_regs(), (long)&nd, 1); ia64_mc_info.imi_rendez_checkin[cpu] = IA64_MCA_RENDEZ_CHECKIN_NOTDONE; /* Enable all interrupts */ @@ -1185,6 +1195,49 @@ all_in: return; } +/* mca_insert_tr + * + * Switch rid when TR reload and needed! + * iord: 1: itr, 2: itr; + * +*/ +static void mca_insert_tr(u64 iord) +{ + + int i; + u64 old_rr; + struct ia64_tr_entry *p; + unsigned long psr; + int cpu = smp_processor_id(); + + psr = ia64_clear_ic(); + for (i = IA64_TR_ALLOC_BASE; i < IA64_TR_ALLOC_MAX; i++) { + p = &__per_cpu_idtrs[cpu][iord-1][i]; + if (p->pte & 0x1) { + old_rr = ia64_get_rr(p->ifa); + if (old_rr != p->rr) { + ia64_set_rr(p->ifa, p->rr); + ia64_srlz_d(); + } + ia64_ptr(iord, p->ifa, p->itir >> 2); + ia64_srlz_i(); + if (iord & 0x1) { + ia64_itr(0x1, i, p->ifa, p->pte, p->itir >> 2); + ia64_srlz_i(); + } + if (iord & 0x2) { + ia64_itr(0x2, i, p->ifa, p->pte, p->itir >> 2); + ia64_srlz_i(); + } + if (old_rr != p->rr) { + ia64_set_rr(p->ifa, old_rr); + ia64_srlz_d(); + } + } + } + ia64_set_psr(psr); +} + /* * ia64_mca_handler * @@ -1212,7 +1265,7 @@ ia64_mca_handler(struct pt_regs *regs, struct switch_stack *sw, int recover, cpu = smp_processor_id(); struct task_struct *previous_current; struct ia64_mca_notify_die nd = - { .sos = sos, .monarch_cpu = &monarch_cpu }; + { .sos = sos, .monarch_cpu = &monarch_cpu, .data = &recover }; static atomic_t mca_count; static cpumask_t mca_cpu; @@ -1228,9 +1281,7 @@ ia64_mca_handler(struct pt_regs *regs, struct switch_stack *sw, previous_current = ia64_mca_modify_original_stack(regs, sw, sos, "MCA"); - if (notify_die(DIE_MCA_MONARCH_ENTER, "MCA", regs, (long)&nd, 0, 0) - == NOTIFY_STOP) - ia64_mca_spin(__FUNCTION__); + NOTIFY_MCA(DIE_MCA_MONARCH_ENTER, regs, (long)&nd, 1); ia64_mc_info.imi_rendez_checkin[cpu] = IA64_MCA_RENDEZ_CHECKIN_CONCURRENT_MCA; if (sos->monarch) { @@ -1244,13 +1295,12 @@ ia64_mca_handler(struct pt_regs *regs, struct switch_stack *sw, * does not work. */ ia64_mca_wakeup_all(); - if (notify_die(DIE_MCA_MONARCH_PROCESS, "MCA", regs, (long)&nd, 0, 0) - == NOTIFY_STOP) - ia64_mca_spin(__FUNCTION__); } else { while (cpu_isset(cpu, mca_cpu)) cpu_relax(); /* spin until monarch wakes us */ - } + } + + NOTIFY_MCA(DIE_MCA_MONARCH_PROCESS, regs, (long)&nd, 1); /* Get the MCA error record and log it */ ia64_mca_log_sal_error_record(SAL_INFO_TYPE_MCA); @@ -1269,15 +1319,14 @@ ia64_mca_handler(struct pt_regs *regs, struct switch_stack *sw, } else { /* Dump buffered message to console */ ia64_mlogbuf_finish(1); -#ifdef CONFIG_KEXEC - atomic_set(&kdump_in_progress, 1); - monarch_cpu = -1; -#endif } - if (notify_die(DIE_MCA_MONARCH_LEAVE, "MCA", regs, (long)&nd, 0, recover) - == NOTIFY_STOP) - ia64_mca_spin(__FUNCTION__); + if (__get_cpu_var(ia64_mca_tr_reload)) { + mca_insert_tr(0x1); /*Reload dynamic itrs*/ + mca_insert_tr(0x2); /*Reload dynamic itrs*/ + } + + NOTIFY_MCA(DIE_MCA_MONARCH_LEAVE, regs, (long)&nd, 1); if (atomic_dec_return(&mca_count) > 0) { int i; @@ -1328,7 +1377,7 @@ ia64_mca_cmc_int_handler(int cmc_irq, void *arg) static DEFINE_SPINLOCK(cmc_history_lock); IA64_MCA_DEBUG("%s: received interrupt vector = %#x on CPU %d\n", - __FUNCTION__, cmc_irq, smp_processor_id()); + __func__, cmc_irq, smp_processor_id()); /* SAL spec states this should run w/ interrupts enabled */ local_irq_enable(); @@ -1598,7 +1647,7 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw, struct ia64_mca_notify_die nd = { .sos = sos, .monarch_cpu = &monarch_cpu }; - (void) notify_die(DIE_INIT_ENTER, "INIT", regs, (long)&nd, 0, 0); + NOTIFY_INIT(DIE_INIT_ENTER, regs, (long)&nd, 0); mprintk(KERN_INFO "Entered OS INIT handler. PSP=%lx cpu=%d monarch=%ld\n", sos->proc_state_param, cpu, sos->monarch); @@ -1614,7 +1663,7 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw, */ if (!sos->monarch && atomic_add_return(1, &slaves) == num_online_cpus()) { mprintk(KERN_WARNING "%s: Promoting cpu %d to monarch.\n", - __FUNCTION__, cpu); + __func__, cpu); atomic_dec(&slaves); sos->monarch = 1; } @@ -1626,7 +1675,7 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw, */ if (sos->monarch && atomic_add_return(1, &monarchs) > 1) { mprintk(KERN_WARNING "%s: Demoting cpu %d to slave.\n", - __FUNCTION__, cpu); + __func__, cpu); atomic_dec(&monarchs); sos->monarch = 0; } @@ -1635,17 +1684,15 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw, ia64_mc_info.imi_rendez_checkin[cpu] = IA64_MCA_RENDEZ_CHECKIN_INIT; while (monarch_cpu == -1) cpu_relax(); /* spin until monarch enters */ - if (notify_die(DIE_INIT_SLAVE_ENTER, "INIT", regs, (long)&nd, 0, 0) - == NOTIFY_STOP) - ia64_mca_spin(__FUNCTION__); - if (notify_die(DIE_INIT_SLAVE_PROCESS, "INIT", regs, (long)&nd, 0, 0) - == NOTIFY_STOP) - ia64_mca_spin(__FUNCTION__); + + NOTIFY_INIT(DIE_INIT_SLAVE_ENTER, regs, (long)&nd, 1); + NOTIFY_INIT(DIE_INIT_SLAVE_PROCESS, regs, (long)&nd, 1); + while (monarch_cpu != -1) cpu_relax(); /* spin until monarch leaves */ - if (notify_die(DIE_INIT_SLAVE_LEAVE, "INIT", regs, (long)&nd, 0, 0) - == NOTIFY_STOP) - ia64_mca_spin(__FUNCTION__); + + NOTIFY_INIT(DIE_INIT_SLAVE_LEAVE, regs, (long)&nd, 1); + mprintk("Slave on cpu %d returning to normal service.\n", cpu); set_curr_task(cpu, previous_current); ia64_mc_info.imi_rendez_checkin[cpu] = IA64_MCA_RENDEZ_CHECKIN_NOTDONE; @@ -1654,9 +1701,7 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw, } monarch_cpu = cpu; - if (notify_die(DIE_INIT_MONARCH_ENTER, "INIT", regs, (long)&nd, 0, 0) - == NOTIFY_STOP) - ia64_mca_spin(__FUNCTION__); + NOTIFY_INIT(DIE_INIT_MONARCH_ENTER, regs, (long)&nd, 1); /* * Wait for a bit. On some machines (e.g., HP's zx2000 and zx6000, INIT can be @@ -1671,12 +1716,9 @@ ia64_init_handler(struct pt_regs *regs, struct switch_stack *sw, * to default_monarch_init_process() above and just print all the * tasks. */ - if (notify_die(DIE_INIT_MONARCH_PROCESS, "INIT", regs, (long)&nd, 0, 0) - == NOTIFY_STOP) - ia64_mca_spin(__FUNCTION__); - if (notify_die(DIE_INIT_MONARCH_LEAVE, "INIT", regs, (long)&nd, 0, 0) - == NOTIFY_STOP) - ia64_mca_spin(__FUNCTION__); + NOTIFY_INIT(DIE_INIT_MONARCH_PROCESS, regs, (long)&nd, 1); + NOTIFY_INIT(DIE_INIT_MONARCH_LEAVE, regs, (long)&nd, 1); + mprintk("\nINIT dump complete. Monarch on cpu %d returning to normal service.\n", cpu); atomic_dec(&monarchs); set_curr_task(cpu, previous_current); @@ -1884,7 +1926,7 @@ ia64_mca_init(void) .priority = 0/* we need to notified last */ }; - IA64_MCA_DEBUG("%s: begin\n", __FUNCTION__); + IA64_MCA_DEBUG("%s: begin\n", __func__); /* Clear the Rendez checkin flag for all cpus */ for(i = 0 ; i < NR_CPUS; i++) @@ -1908,7 +1950,7 @@ ia64_mca_init(void) printk(KERN_INFO "Increasing MCA rendezvous timeout from " "%ld to %ld milliseconds\n", timeout, isrv.v0); timeout = isrv.v0; - (void) notify_die(DIE_MCA_NEW_TIMEOUT, "MCA", NULL, timeout, 0, 0); + NOTIFY_MCA(DIE_MCA_NEW_TIMEOUT, NULL, timeout, 0); continue; } printk(KERN_ERR "Failed to register rendezvous interrupt " @@ -1928,7 +1970,7 @@ ia64_mca_init(void) return; } - IA64_MCA_DEBUG("%s: registered MCA rendezvous spinloop and wakeup mech.\n", __FUNCTION__); + IA64_MCA_DEBUG("%s: registered MCA rendezvous spinloop and wakeup mech.\n", __func__); ia64_mc_info.imi_mca_handler = ia64_tpa(mca_hldlr_ptr->fp); /* @@ -1949,7 +1991,7 @@ ia64_mca_init(void) return; } - IA64_MCA_DEBUG("%s: registered OS MCA handler with SAL at 0x%lx, gp = 0x%lx\n", __FUNCTION__, + IA64_MCA_DEBUG("%s: registered OS MCA handler with SAL at 0x%lx, gp = 0x%lx\n", __func__, ia64_mc_info.imi_mca_handler, ia64_tpa(mca_hldlr_ptr->gp)); /* @@ -1961,7 +2003,7 @@ ia64_mca_init(void) ia64_mc_info.imi_slave_init_handler = ia64_tpa(init_hldlr_ptr_slave->fp); ia64_mc_info.imi_slave_init_handler_size = 0; - IA64_MCA_DEBUG("%s: OS INIT handler at %lx\n", __FUNCTION__, + IA64_MCA_DEBUG("%s: OS INIT handler at %lx\n", __func__, ia64_mc_info.imi_monarch_init_handler); /* Register the os init handler with SAL */ @@ -1982,7 +2024,7 @@ ia64_mca_init(void) return; } - IA64_MCA_DEBUG("%s: registered OS INIT handler with SAL\n", __FUNCTION__); + IA64_MCA_DEBUG("%s: registered OS INIT handler with SAL\n", __func__); /* * Configure the CMCI/P vector and handler. Interrupts for CMC are @@ -2042,7 +2084,7 @@ ia64_mca_late_init(void) cmc_polling_enabled = 0; schedule_work(&cmc_enable_work); - IA64_MCA_DEBUG("%s: CMCI/P setup and enabled.\n", __FUNCTION__); + IA64_MCA_DEBUG("%s: CMCI/P setup and enabled.\n", __func__); #ifdef CONFIG_ACPI /* Setup the CPEI/P vector and handler */ @@ -2065,17 +2107,17 @@ ia64_mca_late_init(void) ia64_cpe_irq = irq; ia64_mca_register_cpev(cpe_vector); IA64_MCA_DEBUG("%s: CPEI/P setup and enabled.\n", - __FUNCTION__); + __func__); return 0; } printk(KERN_ERR "%s: Failed to find irq for CPE " "interrupt handler, vector %d\n", - __FUNCTION__, cpe_vector); + __func__, cpe_vector); } /* If platform doesn't support CPEI, get the timer going. */ if (cpe_poll_enabled) { ia64_mca_cpe_poll(0UL); - IA64_MCA_DEBUG("%s: CPEP setup and enabled.\n", __FUNCTION__); + IA64_MCA_DEBUG("%s: CPEP setup and enabled.\n", __func__); } } #endif diff --git a/arch/ia64/kernel/mca_asm.S b/arch/ia64/kernel/mca_asm.S index 8bc7d259e0c6..a06d46548ff9 100644 --- a/arch/ia64/kernel/mca_asm.S +++ b/arch/ia64/kernel/mca_asm.S @@ -219,8 +219,13 @@ ia64_reload_tr: mov r20=IA64_TR_CURRENT_STACK ;; itr.d dtr[r20]=r16 + GET_THIS_PADDR(r2, ia64_mca_tr_reload) + mov r18 = 1 ;; srlz.d + ;; + st8 [r2] =r18 + ;; done_tlb_purge_and_reload: diff --git a/arch/ia64/kernel/minstate.h b/arch/ia64/kernel/minstate.h index c9ac8bada786..7c548ac52bbc 100644 --- a/arch/ia64/kernel/minstate.h +++ b/arch/ia64/kernel/minstate.h @@ -3,6 +3,18 @@ #include "entry.h" +#ifdef CONFIG_VIRT_CPU_ACCOUNTING +/* read ar.itc in advance, and use it before leaving bank 0 */ +#define ACCOUNT_GET_STAMP \ +(pUStk) mov.m r20=ar.itc; +#define ACCOUNT_SYS_ENTER \ +(pUStk) br.call.spnt rp=account_sys_enter \ + ;; +#else +#define ACCOUNT_GET_STAMP +#define ACCOUNT_SYS_ENTER +#endif + /* * DO_SAVE_MIN switches to the kernel stacks (if necessary) and saves * the minimum state necessary that allows us to turn psr.ic back @@ -122,11 +134,13 @@ ;; \ .mem.offset 0,0; st8.spill [r16]=r2,16; \ .mem.offset 8,0; st8.spill [r17]=r3,16; \ + ACCOUNT_GET_STAMP \ adds r2=IA64_PT_REGS_R16_OFFSET,r1; \ ;; \ EXTRA; \ movl r1=__gp; /* establish kernel global pointer */ \ ;; \ + ACCOUNT_SYS_ENTER \ bsw.1; /* switch back to bank 1 (must be last in insn group) */ \ ;; diff --git a/arch/ia64/kernel/module.c b/arch/ia64/kernel/module.c index e58f4367cf11..e83e2ea3b3e0 100644 --- a/arch/ia64/kernel/module.c +++ b/arch/ia64/kernel/module.c @@ -493,7 +493,7 @@ module_frob_arch_sections (Elf_Ehdr *ehdr, Elf_Shdr *sechdrs, char *secstrings, mod->arch.opd->sh_addralign = 8; mod->arch.opd->sh_size = fdescs * sizeof(struct fdesc); DEBUGP("%s: core.plt=%lx, init.plt=%lx, got=%lx, fdesc=%lx\n", - __FUNCTION__, mod->arch.core_plt->sh_size, mod->arch.init_plt->sh_size, + __func__, mod->arch.core_plt->sh_size, mod->arch.init_plt->sh_size, mod->arch.got->sh_size, mod->arch.opd->sh_size); return 0; } @@ -585,7 +585,7 @@ get_plt (struct module *mod, const struct insn *insn, uint64_t value, int *okp) #if ARCH_MODULE_DEBUG if (plt_target(plt) != target_ip) { printk("%s: mistargeted PLT: wanted %lx, got %lx\n", - __FUNCTION__, target_ip, plt_target(plt)); + __func__, target_ip, plt_target(plt)); *okp = 0; return 0; } @@ -703,7 +703,7 @@ do_reloc (struct module *mod, uint8_t r_type, Elf64_Sym *sym, uint64_t addend, if (r_type == R_IA64_PCREL21BI) { if (!is_internal(mod, val)) { printk(KERN_ERR "%s: %s reloc against non-local symbol (%lx)\n", - __FUNCTION__, reloc_name[r_type], val); + __func__, reloc_name[r_type], val); return -ENOEXEC; } format = RF_INSN21B; @@ -737,7 +737,7 @@ do_reloc (struct module *mod, uint8_t r_type, Elf64_Sym *sym, uint64_t addend, case R_IA64_LDXMOV: if (gp_addressable(mod, val)) { /* turn "ld8" into "mov": */ - DEBUGP("%s: patching ld8 at %p to mov\n", __FUNCTION__, location); + DEBUGP("%s: patching ld8 at %p to mov\n", __func__, location); ia64_patch((u64) location, 0x1fff80fe000UL, 0x10000000000UL); } return 0; @@ -771,7 +771,7 @@ do_reloc (struct module *mod, uint8_t r_type, Elf64_Sym *sym, uint64_t addend, if (!ok) return -ENOEXEC; - DEBUGP("%s: [%p]<-%016lx = %s(%lx)\n", __FUNCTION__, location, val, + DEBUGP("%s: [%p]<-%016lx = %s(%lx)\n", __func__, location, val, reloc_name[r_type] ? reloc_name[r_type] : "?", sym->st_value + addend); switch (format) { @@ -807,7 +807,7 @@ apply_relocate_add (Elf64_Shdr *sechdrs, const char *strtab, unsigned int symind Elf64_Shdr *target_sec; int ret; - DEBUGP("%s: applying section %u (%u relocs) to %u\n", __FUNCTION__, + DEBUGP("%s: applying section %u (%u relocs) to %u\n", __func__, relsec, n, sechdrs[relsec].sh_info); target_sec = sechdrs + sechdrs[relsec].sh_info; @@ -835,7 +835,7 @@ apply_relocate_add (Elf64_Shdr *sechdrs, const char *strtab, unsigned int symind gp = mod->core_size / 2; gp = (uint64_t) mod->module_core + ((gp + 7) & -8); mod->arch.gp = gp; - DEBUGP("%s: placing gp at 0x%lx\n", __FUNCTION__, gp); + DEBUGP("%s: placing gp at 0x%lx\n", __func__, gp); } for (i = 0; i < n; i++) { @@ -903,7 +903,7 @@ register_unwind_table (struct module *mod) init = start + num_core; } - DEBUGP("%s: name=%s, gp=%lx, num_init=%lu, num_core=%lu\n", __FUNCTION__, + DEBUGP("%s: name=%s, gp=%lx, num_init=%lu, num_core=%lu\n", __func__, mod->name, mod->arch.gp, num_init, num_core); /* @@ -912,13 +912,13 @@ register_unwind_table (struct module *mod) if (num_core > 0) { mod->arch.core_unw_table = unw_add_unwind_table(mod->name, 0, mod->arch.gp, core, core + num_core); - DEBUGP("%s: core: handle=%p [%p-%p)\n", __FUNCTION__, + DEBUGP("%s: core: handle=%p [%p-%p)\n", __func__, mod->arch.core_unw_table, core, core + num_core); } if (num_init > 0) { mod->arch.init_unw_table = unw_add_unwind_table(mod->name, 0, mod->arch.gp, init, init + num_init); - DEBUGP("%s: init: handle=%p [%p-%p)\n", __FUNCTION__, + DEBUGP("%s: init: handle=%p [%p-%p)\n", __func__, mod->arch.init_unw_table, init, init + num_init); } } @@ -926,7 +926,7 @@ register_unwind_table (struct module *mod) int module_finalize (const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs, struct module *mod) { - DEBUGP("%s: init: entry=%p\n", __FUNCTION__, mod->init); + DEBUGP("%s: init: entry=%p\n", __func__, mod->init); if (mod->arch.unwind) register_unwind_table(mod); return 0; diff --git a/arch/ia64/kernel/msi_ia64.c b/arch/ia64/kernel/msi_ia64.c index e86d02959794..60c6ef67ebb2 100644 --- a/arch/ia64/kernel/msi_ia64.c +++ b/arch/ia64/kernel/msi_ia64.c @@ -57,7 +57,7 @@ static void ia64_set_msi_irq_affinity(unsigned int irq, cpumask_t cpu_mask) if (!cpu_online(cpu)) return; - if (reassign_irq_vector(irq, cpu)) + if (irq_prepare_move(irq, cpu)) return; read_msi_msg(irq, &msg); @@ -119,6 +119,7 @@ void ia64_teardown_msi_irq(unsigned int irq) static void ia64_ack_msi_irq(unsigned int irq) { + irq_complete_move(irq); move_native_irq(irq); ia64_eoi(); } diff --git a/arch/ia64/kernel/numa.c b/arch/ia64/kernel/numa.c index a78b45f5fe2f..c93420c97409 100644 --- a/arch/ia64/kernel/numa.c +++ b/arch/ia64/kernel/numa.c @@ -73,7 +73,7 @@ void __init build_cpu_to_node_map(void) for(node=0; node < MAX_NUMNODES; node++) cpus_clear(node_to_cpu_mask[node]); - for(cpu = 0; cpu < NR_CPUS; ++cpu) { + for_each_possible_early_cpu(cpu) { node = -1; for (i = 0; i < NR_CPUS; ++i) if (cpu_physical_id(cpu) == node_cpuid[i].phys_id) { diff --git a/arch/ia64/kernel/patch.c b/arch/ia64/kernel/patch.c index 2cb9425e0421..e0dca8743dbb 100644 --- a/arch/ia64/kernel/patch.c +++ b/arch/ia64/kernel/patch.c @@ -135,10 +135,10 @@ ia64_patch_mckinley_e9 (unsigned long start, unsigned long end) while (offp < (s32 *) end) { wp = (u64 *) ia64_imva((char *) offp + *offp); - wp[0] = 0x0000000100000000UL; /* nop.m 0; nop.i 0; nop.i 0 */ - wp[1] = 0x0004000000000200UL; - wp[2] = 0x0000000100000011UL; /* nop.m 0; nop.i 0; br.ret.sptk.many b6 */ - wp[3] = 0x0084006880000200UL; + wp[0] = 0x0000000100000011UL; /* nop.m 0; nop.i 0; br.ret.sptk.many b6 */ + wp[1] = 0x0084006880000200UL; + wp[2] = 0x0000000100000000UL; /* nop.m 0; nop.i 0; nop.i 0 */ + wp[3] = 0x0004000000000200UL; ia64_fc(wp); ia64_fc(wp + 2); ++offp; } diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c index f6b99719f10f..c8e403752a0c 100644 --- a/arch/ia64/kernel/perfmon.c +++ b/arch/ia64/kernel/perfmon.c @@ -227,12 +227,12 @@ #ifdef PFM_DEBUGGING #define DPRINT(a) \ do { \ - if (unlikely(pfm_sysctl.debug >0)) { printk("%s.%d: CPU%d [%d] ", __FUNCTION__, __LINE__, smp_processor_id(), task_pid_nr(current)); printk a; } \ + if (unlikely(pfm_sysctl.debug >0)) { printk("%s.%d: CPU%d [%d] ", __func__, __LINE__, smp_processor_id(), task_pid_nr(current)); printk a; } \ } while (0) #define DPRINT_ovfl(a) \ do { \ - if (unlikely(pfm_sysctl.debug > 0 && pfm_sysctl.debug_ovfl >0)) { printk("%s.%d: CPU%d [%d] ", __FUNCTION__, __LINE__, smp_processor_id(), task_pid_nr(current)); printk a; } \ + if (unlikely(pfm_sysctl.debug > 0 && pfm_sysctl.debug_ovfl >0)) { printk("%s.%d: CPU%d [%d] ", __func__, __LINE__, smp_processor_id(), task_pid_nr(current)); printk a; } \ } while (0) #endif @@ -4204,10 +4204,10 @@ pfm_check_task_exist(pfm_context_t *ctx) do_each_thread (g, t) { if (t->thread.pfm_context == ctx) { ret = 0; - break; + goto out; } } while_each_thread (g, t); - +out: read_unlock(&tasklist_lock); DPRINT(("pfm_check_task_exist: ret=%d ctx=%p\n", ret, ctx)); @@ -5511,7 +5511,7 @@ stop_monitoring: } static int -pfm_do_interrupt_handler(int irq, void *arg, struct pt_regs *regs) +pfm_do_interrupt_handler(void *arg, struct pt_regs *regs) { struct task_struct *task; pfm_context_t *ctx; @@ -5591,7 +5591,7 @@ pfm_interrupt_handler(int irq, void *arg) start_cycles = ia64_get_itc(); - ret = pfm_do_interrupt_handler(irq, arg, regs); + ret = pfm_do_interrupt_handler(arg, regs); total_cycles = ia64_get_itc(); diff --git a/arch/ia64/kernel/perfmon_default_smpl.c b/arch/ia64/kernel/perfmon_default_smpl.c index a7af1cb419f9..5f637bbfcccd 100644 --- a/arch/ia64/kernel/perfmon_default_smpl.c +++ b/arch/ia64/kernel/perfmon_default_smpl.c @@ -24,12 +24,12 @@ MODULE_LICENSE("GPL"); #ifdef DEFAULT_DEBUG #define DPRINT(a) \ do { \ - if (unlikely(pfm_sysctl.debug >0)) { printk("%s.%d: CPU%d ", __FUNCTION__, __LINE__, smp_processor_id()); printk a; } \ + if (unlikely(pfm_sysctl.debug >0)) { printk("%s.%d: CPU%d ", __func__, __LINE__, smp_processor_id()); printk a; } \ } while (0) #define DPRINT_ovfl(a) \ do { \ - if (unlikely(pfm_sysctl.debug > 0 && pfm_sysctl.debug_ovfl >0)) { printk("%s.%d: CPU%d ", __FUNCTION__, __LINE__, smp_processor_id()); printk a; } \ + if (unlikely(pfm_sysctl.debug > 0 && pfm_sysctl.debug_ovfl >0)) { printk("%s.%d: CPU%d ", __func__, __LINE__, smp_processor_id()); printk a; } \ } while (0) #else diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c index 49937a383b23..a5ea817cbcbf 100644 --- a/arch/ia64/kernel/process.c +++ b/arch/ia64/kernel/process.c @@ -625,21 +625,6 @@ do_dump_fpu (struct unw_frame_info *info, void *arg) do_dump_task_fpu(current, info, arg); } -int -dump_task_regs(struct task_struct *task, elf_gregset_t *regs) -{ - struct unw_frame_info tcore_info; - - if (current == task) { - unw_init_running(do_copy_regs, regs); - } else { - memset(&tcore_info, 0, sizeof(tcore_info)); - unw_init_from_blocked_task(&tcore_info, task); - do_copy_task_regs(task, &tcore_info, regs); - } - return 1; -} - void ia64_elf_core_copy_regs (struct pt_regs *pt, elf_gregset_t dst) { @@ -647,21 +632,6 @@ ia64_elf_core_copy_regs (struct pt_regs *pt, elf_gregset_t dst) } int -dump_task_fpu (struct task_struct *task, elf_fpregset_t *dst) -{ - struct unw_frame_info tcore_info; - - if (current == task) { - unw_init_running(do_dump_fpu, dst); - } else { - memset(&tcore_info, 0, sizeof(tcore_info)); - unw_init_from_blocked_task(&tcore_info, task); - do_dump_task_fpu(task, &tcore_info, dst); - } - return 1; -} - -int dump_fpu (struct pt_regs *pt, elf_fpregset_t dst) { unw_init_running(do_dump_fpu, dst); diff --git a/arch/ia64/kernel/ptrace.c b/arch/ia64/kernel/ptrace.c index 331d6768b5d5..2a9943b5947f 100644 --- a/arch/ia64/kernel/ptrace.c +++ b/arch/ia64/kernel/ptrace.c @@ -3,6 +3,9 @@ * * Copyright (C) 1999-2005 Hewlett-Packard Co * David Mosberger-Tang <davidm@hpl.hp.com> + * Copyright (C) 2006 Intel Co + * 2006-08-12 - IA64 Native Utrace implementation support added by + * Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com> * * Derived from the x86 and Alpha versions. */ @@ -17,6 +20,8 @@ #include <linux/security.h> #include <linux/audit.h> #include <linux/signal.h> +#include <linux/regset.h> +#include <linux/elf.h> #include <asm/pgtable.h> #include <asm/processor.h> @@ -698,52 +703,6 @@ thread_matches (struct task_struct *thread, unsigned long addr) } /* - * GDB apparently wants to be able to read the register-backing store - * of any thread when attached to a given process. If we are peeking - * or poking an address that happens to reside in the kernel-backing - * store of another thread, we need to attach to that thread, because - * otherwise we end up accessing stale data. - * - * task_list_lock must be read-locked before calling this routine! - */ -static struct task_struct * -find_thread_for_addr (struct task_struct *child, unsigned long addr) -{ - struct task_struct *p; - struct mm_struct *mm; - struct list_head *this, *next; - int mm_users; - - if (!(mm = get_task_mm(child))) - return child; - - /* -1 because of our get_task_mm(): */ - mm_users = atomic_read(&mm->mm_users) - 1; - if (mm_users <= 1) - goto out; /* not multi-threaded */ - - /* - * Traverse the current process' children list. Every task that - * one attaches to becomes a child. And it is only attached children - * of the debugger that are of interest (ptrace_check_attach checks - * for this). - */ - list_for_each_safe(this, next, ¤t->children) { - p = list_entry(this, struct task_struct, sibling); - if (p->tgid != child->tgid) - continue; - if (thread_matches(p, addr)) { - child = p; - goto out; - } - } - - out: - mmput(mm); - return child; -} - -/* * Write f32-f127 back to task->thread.fph if it has been modified. */ inline void @@ -786,25 +745,6 @@ ia64_sync_fph (struct task_struct *task) psr->dfh = 1; } -static int -access_fr (struct unw_frame_info *info, int regnum, int hi, - unsigned long *data, int write_access) -{ - struct ia64_fpreg fpval; - int ret; - - ret = unw_get_fr(info, regnum, &fpval); - if (ret < 0) - return ret; - - if (write_access) { - fpval.u.bits[hi] = *data; - ret = unw_set_fr(info, regnum, fpval); - } else - *data = fpval.u.bits[hi]; - return ret; -} - /* * Change the machine-state of CHILD such that it will return via the normal * kernel exit-path, rather than the syscall-exit path. @@ -826,14 +766,14 @@ convert_to_non_syscall (struct task_struct *child, struct pt_regs *pt, if ((long)((unsigned long)child + IA64_STK_OFFSET - sp) < IA64_PT_REGS_SIZE) { dprintk("ptrace.%s: ran off the top of the kernel " - "stack\n", __FUNCTION__); + "stack\n", __func__); return; } if (unw_get_pr (&prev_info, &pr) < 0) { unw_get_rp(&prev_info, &ip); dprintk("ptrace.%s: failed to read " "predicate register (ip=0x%lx)\n", - __FUNCTION__, ip); + __func__, ip); return; } if (unw_is_intr_frame(&info) @@ -906,330 +846,7 @@ access_nat_bits (struct task_struct *child, struct pt_regs *pt, static int access_uarea (struct task_struct *child, unsigned long addr, - unsigned long *data, int write_access) -{ - unsigned long *ptr, regnum, urbs_end, rnat_addr, cfm; - struct switch_stack *sw; - struct pt_regs *pt; -# define pt_reg_addr(pt, reg) ((void *) \ - ((unsigned long) (pt) \ - + offsetof(struct pt_regs, reg))) - - - pt = task_pt_regs(child); - sw = (struct switch_stack *) (child->thread.ksp + 16); - - if ((addr & 0x7) != 0) { - dprintk("ptrace: unaligned register address 0x%lx\n", addr); - return -1; - } - - if (addr < PT_F127 + 16) { - /* accessing fph */ - if (write_access) - ia64_sync_fph(child); - else - ia64_flush_fph(child); - ptr = (unsigned long *) - ((unsigned long) &child->thread.fph + addr); - } else if ((addr >= PT_F10) && (addr < PT_F11 + 16)) { - /* scratch registers untouched by kernel (saved in pt_regs) */ - ptr = pt_reg_addr(pt, f10) + (addr - PT_F10); - } else if (addr >= PT_F12 && addr < PT_F15 + 16) { - /* - * Scratch registers untouched by kernel (saved in - * switch_stack). - */ - ptr = (unsigned long *) ((long) sw - + (addr - PT_NAT_BITS - 32)); - } else if (addr < PT_AR_LC + 8) { - /* preserved state: */ - struct unw_frame_info info; - char nat = 0; - int ret; - - unw_init_from_blocked_task(&info, child); - if (unw_unwind_to_user(&info) < 0) - return -1; - - switch (addr) { - case PT_NAT_BITS: - return access_nat_bits(child, pt, &info, - data, write_access); - - case PT_R4: case PT_R5: case PT_R6: case PT_R7: - if (write_access) { - /* read NaT bit first: */ - unsigned long dummy; - - ret = unw_get_gr(&info, (addr - PT_R4)/8 + 4, - &dummy, &nat); - if (ret < 0) - return ret; - } - return unw_access_gr(&info, (addr - PT_R4)/8 + 4, data, - &nat, write_access); - - case PT_B1: case PT_B2: case PT_B3: - case PT_B4: case PT_B5: - return unw_access_br(&info, (addr - PT_B1)/8 + 1, data, - write_access); - - case PT_AR_EC: - return unw_access_ar(&info, UNW_AR_EC, data, - write_access); - - case PT_AR_LC: - return unw_access_ar(&info, UNW_AR_LC, data, - write_access); - - default: - if (addr >= PT_F2 && addr < PT_F5 + 16) - return access_fr(&info, (addr - PT_F2)/16 + 2, - (addr & 8) != 0, data, - write_access); - else if (addr >= PT_F16 && addr < PT_F31 + 16) - return access_fr(&info, - (addr - PT_F16)/16 + 16, - (addr & 8) != 0, - data, write_access); - else { - dprintk("ptrace: rejecting access to register " - "address 0x%lx\n", addr); - return -1; - } - } - } else if (addr < PT_F9+16) { - /* scratch state */ - switch (addr) { - case PT_AR_BSP: - /* - * By convention, we use PT_AR_BSP to refer to - * the end of the user-level backing store. - * Use ia64_rse_skip_regs(PT_AR_BSP, -CFM.sof) - * to get the real value of ar.bsp at the time - * the kernel was entered. - * - * Furthermore, when changing the contents of - * PT_AR_BSP (or PT_CFM) we MUST copy any - * users-level stacked registers that are - * stored on the kernel stack back to - * user-space because otherwise, we might end - * up clobbering kernel stacked registers. - * Also, if this happens while the task is - * blocked in a system call, which convert the - * state such that the non-system-call exit - * path is used. This ensures that the proper - * state will be picked up when resuming - * execution. However, it *also* means that - * once we write PT_AR_BSP/PT_CFM, it won't be - * possible to modify the syscall arguments of - * the pending system call any longer. This - * shouldn't be an issue because modifying - * PT_AR_BSP/PT_CFM generally implies that - * we're either abandoning the pending system - * call or that we defer it's re-execution - * (e.g., due to GDB doing an inferior - * function call). - */ - urbs_end = ia64_get_user_rbs_end(child, pt, &cfm); - if (write_access) { - if (*data != urbs_end) { - if (ia64_sync_user_rbs(child, sw, - pt->ar_bspstore, - urbs_end) < 0) - return -1; - if (in_syscall(pt)) - convert_to_non_syscall(child, - pt, - cfm); - /* - * Simulate user-level write - * of ar.bsp: - */ - pt->loadrs = 0; - pt->ar_bspstore = *data; - } - } else - *data = urbs_end; - return 0; - - case PT_CFM: - urbs_end = ia64_get_user_rbs_end(child, pt, &cfm); - if (write_access) { - if (((cfm ^ *data) & PFM_MASK) != 0) { - if (ia64_sync_user_rbs(child, sw, - pt->ar_bspstore, - urbs_end) < 0) - return -1; - if (in_syscall(pt)) - convert_to_non_syscall(child, - pt, - cfm); - pt->cr_ifs = ((pt->cr_ifs & ~PFM_MASK) - | (*data & PFM_MASK)); - } - } else - *data = cfm; - return 0; - - case PT_CR_IPSR: - if (write_access) { - unsigned long tmp = *data; - /* psr.ri==3 is a reserved value: SDM 2:25 */ - if ((tmp & IA64_PSR_RI) == IA64_PSR_RI) - tmp &= ~IA64_PSR_RI; - pt->cr_ipsr = ((tmp & IPSR_MASK) - | (pt->cr_ipsr & ~IPSR_MASK)); - } else - *data = (pt->cr_ipsr & IPSR_MASK); - return 0; - - case PT_AR_RSC: - if (write_access) - pt->ar_rsc = *data | (3 << 2); /* force PL3 */ - else - *data = pt->ar_rsc; - return 0; - - case PT_AR_RNAT: - urbs_end = ia64_get_user_rbs_end(child, pt, NULL); - rnat_addr = (long) ia64_rse_rnat_addr((long *) - urbs_end); - if (write_access) - return ia64_poke(child, sw, urbs_end, - rnat_addr, *data); - else - return ia64_peek(child, sw, urbs_end, - rnat_addr, data); - - case PT_R1: - ptr = pt_reg_addr(pt, r1); - break; - case PT_R2: case PT_R3: - ptr = pt_reg_addr(pt, r2) + (addr - PT_R2); - break; - case PT_R8: case PT_R9: case PT_R10: case PT_R11: - ptr = pt_reg_addr(pt, r8) + (addr - PT_R8); - break; - case PT_R12: case PT_R13: - ptr = pt_reg_addr(pt, r12) + (addr - PT_R12); - break; - case PT_R14: - ptr = pt_reg_addr(pt, r14); - break; - case PT_R15: - ptr = pt_reg_addr(pt, r15); - break; - case PT_R16: case PT_R17: case PT_R18: case PT_R19: - case PT_R20: case PT_R21: case PT_R22: case PT_R23: - case PT_R24: case PT_R25: case PT_R26: case PT_R27: - case PT_R28: case PT_R29: case PT_R30: case PT_R31: - ptr = pt_reg_addr(pt, r16) + (addr - PT_R16); - break; - case PT_B0: - ptr = pt_reg_addr(pt, b0); - break; - case PT_B6: - ptr = pt_reg_addr(pt, b6); - break; - case PT_B7: - ptr = pt_reg_addr(pt, b7); - break; - case PT_F6: case PT_F6+8: case PT_F7: case PT_F7+8: - case PT_F8: case PT_F8+8: case PT_F9: case PT_F9+8: - ptr = pt_reg_addr(pt, f6) + (addr - PT_F6); - break; - case PT_AR_BSPSTORE: - ptr = pt_reg_addr(pt, ar_bspstore); - break; - case PT_AR_UNAT: - ptr = pt_reg_addr(pt, ar_unat); - break; - case PT_AR_PFS: - ptr = pt_reg_addr(pt, ar_pfs); - break; - case PT_AR_CCV: - ptr = pt_reg_addr(pt, ar_ccv); - break; - case PT_AR_FPSR: - ptr = pt_reg_addr(pt, ar_fpsr); - break; - case PT_CR_IIP: - ptr = pt_reg_addr(pt, cr_iip); - break; - case PT_PR: - ptr = pt_reg_addr(pt, pr); - break; - /* scratch register */ - - default: - /* disallow accessing anything else... */ - dprintk("ptrace: rejecting access to register " - "address 0x%lx\n", addr); - return -1; - } - } else if (addr <= PT_AR_SSD) { - ptr = pt_reg_addr(pt, ar_csd) + (addr - PT_AR_CSD); - } else { - /* access debug registers */ - - if (addr >= PT_IBR) { - regnum = (addr - PT_IBR) >> 3; - ptr = &child->thread.ibr[0]; - } else { - regnum = (addr - PT_DBR) >> 3; - ptr = &child->thread.dbr[0]; - } - - if (regnum >= 8) { - dprintk("ptrace: rejecting access to register " - "address 0x%lx\n", addr); - return -1; - } -#ifdef CONFIG_PERFMON - /* - * Check if debug registers are used by perfmon. This - * test must be done once we know that we can do the - * operation, i.e. the arguments are all valid, but - * before we start modifying the state. - * - * Perfmon needs to keep a count of how many processes - * are trying to modify the debug registers for system - * wide monitoring sessions. - * - * We also include read access here, because they may - * cause the PMU-installed debug register state - * (dbr[], ibr[]) to be reset. The two arrays are also - * used by perfmon, but we do not use - * IA64_THREAD_DBG_VALID. The registers are restored - * by the PMU context switch code. - */ - if (pfm_use_debug_registers(child)) return -1; -#endif - - if (!(child->thread.flags & IA64_THREAD_DBG_VALID)) { - child->thread.flags |= IA64_THREAD_DBG_VALID; - memset(child->thread.dbr, 0, - sizeof(child->thread.dbr)); - memset(child->thread.ibr, 0, - sizeof(child->thread.ibr)); - } - - ptr += regnum; - - if ((regnum & 1) && write_access) { - /* don't let the user set kernel-level breakpoints: */ - *ptr = *data & ~(7UL << 56); - return 0; - } - } - if (write_access) - *ptr = *data; - else - *data = *ptr; - return 0; -} + unsigned long *data, int write_access); static long ptrace_getregs (struct task_struct *child, struct pt_all_user_regs __user *ppr) @@ -1521,215 +1138,97 @@ ptrace_setregs (struct task_struct *child, struct pt_all_user_regs __user *ppr) return ret; } -/* - * Called by kernel/ptrace.c when detaching.. - * - * Make sure the single step bit is not set. - */ void -ptrace_disable (struct task_struct *child) +user_enable_single_step (struct task_struct *child) { struct ia64_psr *child_psr = ia64_psr(task_pt_regs(child)); - /* make sure the single step/taken-branch trap bits are not set: */ - clear_tsk_thread_flag(child, TIF_SINGLESTEP); - child_psr->ss = 0; - child_psr->tb = 0; + set_tsk_thread_flag(child, TIF_SINGLESTEP); + child_psr->ss = 1; } -asmlinkage long -sys_ptrace (long request, pid_t pid, unsigned long addr, unsigned long data) +void +user_enable_block_step (struct task_struct *child) { - struct pt_regs *pt; - unsigned long urbs_end, peek_or_poke; - struct task_struct *child; - struct switch_stack *sw; - long ret; - struct unw_frame_info info; + struct ia64_psr *child_psr = ia64_psr(task_pt_regs(child)); - lock_kernel(); - ret = -EPERM; - if (request == PTRACE_TRACEME) { - ret = ptrace_traceme(); - goto out; - } + set_tsk_thread_flag(child, TIF_SINGLESTEP); + child_psr->tb = 1; +} - peek_or_poke = (request == PTRACE_PEEKTEXT - || request == PTRACE_PEEKDATA - || request == PTRACE_POKETEXT - || request == PTRACE_POKEDATA); - ret = -ESRCH; - read_lock(&tasklist_lock); - { - child = find_task_by_pid(pid); - if (child) { - if (peek_or_poke) - child = find_thread_for_addr(child, addr); - get_task_struct(child); - } - } - read_unlock(&tasklist_lock); - if (!child) - goto out; - ret = -EPERM; - if (pid == 1) /* no messing around with init! */ - goto out_tsk; - - if (request == PTRACE_ATTACH) { - ret = ptrace_attach(child); - if (!ret) - arch_ptrace_attach(child); - goto out_tsk; - } +void +user_disable_single_step (struct task_struct *child) +{ + struct ia64_psr *child_psr = ia64_psr(task_pt_regs(child)); - ret = ptrace_check_attach(child, request == PTRACE_KILL); - if (ret < 0) - goto out_tsk; + /* make sure the single step/taken-branch trap bits are not set: */ + clear_tsk_thread_flag(child, TIF_SINGLESTEP); + child_psr->ss = 0; + child_psr->tb = 0; +} - pt = task_pt_regs(child); - sw = (struct switch_stack *) (child->thread.ksp + 16); +/* + * Called by kernel/ptrace.c when detaching.. + * + * Make sure the single step bit is not set. + */ +void +ptrace_disable (struct task_struct *child) +{ + user_disable_single_step(child); +} +long +arch_ptrace (struct task_struct *child, long request, long addr, long data) +{ switch (request) { - case PTRACE_PEEKTEXT: - case PTRACE_PEEKDATA: + case PTRACE_PEEKTEXT: + case PTRACE_PEEKDATA: /* read word at location addr */ - urbs_end = ia64_get_user_rbs_end(child, pt, NULL); - ret = ia64_peek(child, sw, urbs_end, addr, &data); - if (ret == 0) { - ret = data; - /* ensure "ret" is not mistaken as an error code: */ - force_successful_syscall_return(); - } - goto out_tsk; - - case PTRACE_POKETEXT: - case PTRACE_POKEDATA: - /* write the word at location addr */ - urbs_end = ia64_get_user_rbs_end(child, pt, NULL); - ret = ia64_poke(child, sw, urbs_end, addr, data); - - /* Make sure user RBS has the latest data */ - unw_init_from_blocked_task(&info, child); - do_sync_rbs(&info, ia64_sync_user_rbs); + if (access_process_vm(child, addr, &data, sizeof(data), 0) + != sizeof(data)) + return -EIO; + /* ensure return value is not mistaken for error code */ + force_successful_syscall_return(); + return data; - goto out_tsk; + /* PTRACE_POKETEXT and PTRACE_POKEDATA is handled + * by the generic ptrace_request(). + */ - case PTRACE_PEEKUSR: + case PTRACE_PEEKUSR: /* read the word at addr in the USER area */ - if (access_uarea(child, addr, &data, 0) < 0) { - ret = -EIO; - goto out_tsk; - } - ret = data; - /* ensure "ret" is not mistaken as an error code */ + if (access_uarea(child, addr, &data, 0) < 0) + return -EIO; + /* ensure return value is not mistaken for error code */ force_successful_syscall_return(); - goto out_tsk; + return data; - case PTRACE_POKEUSR: + case PTRACE_POKEUSR: /* write the word at addr in the USER area */ - if (access_uarea(child, addr, &data, 1) < 0) { - ret = -EIO; - goto out_tsk; - } - ret = 0; - goto out_tsk; + if (access_uarea(child, addr, &data, 1) < 0) + return -EIO; + return 0; - case PTRACE_OLD_GETSIGINFO: + case PTRACE_OLD_GETSIGINFO: /* for backwards-compatibility */ - ret = ptrace_request(child, PTRACE_GETSIGINFO, addr, data); - goto out_tsk; + return ptrace_request(child, PTRACE_GETSIGINFO, addr, data); - case PTRACE_OLD_SETSIGINFO: + case PTRACE_OLD_SETSIGINFO: /* for backwards-compatibility */ - ret = ptrace_request(child, PTRACE_SETSIGINFO, addr, data); - goto out_tsk; - - case PTRACE_SYSCALL: - /* continue and stop at next (return from) syscall */ - case PTRACE_CONT: - /* restart after signal. */ - ret = -EIO; - if (!valid_signal(data)) - goto out_tsk; - if (request == PTRACE_SYSCALL) - set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - else - clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - child->exit_code = data; + return ptrace_request(child, PTRACE_SETSIGINFO, addr, data); - /* - * Make sure the single step/taken-branch trap bits - * are not set: - */ - clear_tsk_thread_flag(child, TIF_SINGLESTEP); - ia64_psr(pt)->ss = 0; - ia64_psr(pt)->tb = 0; + case PTRACE_GETREGS: + return ptrace_getregs(child, + (struct pt_all_user_regs __user *) data); - wake_up_process(child); - ret = 0; - goto out_tsk; + case PTRACE_SETREGS: + return ptrace_setregs(child, + (struct pt_all_user_regs __user *) data); - case PTRACE_KILL: - /* - * Make the child exit. Best I can do is send it a - * sigkill. Perhaps it should be put in the status - * that it wants to exit. - */ - if (child->exit_state == EXIT_ZOMBIE) - /* already dead */ - goto out_tsk; - child->exit_code = SIGKILL; - - ptrace_disable(child); - wake_up_process(child); - ret = 0; - goto out_tsk; - - case PTRACE_SINGLESTEP: - /* let child execute for one instruction */ - case PTRACE_SINGLEBLOCK: - ret = -EIO; - if (!valid_signal(data)) - goto out_tsk; - - clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - set_tsk_thread_flag(child, TIF_SINGLESTEP); - if (request == PTRACE_SINGLESTEP) { - ia64_psr(pt)->ss = 1; - } else { - ia64_psr(pt)->tb = 1; - } - child->exit_code = data; - - /* give it a chance to run. */ - wake_up_process(child); - ret = 0; - goto out_tsk; - - case PTRACE_DETACH: - /* detach a process that was attached. */ - ret = ptrace_detach(child, data); - goto out_tsk; - - case PTRACE_GETREGS: - ret = ptrace_getregs(child, - (struct pt_all_user_regs __user *) data); - goto out_tsk; - - case PTRACE_SETREGS: - ret = ptrace_setregs(child, - (struct pt_all_user_regs __user *) data); - goto out_tsk; - - default: - ret = ptrace_request(child, request, addr, data); - goto out_tsk; + default: + return ptrace_request(child, request, addr, data); } - out_tsk: - put_task_struct(child); - out: - unlock_kernel(); - return ret; } @@ -1811,3 +1310,892 @@ syscall_trace_leave (long arg0, long arg1, long arg2, long arg3, if (test_thread_flag(TIF_RESTORE_RSE)) ia64_sync_krbs(); } + +/* Utrace implementation starts here */ +struct regset_get { + void *kbuf; + void __user *ubuf; +}; + +struct regset_set { + const void *kbuf; + const void __user *ubuf; +}; + +struct regset_getset { + struct task_struct *target; + const struct user_regset *regset; + union { + struct regset_get get; + struct regset_set set; + } u; + unsigned int pos; + unsigned int count; + int ret; +}; + +static int +access_elf_gpreg(struct task_struct *target, struct unw_frame_info *info, + unsigned long addr, unsigned long *data, int write_access) +{ + struct pt_regs *pt; + unsigned long *ptr = NULL; + int ret; + char nat = 0; + + pt = task_pt_regs(target); + switch (addr) { + case ELF_GR_OFFSET(1): + ptr = &pt->r1; + break; + case ELF_GR_OFFSET(2): + case ELF_GR_OFFSET(3): + ptr = (void *)&pt->r2 + (addr - ELF_GR_OFFSET(2)); + break; + case ELF_GR_OFFSET(4) ... ELF_GR_OFFSET(7): + if (write_access) { + /* read NaT bit first: */ + unsigned long dummy; + + ret = unw_get_gr(info, addr/8, &dummy, &nat); + if (ret < 0) + return ret; + } + return unw_access_gr(info, addr/8, data, &nat, write_access); + case ELF_GR_OFFSET(8) ... ELF_GR_OFFSET(11): + ptr = (void *)&pt->r8 + addr - ELF_GR_OFFSET(8); + break; + case ELF_GR_OFFSET(12): + case ELF_GR_OFFSET(13): + ptr = (void *)&pt->r12 + addr - ELF_GR_OFFSET(12); + break; + case ELF_GR_OFFSET(14): + ptr = &pt->r14; + break; + case ELF_GR_OFFSET(15): + ptr = &pt->r15; + } + if (write_access) + *ptr = *data; + else + *data = *ptr; + return 0; +} + +static int +access_elf_breg(struct task_struct *target, struct unw_frame_info *info, + unsigned long addr, unsigned long *data, int write_access) +{ + struct pt_regs *pt; + unsigned long *ptr = NULL; + + pt = task_pt_regs(target); + switch (addr) { + case ELF_BR_OFFSET(0): + ptr = &pt->b0; + break; + case ELF_BR_OFFSET(1) ... ELF_BR_OFFSET(5): + return unw_access_br(info, (addr - ELF_BR_OFFSET(0))/8, + data, write_access); + case ELF_BR_OFFSET(6): + ptr = &pt->b6; + break; + case ELF_BR_OFFSET(7): + ptr = &pt->b7; + } + if (write_access) + *ptr = *data; + else + *data = *ptr; + return 0; +} + +static int +access_elf_areg(struct task_struct *target, struct unw_frame_info *info, + unsigned long addr, unsigned long *data, int write_access) +{ + struct pt_regs *pt; + unsigned long cfm, urbs_end; + unsigned long *ptr = NULL; + + pt = task_pt_regs(target); + if (addr >= ELF_AR_RSC_OFFSET && addr <= ELF_AR_SSD_OFFSET) { + switch (addr) { + case ELF_AR_RSC_OFFSET: + /* force PL3 */ + if (write_access) + pt->ar_rsc = *data | (3 << 2); + else + *data = pt->ar_rsc; + return 0; + case ELF_AR_BSP_OFFSET: + /* + * By convention, we use PT_AR_BSP to refer to + * the end of the user-level backing store. + * Use ia64_rse_skip_regs(PT_AR_BSP, -CFM.sof) + * to get the real value of ar.bsp at the time + * the kernel was entered. + * + * Furthermore, when changing the contents of + * PT_AR_BSP (or PT_CFM) while the task is + * blocked in a system call, convert the state + * so that the non-system-call exit + * path is used. This ensures that the proper + * state will be picked up when resuming + * execution. However, it *also* means that + * once we write PT_AR_BSP/PT_CFM, it won't be + * possible to modify the syscall arguments of + * the pending system call any longer. This + * shouldn't be an issue because modifying + * PT_AR_BSP/PT_CFM generally implies that + * we're either abandoning the pending system + * call or that we defer it's re-execution + * (e.g., due to GDB doing an inferior + * function call). + */ + urbs_end = ia64_get_user_rbs_end(target, pt, &cfm); + if (write_access) { + if (*data != urbs_end) { + if (in_syscall(pt)) + convert_to_non_syscall(target, + pt, + cfm); + /* + * Simulate user-level write + * of ar.bsp: + */ + pt->loadrs = 0; + pt->ar_bspstore = *data; + } + } else + *data = urbs_end; + return 0; + case ELF_AR_BSPSTORE_OFFSET: + ptr = &pt->ar_bspstore; + break; + case ELF_AR_RNAT_OFFSET: + ptr = &pt->ar_rnat; + break; + case ELF_AR_CCV_OFFSET: + ptr = &pt->ar_ccv; + break; + case ELF_AR_UNAT_OFFSET: + ptr = &pt->ar_unat; + break; + case ELF_AR_FPSR_OFFSET: + ptr = &pt->ar_fpsr; + break; + case ELF_AR_PFS_OFFSET: + ptr = &pt->ar_pfs; + break; + case ELF_AR_LC_OFFSET: + return unw_access_ar(info, UNW_AR_LC, data, + write_access); + case ELF_AR_EC_OFFSET: + return unw_access_ar(info, UNW_AR_EC, data, + write_access); + case ELF_AR_CSD_OFFSET: + ptr = &pt->ar_csd; + break; + case ELF_AR_SSD_OFFSET: + ptr = &pt->ar_ssd; + } + } else if (addr >= ELF_CR_IIP_OFFSET && addr <= ELF_CR_IPSR_OFFSET) { + switch (addr) { + case ELF_CR_IIP_OFFSET: + ptr = &pt->cr_iip; + break; + case ELF_CFM_OFFSET: + urbs_end = ia64_get_user_rbs_end(target, pt, &cfm); + if (write_access) { + if (((cfm ^ *data) & PFM_MASK) != 0) { + if (in_syscall(pt)) + convert_to_non_syscall(target, + pt, + cfm); + pt->cr_ifs = ((pt->cr_ifs & ~PFM_MASK) + | (*data & PFM_MASK)); + } + } else + *data = cfm; + return 0; + case ELF_CR_IPSR_OFFSET: + if (write_access) { + unsigned long tmp = *data; + /* psr.ri==3 is a reserved value: SDM 2:25 */ + if ((tmp & IA64_PSR_RI) == IA64_PSR_RI) + tmp &= ~IA64_PSR_RI; + pt->cr_ipsr = ((tmp & IPSR_MASK) + | (pt->cr_ipsr & ~IPSR_MASK)); + } else + *data = (pt->cr_ipsr & IPSR_MASK); + return 0; + } + } else if (addr == ELF_NAT_OFFSET) + return access_nat_bits(target, pt, info, + data, write_access); + else if (addr == ELF_PR_OFFSET) + ptr = &pt->pr; + else + return -1; + + if (write_access) + *ptr = *data; + else + *data = *ptr; + + return 0; +} + +static int +access_elf_reg(struct task_struct *target, struct unw_frame_info *info, + unsigned long addr, unsigned long *data, int write_access) +{ + if (addr >= ELF_GR_OFFSET(1) && addr <= ELF_GR_OFFSET(15)) + return access_elf_gpreg(target, info, addr, data, write_access); + else if (addr >= ELF_BR_OFFSET(0) && addr <= ELF_BR_OFFSET(7)) + return access_elf_breg(target, info, addr, data, write_access); + else + return access_elf_areg(target, info, addr, data, write_access); +} + +void do_gpregs_get(struct unw_frame_info *info, void *arg) +{ + struct pt_regs *pt; + struct regset_getset *dst = arg; + elf_greg_t tmp[16]; + unsigned int i, index, min_copy; + + if (unw_unwind_to_user(info) < 0) + return; + + /* + * coredump format: + * r0-r31 + * NaT bits (for r0-r31; bit N == 1 iff rN is a NaT) + * predicate registers (p0-p63) + * b0-b7 + * ip cfm user-mask + * ar.rsc ar.bsp ar.bspstore ar.rnat + * ar.ccv ar.unat ar.fpsr ar.pfs ar.lc ar.ec + */ + + + /* Skip r0 */ + if (dst->count > 0 && dst->pos < ELF_GR_OFFSET(1)) { + dst->ret = user_regset_copyout_zero(&dst->pos, &dst->count, + &dst->u.get.kbuf, + &dst->u.get.ubuf, + 0, ELF_GR_OFFSET(1)); + if (dst->ret || dst->count == 0) + return; + } + + /* gr1 - gr15 */ + if (dst->count > 0 && dst->pos < ELF_GR_OFFSET(16)) { + index = (dst->pos - ELF_GR_OFFSET(1)) / sizeof(elf_greg_t); + min_copy = ELF_GR_OFFSET(16) > (dst->pos + dst->count) ? + (dst->pos + dst->count) : ELF_GR_OFFSET(16); + for (i = dst->pos; i < min_copy; i += sizeof(elf_greg_t), + index++) + if (access_elf_reg(dst->target, info, i, + &tmp[index], 0) < 0) { + dst->ret = -EIO; + return; + } + dst->ret = user_regset_copyout(&dst->pos, &dst->count, + &dst->u.get.kbuf, &dst->u.get.ubuf, tmp, + ELF_GR_OFFSET(1), ELF_GR_OFFSET(16)); + if (dst->ret || dst->count == 0) + return; + } + + /* r16-r31 */ + if (dst->count > 0 && dst->pos < ELF_NAT_OFFSET) { + pt = task_pt_regs(dst->target); + dst->ret = user_regset_copyout(&dst->pos, &dst->count, + &dst->u.get.kbuf, &dst->u.get.ubuf, &pt->r16, + ELF_GR_OFFSET(16), ELF_NAT_OFFSET); + if (dst->ret || dst->count == 0) + return; + } + + /* nat, pr, b0 - b7 */ + if (dst->count > 0 && dst->pos < ELF_CR_IIP_OFFSET) { + index = (dst->pos - ELF_NAT_OFFSET) / sizeof(elf_greg_t); + min_copy = ELF_CR_IIP_OFFSET > (dst->pos + dst->count) ? + (dst->pos + dst->count) : ELF_CR_IIP_OFFSET; + for (i = dst->pos; i < min_copy; i += sizeof(elf_greg_t), + index++) + if (access_elf_reg(dst->target, info, i, + &tmp[index], 0) < 0) { + dst->ret = -EIO; + return; + } + dst->ret = user_regset_copyout(&dst->pos, &dst->count, + &dst->u.get.kbuf, &dst->u.get.ubuf, tmp, + ELF_NAT_OFFSET, ELF_CR_IIP_OFFSET); + if (dst->ret || dst->count == 0) + return; + } + + /* ip cfm psr ar.rsc ar.bsp ar.bspstore ar.rnat + * ar.ccv ar.unat ar.fpsr ar.pfs ar.lc ar.ec ar.csd ar.ssd + */ + if (dst->count > 0 && dst->pos < (ELF_AR_END_OFFSET)) { + index = (dst->pos - ELF_CR_IIP_OFFSET) / sizeof(elf_greg_t); + min_copy = ELF_AR_END_OFFSET > (dst->pos + dst->count) ? + (dst->pos + dst->count) : ELF_AR_END_OFFSET; + for (i = dst->pos; i < min_copy; i += sizeof(elf_greg_t), + index++) + if (access_elf_reg(dst->target, info, i, + &tmp[index], 0) < 0) { + dst->ret = -EIO; + return; + } + dst->ret = user_regset_copyout(&dst->pos, &dst->count, + &dst->u.get.kbuf, &dst->u.get.ubuf, tmp, + ELF_CR_IIP_OFFSET, ELF_AR_END_OFFSET); + } +} + +void do_gpregs_set(struct unw_frame_info *info, void *arg) +{ + struct pt_regs *pt; + struct regset_getset *dst = arg; + elf_greg_t tmp[16]; + unsigned int i, index; + + if (unw_unwind_to_user(info) < 0) + return; + + /* Skip r0 */ + if (dst->count > 0 && dst->pos < ELF_GR_OFFSET(1)) { + dst->ret = user_regset_copyin_ignore(&dst->pos, &dst->count, + &dst->u.set.kbuf, + &dst->u.set.ubuf, + 0, ELF_GR_OFFSET(1)); + if (dst->ret || dst->count == 0) + return; + } + + /* gr1-gr15 */ + if (dst->count > 0 && dst->pos < ELF_GR_OFFSET(16)) { + i = dst->pos; + index = (dst->pos - ELF_GR_OFFSET(1)) / sizeof(elf_greg_t); + dst->ret = user_regset_copyin(&dst->pos, &dst->count, + &dst->u.set.kbuf, &dst->u.set.ubuf, tmp, + ELF_GR_OFFSET(1), ELF_GR_OFFSET(16)); + if (dst->ret) + return; + for ( ; i < dst->pos; i += sizeof(elf_greg_t), index++) + if (access_elf_reg(dst->target, info, i, + &tmp[index], 1) < 0) { + dst->ret = -EIO; + return; + } + if (dst->count == 0) + return; + } + + /* gr16-gr31 */ + if (dst->count > 0 && dst->pos < ELF_NAT_OFFSET) { + pt = task_pt_regs(dst->target); + dst->ret = user_regset_copyin(&dst->pos, &dst->count, + &dst->u.set.kbuf, &dst->u.set.ubuf, &pt->r16, + ELF_GR_OFFSET(16), ELF_NAT_OFFSET); + if (dst->ret || dst->count == 0) + return; + } + + /* nat, pr, b0 - b7 */ + if (dst->count > 0 && dst->pos < ELF_CR_IIP_OFFSET) { + i = dst->pos; + index = (dst->pos - ELF_NAT_OFFSET) / sizeof(elf_greg_t); + dst->ret = user_regset_copyin(&dst->pos, &dst->count, + &dst->u.set.kbuf, &dst->u.set.ubuf, tmp, + ELF_NAT_OFFSET, ELF_CR_IIP_OFFSET); + if (dst->ret) + return; + for (; i < dst->pos; i += sizeof(elf_greg_t), index++) + if (access_elf_reg(dst->target, info, i, + &tmp[index], 1) < 0) { + dst->ret = -EIO; + return; + } + if (dst->count == 0) + return; + } + + /* ip cfm psr ar.rsc ar.bsp ar.bspstore ar.rnat + * ar.ccv ar.unat ar.fpsr ar.pfs ar.lc ar.ec ar.csd ar.ssd + */ + if (dst->count > 0 && dst->pos < (ELF_AR_END_OFFSET)) { + i = dst->pos; + index = (dst->pos - ELF_CR_IIP_OFFSET) / sizeof(elf_greg_t); + dst->ret = user_regset_copyin(&dst->pos, &dst->count, + &dst->u.set.kbuf, &dst->u.set.ubuf, tmp, + ELF_CR_IIP_OFFSET, ELF_AR_END_OFFSET); + if (dst->ret) + return; + for ( ; i < dst->pos; i += sizeof(elf_greg_t), index++) + if (access_elf_reg(dst->target, info, i, + &tmp[index], 1) < 0) { + dst->ret = -EIO; + return; + } + } +} + +#define ELF_FP_OFFSET(i) (i * sizeof(elf_fpreg_t)) + +void do_fpregs_get(struct unw_frame_info *info, void *arg) +{ + struct regset_getset *dst = arg; + struct task_struct *task = dst->target; + elf_fpreg_t tmp[30]; + int index, min_copy, i; + + if (unw_unwind_to_user(info) < 0) + return; + + /* Skip pos 0 and 1 */ + if (dst->count > 0 && dst->pos < ELF_FP_OFFSET(2)) { + dst->ret = user_regset_copyout_zero(&dst->pos, &dst->count, + &dst->u.get.kbuf, + &dst->u.get.ubuf, + 0, ELF_FP_OFFSET(2)); + if (dst->count == 0 || dst->ret) + return; + } + + /* fr2-fr31 */ + if (dst->count > 0 && dst->pos < ELF_FP_OFFSET(32)) { + index = (dst->pos - ELF_FP_OFFSET(2)) / sizeof(elf_fpreg_t); + + min_copy = min(((unsigned int)ELF_FP_OFFSET(32)), + dst->pos + dst->count); + for (i = dst->pos; i < min_copy; i += sizeof(elf_fpreg_t), + index++) + if (unw_get_fr(info, i / sizeof(elf_fpreg_t), + &tmp[index])) { + dst->ret = -EIO; + return; + } + dst->ret = user_regset_copyout(&dst->pos, &dst->count, + &dst->u.get.kbuf, &dst->u.get.ubuf, tmp, + ELF_FP_OFFSET(2), ELF_FP_OFFSET(32)); + if (dst->count == 0 || dst->ret) + return; + } + + /* fph */ + if (dst->count > 0) { + ia64_flush_fph(dst->target); + if (task->thread.flags & IA64_THREAD_FPH_VALID) + dst->ret = user_regset_copyout( + &dst->pos, &dst->count, + &dst->u.get.kbuf, &dst->u.get.ubuf, + &dst->target->thread.fph, + ELF_FP_OFFSET(32), -1); + else + /* Zero fill instead. */ + dst->ret = user_regset_copyout_zero( + &dst->pos, &dst->count, + &dst->u.get.kbuf, &dst->u.get.ubuf, + ELF_FP_OFFSET(32), -1); + } +} + +void do_fpregs_set(struct unw_frame_info *info, void *arg) +{ + struct regset_getset *dst = arg; + elf_fpreg_t fpreg, tmp[30]; + int index, start, end; + + if (unw_unwind_to_user(info) < 0) + return; + + /* Skip pos 0 and 1 */ + if (dst->count > 0 && dst->pos < ELF_FP_OFFSET(2)) { + dst->ret = user_regset_copyin_ignore(&dst->pos, &dst->count, + &dst->u.set.kbuf, + &dst->u.set.ubuf, + 0, ELF_FP_OFFSET(2)); + if (dst->count == 0 || dst->ret) + return; + } + + /* fr2-fr31 */ + if (dst->count > 0 && dst->pos < ELF_FP_OFFSET(32)) { + start = dst->pos; + end = min(((unsigned int)ELF_FP_OFFSET(32)), + dst->pos + dst->count); + dst->ret = user_regset_copyin(&dst->pos, &dst->count, + &dst->u.set.kbuf, &dst->u.set.ubuf, tmp, + ELF_FP_OFFSET(2), ELF_FP_OFFSET(32)); + if (dst->ret) + return; + + if (start & 0xF) { /* only write high part */ + if (unw_get_fr(info, start / sizeof(elf_fpreg_t), + &fpreg)) { + dst->ret = -EIO; + return; + } + tmp[start / sizeof(elf_fpreg_t) - 2].u.bits[0] + = fpreg.u.bits[0]; + start &= ~0xFUL; + } + if (end & 0xF) { /* only write low part */ + if (unw_get_fr(info, end / sizeof(elf_fpreg_t), + &fpreg)) { + dst->ret = -EIO; + return; + } + tmp[end / sizeof(elf_fpreg_t) - 2].u.bits[1] + = fpreg.u.bits[1]; + end = (end + 0xF) & ~0xFUL; + } + + for ( ; start < end ; start += sizeof(elf_fpreg_t)) { + index = start / sizeof(elf_fpreg_t); + if (unw_set_fr(info, index, tmp[index - 2])) { + dst->ret = -EIO; + return; + } + } + if (dst->ret || dst->count == 0) + return; + } + + /* fph */ + if (dst->count > 0 && dst->pos < ELF_FP_OFFSET(128)) { + ia64_sync_fph(dst->target); + dst->ret = user_regset_copyin(&dst->pos, &dst->count, + &dst->u.set.kbuf, + &dst->u.set.ubuf, + &dst->target->thread.fph, + ELF_FP_OFFSET(32), -1); + } +} + +static int +do_regset_call(void (*call)(struct unw_frame_info *, void *), + struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + const void *kbuf, const void __user *ubuf) +{ + struct regset_getset info = { .target = target, .regset = regset, + .pos = pos, .count = count, + .u.set = { .kbuf = kbuf, .ubuf = ubuf }, + .ret = 0 }; + + if (target == current) + unw_init_running(call, &info); + else { + struct unw_frame_info ufi; + memset(&ufi, 0, sizeof(ufi)); + unw_init_from_blocked_task(&ufi, target); + (*call)(&ufi, &info); + } + + return info.ret; +} + +static int +gpregs_get(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + void *kbuf, void __user *ubuf) +{ + return do_regset_call(do_gpregs_get, target, regset, pos, count, + kbuf, ubuf); +} + +static int gpregs_set(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + const void *kbuf, const void __user *ubuf) +{ + return do_regset_call(do_gpregs_set, target, regset, pos, count, + kbuf, ubuf); +} + +static void do_gpregs_writeback(struct unw_frame_info *info, void *arg) +{ + do_sync_rbs(info, ia64_sync_user_rbs); +} + +/* + * This is called to write back the register backing store. + * ptrace does this before it stops, so that a tracer reading the user + * memory after the thread stops will get the current register data. + */ +static int +gpregs_writeback(struct task_struct *target, + const struct user_regset *regset, + int now) +{ + if (test_and_set_tsk_thread_flag(target, TIF_RESTORE_RSE)) + return 0; + tsk_set_notify_resume(target); + return do_regset_call(do_gpregs_writeback, target, regset, 0, 0, + NULL, NULL); +} + +static int +fpregs_active(struct task_struct *target, const struct user_regset *regset) +{ + return (target->thread.flags & IA64_THREAD_FPH_VALID) ? 128 : 32; +} + +static int fpregs_get(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + void *kbuf, void __user *ubuf) +{ + return do_regset_call(do_fpregs_get, target, regset, pos, count, + kbuf, ubuf); +} + +static int fpregs_set(struct task_struct *target, + const struct user_regset *regset, + unsigned int pos, unsigned int count, + const void *kbuf, const void __user *ubuf) +{ + return do_regset_call(do_fpregs_set, target, regset, pos, count, + kbuf, ubuf); +} + +static int +access_uarea(struct task_struct *child, unsigned long addr, + unsigned long *data, int write_access) +{ + unsigned int pos = -1; /* an invalid value */ + int ret; + unsigned long *ptr, regnum; + + if ((addr & 0x7) != 0) { + dprintk("ptrace: unaligned register address 0x%lx\n", addr); + return -1; + } + if ((addr >= PT_NAT_BITS + 8 && addr < PT_F2) || + (addr >= PT_R7 + 8 && addr < PT_B1) || + (addr >= PT_AR_LC + 8 && addr < PT_CR_IPSR) || + (addr >= PT_AR_SSD + 8 && addr < PT_DBR)) { + dprintk("ptrace: rejecting access to register " + "address 0x%lx\n", addr); + return -1; + } + + switch (addr) { + case PT_F32 ... (PT_F127 + 15): + pos = addr - PT_F32 + ELF_FP_OFFSET(32); + break; + case PT_F2 ... (PT_F5 + 15): + pos = addr - PT_F2 + ELF_FP_OFFSET(2); + break; + case PT_F10 ... (PT_F31 + 15): + pos = addr - PT_F10 + ELF_FP_OFFSET(10); + break; + case PT_F6 ... (PT_F9 + 15): + pos = addr - PT_F6 + ELF_FP_OFFSET(6); + break; + } + + if (pos != -1) { + if (write_access) + ret = fpregs_set(child, NULL, pos, + sizeof(unsigned long), data, NULL); + else + ret = fpregs_get(child, NULL, pos, + sizeof(unsigned long), data, NULL); + if (ret != 0) + return -1; + return 0; + } + + switch (addr) { + case PT_NAT_BITS: + pos = ELF_NAT_OFFSET; + break; + case PT_R4 ... PT_R7: + pos = addr - PT_R4 + ELF_GR_OFFSET(4); + break; + case PT_B1 ... PT_B5: + pos = addr - PT_B1 + ELF_BR_OFFSET(1); + break; + case PT_AR_EC: + pos = ELF_AR_EC_OFFSET; + break; + case PT_AR_LC: + pos = ELF_AR_LC_OFFSET; + break; + case PT_CR_IPSR: + pos = ELF_CR_IPSR_OFFSET; + break; + case PT_CR_IIP: + pos = ELF_CR_IIP_OFFSET; + break; + case PT_CFM: + pos = ELF_CFM_OFFSET; + break; + case PT_AR_UNAT: + pos = ELF_AR_UNAT_OFFSET; + break; + case PT_AR_PFS: + pos = ELF_AR_PFS_OFFSET; + break; + case PT_AR_RSC: + pos = ELF_AR_RSC_OFFSET; + break; + case PT_AR_RNAT: + pos = ELF_AR_RNAT_OFFSET; + break; + case PT_AR_BSPSTORE: + pos = ELF_AR_BSPSTORE_OFFSET; + break; + case PT_PR: + pos = ELF_PR_OFFSET; + break; + case PT_B6: + pos = ELF_BR_OFFSET(6); + break; + case PT_AR_BSP: + pos = ELF_AR_BSP_OFFSET; + break; + case PT_R1 ... PT_R3: + pos = addr - PT_R1 + ELF_GR_OFFSET(1); + break; + case PT_R12 ... PT_R15: + pos = addr - PT_R12 + ELF_GR_OFFSET(12); + break; + case PT_R8 ... PT_R11: + pos = addr - PT_R8 + ELF_GR_OFFSET(8); + break; + case PT_R16 ... PT_R31: + pos = addr - PT_R16 + ELF_GR_OFFSET(16); + break; + case PT_AR_CCV: + pos = ELF_AR_CCV_OFFSET; + break; + case PT_AR_FPSR: + pos = ELF_AR_FPSR_OFFSET; + break; + case PT_B0: + pos = ELF_BR_OFFSET(0); + break; + case PT_B7: + pos = ELF_BR_OFFSET(7); + break; + case PT_AR_CSD: + pos = ELF_AR_CSD_OFFSET; + break; + case PT_AR_SSD: + pos = ELF_AR_SSD_OFFSET; + break; + } + + if (pos != -1) { + if (write_access) + ret = gpregs_set(child, NULL, pos, + sizeof(unsigned long), data, NULL); + else + ret = gpregs_get(child, NULL, pos, + sizeof(unsigned long), data, NULL); + if (ret != 0) + return -1; + return 0; + } + + /* access debug registers */ + if (addr >= PT_IBR) { + regnum = (addr - PT_IBR) >> 3; + ptr = &child->thread.ibr[0]; + } else { + regnum = (addr - PT_DBR) >> 3; + ptr = &child->thread.dbr[0]; + } + + if (regnum >= 8) { + dprintk("ptrace: rejecting access to register " + "address 0x%lx\n", addr); + return -1; + } +#ifdef CONFIG_PERFMON + /* + * Check if debug registers are used by perfmon. This + * test must be done once we know that we can do the + * operation, i.e. the arguments are all valid, but + * before we start modifying the state. + * + * Perfmon needs to keep a count of how many processes + * are trying to modify the debug registers for system + * wide monitoring sessions. + * + * We also include read access here, because they may + * cause the PMU-installed debug register state + * (dbr[], ibr[]) to be reset. The two arrays are also + * used by perfmon, but we do not use + * IA64_THREAD_DBG_VALID. The registers are restored + * by the PMU context switch code. + */ + if (pfm_use_debug_registers(child)) + return -1; +#endif + + if (!(child->thread.flags & IA64_THREAD_DBG_VALID)) { + child->thread.flags |= IA64_THREAD_DBG_VALID; + memset(child->thread.dbr, 0, + sizeof(child->thread.dbr)); + memset(child->thread.ibr, 0, + sizeof(child->thread.ibr)); + } + + ptr += regnum; + + if ((regnum & 1) && write_access) { + /* don't let the user set kernel-level breakpoints: */ + *ptr = *data & ~(7UL << 56); + return 0; + } + if (write_access) + *ptr = *data; + else + *data = *ptr; + return 0; +} + +static const struct user_regset native_regsets[] = { + { + .core_note_type = NT_PRSTATUS, + .n = ELF_NGREG, + .size = sizeof(elf_greg_t), .align = sizeof(elf_greg_t), + .get = gpregs_get, .set = gpregs_set, + .writeback = gpregs_writeback + }, + { + .core_note_type = NT_PRFPREG, + .n = ELF_NFPREG, + .size = sizeof(elf_fpreg_t), .align = sizeof(elf_fpreg_t), + .get = fpregs_get, .set = fpregs_set, .active = fpregs_active + }, +}; + +static const struct user_regset_view user_ia64_view = { + .name = "ia64", + .e_machine = EM_IA_64, + .regsets = native_regsets, .n = ARRAY_SIZE(native_regsets) +}; + +const struct user_regset_view *task_user_regset_view(struct task_struct *tsk) +{ +#ifdef CONFIG_IA32_SUPPORT + extern const struct user_regset_view user_ia32_view; + if (IS_IA32_PROCESS(task_pt_regs(tsk))) + return &user_ia32_view; +#endif + return &user_ia64_view; +} diff --git a/arch/ia64/kernel/sal.c b/arch/ia64/kernel/sal.c index f44fe8412162..a3022dc48ef8 100644 --- a/arch/ia64/kernel/sal.c +++ b/arch/ia64/kernel/sal.c @@ -109,6 +109,13 @@ check_versions (struct ia64_sal_systab *systab) sal_revision = SAL_VERSION_CODE(2, 8); sal_version = SAL_VERSION_CODE(0, 0); } + + if (ia64_platform_is("sn2") && (sal_revision == SAL_VERSION_CODE(2, 9))) + /* + * SGI Altix has hard-coded version 2.9 in their prom + * but they actually implement 3.2, so let's fix it here. + */ + sal_revision = SAL_VERSION_CODE(3, 2); } static void __init diff --git a/arch/ia64/kernel/salinfo.c b/arch/ia64/kernel/salinfo.c index 779c3cca206c..b11bb50a197a 100644 --- a/arch/ia64/kernel/salinfo.c +++ b/arch/ia64/kernel/salinfo.c @@ -44,8 +44,8 @@ #include <linux/smp.h> #include <linux/timer.h> #include <linux/vmalloc.h> +#include <linux/semaphore.h> -#include <asm/semaphore.h> #include <asm/sal.h> #include <asm/uaccess.h> diff --git a/arch/ia64/kernel/semaphore.c b/arch/ia64/kernel/semaphore.c deleted file mode 100644 index 2724ef3fbae2..000000000000 --- a/arch/ia64/kernel/semaphore.c +++ /dev/null @@ -1,165 +0,0 @@ -/* - * IA-64 semaphore implementation (derived from x86 version). - * - * Copyright (C) 1999-2000, 2002 Hewlett-Packard Co - * David Mosberger-Tang <davidm@hpl.hp.com> - */ - -/* - * Semaphores are implemented using a two-way counter: The "count" - * variable is decremented for each process that tries to acquire the - * semaphore, while the "sleepers" variable is a count of such - * acquires. - * - * Notably, the inline "up()" and "down()" functions can efficiently - * test if they need to do any extra work (up needs to do something - * only if count was negative before the increment operation. - * - * "sleeping" and the contention routine ordering is protected - * by the spinlock in the semaphore's waitqueue head. - * - * Note that these functions are only called when there is contention - * on the lock, and as such all this is the "non-critical" part of the - * whole semaphore business. The critical part is the inline stuff in - * <asm/semaphore.h> where we want to avoid any extra jumps and calls. - */ -#include <linux/sched.h> -#include <linux/init.h> - -#include <asm/errno.h> -#include <asm/semaphore.h> - -/* - * Logic: - * - Only on a boundary condition do we need to care. When we go - * from a negative count to a non-negative, we wake people up. - * - When we go from a non-negative count to a negative do we - * (a) synchronize with the "sleepers" count and (b) make sure - * that we're on the wakeup list before we synchronize so that - * we cannot lose wakeup events. - */ - -void -__up (struct semaphore *sem) -{ - wake_up(&sem->wait); -} - -void __sched __down (struct semaphore *sem) -{ - struct task_struct *tsk = current; - DECLARE_WAITQUEUE(wait, tsk); - unsigned long flags; - - tsk->state = TASK_UNINTERRUPTIBLE; - spin_lock_irqsave(&sem->wait.lock, flags); - add_wait_queue_exclusive_locked(&sem->wait, &wait); - - sem->sleepers++; - for (;;) { - int sleepers = sem->sleepers; - - /* - * Add "everybody else" into it. They aren't - * playing, because we own the spinlock in - * the wait_queue_head. - */ - if (!atomic_add_negative(sleepers - 1, &sem->count)) { - sem->sleepers = 0; - break; - } - sem->sleepers = 1; /* us - see -1 above */ - spin_unlock_irqrestore(&sem->wait.lock, flags); - - schedule(); - - spin_lock_irqsave(&sem->wait.lock, flags); - tsk->state = TASK_UNINTERRUPTIBLE; - } - remove_wait_queue_locked(&sem->wait, &wait); - wake_up_locked(&sem->wait); - spin_unlock_irqrestore(&sem->wait.lock, flags); - tsk->state = TASK_RUNNING; -} - -int __sched __down_interruptible (struct semaphore * sem) -{ - int retval = 0; - struct task_struct *tsk = current; - DECLARE_WAITQUEUE(wait, tsk); - unsigned long flags; - - tsk->state = TASK_INTERRUPTIBLE; - spin_lock_irqsave(&sem->wait.lock, flags); - add_wait_queue_exclusive_locked(&sem->wait, &wait); - - sem->sleepers ++; - for (;;) { - int sleepers = sem->sleepers; - - /* - * With signals pending, this turns into - * the trylock failure case - we won't be - * sleeping, and we* can't get the lock as - * it has contention. Just correct the count - * and exit. - */ - if (signal_pending(current)) { - retval = -EINTR; - sem->sleepers = 0; - atomic_add(sleepers, &sem->count); - break; - } - - /* - * Add "everybody else" into it. They aren't - * playing, because we own the spinlock in - * wait_queue_head. The "-1" is because we're - * still hoping to get the semaphore. - */ - if (!atomic_add_negative(sleepers - 1, &sem->count)) { - sem->sleepers = 0; - break; - } - sem->sleepers = 1; /* us - see -1 above */ - spin_unlock_irqrestore(&sem->wait.lock, flags); - - schedule(); - - spin_lock_irqsave(&sem->wait.lock, flags); - tsk->state = TASK_INTERRUPTIBLE; - } - remove_wait_queue_locked(&sem->wait, &wait); - wake_up_locked(&sem->wait); - spin_unlock_irqrestore(&sem->wait.lock, flags); - - tsk->state = TASK_RUNNING; - return retval; -} - -/* - * Trylock failed - make sure we correct for having decremented the - * count. - */ -int -__down_trylock (struct semaphore *sem) -{ - unsigned long flags; - int sleepers; - - spin_lock_irqsave(&sem->wait.lock, flags); - sleepers = sem->sleepers + 1; - sem->sleepers = 0; - - /* - * Add "everybody else" and us into it. They aren't - * playing, because we own the spinlock in the - * wait_queue_head. - */ - if (!atomic_add_negative(sleepers, &sem->count)) { - wake_up_locked(&sem->wait); - } - - spin_unlock_irqrestore(&sem->wait.lock, flags); - return 1; -} diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c index ebd1a09f3201..5015ca1275ca 100644 --- a/arch/ia64/kernel/setup.c +++ b/arch/ia64/kernel/setup.c @@ -59,6 +59,7 @@ #include <asm/setup.h> #include <asm/smp.h> #include <asm/system.h> +#include <asm/tlbflush.h> #include <asm/unistd.h> #include <asm/hpsim.h> @@ -176,6 +177,29 @@ filter_rsvd_memory (unsigned long start, unsigned long end, void *arg) return 0; } +/* + * Similar to "filter_rsvd_memory()", but the reserved memory ranges + * are not filtered out. + */ +int __init +filter_memory(unsigned long start, unsigned long end, void *arg) +{ + void (*func)(unsigned long, unsigned long, int); + +#if IGNORE_PFN0 + if (start == PAGE_OFFSET) { + printk(KERN_WARNING "warning: skipping physical page 0\n"); + start += PAGE_SIZE; + if (start >= end) + return 0; + } +#endif + func = arg; + if (start < end) + call_pernode_memory(__pa(start), end - start, func); + return 0; +} + static void __init sort_regions (struct rsvd_region *rsvd_region, int max) { @@ -493,6 +517,8 @@ setup_arch (char **cmdline_p) acpi_table_init(); # ifdef CONFIG_ACPI_NUMA acpi_numa_init(); + per_cpu_scan_finalize((cpus_weight(early_cpu_possible_map) == 0 ? + 32 : cpus_weight(early_cpu_possible_map)), additional_cpus); # endif #else # ifdef CONFIG_SMP @@ -690,7 +716,7 @@ get_model_name(__u8 family, __u8 model) if (overflow++ == 0) printk(KERN_ERR "%s: Table overflow. Some processor model information will be missing\n", - __FUNCTION__); + __func__); return "Unknown"; } @@ -785,7 +811,7 @@ get_max_cacheline_size (void) status = ia64_pal_cache_summary(&levels, &unique_caches); if (status != 0) { printk(KERN_ERR "%s: ia64_pal_cache_summary() failed (status=%ld)\n", - __FUNCTION__, status); + __func__, status); max = SMP_CACHE_BYTES; /* Safest setup for "flush_icache_range()" */ ia64_i_cache_stride_shift = I_CACHE_STRIDE_SHIFT; @@ -798,7 +824,7 @@ get_max_cacheline_size (void) if (status != 0) { printk(KERN_ERR "%s: ia64_pal_cache_config_info(l=%lu, 2) failed (status=%ld)\n", - __FUNCTION__, l, status); + __func__, l, status); max = SMP_CACHE_BYTES; /* The safest setup for "flush_icache_range()" */ cci.pcci_stride = I_CACHE_STRIDE_SHIFT; @@ -814,7 +840,7 @@ get_max_cacheline_size (void) if (status != 0) { printk(KERN_ERR "%s: ia64_pal_cache_config_info(l=%lu, 1) failed (status=%ld)\n", - __FUNCTION__, l, status); + __func__, l, status); /* The safest setup for "flush_icache_range()" */ cci.pcci_stride = I_CACHE_STRIDE_SHIFT; } @@ -946,9 +972,10 @@ cpu_init (void) #endif /* set ia64_ctx.max_rid to the maximum RID that is supported by all CPUs: */ - if (ia64_pal_vm_summary(NULL, &vmi) == 0) + if (ia64_pal_vm_summary(NULL, &vmi) == 0) { max_ctx = (1U << (vmi.pal_vm_info_2_s.rid_size - 3)) - 1; - else { + setup_ptcg_sem(vmi.pal_vm_info_2_s.max_purges, NPTCG_FROM_PAL); + } else { printk(KERN_WARNING "cpu_init: PAL VM summary failed, assuming 18 RID bits\n"); max_ctx = (1U << 15) - 1; /* use architected minimum */ } diff --git a/arch/ia64/kernel/signal.c b/arch/ia64/kernel/signal.c index 309da3567bc8..5740296c35af 100644 --- a/arch/ia64/kernel/signal.c +++ b/arch/ia64/kernel/signal.c @@ -342,15 +342,33 @@ setup_frame (int sig, struct k_sigaction *ka, siginfo_t *info, sigset_t *set, new_sp = scr->pt.r12; tramp_addr = (unsigned long) __kernel_sigtramp; - if ((ka->sa.sa_flags & SA_ONSTACK) && sas_ss_flags(new_sp) == 0) { - new_sp = current->sas_ss_sp + current->sas_ss_size; - /* - * We need to check for the register stack being on the signal stack - * separately, because it's switched separately (memory stack is switched - * in the kernel, register stack is switched in the signal trampoline). - */ - if (!rbs_on_sig_stack(scr->pt.ar_bspstore)) - new_rbs = (current->sas_ss_sp + sizeof(long) - 1) & ~(sizeof(long) - 1); + if (ka->sa.sa_flags & SA_ONSTACK) { + int onstack = sas_ss_flags(new_sp); + + if (onstack == 0) { + new_sp = current->sas_ss_sp + current->sas_ss_size; + /* + * We need to check for the register stack being on the + * signal stack separately, because it's switched + * separately (memory stack is switched in the kernel, + * register stack is switched in the signal trampoline). + */ + if (!rbs_on_sig_stack(scr->pt.ar_bspstore)) + new_rbs = ALIGN(current->sas_ss_sp, + sizeof(long)); + } else if (onstack == SS_ONSTACK) { + unsigned long check_sp; + + /* + * If we are on the alternate signal stack and would + * overflow it, don't. Return an always-bogus address + * instead so we will die with SIGSEGV. + */ + check_sp = (new_sp - sizeof(*frame)) & -STACK_ALIGN; + if (!likely(on_sig_stack(check_sp))) + return force_sigsegv_info(sig, (void __user *) + check_sp); + } } frame = (void __user *) ((new_sp - sizeof(*frame)) & -STACK_ALIGN); diff --git a/arch/ia64/kernel/smp.c b/arch/ia64/kernel/smp.c index 4e446aa5f4ac..9a9d4c489330 100644 --- a/arch/ia64/kernel/smp.c +++ b/arch/ia64/kernel/smp.c @@ -213,6 +213,19 @@ send_IPI_allbutself (int op) * Called with preemption disabled. */ static inline void +send_IPI_mask(cpumask_t mask, int op) +{ + unsigned int cpu; + + for_each_cpu_mask(cpu, mask) { + send_IPI_single(cpu, op); + } +} + +/* + * Called with preemption disabled. + */ +static inline void send_IPI_all (int op) { int i; @@ -401,6 +414,75 @@ smp_call_function_single (int cpuid, void (*func) (void *info), void *info, int } EXPORT_SYMBOL(smp_call_function_single); +/** + * smp_call_function_mask(): Run a function on a set of other CPUs. + * <mask> The set of cpus to run on. Must not include the current cpu. + * <func> The function to run. This must be fast and non-blocking. + * <info> An arbitrary pointer to pass to the function. + * <wait> If true, wait (atomically) until function + * has completed on other CPUs. + * + * Returns 0 on success, else a negative status code. + * + * If @wait is true, then returns once @func has returned; otherwise + * it returns just before the target cpu calls @func. + * + * You must not call this function with disabled interrupts or from a + * hardware interrupt handler or from a bottom half handler. + */ +int smp_call_function_mask(cpumask_t mask, + void (*func)(void *), void *info, + int wait) +{ + struct call_data_struct data; + cpumask_t allbutself; + int cpus; + + spin_lock(&call_lock); + allbutself = cpu_online_map; + cpu_clear(smp_processor_id(), allbutself); + + cpus_and(mask, mask, allbutself); + cpus = cpus_weight(mask); + if (!cpus) { + spin_unlock(&call_lock); + return 0; + } + + /* Can deadlock when called with interrupts disabled */ + WARN_ON(irqs_disabled()); + + data.func = func; + data.info = info; + atomic_set(&data.started, 0); + data.wait = wait; + if (wait) + atomic_set(&data.finished, 0); + + call_data = &data; + mb(); /* ensure store to call_data precedes setting of IPI_CALL_FUNC*/ + + /* Send a message to other CPUs */ + if (cpus_equal(mask, allbutself)) + send_IPI_allbutself(IPI_CALL_FUNC); + else + send_IPI_mask(mask, IPI_CALL_FUNC); + + /* Wait for response */ + while (atomic_read(&data.started) != cpus) + cpu_relax(); + + if (wait) + while (atomic_read(&data.finished) != cpus) + cpu_relax(); + call_data = NULL; + + spin_unlock(&call_lock); + return 0; + +} +EXPORT_SYMBOL(smp_call_function_mask); + /* * this function sends a 'generic call function' IPI to all other CPUs * in the system. diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c index 32ee5979a042..16483be18c0b 100644 --- a/arch/ia64/kernel/smpboot.c +++ b/arch/ia64/kernel/smpboot.c @@ -400,9 +400,9 @@ smp_callin (void) /* Setup the per cpu irq handling data structures */ __setup_vector_irq(cpuid); cpu_set(cpuid, cpu_online_map); - unlock_ipi_calllock(); per_cpu(cpu_state, cpuid) = CPU_ONLINE; spin_unlock(&vector_lock); + unlock_ipi_calllock(); smp_setup_percpu_timer(); diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c index 17fda5293c67..48e15a51782f 100644 --- a/arch/ia64/kernel/time.c +++ b/arch/ia64/kernel/time.c @@ -59,6 +59,84 @@ static struct clocksource clocksource_itc = { }; static struct clocksource *itc_clocksource; +#ifdef CONFIG_VIRT_CPU_ACCOUNTING + +#include <linux/kernel_stat.h> + +extern cputime_t cycle_to_cputime(u64 cyc); + +/* + * Called from the context switch with interrupts disabled, to charge all + * accumulated times to the current process, and to prepare accounting on + * the next process. + */ +void ia64_account_on_switch(struct task_struct *prev, struct task_struct *next) +{ + struct thread_info *pi = task_thread_info(prev); + struct thread_info *ni = task_thread_info(next); + cputime_t delta_stime, delta_utime; + __u64 now; + + now = ia64_get_itc(); + + delta_stime = cycle_to_cputime(pi->ac_stime + (now - pi->ac_stamp)); + account_system_time(prev, 0, delta_stime); + account_system_time_scaled(prev, delta_stime); + + if (pi->ac_utime) { + delta_utime = cycle_to_cputime(pi->ac_utime); + account_user_time(prev, delta_utime); + account_user_time_scaled(prev, delta_utime); + } + + pi->ac_stamp = ni->ac_stamp = now; + ni->ac_stime = ni->ac_utime = 0; +} + +/* + * Account time for a transition between system, hard irq or soft irq state. + * Note that this function is called with interrupts enabled. + */ +void account_system_vtime(struct task_struct *tsk) +{ + struct thread_info *ti = task_thread_info(tsk); + unsigned long flags; + cputime_t delta_stime; + __u64 now; + + local_irq_save(flags); + + now = ia64_get_itc(); + + delta_stime = cycle_to_cputime(ti->ac_stime + (now - ti->ac_stamp)); + account_system_time(tsk, 0, delta_stime); + account_system_time_scaled(tsk, delta_stime); + ti->ac_stime = 0; + + ti->ac_stamp = now; + + local_irq_restore(flags); +} + +/* + * Called from the timer interrupt handler to charge accumulated user time + * to the current process. Must be called with interrupts disabled. + */ +void account_process_tick(struct task_struct *p, int user_tick) +{ + struct thread_info *ti = task_thread_info(p); + cputime_t delta_utime; + + if (ti->ac_utime) { + delta_utime = cycle_to_cputime(ti->ac_utime); + account_user_time(p, delta_utime); + account_user_time_scaled(p, delta_utime); + ti->ac_utime = 0; + } +} + +#endif /* CONFIG_VIRT_CPU_ACCOUNTING */ + static irqreturn_t timer_interrupt (int irq, void *dev_id) { diff --git a/arch/ia64/kernel/unaligned.c b/arch/ia64/kernel/unaligned.c index 52f70bbc192a..ff0e7c10faa7 100644 --- a/arch/ia64/kernel/unaligned.c +++ b/arch/ia64/kernel/unaligned.c @@ -13,6 +13,7 @@ * 2001/08/13 Correct size of extended floats (float_fsz) from 16 to 10 bytes. * 2001/01/17 Add support emulation of unaligned kernel accesses. */ +#include <linux/jiffies.h> #include <linux/kernel.h> #include <linux/sched.h> #include <linux/tty.h> @@ -28,7 +29,7 @@ extern int die_if_kernel(char *str, struct pt_regs *regs, long err); #undef DEBUG_UNALIGNED_TRAP #ifdef DEBUG_UNALIGNED_TRAP -# define DPRINT(a...) do { printk("%s %u: ", __FUNCTION__, __LINE__); printk (a); } while (0) +# define DPRINT(a...) do { printk("%s %u: ", __func__, __LINE__); printk (a); } while (0) # define DDUMP(str,vp,len) dump(str, vp, len) static void @@ -674,7 +675,7 @@ emulate_load_updates (update_t type, load_store_t ld, struct pt_regs *regs, unsi * just in case. */ if (ld.x6_op == 1 || ld.x6_op == 3) { - printk(KERN_ERR "%s: register update on speculative load, error\n", __FUNCTION__); + printk(KERN_ERR "%s: register update on speculative load, error\n", __func__); if (die_if_kernel("unaligned reference on speculative load with register update\n", regs, 30)) return; @@ -1104,7 +1105,7 @@ emulate_load_floatpair (unsigned long ifa, load_store_t ld, struct pt_regs *regs */ if (ld.x6_op == 1 || ld.x6_op == 3) printk(KERN_ERR "%s: register update on speculative load pair, error\n", - __FUNCTION__); + __func__); setreg(ld.r3, ifa, 0, regs); } @@ -1290,7 +1291,7 @@ within_logging_rate_limit (void) { static unsigned long count, last_time; - if (jiffies - last_time > 5*HZ) + if (time_after(jiffies, last_time + 5 * HZ)) count = 0; if (count < 5) { last_time = jiffies; diff --git a/arch/ia64/kernel/unwind.c b/arch/ia64/kernel/unwind.c index c1bdb5131814..67810b77d998 100644 --- a/arch/ia64/kernel/unwind.c +++ b/arch/ia64/kernel/unwind.c @@ -257,7 +257,7 @@ pt_regs_off (unsigned long reg) off = unw.pt_regs_offsets[reg]; if (off < 0) { - UNW_DPRINT(0, "unwind.%s: bad scratch reg r%lu\n", __FUNCTION__, reg); + UNW_DPRINT(0, "unwind.%s: bad scratch reg r%lu\n", __func__, reg); off = 0; } return (unsigned long) off; @@ -268,13 +268,13 @@ get_scratch_regs (struct unw_frame_info *info) { if (!info->pt) { /* This should not happen with valid unwind info. */ - UNW_DPRINT(0, "unwind.%s: bad unwind info: resetting info->pt\n", __FUNCTION__); + UNW_DPRINT(0, "unwind.%s: bad unwind info: resetting info->pt\n", __func__); if (info->flags & UNW_FLAG_INTERRUPT_FRAME) info->pt = (unsigned long) ((struct pt_regs *) info->psp - 1); else info->pt = info->sp - 16; } - UNW_DPRINT(3, "unwind.%s: sp 0x%lx pt 0x%lx\n", __FUNCTION__, info->sp, info->pt); + UNW_DPRINT(3, "unwind.%s: sp 0x%lx pt 0x%lx\n", __func__, info->sp, info->pt); return (struct pt_regs *) info->pt; } @@ -294,7 +294,7 @@ unw_access_gr (struct unw_frame_info *info, int regnum, unsigned long *val, char return 0; } UNW_DPRINT(0, "unwind.%s: trying to access non-existent r%u\n", - __FUNCTION__, regnum); + __func__, regnum); return -1; } @@ -341,7 +341,7 @@ unw_access_gr (struct unw_frame_info *info, int regnum, unsigned long *val, char { UNW_DPRINT(0, "unwind.%s: %p outside of regstk " "[0x%lx-0x%lx)\n", - __FUNCTION__, (void *) addr, + __func__, (void *) addr, info->regstk.limit, info->regstk.top); return -1; @@ -374,7 +374,7 @@ unw_access_gr (struct unw_frame_info *info, int regnum, unsigned long *val, char || (unsigned long) addr >= info->regstk.top) { UNW_DPRINT(0, "unwind.%s: ignoring attempt to access register outside " - "of rbs\n", __FUNCTION__); + "of rbs\n", __func__); return -1; } if ((unsigned long) nat_addr >= info->regstk.top) @@ -385,7 +385,7 @@ unw_access_gr (struct unw_frame_info *info, int regnum, unsigned long *val, char if (write) { if (read_only(addr)) { UNW_DPRINT(0, "unwind.%s: ignoring attempt to write read-only location\n", - __FUNCTION__); + __func__); } else { *addr = *val; if (*nat) @@ -427,13 +427,13 @@ unw_access_br (struct unw_frame_info *info, int regnum, unsigned long *val, int default: UNW_DPRINT(0, "unwind.%s: trying to access non-existent b%u\n", - __FUNCTION__, regnum); + __func__, regnum); return -1; } if (write) if (read_only(addr)) { UNW_DPRINT(0, "unwind.%s: ignoring attempt to write read-only location\n", - __FUNCTION__); + __func__); } else *addr = *val; else @@ -450,7 +450,7 @@ unw_access_fr (struct unw_frame_info *info, int regnum, struct ia64_fpreg *val, if ((unsigned) (regnum - 2) >= 126) { UNW_DPRINT(0, "unwind.%s: trying to access non-existent f%u\n", - __FUNCTION__, regnum); + __func__, regnum); return -1; } @@ -482,7 +482,7 @@ unw_access_fr (struct unw_frame_info *info, int regnum, struct ia64_fpreg *val, if (write) if (read_only(addr)) { UNW_DPRINT(0, "unwind.%s: ignoring attempt to write read-only location\n", - __FUNCTION__); + __func__); } else *addr = *val; else @@ -572,14 +572,14 @@ unw_access_ar (struct unw_frame_info *info, int regnum, unsigned long *val, int default: UNW_DPRINT(0, "unwind.%s: trying to access non-existent ar%u\n", - __FUNCTION__, regnum); + __func__, regnum); return -1; } if (write) { if (read_only(addr)) { UNW_DPRINT(0, "unwind.%s: ignoring attempt to write read-only location\n", - __FUNCTION__); + __func__); } else *addr = *val; } else @@ -600,7 +600,7 @@ unw_access_pr (struct unw_frame_info *info, unsigned long *val, int write) if (write) { if (read_only(addr)) { UNW_DPRINT(0, "unwind.%s: ignoring attempt to write read-only location\n", - __FUNCTION__); + __func__); } else *addr = *val; } else @@ -699,7 +699,7 @@ decode_abreg (unsigned char abreg, int memory) default: break; } - UNW_DPRINT(0, "unwind.%s: bad abreg=0x%x\n", __FUNCTION__, abreg); + UNW_DPRINT(0, "unwind.%s: bad abreg=0x%x\n", __func__, abreg); return UNW_REG_LC; } @@ -739,7 +739,7 @@ spill_next_when (struct unw_reg_info **regp, struct unw_reg_info *lim, unw_word return; } } - UNW_DPRINT(0, "unwind.%s: excess spill!\n", __FUNCTION__); + UNW_DPRINT(0, "unwind.%s: excess spill!\n", __func__); } static inline void @@ -855,11 +855,11 @@ desc_abi (unsigned char abi, unsigned char context, struct unw_state_record *sr) { if (abi == 3 && context == 'i') { sr->flags |= UNW_FLAG_INTERRUPT_FRAME; - UNW_DPRINT(3, "unwind.%s: interrupt frame\n", __FUNCTION__); + UNW_DPRINT(3, "unwind.%s: interrupt frame\n", __func__); } else UNW_DPRINT(0, "unwind%s: ignoring unwabi(abi=0x%x,context=0x%x)\n", - __FUNCTION__, abi, context); + __func__, abi, context); } static inline void @@ -1347,7 +1347,7 @@ script_emit (struct unw_script *script, struct unw_insn insn) { if (script->count >= UNW_MAX_SCRIPT_LEN) { UNW_DPRINT(0, "unwind.%s: script exceeds maximum size of %u instructions!\n", - __FUNCTION__, UNW_MAX_SCRIPT_LEN); + __func__, UNW_MAX_SCRIPT_LEN); return; } script->insn[script->count++] = insn; @@ -1389,7 +1389,7 @@ emit_nat_info (struct unw_state_record *sr, int i, struct unw_script *script) default: UNW_DPRINT(0, "unwind.%s: don't know how to emit nat info for where = %u\n", - __FUNCTION__, r->where); + __func__, r->where); return; } insn.opc = opc; @@ -1446,7 +1446,7 @@ compile_reg (struct unw_state_record *sr, int i, struct unw_script *script) val = offsetof(struct pt_regs, f6) + 16*(rval - 6); else UNW_DPRINT(0, "unwind.%s: kernel may not touch f%lu\n", - __FUNCTION__, rval); + __func__, rval); } break; @@ -1474,7 +1474,7 @@ compile_reg (struct unw_state_record *sr, int i, struct unw_script *script) default: UNW_DPRINT(0, "unwind%s: register %u has unexpected `where' value of %u\n", - __FUNCTION__, i, r->where); + __func__, i, r->where); break; } insn.opc = opc; @@ -1547,10 +1547,10 @@ build_script (struct unw_frame_info *info) r->when = UNW_WHEN_NEVER; sr.pr_val = info->pr; - UNW_DPRINT(3, "unwind.%s: ip 0x%lx\n", __FUNCTION__, ip); + UNW_DPRINT(3, "unwind.%s: ip 0x%lx\n", __func__, ip); script = script_new(ip); if (!script) { - UNW_DPRINT(0, "unwind.%s: failed to create unwind script\n", __FUNCTION__); + UNW_DPRINT(0, "unwind.%s: failed to create unwind script\n", __func__); STAT(unw.stat.script.build_time += ia64_get_itc() - start); return NULL; } @@ -1569,7 +1569,7 @@ build_script (struct unw_frame_info *info) if (!e) { /* no info, return default unwinder (leaf proc, no mem stack, no saved regs) */ UNW_DPRINT(1, "unwind.%s: no unwind info for ip=0x%lx (prev ip=0x%lx)\n", - __FUNCTION__, ip, unw.cache[info->prev_script].ip); + __func__, ip, unw.cache[info->prev_script].ip); sr.curr.reg[UNW_REG_RP].where = UNW_WHERE_BR; sr.curr.reg[UNW_REG_RP].when = -1; sr.curr.reg[UNW_REG_RP].val = 0; @@ -1618,13 +1618,13 @@ build_script (struct unw_frame_info *info) sr.curr.reg[UNW_REG_RP].when = -1; sr.curr.reg[UNW_REG_RP].val = sr.return_link_reg; UNW_DPRINT(1, "unwind.%s: using default for rp at ip=0x%lx where=%d val=0x%lx\n", - __FUNCTION__, ip, sr.curr.reg[UNW_REG_RP].where, + __func__, ip, sr.curr.reg[UNW_REG_RP].where, sr.curr.reg[UNW_REG_RP].val); } #ifdef UNW_DEBUG UNW_DPRINT(1, "unwind.%s: state record for func 0x%lx, t=%u:\n", - __FUNCTION__, table->segment_base + e->start_offset, sr.when_target); + __func__, table->segment_base + e->start_offset, sr.when_target); for (r = sr.curr.reg; r < sr.curr.reg + UNW_NUM_REGS; ++r) { if (r->where != UNW_WHERE_NONE || r->when != UNW_WHEN_NEVER) { UNW_DPRINT(1, " %s <- ", unw.preg_name[r - sr.curr.reg]); @@ -1746,7 +1746,7 @@ run_script (struct unw_script *script, struct unw_frame_info *state) } else { s[dst] = 0; UNW_DPRINT(0, "unwind.%s: no state->pt, dst=%ld, val=%ld\n", - __FUNCTION__, dst, val); + __func__, dst, val); } break; @@ -1756,7 +1756,7 @@ run_script (struct unw_script *script, struct unw_frame_info *state) else { s[dst] = 0; UNW_DPRINT(0, "unwind.%s: UNW_INSN_MOVE_CONST bad val=%ld\n", - __FUNCTION__, val); + __func__, val); } break; @@ -1791,7 +1791,7 @@ run_script (struct unw_script *script, struct unw_frame_info *state) || s[val] < TASK_SIZE) { UNW_DPRINT(0, "unwind.%s: rejecting bad psp=0x%lx\n", - __FUNCTION__, s[val]); + __func__, s[val]); break; } #endif @@ -1825,7 +1825,7 @@ find_save_locs (struct unw_frame_info *info) if ((info->ip & (local_cpu_data->unimpl_va_mask | 0xf)) || info->ip < TASK_SIZE) { /* don't let obviously bad addresses pollute the cache */ /* FIXME: should really be level 0 but it occurs too often. KAO */ - UNW_DPRINT(1, "unwind.%s: rejecting bad ip=0x%lx\n", __FUNCTION__, info->ip); + UNW_DPRINT(1, "unwind.%s: rejecting bad ip=0x%lx\n", __func__, info->ip); info->rp_loc = NULL; return -1; } @@ -1838,7 +1838,7 @@ find_save_locs (struct unw_frame_info *info) spin_unlock_irqrestore(&unw.lock, flags); UNW_DPRINT(0, "unwind.%s: failed to locate/build unwind script for ip %lx\n", - __FUNCTION__, info->ip); + __func__, info->ip); return -1; } have_write_lock = 1; @@ -1882,21 +1882,21 @@ unw_unwind (struct unw_frame_info *info) if (!unw_valid(info, info->rp_loc)) { /* FIXME: should really be level 0 but it occurs too often. KAO */ UNW_DPRINT(1, "unwind.%s: failed to locate return link (ip=0x%lx)!\n", - __FUNCTION__, info->ip); + __func__, info->ip); STAT(unw.stat.api.unwind_time += ia64_get_itc() - start; local_irq_restore(flags)); return -1; } /* restore the ip */ ip = info->ip = *info->rp_loc; if (ip < GATE_ADDR) { - UNW_DPRINT(2, "unwind.%s: reached user-space (ip=0x%lx)\n", __FUNCTION__, ip); + UNW_DPRINT(2, "unwind.%s: reached user-space (ip=0x%lx)\n", __func__, ip); STAT(unw.stat.api.unwind_time += ia64_get_itc() - start; local_irq_restore(flags)); return -1; } /* validate the previous stack frame pointer */ if (!unw_valid(info, info->pfs_loc)) { - UNW_DPRINT(0, "unwind.%s: failed to locate ar.pfs!\n", __FUNCTION__); + UNW_DPRINT(0, "unwind.%s: failed to locate ar.pfs!\n", __func__); STAT(unw.stat.api.unwind_time += ia64_get_itc() - start; local_irq_restore(flags)); return -1; } @@ -1912,13 +1912,13 @@ unw_unwind (struct unw_frame_info *info) num_regs = *info->cfm_loc & 0x7f; /* size of frame */ info->pfs_loc = (unsigned long *) (info->pt + offsetof(struct pt_regs, ar_pfs)); - UNW_DPRINT(3, "unwind.%s: interrupt_frame pt 0x%lx\n", __FUNCTION__, info->pt); + UNW_DPRINT(3, "unwind.%s: interrupt_frame pt 0x%lx\n", __func__, info->pt); } else num_regs = (*info->cfm_loc >> 7) & 0x7f; /* size of locals */ info->bsp = (unsigned long) ia64_rse_skip_regs((unsigned long *) info->bsp, -num_regs); if (info->bsp < info->regstk.limit || info->bsp > info->regstk.top) { UNW_DPRINT(0, "unwind.%s: bsp (0x%lx) out of range [0x%lx-0x%lx]\n", - __FUNCTION__, info->bsp, info->regstk.limit, info->regstk.top); + __func__, info->bsp, info->regstk.limit, info->regstk.top); STAT(unw.stat.api.unwind_time += ia64_get_itc() - start; local_irq_restore(flags)); return -1; } @@ -1927,14 +1927,14 @@ unw_unwind (struct unw_frame_info *info) info->sp = info->psp; if (info->sp < info->memstk.top || info->sp > info->memstk.limit) { UNW_DPRINT(0, "unwind.%s: sp (0x%lx) out of range [0x%lx-0x%lx]\n", - __FUNCTION__, info->sp, info->memstk.top, info->memstk.limit); + __func__, info->sp, info->memstk.top, info->memstk.limit); STAT(unw.stat.api.unwind_time += ia64_get_itc() - start; local_irq_restore(flags)); return -1; } if (info->ip == prev_ip && info->sp == prev_sp && info->bsp == prev_bsp) { UNW_DPRINT(0, "unwind.%s: ip, sp, bsp unchanged; stopping here (ip=0x%lx)\n", - __FUNCTION__, ip); + __func__, ip); STAT(unw.stat.api.unwind_time += ia64_get_itc() - start; local_irq_restore(flags)); return -1; } @@ -1961,7 +1961,7 @@ unw_unwind_to_user (struct unw_frame_info *info) if ((long)((unsigned long)info->task + IA64_STK_OFFSET - sp) < IA64_PT_REGS_SIZE) { UNW_DPRINT(0, "unwind.%s: ran off the top of the kernel stack\n", - __FUNCTION__); + __func__); break; } if (unw_is_intr_frame(info) && @@ -1971,13 +1971,13 @@ unw_unwind_to_user (struct unw_frame_info *info) unw_get_rp(info, &ip); UNW_DPRINT(0, "unwind.%s: failed to read " "predicate register (ip=0x%lx)\n", - __FUNCTION__, ip); + __func__, ip); return -1; } } while (unw_unwind(info) >= 0); unw_get_ip(info, &ip); UNW_DPRINT(0, "unwind.%s: failed to unwind to user-level (ip=0x%lx)\n", - __FUNCTION__, ip); + __func__, ip); return -1; } EXPORT_SYMBOL(unw_unwind_to_user); @@ -2028,7 +2028,7 @@ init_frame_info (struct unw_frame_info *info, struct task_struct *t, " pr 0x%lx\n" " sw 0x%lx\n" " sp 0x%lx\n", - __FUNCTION__, (unsigned long) t, rbslimit, rbstop, stktop, stklimit, + __func__, (unsigned long) t, rbslimit, rbstop, stktop, stklimit, info->pr, (unsigned long) info->sw, info->sp); STAT(unw.stat.api.init_time += ia64_get_itc() - start; local_irq_restore(flags)); } @@ -2047,7 +2047,7 @@ unw_init_frame_info (struct unw_frame_info *info, struct task_struct *t, struct " bsp 0x%lx\n" " sol 0x%lx\n" " ip 0x%lx\n", - __FUNCTION__, info->bsp, sol, info->ip); + __func__, info->bsp, sol, info->ip); find_save_locs(info); } @@ -2058,7 +2058,7 @@ unw_init_from_blocked_task (struct unw_frame_info *info, struct task_struct *t) { struct switch_stack *sw = (struct switch_stack *) (t->thread.ksp + 16); - UNW_DPRINT(1, "unwind.%s\n", __FUNCTION__); + UNW_DPRINT(1, "unwind.%s\n", __func__); unw_init_frame_info(info, t, sw); } EXPORT_SYMBOL(unw_init_from_blocked_task); @@ -2088,7 +2088,7 @@ unw_add_unwind_table (const char *name, unsigned long segment_base, unsigned lon if (end - start <= 0) { UNW_DPRINT(0, "unwind.%s: ignoring attempt to insert empty unwind table\n", - __FUNCTION__); + __func__); return NULL; } @@ -2119,14 +2119,14 @@ unw_remove_unwind_table (void *handle) if (!handle) { UNW_DPRINT(0, "unwind.%s: ignoring attempt to remove non-existent unwind table\n", - __FUNCTION__); + __func__); return; } table = handle; if (table == &unw.kernel_table) { UNW_DPRINT(0, "unwind.%s: sorry, freeing the kernel's unwind table is a " - "no-can-do!\n", __FUNCTION__); + "no-can-do!\n", __func__); return; } @@ -2139,7 +2139,7 @@ unw_remove_unwind_table (void *handle) break; if (!prev) { UNW_DPRINT(0, "unwind.%s: failed to find unwind table %p\n", - __FUNCTION__, (void *) table); + __func__, (void *) table); spin_unlock_irqrestore(&unw.lock, flags); return; } @@ -2185,7 +2185,7 @@ create_gate_table (void) } if (!punw) { - printk("%s: failed to find gate DSO's unwind table!\n", __FUNCTION__); + printk("%s: failed to find gate DSO's unwind table!\n", __func__); return 0; } @@ -2202,7 +2202,7 @@ create_gate_table (void) unw.gate_table = kmalloc(size, GFP_KERNEL); if (!unw.gate_table) { unw.gate_table_size = 0; - printk(KERN_ERR "%s: unable to create unwind data for gate page!\n", __FUNCTION__); + printk(KERN_ERR "%s: unable to create unwind data for gate page!\n", __func__); return 0; } unw.gate_table_size = size; |