Use helpers to obtain task pid in printks (arch code)
Alexey Dobriyan [Fri, 19 Oct 2007 06:40:41 +0000 (23:40 -0700)]
One of the easiest things to isolate is the pid printed in kernel log.
There was a patch, that made this for arch-independent code, this one makes
so for arch/xxx files.

It took some time to cross-compile it, but hopefully these are all the
printks in arch code.

Signed-off-by: Alexey Dobriyan <adobriyan@openvz.org>
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

37 files changed:
arch/alpha/kernel/semaphore.c
arch/alpha/kernel/traps.c
arch/alpha/mm/fault.c
arch/arm/kernel/process.c
arch/arm/kernel/ptrace.c
arch/arm/kernel/traps.c
arch/arm/mm/alignment.c
arch/ia64/ia32/sys_ia32.c
arch/ia64/kernel/perfmon.c
arch/ia64/kernel/perfmon_default_smpl.c
arch/ia64/kernel/process.c
arch/ia64/kernel/traps.c
arch/ia64/kernel/unaligned.c
arch/m32r/kernel/traps.c
arch/m68k/kernel/traps.c
arch/mips/kernel/traps.c
arch/parisc/kernel/traps.c
arch/parisc/kernel/unaligned.c
arch/parisc/mm/fault.c
arch/powerpc/kernel/process.c
arch/powerpc/kernel/traps.c
arch/s390/kernel/process.c
arch/sh/kernel/process.c
arch/sh/kernel/signal.c
arch/sh/kernel/traps.c
arch/sh64/kernel/traps.c
arch/sh64/mm/fault.c
arch/sparc/kernel/ptrace.c
arch/sparc/kernel/sys_sparc.c
arch/sparc/kernel/traps.c
arch/sparc64/kernel/traps.c
arch/um/sys-x86_64/sysrq.c
arch/x86/kernel/process_32.c
arch/x86/kernel/signal_32.c
arch/x86/kernel/traps_32.c
arch/x86/mm/fault_32.c
arch/xtensa/kernel/traps.c

index 8c8aaa2..8d2982a 100644 (file)
@@ -69,7 +69,7 @@ __down_failed(struct semaphore *sem)
 
 #ifdef CONFIG_DEBUG_SEMAPHORE
        printk("%s(%d): down failed(%p)\n",
-              tsk->comm, tsk->pid, sem);
+              tsk->comm, task_pid_nr(tsk), sem);
 #endif
 
        tsk->state = TASK_UNINTERRUPTIBLE;
@@ -98,7 +98,7 @@ __down_failed(struct semaphore *sem)
 
 #ifdef CONFIG_DEBUG_SEMAPHORE
        printk("%s(%d): down acquired(%p)\n",
-              tsk->comm, tsk->pid, sem);
+              tsk->comm, task_pid_nr(tsk), sem);
 #endif
 }
 
@@ -111,7 +111,7 @@ __down_failed_interruptible(struct semaphore *sem)
 
 #ifdef CONFIG_DEBUG_SEMAPHORE
        printk("%s(%d): down failed(%p)\n",
-              tsk->comm, tsk->pid, sem);
+              tsk->comm, task_pid_nr(tsk), sem);
 #endif
 
        tsk->state = TASK_INTERRUPTIBLE;
@@ -139,7 +139,7 @@ __down_failed_interruptible(struct semaphore *sem)
 
 #ifdef CONFIG_DEBUG_SEMAPHORE
        printk("%s(%d): down %s(%p)\n",
-              current->comm, current->pid,
+              current->comm, task_pid_nr(current),
               (ret < 0 ? "interrupted" : "acquired"), sem);
 #endif
        return ret;
@@ -168,7 +168,7 @@ down(struct semaphore *sem)
 #endif
 #ifdef CONFIG_DEBUG_SEMAPHORE
        printk("%s(%d): down(%p) <count=%d> from %p\n",
-              current->comm, current->pid, sem,
+              current->comm, task_pid_nr(current), sem,
               atomic_read(&sem->count), __builtin_return_address(0));
 #endif
        __down(sem);
@@ -182,7 +182,7 @@ down_interruptible(struct semaphore *sem)
 #endif
 #ifdef CONFIG_DEBUG_SEMAPHORE
        printk("%s(%d): down(%p) <count=%d> from %p\n",
-              current->comm, current->pid, sem,
+              current->comm, task_pid_nr(current), sem,
               atomic_read(&sem->count), __builtin_return_address(0));
 #endif
        return __down_interruptible(sem);
@@ -201,7 +201,7 @@ down_trylock(struct semaphore *sem)
 
 #ifdef CONFIG_DEBUG_SEMAPHORE
        printk("%s(%d): down_trylock %s from %p\n",
-              current->comm, current->pid,
+              current->comm, task_pid_nr(current),
               ret ? "failed" : "acquired",
               __builtin_return_address(0));
 #endif
@@ -217,7 +217,7 @@ up(struct semaphore *sem)
 #endif
 #ifdef CONFIG_DEBUG_SEMAPHORE
        printk("%s(%d): up(%p) <count=%d> from %p\n",
-              current->comm, current->pid, sem,
+              current->comm, task_pid_nr(current), sem,
               atomic_read(&sem->count), __builtin_return_address(0));
 #endif
        __up(sem);
index ec0f05e..2dc7f9f 100644 (file)
@@ -182,7 +182,7 @@ die_if_kernel(char * str, struct pt_regs *regs, long err, unsigned long *r9_15)
 #ifdef CONFIG_SMP
        printk("CPU %d ", hard_smp_processor_id());
 #endif
-       printk("%s(%d): %s %ld\n", current->comm, current->pid, str, err);
+       printk("%s(%d): %s %ld\n", current->comm, task_pid_nr(current), str, err);
        dik_show_regs(regs, r9_15);
        add_taint(TAINT_DIE);
        dik_show_trace((unsigned long *)(regs+1));
@@ -646,7 +646,7 @@ got_exception:
        lock_kernel();
 
        printk("%s(%d): unhandled unaligned exception\n",
-              current->comm, current->pid);
+              current->comm, task_pid_nr(current));
 
        printk("pc = [<%016lx>]  ra = [<%016lx>]  ps = %04lx\n",
               pc, una_reg(26), regs->ps);
@@ -786,7 +786,7 @@ do_entUnaUser(void __user * va, unsigned long opcode,
                }
                if (++cnt < 5) {
                        printk("%s(%d): unaligned trap at %016lx: %p %lx %ld\n",
-                              current->comm, current->pid,
+                              current->comm, task_pid_nr(current),
                               regs->pc - 4, va, opcode, reg);
                }
                last_time = jiffies;
index e0593e6..4829f96 100644 (file)
@@ -194,7 +194,7 @@ do_page_fault(unsigned long address, unsigned long mmcsr,
                goto survive;
        }
        printk(KERN_ALERT "VM: killing process %s(%d)\n",
-              current->comm, current->pid);
+              current->comm, task_pid_nr(current));
        if (!user_mode(regs))
                goto no_context;
        do_group_exit(SIGKILL);
index 93b7f8e..4f1a031 100644 (file)
@@ -265,7 +265,7 @@ void __show_regs(struct pt_regs *regs)
 void show_regs(struct pt_regs * regs)
 {
        printk("\n");
-       printk("Pid: %d, comm: %20s\n", current->pid, current->comm);
+       printk("Pid: %d, comm: %20s\n", task_pid_nr(current), current->comm);
        __show_regs(regs);
        __backtrace();
 }
index 5feee72..4b05dc5 100644 (file)
@@ -382,16 +382,16 @@ static void clear_breakpoint(struct task_struct *task, struct debug_entry *bp)
 
                if (ret != 2 || old_insn.thumb != BREAKINST_THUMB)
                        printk(KERN_ERR "%s:%d: corrupted Thumb breakpoint at "
-                               "0x%08lx (0x%04x)\n", task->comm, task->pid,
-                               addr, old_insn.thumb);
+                               "0x%08lx (0x%04x)\n", task->comm,
+                               task_pid_nr(task), addr, old_insn.thumb);
        } else {
                ret = swap_insn(task, addr & ~3, &old_insn.arm,
                                &bp->insn.arm, 4);
 
                if (ret != 4 || old_insn.arm != BREAKINST_ARM)
                        printk(KERN_ERR "%s:%d: corrupted ARM breakpoint at "
-                               "0x%08lx (0x%08x)\n", task->comm, task->pid,
-                               addr, old_insn.arm);
+                               "0x%08lx (0x%08x)\n", task->comm,
+                               task_pid_nr(task), addr, old_insn.arm);
        }
 }
 
index 8ad4761..4764bd9 100644 (file)
@@ -223,7 +223,7 @@ static void __die(const char *str, int err, struct thread_info *thread, struct p
        print_modules();
        __show_regs(regs);
        printk("Process %s (pid: %d, stack limit = 0x%p)\n",
-               tsk->comm, tsk->pid, thread + 1);
+               tsk->comm, task_pid_nr(tsk), thread + 1);
 
        if (!user_mode(regs) || in_interrupt()) {
                dump_mem("Stack: ", regs->ARM_sp,
@@ -337,7 +337,7 @@ asmlinkage void __exception do_undefinstr(struct pt_regs *regs)
 #ifdef CONFIG_DEBUG_USER
        if (user_debug & UDBG_UNDEFINED) {
                printk(KERN_INFO "%s (%d): undefined instruction: pc=%p\n",
-                       current->comm, current->pid, pc);
+                       current->comm, task_pid_nr(current), pc);
                dump_instr(regs);
        }
 #endif
@@ -388,7 +388,7 @@ static int bad_syscall(int n, struct pt_regs *regs)
 #ifdef CONFIG_DEBUG_USER
        if (user_debug & UDBG_SYSCALL) {
                printk(KERN_ERR "[%d] %s: obsolete system call %08x.\n",
-                       current->pid, current->comm, n);
+                       task_pid_nr(current), current->comm, n);
                dump_instr(regs);
        }
 #endif
@@ -565,7 +565,7 @@ asmlinkage int arm_syscall(int no, struct pt_regs *regs)
         */
        if (user_debug & UDBG_SYSCALL) {
                printk("[%d] %s: arm syscall %d\n",
-                      current->pid, current->comm, no);
+                      task_pid_nr(current), current->comm, no);
                dump_instr(regs);
                if (user_mode(regs)) {
                        __show_regs(regs);
@@ -642,7 +642,7 @@ baddataabort(int code, unsigned long instr, struct pt_regs *regs)
 #ifdef CONFIG_DEBUG_USER
        if (user_debug & UDBG_BADABORT) {
                printk(KERN_ERR "[%d] %s: bad data abort: code %d instr 0x%08lx\n",
-                       current->pid, current->comm, code, instr);
+                       task_pid_nr(current), current->comm, code, instr);
                dump_instr(regs);
                show_pte(current->mm, addr);
        }
index 074b7cb..e162cca 100644 (file)
@@ -757,7 +757,7 @@ do_alignment(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
        if (ai_usermode & 1)
                printk("Alignment trap: %s (%d) PC=0x%08lx Instr=0x%0*lx "
                       "Address=0x%08lx FSR 0x%03x\n", current->comm,
-                       current->pid, instrptr,
+                       task_pid_nr(current), instrptr,
                        thumb_mode(regs) ? 4 : 8,
                        thumb_mode(regs) ? tinstr : instr,
                        addr, fsr);
index a3405b3..d025a22 100644 (file)
@@ -773,7 +773,7 @@ emulate_mmap (struct file *file, unsigned long start, unsigned long len, int pro
                        if (flags & MAP_SHARED)
                                printk(KERN_INFO
                                       "%s(%d): emulate_mmap() can't share head (addr=0x%lx)\n",
-                                      current->comm, current->pid, start);
+                                      current->comm, task_pid_nr(current), start);
                        ret = mmap_subpage(file, start, min(PAGE_ALIGN(start), end), prot, flags,
                                           off);
                        if (IS_ERR((void *) ret))
@@ -786,7 +786,7 @@ emulate_mmap (struct file *file, unsigned long start, unsigned long len, int pro
                        if (flags & MAP_SHARED)
                                printk(KERN_INFO
                                       "%s(%d): emulate_mmap() can't share tail (end=0x%lx)\n",
-                                      current->comm, current->pid, end);
+                                      current->comm, task_pid_nr(current), end);
                        ret = mmap_subpage(file, max(start, PAGE_START(end)), end, prot, flags,
                                           (off + len) - offset_in_page(end));
                        if (IS_ERR((void *) ret))
@@ -816,7 +816,7 @@ emulate_mmap (struct file *file, unsigned long start, unsigned long len, int pro
 
        if ((flags & MAP_SHARED) && !is_congruent)
                printk(KERN_INFO "%s(%d): emulate_mmap() can't share contents of incongruent mmap "
-                      "(addr=0x%lx,off=0x%llx)\n", current->comm, current->pid, start, off);
+                      "(addr=0x%lx,off=0x%llx)\n", current->comm, task_pid_nr(current), start, off);
 
        DBG("mmap_body: mapping [0x%lx-0x%lx) %s with poff 0x%llx\n", pstart, pend,
            is_congruent ? "congruent" : "not congruent", poff);
index f55fa07..59169bf 100644 (file)
  */
 #define PROTECT_CTX(c, f) \
        do {  \
-               DPRINT(("spinlock_irq_save ctx %p by [%d]\n", c, current->pid)); \
+               DPRINT(("spinlock_irq_save ctx %p by [%d]\n", c, task_pid_nr(current))); \
                spin_lock_irqsave(&(c)->ctx_lock, f); \
-               DPRINT(("spinlocked ctx %p  by [%d]\n", c, current->pid)); \
+               DPRINT(("spinlocked ctx %p  by [%d]\n", c, task_pid_nr(current))); \
        } while(0)
 
 #define UNPROTECT_CTX(c, f) \
        do { \
-               DPRINT(("spinlock_irq_restore ctx %p by [%d]\n", c, current->pid)); \
+               DPRINT(("spinlock_irq_restore ctx %p by [%d]\n", c, task_pid_nr(current))); \
                spin_unlock_irqrestore(&(c)->ctx_lock, f); \
        } while(0)
 
 #ifdef PFM_DEBUGGING
 #define DPRINT(a) \
        do { \
-               if (unlikely(pfm_sysctl.debug >0)) { printk("%s.%d: CPU%d [%d] ", __FUNCTION__, __LINE__, smp_processor_id(), current->pid); printk a; } \
+               if (unlikely(pfm_sysctl.debug >0)) { printk("%s.%d: CPU%d [%d] ", __FUNCTION__, __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(), current->pid); printk a; } \
+               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; } \
        } while (0)
 #endif
 
@@ -913,7 +913,7 @@ pfm_mask_monitoring(struct task_struct *task)
        unsigned long mask, val, ovfl_mask;
        int i;
 
-       DPRINT_ovfl(("masking monitoring for [%d]\n", task->pid));
+       DPRINT_ovfl(("masking monitoring for [%d]\n", task_pid_nr(task)));
 
        ovfl_mask = pmu_conf->ovfl_val;
        /*
@@ -992,12 +992,12 @@ pfm_restore_monitoring(struct task_struct *task)
        ovfl_mask = pmu_conf->ovfl_val;
 
        if (task != current) {
-               printk(KERN_ERR "perfmon.%d: invalid task[%d] current[%d]\n", __LINE__, task->pid, current->pid);
+               printk(KERN_ERR "perfmon.%d: invalid task[%d] current[%d]\n", __LINE__, task_pid_nr(task), task_pid_nr(current));
                return;
        }
        if (ctx->ctx_state != PFM_CTX_MASKED) {
                printk(KERN_ERR "perfmon.%d: task[%d] current[%d] invalid state=%d\n", __LINE__,
-                       task->pid, current->pid, ctx->ctx_state);
+                       task_pid_nr(task), task_pid_nr(current), ctx->ctx_state);
                return;
        }
        psr = pfm_get_psr();
@@ -1051,7 +1051,8 @@ pfm_restore_monitoring(struct task_struct *task)
                if ((mask & 0x1) == 0UL) continue;
                ctx->th_pmcs[i] = ctx->ctx_pmcs[i];
                ia64_set_pmc(i, ctx->th_pmcs[i]);
-               DPRINT(("[%d] pmc[%d]=0x%lx\n", task->pid, i, ctx->th_pmcs[i]));
+               DPRINT(("[%d] pmc[%d]=0x%lx\n",
+                                       task_pid_nr(task), i, ctx->th_pmcs[i]));
        }
        ia64_srlz_d();
 
@@ -1370,7 +1371,7 @@ pfm_reserve_session(struct task_struct *task, int is_syswide, unsigned int cpu)
 
 error_conflict:
        DPRINT(("system wide not possible, conflicting session [%d] on CPU%d\n",
-               pfm_sessions.pfs_sys_session[cpu]->pid,
+               task_pid_nr(pfm_sessions.pfs_sys_session[cpu]),
                cpu));
 abort:
        UNLOCK_PFS(flags);
@@ -1442,7 +1443,7 @@ pfm_remove_smpl_mapping(struct task_struct *task, void *vaddr, unsigned long siz
 
        /* sanity checks */
        if (task->mm == NULL || size == 0UL || vaddr == NULL) {
-               printk(KERN_ERR "perfmon: pfm_remove_smpl_mapping [%d] invalid context mm=%p\n", task->pid, task->mm);
+               printk(KERN_ERR "perfmon: pfm_remove_smpl_mapping [%d] invalid context mm=%p\n", task_pid_nr(task), task->mm);
                return -EINVAL;
        }
 
@@ -1459,7 +1460,7 @@ pfm_remove_smpl_mapping(struct task_struct *task, void *vaddr, unsigned long siz
 
        up_write(&task->mm->mmap_sem);
        if (r !=0) {
-               printk(KERN_ERR "perfmon: [%d] unable to unmap sampling buffer @%p size=%lu\n", task->pid, vaddr, size);
+               printk(KERN_ERR "perfmon: [%d] unable to unmap sampling buffer @%p size=%lu\n", task_pid_nr(task), vaddr, size);
        }
 
        DPRINT(("do_unmap(%p, %lu)=%d\n", vaddr, size, r));
@@ -1501,7 +1502,7 @@ pfm_free_smpl_buffer(pfm_context_t *ctx)
        return 0;
 
 invalid_free:
-       printk(KERN_ERR "perfmon: pfm_free_smpl_buffer [%d] no buffer\n", current->pid);
+       printk(KERN_ERR "perfmon: pfm_free_smpl_buffer [%d] no buffer\n", task_pid_nr(current));
        return -EINVAL;
 }
 #endif
@@ -1547,13 +1548,13 @@ pfm_read(struct file *filp, char __user *buf, size_t size, loff_t *ppos)
        unsigned long flags;
        DECLARE_WAITQUEUE(wait, current);
        if (PFM_IS_FILE(filp) == 0) {
-               printk(KERN_ERR "perfmon: pfm_poll: bad magic [%d]\n", current->pid);
+               printk(KERN_ERR "perfmon: pfm_poll: bad magic [%d]\n", task_pid_nr(current));
                return -EINVAL;
        }
 
        ctx = (pfm_context_t *)filp->private_data;
        if (ctx == NULL) {
-               printk(KERN_ERR "perfmon: pfm_read: NULL ctx [%d]\n", current->pid);
+               printk(KERN_ERR "perfmon: pfm_read: NULL ctx [%d]\n", task_pid_nr(current));
                return -EINVAL;
        }
 
@@ -1607,7 +1608,7 @@ pfm_read(struct file *filp, char __user *buf, size_t size, loff_t *ppos)
 
                PROTECT_CTX(ctx, flags);
        }
-       DPRINT(("[%d] back to running ret=%ld\n", current->pid, ret));
+       DPRINT(("[%d] back to running ret=%ld\n", task_pid_nr(current), ret));
        set_current_state(TASK_RUNNING);
        remove_wait_queue(&ctx->ctx_msgq_wait, &wait);
 
@@ -1616,7 +1617,7 @@ pfm_read(struct file *filp, char __user *buf, size_t size, loff_t *ppos)
        ret = -EINVAL;
        msg = pfm_get_next_msg(ctx);
        if (msg == NULL) {
-               printk(KERN_ERR "perfmon: pfm_read no msg for ctx=%p [%d]\n", ctx, current->pid);
+               printk(KERN_ERR "perfmon: pfm_read no msg for ctx=%p [%d]\n", ctx, task_pid_nr(current));
                goto abort_locked;
        }
 
@@ -1647,13 +1648,13 @@ pfm_poll(struct file *filp, poll_table * wait)
        unsigned int mask = 0;
 
        if (PFM_IS_FILE(filp) == 0) {
-               printk(KERN_ERR "perfmon: pfm_poll: bad magic [%d]\n", current->pid);
+               printk(KERN_ERR "perfmon: pfm_poll: bad magic [%d]\n", task_pid_nr(current));
                return 0;
        }
 
        ctx = (pfm_context_t *)filp->private_data;
        if (ctx == NULL) {
-               printk(KERN_ERR "perfmon: pfm_poll: NULL ctx [%d]\n", current->pid);
+               printk(KERN_ERR "perfmon: pfm_poll: NULL ctx [%d]\n", task_pid_nr(current));
                return 0;
        }
 
@@ -1692,7 +1693,7 @@ pfm_do_fasync(int fd, struct file *filp, pfm_context_t *ctx, int on)
        ret = fasync_helper (fd, filp, on, &ctx->ctx_async_queue);
 
        DPRINT(("pfm_fasync called by [%d] on ctx_fd=%d on=%d async_queue=%p ret=%d\n",
-               current->pid,
+               task_pid_nr(current),
                fd,
                on,
                ctx->ctx_async_queue, ret));
@@ -1707,13 +1708,13 @@ pfm_fasync(int fd, struct file *filp, int on)
        int ret;
 
        if (PFM_IS_FILE(filp) == 0) {
-               printk(KERN_ERR "perfmon: pfm_fasync bad magic [%d]\n", current->pid);
+               printk(KERN_ERR "perfmon: pfm_fasync bad magic [%d]\n", task_pid_nr(current));
                return -EBADF;
        }
 
        ctx = (pfm_context_t *)filp->private_data;
        if (ctx == NULL) {
-               printk(KERN_ERR "perfmon: pfm_fasync NULL ctx [%d]\n", current->pid);
+               printk(KERN_ERR "perfmon: pfm_fasync NULL ctx [%d]\n", task_pid_nr(current));
                return -EBADF;
        }
        /*
@@ -1759,7 +1760,7 @@ pfm_syswide_force_stop(void *info)
        if (owner != ctx->ctx_task) {
                printk(KERN_ERR "perfmon: pfm_syswide_force_stop CPU%d unexpected owner [%d] instead of [%d]\n",
                        smp_processor_id(),
-                       owner->pid, ctx->ctx_task->pid);
+                       task_pid_nr(owner), task_pid_nr(ctx->ctx_task));
                return;
        }
        if (GET_PMU_CTX() != ctx) {
@@ -1769,7 +1770,7 @@ pfm_syswide_force_stop(void *info)
                return;
        }
 
-       DPRINT(("on CPU%d forcing system wide stop for [%d]\n", smp_processor_id(), ctx->ctx_task->pid));       
+       DPRINT(("on CPU%d forcing system wide stop for [%d]\n", smp_processor_id(), task_pid_nr(ctx->ctx_task)));
        /*
         * the context is already protected in pfm_close(), we simply
         * need to mask interrupts to avoid a PMU interrupt race on
@@ -1821,7 +1822,7 @@ pfm_flush(struct file *filp, fl_owner_t id)
 
        ctx = (pfm_context_t *)filp->private_data;
        if (ctx == NULL) {
-               printk(KERN_ERR "perfmon: pfm_flush: NULL ctx [%d]\n", current->pid);
+               printk(KERN_ERR "perfmon: pfm_flush: NULL ctx [%d]\n", task_pid_nr(current));
                return -EBADF;
        }
 
@@ -1969,7 +1970,7 @@ pfm_close(struct inode *inode, struct file *filp)
        
        ctx = (pfm_context_t *)filp->private_data;
        if (ctx == NULL) {
-               printk(KERN_ERR "perfmon: pfm_close: NULL ctx [%d]\n", current->pid);
+               printk(KERN_ERR "perfmon: pfm_close: NULL ctx [%d]\n", task_pid_nr(current));
                return -EBADF;
        }
 
@@ -2066,7 +2067,7 @@ pfm_close(struct inode *inode, struct file *filp)
                 */
                ctx->ctx_state = PFM_CTX_ZOMBIE;
 
-               DPRINT(("zombie ctx for [%d]\n", task->pid));
+               DPRINT(("zombie ctx for [%d]\n", task_pid_nr(task)));
                /*
                 * cannot free the context on the spot. deferred until
                 * the task notices the ZOMBIE state
@@ -2472,7 +2473,7 @@ pfm_setup_buffer_fmt(struct task_struct *task, struct file *filp, pfm_context_t
        /* invoke and lock buffer format, if found */
        fmt = pfm_find_buffer_fmt(arg->ctx_smpl_buf_id);
        if (fmt == NULL) {
-               DPRINT(("[%d] cannot find buffer format\n", task->pid));
+               DPRINT(("[%d] cannot find buffer format\n", task_pid_nr(task)));
                return -EINVAL;
        }
 
@@ -2483,7 +2484,7 @@ pfm_setup_buffer_fmt(struct task_struct *task, struct file *filp, pfm_context_t
 
        ret = pfm_buf_fmt_validate(fmt, task, ctx_flags, cpu, fmt_arg);
 
-       DPRINT(("[%d] after validate(0x%x,%d,%p)=%d\n", task->pid, ctx_flags, cpu, fmt_arg, ret));
+       DPRINT(("[%d] after validate(0x%x,%d,%p)=%d\n", task_pid_nr(task), ctx_flags, cpu, fmt_arg, ret));
 
        if (ret) goto error;
 
@@ -2605,23 +2606,23 @@ pfm_task_incompatible(pfm_context_t *ctx, struct task_struct *task)
         * no kernel task or task not owner by caller
         */
        if (task->mm == NULL) {
-               DPRINT(("task [%d] has not memory context (kernel thread)\n", task->pid));
+               DPRINT(("task [%d] has not memory context (kernel thread)\n", task_pid_nr(task)));
                return -EPERM;
        }
        if (pfm_bad_permissions(task)) {
-               DPRINT(("no permission to attach to  [%d]\n", task->pid));
+               DPRINT(("no permission to attach to  [%d]\n", task_pid_nr(task)));
                return -EPERM;
        }
        /*
         * cannot block in self-monitoring mode
         */
        if (CTX_OVFL_NOBLOCK(ctx) == 0 && task == current) {
-               DPRINT(("cannot load a blocking context on self for [%d]\n", task->pid));
+               DPRINT(("cannot load a blocking context on self for [%d]\n", task_pid_nr(task)));
                return -EINVAL;
        }
 
        if (task->exit_state == EXIT_ZOMBIE) {
-               DPRINT(("cannot attach to  zombie task [%d]\n", task->pid));
+               DPRINT(("cannot attach to  zombie task [%d]\n", task_pid_nr(task)));
                return -EBUSY;
        }
 
@@ -2631,7 +2632,7 @@ pfm_task_incompatible(pfm_context_t *ctx, struct task_struct *task)
        if (task == current) return 0;
 
        if ((task->state != TASK_STOPPED) && (task->state != TASK_TRACED)) {
-               DPRINT(("cannot attach to non-stopped task [%d] state=%ld\n", task->pid, task->state));
+               DPRINT(("cannot attach to non-stopped task [%d] state=%ld\n", task_pid_nr(task), task->state));
                return -EBUSY;
        }
        /*
@@ -3512,7 +3513,7 @@ pfm_use_debug_registers(struct task_struct *task)
 
        if (pmu_conf->use_rr_dbregs == 0) return 0;
 
-       DPRINT(("called for [%d]\n", task->pid));
+       DPRINT(("called for [%d]\n", task_pid_nr(task)));
 
        /*
         * do it only once
@@ -3543,7 +3544,7 @@ pfm_use_debug_registers(struct task_struct *task)
        DPRINT(("ptrace_use_dbregs=%u  sys_use_dbregs=%u by [%d] ret = %d\n",
                  pfm_sessions.pfs_ptrace_use_dbregs,
                  pfm_sessions.pfs_sys_use_dbregs,
-                 task->pid, ret));
+                 task_pid_nr(task), ret));
 
        UNLOCK_PFS(flags);
 
@@ -3568,7 +3569,7 @@ pfm_release_debug_registers(struct task_struct *task)
 
        LOCK_PFS(flags);
        if (pfm_sessions.pfs_ptrace_use_dbregs == 0) {
-               printk(KERN_ERR "perfmon: invalid release for [%d] ptrace_use_dbregs=0\n", task->pid);
+               printk(KERN_ERR "perfmon: invalid release for [%d] ptrace_use_dbregs=0\n", task_pid_nr(task));
                ret = -1;
        }  else {
                pfm_sessions.pfs_ptrace_use_dbregs--;
@@ -3620,7 +3621,7 @@ pfm_restart(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 
        /* sanity check */
        if (unlikely(task == NULL)) {
-               printk(KERN_ERR "perfmon: [%d] pfm_restart no task\n", current->pid);
+               printk(KERN_ERR "perfmon: [%d] pfm_restart no task\n", task_pid_nr(current));
                return -EINVAL;
        }
 
@@ -3629,7 +3630,7 @@ pfm_restart(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
                fmt = ctx->ctx_buf_fmt;
 
                DPRINT(("restarting self %d ovfl=0x%lx\n",
-                       task->pid,
+                       task_pid_nr(task),
                        ctx->ctx_ovfl_regs[0]));
 
                if (CTX_HAS_SMPL(ctx)) {
@@ -3653,11 +3654,11 @@ pfm_restart(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
                                pfm_reset_regs(ctx, ctx->ctx_ovfl_regs, PFM_PMD_LONG_RESET);
 
                        if (rst_ctrl.bits.mask_monitoring == 0) {
-                               DPRINT(("resuming monitoring for [%d]\n", task->pid));
+                               DPRINT(("resuming monitoring for [%d]\n", task_pid_nr(task)));
 
                                if (state == PFM_CTX_MASKED) pfm_restore_monitoring(task);
                        } else {
-                               DPRINT(("keeping monitoring stopped for [%d]\n", task->pid));
+                               DPRINT(("keeping monitoring stopped for [%d]\n", task_pid_nr(task)));
 
                                // cannot use pfm_stop_monitoring(task, regs);
                        }
@@ -3714,10 +3715,10 @@ pfm_restart(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
         * "self-monitoring".
         */
        if (CTX_OVFL_NOBLOCK(ctx) == 0 && state == PFM_CTX_MASKED) {
-               DPRINT(("unblocking [%d] \n", task->pid));
+               DPRINT(("unblocking [%d] \n", task_pid_nr(task)));
                complete(&ctx->ctx_restart_done);
        } else {
-               DPRINT(("[%d] armed exit trap\n", task->pid));
+               DPRINT(("[%d] armed exit trap\n", task_pid_nr(task)));
 
                ctx->ctx_fl_trap_reason = PFM_TRAP_REASON_RESET;
 
@@ -3805,7 +3806,7 @@ pfm_write_ibr_dbr(int mode, pfm_context_t *ctx, void *arg, int count, struct pt_
         * don't bother if we are loaded and task is being debugged
         */
        if (is_loaded && (thread->flags & IA64_THREAD_DBG_VALID) != 0) {
-               DPRINT(("debug registers already in use for [%d]\n", task->pid));
+               DPRINT(("debug registers already in use for [%d]\n", task_pid_nr(task)));
                return -EBUSY;
        }
 
@@ -3846,7 +3847,7 @@ pfm_write_ibr_dbr(int mode, pfm_context_t *ctx, void *arg, int count, struct pt_
         * is shared by all processes running on it
         */
        if (first_time && can_access_pmu) {
-               DPRINT(("[%d] clearing ibrs, dbrs\n", task->pid));
+               DPRINT(("[%d] clearing ibrs, dbrs\n", task_pid_nr(task)));
                for (i=0; i < pmu_conf->num_ibrs; i++) {
                        ia64_set_ibr(i, 0UL);
                        ia64_dv_serialize_instruction();
@@ -4035,7 +4036,7 @@ pfm_stop(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
                return -EBUSY;
        }
        DPRINT(("task [%d] ctx_state=%d is_system=%d\n",
-               PFM_CTX_TASK(ctx)->pid,
+               task_pid_nr(PFM_CTX_TASK(ctx)),
                state,
                is_system));
        /*
@@ -4093,7 +4094,7 @@ pfm_stop(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
                 * monitoring disabled in kernel at next reschedule
                 */
                ctx->ctx_saved_psr_up = 0;
-               DPRINT(("task=[%d]\n", task->pid));
+               DPRINT(("task=[%d]\n", task_pid_nr(task)));
        }
        return 0;
 }
@@ -4298,11 +4299,12 @@ pfm_context_load(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 
                if (is_system) {
                        if (pfm_sessions.pfs_ptrace_use_dbregs) {
-                               DPRINT(("cannot load [%d] dbregs in use\n", task->pid));
+                               DPRINT(("cannot load [%d] dbregs in use\n",
+                                                       task_pid_nr(task)));
                                ret = -EBUSY;
                        } else {
                                pfm_sessions.pfs_sys_use_dbregs++;
-                               DPRINT(("load [%d] increased sys_use_dbreg=%u\n", task->pid, pfm_sessions.pfs_sys_use_dbregs));
+                               DPRINT(("load [%d] increased sys_use_dbreg=%u\n", task_pid_nr(task), pfm_sessions.pfs_sys_use_dbregs));
                                set_dbregs = 1;
                        }
                }
@@ -4394,7 +4396,7 @@ pfm_context_load(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
 
                        /* allow user level control */
                        ia64_psr(regs)->sp = 0;
-                       DPRINT(("clearing psr.sp for [%d]\n", task->pid));
+                       DPRINT(("clearing psr.sp for [%d]\n", task_pid_nr(task)));
 
                        SET_LAST_CPU(ctx, smp_processor_id());
                        INC_ACTIVATION();
@@ -4429,7 +4431,7 @@ pfm_context_load(pfm_context_t *ctx, void *arg, int count, struct pt_regs *regs)
                 */
                SET_PMU_OWNER(task, ctx);
 
-               DPRINT(("context loaded on PMU for [%d]\n", task->pid));
+               DPRINT(("context loaded on PMU for [%d]\n", task_pid_nr(task)));
        } else {
                /*
                 * when not current, task MUST be stopped, so this is safe
@@ -4493,7 +4495,7 @@ pfm_context_unload(pfm_context_t *ctx, void *arg, int count, struct pt_regs *reg
        int prev_state, is_system;
        int ret;
 
-       DPRINT(("ctx_state=%d task [%d]\n", ctx->ctx_state, task ? task->pid : -1));
+       DPRINT(("ctx_state=%d task [%d]\n", ctx->ctx_state, task ? task_pid_nr(task) : -1));
 
        prev_state = ctx->ctx_state;
        is_system  = ctx->ctx_fl_system;
@@ -4568,7 +4570,7 @@ pfm_context_unload(pfm_context_t *ctx, void *arg, int count, struct pt_regs *reg
                 */
                ia64_psr(regs)->sp = 1;
 
-               DPRINT(("setting psr.sp for [%d]\n", task->pid));
+               DPRINT(("setting psr.sp for [%d]\n", task_pid_nr(task)));
        }
        /*
         * save PMDs to context
@@ -4608,7 +4610,7 @@ pfm_context_unload(pfm_context_t *ctx, void *arg, int count, struct pt_regs *reg
        ctx->ctx_fl_can_restart  = 0;
        ctx->ctx_fl_going_zombie = 0;
 
-       DPRINT(("disconnected [%d] from context\n", task->pid));
+       DPRINT(("disconnected [%d] from context\n", task_pid_nr(task)));
 
        return 0;
 }
@@ -4631,7 +4633,7 @@ pfm_exit_thread(struct task_struct *task)
 
        PROTECT_CTX(ctx, flags);
 
-       DPRINT(("state=%d task [%d]\n", ctx->ctx_state, task->pid));
+       DPRINT(("state=%d task [%d]\n", ctx->ctx_state, task_pid_nr(task)));
 
        state = ctx->ctx_state;
        switch(state) {
@@ -4640,13 +4642,13 @@ pfm_exit_thread(struct task_struct *task)
                         * only comes to this function if pfm_context is not NULL, i.e., cannot
                         * be in unloaded state
                         */
-                       printk(KERN_ERR "perfmon: pfm_exit_thread [%d] ctx unloaded\n", task->pid);
+                       printk(KERN_ERR "perfmon: pfm_exit_thread [%d] ctx unloaded\n", task_pid_nr(task));
                        break;
                case PFM_CTX_LOADED:
                case PFM_CTX_MASKED:
                        ret = pfm_context_unload(ctx, NULL, 0, regs);
                        if (ret) {
-                               printk(KERN_ERR "perfmon: pfm_exit_thread [%d] state=%d unload failed %d\n", task->pid, state, ret);
+                               printk(KERN_ERR "perfmon: pfm_exit_thread [%d] state=%d unload failed %d\n", task_pid_nr(task), state, ret);
                        }
                        DPRINT(("ctx unloaded for current state was %d\n", state));
 
@@ -4655,12 +4657,12 @@ pfm_exit_thread(struct task_struct *task)
                case PFM_CTX_ZOMBIE:
                        ret = pfm_context_unload(ctx, NULL, 0, regs);
                        if (ret) {
-                               printk(KERN_ERR "perfmon: pfm_exit_thread [%d] state=%d unload failed %d\n", task->pid, state, ret);
+                               printk(KERN_ERR "perfmon: pfm_exit_thread [%d] state=%d unload failed %d\n", task_pid_nr(task), state, ret);
                        }
                        free_ok = 1;
                        break;
                default:
-                       printk(KERN_ERR "perfmon: pfm_exit_thread [%d] unexpected state=%d\n", task->pid, state);
+                       printk(KERN_ERR "perfmon: pfm_exit_thread [%d] unexpected state=%d\n", task_pid_nr(task), state);
                        break;
        }
        UNPROTECT_CTX(ctx, flags);
@@ -4744,7 +4746,7 @@ recheck:
        DPRINT(("context %d state=%d [%d] task_state=%ld must_stop=%d\n",
                ctx->ctx_fd,
                state,
-               task->pid,
+               task_pid_nr(task),
                task->state, PFM_CMD_STOPPED(cmd)));
 
        /*
@@ -4791,7 +4793,7 @@ recheck:
         */
        if (PFM_CMD_STOPPED(cmd)) {
                if ((task->state != TASK_STOPPED) && (task->state != TASK_TRACED)) {
-                       DPRINT(("[%d] task not in stopped state\n", task->pid));
+                       DPRINT(("[%d] task not in stopped state\n", task_pid_nr(task)));
                        return -EBUSY;
                }
                /*
@@ -4884,7 +4886,7 @@ restart_args:
         * limit abuse to min page size
         */
        if (unlikely(sz > PFM_MAX_ARGSIZE)) {
-               printk(KERN_ERR "perfmon: [%d] argument too big %lu\n", current->pid, sz);
+               printk(KERN_ERR "perfmon: [%d] argument too big %lu\n", task_pid_nr(current), sz);
                return -E2BIG;
        }
 
@@ -5031,11 +5033,11 @@ pfm_context_force_terminate(pfm_context_t *ctx, struct pt_regs *regs)
 {
        int ret;
 
-       DPRINT(("entering for [%d]\n", current->pid));
+       DPRINT(("entering for [%d]\n", task_pid_nr(current)));
 
        ret = pfm_context_unload(ctx, NULL, 0, regs);
        if (ret) {
-               printk(KERN_ERR "pfm_context_force_terminate: [%d] unloaded failed with %d\n", current->pid, ret);
+               printk(KERN_ERR "pfm_context_force_terminate: [%d] unloaded failed with %d\n", task_pid_nr(current), ret);
        }
 
        /*
@@ -5072,7 +5074,7 @@ pfm_handle_work(void)
 
        ctx = PFM_GET_CTX(current);
        if (ctx == NULL) {
-               printk(KERN_ERR "perfmon: [%d] has no PFM context\n", current->pid);
+               printk(KERN_ERR "perfmon: [%d] has no PFM context\n", task_pid_nr(current));
                return;
        }
 
@@ -5269,7 +5271,7 @@ pfm_overflow_handler(struct task_struct *task, pfm_context_t *ctx, u64 pmc0, str
        DPRINT_ovfl(("pmc0=0x%lx pid=%d iip=0x%lx, %s "
                     "used_pmds=0x%lx\n",
                        pmc0,
-                       task ? task->pid: -1,
+                       task ? task_pid_nr(task): -1,
                        (regs ? regs->cr_iip : 0),
                        CTX_OVFL_NOBLOCK(ctx) ? "nonblocking" : "blocking",
                        ctx->ctx_used_pmds[0]));
@@ -5458,7 +5460,7 @@ pfm_overflow_handler(struct task_struct *task, pfm_context_t *ctx, u64 pmc0, str
        }
 
        DPRINT_ovfl(("owner [%d] pending=%ld reason=%u ovfl_pmds=0x%lx ovfl_notify=0x%lx masked=%d\n",
-                       GET_PMU_OWNER() ? GET_PMU_OWNER()->pid : -1,
+                       GET_PMU_OWNER() ? task_pid_nr(GET_PMU_OWNER()) : -1,
                        PFM_GET_WORK_PENDING(task),
                        ctx->ctx_fl_trap_reason,
                        ovfl_pmds,
@@ -5483,7 +5485,7 @@ pfm_overflow_handler(struct task_struct *task, pfm_context_t *ctx, u64 pmc0, str
 sanity_check:
        printk(KERN_ERR "perfmon: CPU%d overflow handler [%d] pmc0=0x%lx\n",
                        smp_processor_id(),
-                       task ? task->pid : -1,
+                       task ? task_pid_nr(task) : -1,
                        pmc0);
        return;
 
@@ -5516,7 +5518,7 @@ stop_monitoring:
         *
         * Overall pretty hairy stuff....
         */
-       DPRINT(("ctx is zombie for [%d], converted to spurious\n", task ? task->pid: -1));
+       DPRINT(("ctx is zombie for [%d], converted to spurious\n", task ? task_pid_nr(task): -1));
        pfm_clear_psr_up();
        ia64_psr(regs)->up = 0;
        ia64_psr(regs)->sp = 1;
@@ -5577,13 +5579,13 @@ pfm_do_interrupt_handler(int irq, void *arg, struct pt_regs *regs)
 
 report_spurious1:
        printk(KERN_INFO "perfmon: spurious overflow interrupt on CPU%d: process %d has no PFM context\n",
-               this_cpu, task->pid);
+               this_cpu, task_pid_nr(task));
        pfm_unfreeze_pmu();
        return -1;
 report_spurious2:
        printk(KERN_INFO "perfmon: spurious overflow interrupt on CPU%d: process %d, invalid flag\n", 
                this_cpu, 
-               task->pid);
+               task_pid_nr(task));
        pfm_unfreeze_pmu();
        return -1;
 }
@@ -5870,7 +5872,8 @@ pfm_force_cleanup(pfm_context_t *ctx, struct pt_regs *regs)
        ia64_psr(regs)->sp = 1;
 
        if (GET_PMU_OWNER() == task) {
-               DPRINT(("cleared ownership for [%d]\n", ctx->ctx_task->pid));
+               DPRINT(("cleared ownership for [%d]\n",
+                                       task_pid_nr(ctx->ctx_task)));
                SET_PMU_OWNER(NULL, NULL);
        }
 
@@ -5882,7 +5885,7 @@ pfm_force_cleanup(pfm_context_t *ctx, struct pt_regs *regs)
        task->thread.pfm_context  = NULL;
        task->thread.flags       &= ~IA64_THREAD_PM_VALID;
 
-       DPRINT(("force cleanup for [%d]\n",  task->pid));
+       DPRINT(("force cleanup for [%d]\n",  task_pid_nr(task)));
 }
 
 
@@ -6426,7 +6429,7 @@ pfm_flush_pmds(struct task_struct *task, pfm_context_t *ctx)
 
                if (PMD_IS_COUNTING(i)) {
                        DPRINT(("[%d] pmd[%d] ctx_pmd=0x%lx hw_pmd=0x%lx\n",
-                               task->pid,
+                               task_pid_nr(task),
                                i,
                                ctx->ctx_pmds[i].val,
                                val & ovfl_val));
@@ -6448,11 +6451,11 @@ pfm_flush_pmds(struct task_struct *task, pfm_context_t *ctx)
                         */
                        if (pmc0 & (1UL << i)) {
                                val += 1 + ovfl_val;
-                               DPRINT(("[%d] pmd[%d] overflowed\n", task->pid, i));
+                               DPRINT(("[%d] pmd[%d] overflowed\n", task_pid_nr(task), i));
                        }
                }
 
-               DPRINT(("[%d] ctx_pmd[%d]=0x%lx  pmd_val=0x%lx\n", task->pid, i, val, pmd_val));
+               DPRINT(("[%d] ctx_pmd[%d]=0x%lx  pmd_val=0x%lx\n", task_pid_nr(task), i, val, pmd_val));
 
                if (is_self) ctx->th_pmds[i] = pmd_val;
 
@@ -6793,14 +6796,14 @@ dump_pmu_state(const char *from)
        printk("CPU%d from %s() current [%d] iip=0x%lx %s\n", 
                this_cpu, 
                from, 
-               current->pid, 
+               task_pid_nr(current),
                regs->cr_iip,
                current->comm);
 
        task = GET_PMU_OWNER();
        ctx  = GET_PMU_CTX();
 
-       printk("->CPU%d owner [%d] ctx=%p\n", this_cpu, task ? task->pid : -1, ctx);
+       printk("->CPU%d owner [%d] ctx=%p\n", this_cpu, task ? task_pid_nr(task) : -1, ctx);
 
        psr = pfm_get_psr();
 
@@ -6848,7 +6851,7 @@ pfm_inherit(struct task_struct *task, struct pt_regs *regs)
 {
        struct thread_struct *thread;
 
-       DPRINT(("perfmon: pfm_inherit clearing state for [%d]\n", task->pid));
+       DPRINT(("perfmon: pfm_inherit clearing state for [%d]\n", task_pid_nr(task)));
 
        thread = &task->thread;
 
index ff80eab..a7af1cb 100644 (file)
@@ -44,11 +44,11 @@ default_validate(struct task_struct *task, unsigned int flags, int cpu, void *da
        int ret = 0;
 
        if (data == NULL) {
-               DPRINT(("[%d] no argument passed\n", task->pid));
+               DPRINT(("[%d] no argument passed\n", task_pid_nr(task)));
                return -EINVAL;
        }
 
-       DPRINT(("[%d] validate flags=0x%x CPU%d\n", task->pid, flags, cpu));
+       DPRINT(("[%d] validate flags=0x%x CPU%d\n", task_pid_nr(task), flags, cpu));
 
        /*
         * must hold at least the buffer header + one minimally sized entry
@@ -88,7 +88,7 @@ default_init(struct task_struct *task, void *buf, unsigned int flags, int cpu, v
        hdr->hdr_count        = 0UL;
 
        DPRINT(("[%d] buffer=%p buf_size=%lu hdr_size=%lu hdr_version=%u cur_offs=%lu\n",
-               task->pid,
+               task_pid_nr(task),
                buf,
                hdr->hdr_buf_size,
                sizeof(*hdr),
@@ -245,7 +245,7 @@ default_restart(struct task_struct *task, pfm_ovfl_ctrl_t *ctrl, void *buf, stru
 static int
 default_exit(struct task_struct *task, void *buf, struct pt_regs *regs)
 {
-       DPRINT(("[%d] exit(%p)\n", task->pid, buf));
+       DPRINT(("[%d] exit(%p)\n", task_pid_nr(task), buf));
        return 0;
 }
 
index c613fc0..2418289 100644 (file)
@@ -105,7 +105,8 @@ show_regs (struct pt_regs *regs)
        unsigned long ip = regs->cr_iip + ia64_psr(regs)->ri;
 
        print_modules();
-       printk("\nPid: %d, CPU %d, comm: %20s\n", current->pid, smp_processor_id(), current->comm);
+       printk("\nPid: %d, CPU %d, comm: %20s\n", task_pid_nr(current),
+                       smp_processor_id(), current->comm);
        printk("psr : %016lx ifs : %016lx ip  : [<%016lx>]    %s\n",
               regs->cr_ipsr, regs->cr_ifs, ip, print_tainted());
        print_symbol("ip is at %s\n", ip);
index 3aeaf15..78d65cb 100644 (file)
@@ -61,7 +61,7 @@ die (const char *str, struct pt_regs *regs, long err)
 
        if (++die.lock_owner_depth < 3) {
                printk("%s[%d]: %s %ld [%d]\n",
-                       current->comm, current->pid, str, err, ++die_counter);
+               current->comm, task_pid_nr(current), str, err, ++die_counter);
                (void) notify_die(DIE_OOPS, (char *)str, regs, err, 255, SIGSEGV);
                show_regs(regs);
        } else
@@ -315,7 +315,7 @@ handle_fpu_swa (int fp_fault, struct pt_regs *regs, unsigned long isr)
                                last.time = current_jiffies + 5 * HZ;
                                printk(KERN_WARNING
                                        "%s(%d): floating-point assist fault at ip %016lx, isr %016lx\n",
-                                       current->comm, current->pid, regs->cr_iip + ia64_psr(regs)->ri, isr);
+                                       current->comm, task_pid_nr(current), regs->cr_iip + ia64_psr(regs)->ri, isr);
                        }
                }
        }
@@ -453,7 +453,7 @@ ia64_fault (unsigned long vector, unsigned long isr, unsigned long ifa,
                if (code == 8) {
 # ifdef CONFIG_IA64_PRINT_HAZARDS
                        printk("%s[%d]: possible hazard @ ip=%016lx (pr = %016lx)\n",
-                              current->comm, current->pid,
+                              current->comm, task_pid_nr(current),
                               regs.cr_iip + ia64_psr(&regs)->ri, regs.pr);
 # endif
                        return;
index fe6aa5a..2173de9 100644 (file)
@@ -1340,7 +1340,8 @@ ia64_handle_unaligned (unsigned long ifa, struct pt_regs *regs)
                        size_t len;
 
                        len = sprintf(buf, "%s(%d): unaligned access to 0x%016lx, "
-                                     "ip=0x%016lx\n\r", current->comm, current->pid,
+                                     "ip=0x%016lx\n\r", current->comm,
+                                     task_pid_nr(current),
                                      ifa, regs->cr_iip + ipsr->ri);
                        /*
                         * Don't call tty_write_message() if we're in the kernel; we might
@@ -1363,7 +1364,7 @@ ia64_handle_unaligned (unsigned long ifa, struct pt_regs *regs)
                                       "administrator\n"
                                       "echo 0 > /proc/sys/kernel/ignore-"
                                       "unaligned-usertrap to re-enable\n",
-                                      current->comm, current->pid);
+                                      current->comm, task_pid_nr(current));
                        }
                }
        } else {
index 97e0b1c..89ba4a0 100644 (file)
@@ -196,7 +196,7 @@ static void show_registers(struct pt_regs *regs)
                printk("SPI: %08lx\n", sp);
        }
        printk("Process %s (pid: %d, process nr: %d, stackpage=%08lx)",
-               current->comm, current->pid, 0xffff & i, 4096+(unsigned long)current);
+               current->comm, task_pid_nr(current), 0xffff & i, 4096+(unsigned long)current);
 
        /*
         * When in-kernel, we also print out the stack and code at the
index 4e2752a..97f556f 100644 (file)
@@ -900,7 +900,7 @@ void show_registers(struct pt_regs *regs)
               regs->d4, regs->d5, regs->a0, regs->a1);
 
        printk("Process %s (pid: %d, task=%p)\n",
-               current->comm, current->pid, current);
+               current->comm, task_pid_nr(current), current);
        addr = (unsigned long)&fp->un;
        printk("Frame format=%X ", regs->format);
        switch (regs->format) {
@@ -1038,7 +1038,7 @@ void bad_super_trap (struct frame *fp)
                                fp->un.fmtb.daddr, space_names[ssw & DFC],
                                fp->ptregs.pc);
        }
-       printk ("Current process id is %d\n", current->pid);
+       printk ("Current process id is %d\n", task_pid_nr(current));
        die_if_kernel("BAD KERNEL TRAP", &fp->ptregs, 0);
 }
 
index 7b78d13..fa50078 100644 (file)
@@ -314,7 +314,7 @@ void show_registers(const struct pt_regs *regs)
        __show_regs(regs);
        print_modules();
        printk("Process %s (pid: %d, threadinfo=%p, task=%p)\n",
-               current->comm, current->pid, current_thread_info(), current);
+               current->comm, task_pid_nr(current), current_thread_info(), current);
        show_stacktrace(current, regs);
        show_code((unsigned int __user *) regs->cp0_epc);
        printk("\n");
index bbf029a..99fd569 100644 (file)
@@ -219,7 +219,7 @@ void die_if_kernel(char *str, struct pt_regs *regs, long err)
                        return; /* STFU */
 
                printk(KERN_CRIT "%s (pid %d): %s (code %ld) at " RFMT "\n",
-                       current->comm, current->pid, str, err, regs->iaoq[0]);
+                       current->comm, task_pid_nr(current), str, err, regs->iaoq[0]);
 #ifdef PRINT_USER_FAULTS
                /* XXX for debugging only */
                show_regs(regs);
@@ -252,7 +252,7 @@ KERN_CRIT "                     ||     ||\n");
        
        if (err)
                printk(KERN_CRIT "%s (pid %d): %s (code %ld)\n",
-                       current->comm, current->pid, str, err);
+                       current->comm, task_pid_nr(current), str, err);
 
        /* Wot's wrong wif bein' racy? */
        if (current->thread.flags & PARISC_KERNEL_DEATH) {
@@ -317,7 +317,7 @@ static void handle_break(struct pt_regs *regs)
        if (unlikely(iir != GDB_BREAK_INSN)) {
                printk(KERN_DEBUG "break %d,%d: pid=%d command='%s'\n",
                        iir & 31, (iir>>13) & ((1<<13)-1),
-                       current->pid, current->comm);
+                       task_pid_nr(current), current->comm);
                show_regs(regs);
        }
 #endif
@@ -747,7 +747,7 @@ void handle_interruption(int code, struct pt_regs *regs)
                if (user_mode(regs)) {
 #ifdef PRINT_USER_FAULTS
                        printk(KERN_DEBUG "\nhandle_interruption() pid=%d command='%s'\n",
-                           current->pid, current->comm);
+                           task_pid_nr(current), current->comm);
                        show_regs(regs);
 #endif
                        /* SIGBUS, for lack of a better one. */
@@ -772,7 +772,7 @@ void handle_interruption(int code, struct pt_regs *regs)
                else
                        printk(KERN_DEBUG "User Fault (long pointer) (fault %d) ",
                               code);
-               printk("pid=%d command='%s'\n", current->pid, current->comm);
+               printk("pid=%d command='%s'\n", task_pid_nr(current), current->comm);
                show_regs(regs);
 #endif
                si.si_signo = SIGSEGV;
index 347bb92..aebf3c1 100644 (file)
@@ -469,7 +469,7 @@ void handle_unaligned(struct pt_regs *regs)
                    && ++unaligned_count < 5) {
                        char buf[256];
                        sprintf(buf, "%s(%d): unaligned access to 0x" RFMT " at ip=0x" RFMT "\n",
-                               current->comm, current->pid, regs->ior, regs->iaoq[0]);
+                               current->comm, task_pid_nr(current), regs->ior, regs->iaoq[0]);
                        printk(KERN_WARNING "%s", buf);
 #ifdef DEBUG_UNALIGNED
                        show_regs(regs);
index 1c091b4..b2e3e9a 100644 (file)
@@ -211,7 +211,7 @@ bad_area:
 #ifdef PRINT_USER_FAULTS
                printk(KERN_DEBUG "\n");
                printk(KERN_DEBUG "do_page_fault() pid=%d command='%s' type=%lu address=0x%08lx\n",
-                   tsk->pid, tsk->comm, code, address);
+                   task_pid_nr(tsk), tsk->comm, code, address);
                if (vma) {
                        printk(KERN_DEBUG "vm_start = 0x%08lx, vm_end = 0x%08lx\n",
                                        vma->vm_start, vma->vm_end);
index ea6ad7a..b9d8837 100644 (file)
@@ -459,7 +459,7 @@ void show_regs(struct pt_regs * regs)
                printk("DAR: "REG", DSISR: "REG"\n", regs->dar, regs->dsisr);
 #endif
        printk("TASK = %p[%d] '%s' THREAD: %p",
-              current, current->pid, current->comm, task_thread_info(current));
+              current, task_pid_nr(current), current->comm, task_thread_info(current));
 
 #ifdef CONFIG_SMP
        printk(" CPU: %d", smp_processor_id());
index 9fb4a68..59c464e 100644 (file)
@@ -881,7 +881,7 @@ void nonrecoverable_exception(struct pt_regs *regs)
 void trace_syscall(struct pt_regs *regs)
 {
        printk("Task: %p(%d), PC: %08lX/%08lX, Syscall: %3ld, Result: %s%ld    %s\n",
-              current, current->pid, regs->nip, regs->link, regs->gpr[0],
+              current, task_pid_nr(current), regs->nip, regs->link, regs->gpr[0],
               regs->ccr&0x10000000?"Error=":"", regs->gpr[3], print_tainted());
 }
 
index abb447a..70c5737 100644 (file)
@@ -166,7 +166,7 @@ void show_regs(struct pt_regs *regs)
 
         printk("CPU:    %d    %s\n", task_thread_info(tsk)->cpu, print_tainted());
         printk("Process %s (pid: %d, task: %p, ksp: %p)\n",
-              current->comm, current->pid, (void *) tsk,
+              current->comm, task_pid_nr(current), (void *) tsk,
               (void *) tsk->thread.ksp);
 
        show_registers(regs);
index b446999..6d7f2b0 100644 (file)
@@ -121,7 +121,7 @@ void machine_power_off(void)
 void show_regs(struct pt_regs * regs)
 {
        printk("\n");
-       printk("Pid : %d, Comm: %20s\n", current->pid, current->comm);
+       printk("Pid : %d, Comm: %20s\n", task_pid_nr(current), current->comm);
        print_symbol("PC is at %s\n", instruction_pointer(regs));
        printk("PC  : %08lx SP  : %08lx SR  : %08lx ",
               regs->pc, regs->regs[15], regs->sr);
index 2f42442..ca754fd 100644 (file)
@@ -382,7 +382,7 @@ static int setup_frame(int sig, struct k_sigaction *ka,
        set_fs(USER_DS);
 
        pr_debug("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
-                current->comm, current->pid, frame, regs->pc, regs->pr);
+                current->comm, task_pid_nr(current), frame, regs->pc, regs->pr);
 
        flush_cache_sigtramp(regs->pr);
 
@@ -462,7 +462,7 @@ static int setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
        set_fs(USER_DS);
 
        pr_debug("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
-                current->comm, current->pid, frame, regs->pc, regs->pr);
+                current->comm, task_pid_nr(current), frame, regs->pc, regs->pr);
 
        flush_cache_sigtramp(regs->pr);
 
index dcb46e7..cf99111 100644 (file)
@@ -95,8 +95,8 @@ void die(const char * str, struct pt_regs * regs, long err)
        print_modules();
        show_regs(regs);
 
-       printk("Process: %s (pid: %d, stack limit = %p)\n",
-              current->comm, current->pid, task_stack_page(current) + 1);
+       printk("Process: %s (pid: %d, stack limit = %p)\n", current->comm,
+                       task_pid_nr(current), task_stack_page(current) + 1);
 
        if (!user_mode(regs) || in_interrupt())
                dump_mem("Stack: ", regs->regs[15], THREAD_SIZE +
@@ -386,7 +386,8 @@ static int handle_unaligned_access(u16 instruction, struct pt_regs *regs)
 
                printk(KERN_NOTICE "Fixing up unaligned userspace access "
                       "in \"%s\" pid=%d pc=0x%p ins=0x%04hx\n",
-                      current->comm,current->pid,(u16*)regs->pc,instruction);
+                      current->comm, task_pid_nr(current),
+                      (u16 *)regs->pc, instruction);
        }
 
        ret = -EFAULT;
index 9d0d58f..c03101f 100644 (file)
@@ -764,7 +764,7 @@ static int misaligned_fixup(struct pt_regs *regs)
                --user_mode_unaligned_fixup_count;
                /* Only do 'count' worth of these reports, to remove a potential DoS against syslog */
                printk("Fixing up unaligned userspace access in \"%s\" pid=%d pc=0x%08x ins=0x%08lx\n",
-                      current->comm, current->pid, (__u32)regs->pc, opcode);
+                      current->comm, task_pid_nr(current), (__u32)regs->pc, opcode);
        } else
 #endif
        if (!user_mode(regs) && (kernel_mode_unaligned_fixup_count > 0)) {
@@ -774,7 +774,7 @@ static int misaligned_fixup(struct pt_regs *regs)
                               (__u32)regs->pc, opcode);
                } else {
                        printk("Fixing up unaligned kernelspace access in \"%s\" pid=%d pc=0x%08x ins=0x%08lx\n",
-                              current->comm, current->pid, (__u32)regs->pc, opcode);
+                              current->comm, task_pid_nr(current), (__u32)regs->pc, opcode);
                }
        }
 
index 7aea586..7c79a1b 100644 (file)
@@ -81,7 +81,7 @@ static inline void print_vma(struct vm_area_struct *vma)
 
 static inline void print_task(struct task_struct *tsk)
 {
-       printk("Task pid %d\n", tsk->pid);
+       printk("Task pid %d\n", task_pid_nr(tsk));
 }
 
 static pte_t *lookup_pte(struct mm_struct *mm, unsigned long address)
@@ -272,7 +272,7 @@ bad_area:
                         * usermode, so only need a few */
                        count++;
                        printk("user mode bad_area address=%08lx pid=%d (%s) pc=%08lx\n",
-                               address, current->pid, current->comm,
+                               address, task_pid_nr(current), current->comm,
                                (unsigned long) regs->pc);
 #if 0
                        show_regs(regs);
index 003f8ee..fe562db 100644 (file)
@@ -155,7 +155,7 @@ static inline void read_sunos_user(struct pt_regs *regs, unsigned long offset,
                /* Rest of them are completely unsupported. */
        default:
                printk("%s [%d]: Wants to read user offset %ld\n",
-                      current->comm, current->pid, offset);
+                      current->comm, task_pid_nr(current), offset);
                pt_error_return(regs, EIO);
                return;
        }
@@ -222,7 +222,7 @@ static inline void write_sunos_user(struct pt_regs *regs, unsigned long offset,
                /* Rest of them are completely unsupported or "no-touch". */
        default:
                printk("%s [%d]: Wants to write user offset %ld\n",
-                      current->comm, current->pid, offset);
+                      current->comm, task_pid_nr(current), offset);
                goto failure;
        }
 success:
index 6c0221e..42bf09d 100644 (file)
@@ -357,7 +357,7 @@ c_sys_nis_syscall (struct pt_regs *regs)
        if (count++ > 5)
                return -ENOSYS;
        printk ("%s[%d]: Unimplemented SPARC system call %d\n",
-               current->comm, current->pid, (int)regs->u_regs[1]);
+               current->comm, task_pid_nr(current), (int)regs->u_regs[1]);
 #ifdef DEBUG_UNIMP_SYSCALL     
        show_regs (regs);
 #endif
index 3bc3bff..d404e79 100644 (file)
@@ -38,7 +38,7 @@ struct trap_trace_entry trapbuf[1024];
 
 void syscall_trace_entry(struct pt_regs *regs)
 {
-       printk("%s[%d]: ", current->comm, current->pid);
+       printk("%s[%d]: ", current->comm, task_pid_nr(current));
        printk("scall<%d> (could be %d)\n", (int) regs->u_regs[UREG_G1],
               (int) regs->u_regs[UREG_I0]);
 }
@@ -99,7 +99,7 @@ void die_if_kernel(char *str, struct pt_regs *regs)
 "              /_| \\__/ |_\\\n"
 "                 \\__U_/\n");
 
-       printk("%s(%d): %s [#%d]\n", current->comm, current->pid, str, ++die_counter);
+       printk("%s(%d): %s [#%d]\n", current->comm, task_pid_nr(current), str, ++die_counter);
        show_regs(regs);
        add_taint(TAINT_DIE);
 
index 34573a5..e9c7e4f 100644 (file)
@@ -2225,7 +2225,7 @@ void die_if_kernel(char *str, struct pt_regs *regs)
 "              /_| \\__/ |_\\\n"
 "                 \\__U_/\n");
 
-       printk("%s(%d): %s [#%d]\n", current->comm, current->pid, str, ++die_counter);
+       printk("%s(%d): %s [#%d]\n", current->comm, task_pid_nr(current), str, ++die_counter);
        notify_die(DIE_OOPS, str, regs, 0, 255, SIGSEGV);
        __asm__ __volatile__("flushw");
        __show_regs(regs);
index ce3e07f..7654440 100644 (file)
@@ -15,8 +15,8 @@ void __show_regs(struct pt_regs * regs)
 {
        printk("\n");
        print_modules();
-       printk("Pid: %d, comm: %.20s %s %s\n",
-              current->pid, current->comm, print_tainted(), init_utsname()->release);
+       printk("Pid: %d, comm: %.20s %s %s\n", task_pid_nr(current),
+               current->comm, print_tainted(), init_utsname()->release);
        printk("RIP: %04lx:[<%016lx>] ", PT_REGS_CS(regs) & 0xffff,
               PT_REGS_RIP(regs));
        printk("\nRSP: %016lx  EFLAGS: %08lx\n", PT_REGS_RSP(regs),
index 097aeaf..044a477 100644 (file)
@@ -301,7 +301,7 @@ void show_regs(struct pt_regs * regs)
        unsigned long d0, d1, d2, d3, d6, d7;
 
        printk("\n");
-       printk("Pid: %d, comm: %20s\n", current->pid, current->comm);
+       printk("Pid: %d, comm: %20s\n", task_pid_nr(current), current->comm);
        printk("EIP: %04x:[<%08lx>] CPU: %d\n",0xffff & regs->xcs,regs->eip, smp_processor_id());
        print_symbol("EIP is at %s\n", regs->eip);
 
index 0d79df3..6dc394b 100644 (file)
@@ -200,8 +200,8 @@ badframe:
        if (show_unhandled_signals && printk_ratelimit())
                printk("%s%s[%d] bad frame in sigreturn frame:%p eip:%lx"
                       " esp:%lx oeax:%lx\n",
-                   current->pid > 1 ? KERN_INFO : KERN_EMERG,
-                   current->comm, current->pid, frame, regs->eip,
+                   task_pid_nr(current) > 1 ? KERN_INFO : KERN_EMERG,
+                   current->comm, task_pid_nr(current), frame, regs->eip,
                    regs->esp, regs->orig_eax);
 
        force_sig(SIGSEGV, current);
index b132d39..1e9d572 100644 (file)
@@ -316,7 +316,7 @@ void show_registers(struct pt_regs *regs)
        printk(KERN_EMERG "ds: %04x   es: %04x   fs: %04x  gs: %04x  ss: %04x\n",
               regs->xds & 0xffff, regs->xes & 0xffff, regs->xfs & 0xffff, gs, ss);
        printk(KERN_EMERG "Process %.*s (pid: %d, ti=%p task=%p task.ti=%p)",
-               TASK_COMM_LEN, current->comm, current->pid,
+               TASK_COMM_LEN, current->comm, task_pid_nr(current),
                current_thread_info(), current, task_thread_info(current));
        /*
         * When in-kernel, we also print out the stack and code at the
@@ -622,7 +622,7 @@ fastcall void __kprobes do_general_protection(struct pt_regs * regs,
            printk_ratelimit())
                printk(KERN_INFO
                    "%s[%d] general protection eip:%lx esp:%lx error:%lx\n",
-                   current->comm, current->pid,
+                   current->comm, task_pid_nr(current),
                    regs->eip, regs->esp, error_code);
 
        force_sig(SIGSEGV, current);
index 4fc5e40..4d3e538 100644 (file)
@@ -471,8 +471,8 @@ bad_area_nosemaphore:
                    printk_ratelimit()) {
                        printk("%s%s[%d]: segfault at %08lx eip %08lx "
                            "esp %08lx error %lx\n",
-                           tsk->pid > 1 ? KERN_INFO : KERN_EMERG,
-                           tsk->comm, tsk->pid, address, regs->eip,
+                           task_pid_nr(tsk) > 1 ? KERN_INFO : KERN_EMERG,
+                           tsk->comm, task_pid_nr(tsk), address, regs->eip,
                            regs->esp, error_code);
                }
                tsk->thread.cr2 = address;
index 8be99c7..397bcd6 100644 (file)
@@ -176,7 +176,7 @@ void do_unhandled(struct pt_regs *regs, unsigned long exccause)
        printk("Caught unhandled exception in '%s' "
               "(pid = %d, pc = %#010lx) - should not happen\n"
               "\tEXCCAUSE is %ld\n",
-              current->comm, current->pid, regs->pc, exccause);
+              current->comm, task_pid_nr(current), regs->pc, exccause);
        force_sig(SIGILL, current);
 }
 
@@ -228,7 +228,7 @@ do_illegal_instruction(struct pt_regs *regs)
        /* If in user mode, send SIGILL signal to current process. */
 
        printk("Illegal Instruction in '%s' (pid = %d, pc = %#010lx)\n",
-           current->comm, current->pid, regs->pc);
+           current->comm, task_pid_nr(current), regs->pc);
        force_sig(SIGILL, current);
 }
 
@@ -254,7 +254,7 @@ do_unaligned_user (struct pt_regs *regs)
        current->thread.error_code = -3;
        printk("Unaligned memory access to %08lx in '%s' "
               "(pid = %d, pc = %#010lx)\n",
-              regs->excvaddr, current->comm, current->pid, regs->pc);
+              regs->excvaddr, current->comm, task_pid_nr(current), regs->pc);
        info.si_signo = SIGBUS;
        info.si_errno = 0;
        info.si_code = BUS_ADRALN;