+/*
+ * GPL HEADER START
+ *
+ * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 only,
+ * as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License version 2 for more details (a copy is included
+ * in the LICENSE file that accompanied this code).
+ *
+ * You should have received a copy of the GNU General Public License
+ * version 2 along with this program; If not, see
+ * http://www.gnu.org/licenses/gpl-2.0.html
+ *
+ * GPL HEADER END
+ */
+/*
+ * Copyright (c) 2008, 2010, Oracle and/or its affiliates. All rights reserved.
+ * Use is subject to license terms.
+ *
+ * Copyright (c) 2012, Intel Corporation.
+ */
+/*
+ * This file is part of Lustre, http://www.lustre.org/
+ * Lustre is a trademark of Sun Microsystems, Inc.
+ */
+
#define DEBUG_SUBSYSTEM S_LNET
#define LUSTRE_TRACEFILE_PRIVATE
-#include <libcfs/libcfs.h>
-#include <libcfs/kp30.h>
+#include <linux/slab.h>
+#include <linux/tty.h>
+#include <linux/poll.h>
+#include <linux/mm.h>
#include "tracefile.h"
-#ifndef get_cpu
-#define get_cpu() smp_processor_id()
-#define put_cpu() do { } while (0)
-#endif
-
-/* three types of trace_data in linux */
-enum {
- TCD_TYPE_PROC = 0,
- TCD_TYPE_SOFTIRQ,
- TCD_TYPE_IRQ,
- TCD_TYPE_MAX
-};
-
/* percents to share the total debug memory for each type */
-static unsigned int pages_factor[TCD_TYPE_MAX] = {
- 80, /* 80% pages for TCD_TYPE_PROC */
- 10, /* 10% pages for TCD_TYPE_SOFTIRQ */
- 10 /* 10% pages for TCD_TYPE_IRQ */
+static unsigned int pages_factor[CFS_TCD_TYPE_MAX] = {
+ 80, /* 80% pages for CFS_TCD_TYPE_PROC */
+ 10, /* 10% pages for CFS_TCD_TYPE_SOFTIRQ */
+ 10 /* 10% pages for CFS_TCD_TYPE_IRQ */
};
-char *trace_console_buffers[NR_CPUS][3];
+char *cfs_trace_console_buffers[NR_CPUS][CFS_TCD_TYPE_MAX];
-struct rw_semaphore tracefile_sem;
+static DECLARE_RWSEM(cfs_tracefile_sem);
-int tracefile_init_arch()
+int cfs_tracefile_init_arch(void)
{
- int i;
- int j;
- struct trace_cpu_data *tcd;
-
- init_rwsem(&tracefile_sem);
+ int i;
+ int j;
+ struct cfs_trace_cpu_data *tcd;
/* initialize trace_data */
- memset(trace_data, 0, sizeof(trace_data));
- for (i = 0; i < TCD_TYPE_MAX; i++) {
- trace_data[i]=kmalloc(sizeof(union trace_data_union)*NR_CPUS,
- GFP_KERNEL);
- if (trace_data[i] == NULL)
+ memset(cfs_trace_data, 0, sizeof(cfs_trace_data));
+ for (i = 0; i < CFS_TCD_TYPE_MAX; i++) {
+ cfs_trace_data[i] =
+ kmalloc_array(num_possible_cpus(),
+ sizeof(union cfs_trace_data_union),
+ GFP_KERNEL);
+ if (!cfs_trace_data[i])
goto out;
-
}
/* arch related info initialized */
- tcd_for_each(tcd, i, j) {
+ cfs_tcd_for_each(tcd, i, j) {
+ spin_lock_init(&tcd->tcd_lock);
tcd->tcd_pages_factor = pages_factor[i];
tcd->tcd_type = i;
tcd->tcd_cpu = j;
for (i = 0; i < num_possible_cpus(); i++)
for (j = 0; j < 3; j++) {
- trace_console_buffers[i][j] =
- kmalloc(TRACE_CONSOLE_BUFFER_SIZE,
+ cfs_trace_console_buffers[i][j] =
+ kmalloc(CFS_TRACE_CONSOLE_BUFFER_SIZE,
GFP_KERNEL);
- if (trace_console_buffers[i][j] == NULL)
+ if (!cfs_trace_console_buffers[i][j])
goto out;
}
return 0;
out:
- tracefile_fini_arch();
- printk(KERN_ERR "lnet: No enough memory\n");
+ cfs_tracefile_fini_arch();
+ pr_err("lnet: Not enough memory\n");
return -ENOMEM;
-
}
-void tracefile_fini_arch()
+void cfs_tracefile_fini_arch(void)
{
- int i;
- int j;
+ int i;
+ int j;
for (i = 0; i < num_possible_cpus(); i++)
- for (j = 0; j < 3; j++)
- if (trace_console_buffers[i][j] != NULL) {
- kfree(trace_console_buffers[i][j]);
- trace_console_buffers[i][j] = NULL;
- }
-
- for (i = 0; trace_data[i] != NULL; i++) {
- kfree(trace_data[i]);
- trace_data[i] = NULL;
- }
-}
-
-void tracefile_read_lock()
-{
- down_read(&tracefile_sem);
-}
+ for (j = 0; j < 3; j++) {
+ kfree(cfs_trace_console_buffers[i][j]);
+ cfs_trace_console_buffers[i][j] = NULL;
+ }
-void tracefile_read_unlock()
-{
- up_read(&tracefile_sem);
+ for (i = 0; cfs_trace_data[i]; i++) {
+ kfree(cfs_trace_data[i]);
+ cfs_trace_data[i] = NULL;
+ }
}
-void tracefile_write_lock()
+void cfs_tracefile_read_lock(void)
{
- down_write(&tracefile_sem);
+ down_read(&cfs_tracefile_sem);
}
-void tracefile_write_unlock()
+void cfs_tracefile_read_unlock(void)
{
- up_write(&tracefile_sem);
+ up_read(&cfs_tracefile_sem);
}
-char *
-trace_get_console_buffer(void)
+void cfs_tracefile_write_lock(void)
{
- int cpu = get_cpu();
- int idx;
-
- if (in_irq()) {
- idx = 0;
- } else if (in_softirq()) {
- idx = 1;
- } else {
- idx = 2;
- }
-
- return trace_console_buffers[cpu][idx];
+ down_write(&cfs_tracefile_sem);
}
-void
-trace_put_console_buffer(char *buffer)
+void cfs_tracefile_write_unlock(void)
{
- put_cpu();
+ up_write(&cfs_tracefile_sem);
}
-struct trace_cpu_data *
-trace_get_tcd(void)
+enum cfs_trace_buf_type cfs_trace_buf_idx_get(void)
{
- int cpu;
-
- cpu = get_cpu();
if (in_irq())
- return &(*trace_data[TCD_TYPE_IRQ])[cpu].tcd;
- else if (in_softirq())
- return &(*trace_data[TCD_TYPE_SOFTIRQ])[cpu].tcd;
- return &(*trace_data[TCD_TYPE_PROC])[cpu].tcd;
+ return CFS_TCD_TYPE_IRQ;
+ if (in_softirq())
+ return CFS_TCD_TYPE_SOFTIRQ;
+ return CFS_TCD_TYPE_PROC;
}
-void
-trace_put_tcd (struct trace_cpu_data *tcd)
+/*
+ * The walking argument indicates the locking comes from all tcd types
+ * iterator and we must lock it and dissable local irqs to avoid deadlocks
+ * with other interrupt locks that might be happening. See LU-1311
+ * for details.
+ */
+int cfs_trace_lock_tcd(struct cfs_trace_cpu_data *tcd, int walking)
+ __acquires(&tcd->tcd_lock)
{
- put_cpu();
-}
-
-int trace_lock_tcd(struct trace_cpu_data *tcd)
-{
- __LASSERT(tcd->tcd_type < TCD_TYPE_MAX);
- if (tcd->tcd_type == TCD_TYPE_IRQ)
- local_irq_disable();
- else if (tcd->tcd_type == TCD_TYPE_SOFTIRQ)
- local_bh_disable();
+ __LASSERT(tcd->tcd_type < CFS_TCD_TYPE_MAX);
+ if (tcd->tcd_type == CFS_TCD_TYPE_IRQ)
+ spin_lock_irqsave(&tcd->tcd_lock, tcd->tcd_lock_flags);
+ else if (tcd->tcd_type == CFS_TCD_TYPE_SOFTIRQ)
+ spin_lock_bh(&tcd->tcd_lock);
+ else if (unlikely(walking))
+ spin_lock_irq(&tcd->tcd_lock);
+ else
+ spin_lock(&tcd->tcd_lock);
return 1;
}
-void trace_unlock_tcd(struct trace_cpu_data *tcd)
+void cfs_trace_unlock_tcd(struct cfs_trace_cpu_data *tcd, int walking)
+ __releases(&tcd->tcd_lock)
{
- __LASSERT(tcd->tcd_type < TCD_TYPE_MAX);
- if (tcd->tcd_type == TCD_TYPE_IRQ)
- local_irq_enable();
- else if (tcd->tcd_type == TCD_TYPE_SOFTIRQ)
- local_bh_enable();
+ __LASSERT(tcd->tcd_type < CFS_TCD_TYPE_MAX);
+ if (tcd->tcd_type == CFS_TCD_TYPE_IRQ)
+ spin_unlock_irqrestore(&tcd->tcd_lock, tcd->tcd_lock_flags);
+ else if (tcd->tcd_type == CFS_TCD_TYPE_SOFTIRQ)
+ spin_unlock_bh(&tcd->tcd_lock);
+ else if (unlikely(walking))
+ spin_unlock_irq(&tcd->tcd_lock);
+ else
+ spin_unlock(&tcd->tcd_lock);
}
-int tcd_owns_tage(struct trace_cpu_data *tcd, struct trace_page *tage)
+int cfs_tcd_owns_tage(struct cfs_trace_cpu_data *tcd,
+ struct cfs_trace_page *tage)
{
/*
* XXX nikita: do NOT call portals_debug_msg() (CDEBUG/ENTRY/EXIT)
}
void
-set_ptldebug_header(struct ptldebug_header *header, int subsys, int mask,
- const int line, unsigned long stack)
+cfs_set_ptldebug_header(struct ptldebug_header *header,
+ struct libcfs_debug_msg_data *msgdata,
+ unsigned long stack)
{
- struct timeval tv;
+ struct timespec64 ts;
- do_gettimeofday(&tv);
+ ktime_get_real_ts64(&ts);
- header->ph_subsys = subsys;
- header->ph_mask = mask;
+ header->ph_subsys = msgdata->msg_subsys;
+ header->ph_mask = msgdata->msg_mask;
header->ph_cpu_id = smp_processor_id();
- header->ph_sec = (__u32)tv.tv_sec;
- header->ph_usec = tv.tv_usec;
+ header->ph_type = cfs_trace_buf_idx_get();
+ /* y2038 safe since all user space treats this as unsigned, but
+ * will overflow in 2106
+ */
+ header->ph_sec = (u32)ts.tv_sec;
+ header->ph_usec = ts.tv_nsec / NSEC_PER_USEC;
header->ph_stack = stack;
header->ph_pid = current->pid;
- header->ph_line_num = line;
-#if defined(__arch_um__) && (LINUX_VERSION_CODE < KERNEL_VERSION(2,4,20))
- header->ph_extern_pid = current->thread.extern_pid;
-#elif defined(__arch_um__) && (LINUX_VERSION_CODE < KERNEL_VERSION(2,5,0))
- header->ph_extern_pid = current->thread.mode.tt.extern_pid;
-#else
+ header->ph_line_num = msgdata->msg_line;
header->ph_extern_pid = 0;
-#endif
- return;
}
-void print_to_console(struct ptldebug_header *hdr, int mask, const char *buf,
- int len, const char *file, const char *fn)
+/**
+ * tty_write_msg - write a message to a certain tty, not just the console.
+ * @tty: the destination tty_struct
+ * @msg: the message to write
+ *
+ * tty_write_message is not exported, so write a same function for it
+ *
+ */
+static void tty_write_msg(struct tty_struct *tty, const char *msg)
{
- char *prefix = "Lustre", *ptype = NULL;
-
- if ((mask & D_EMERG) != 0) {
- prefix = "LustreError";
- ptype = KERN_EMERG;
- } else if ((mask & D_ERROR) != 0) {
- prefix = "LustreError";
- ptype = KERN_ERR;
- } else if ((mask & D_WARNING) != 0) {
- prefix = "Lustre";
- ptype = KERN_WARNING;
- } else if ((mask & (D_CONSOLE | libcfs_printk)) != 0) {
- prefix = "Lustre";
- ptype = KERN_INFO;
- }
-
- if ((mask & D_CONSOLE) != 0) {
- printk("%s%s: %.*s", ptype, prefix, len, buf);
- } else {
- printk("%s%s: %d:%d:(%s:%d:%s()) %.*s", ptype, prefix, hdr->ph_pid,
- hdr->ph_extern_pid, file, hdr->ph_line_num, fn, len, buf);
- }
- return;
+ mutex_lock(&tty->atomic_write_lock);
+ tty_lock(tty);
+ if (tty->ops->write && tty->count > 0)
+ tty->ops->write(tty, msg, strlen(msg));
+ tty_unlock(tty);
+ mutex_unlock(&tty->atomic_write_lock);
+ wake_up_interruptible_poll(&tty->write_wait, POLLOUT);
}
-int trace_max_debug_mb(void)
+static void cfs_tty_write_message(const char *prefix, int mask, const char *msg)
{
- int total_mb = (num_physpages >> (20 - CFS_PAGE_SHIFT));
-
- return MAX(512, (total_mb * 80)/100);
+ struct tty_struct *tty;
+
+ tty = get_current_tty();
+ if (!tty)
+ return;
+
+ tty_write_msg(tty, prefix);
+ if ((mask & D_EMERG) || (mask & D_ERROR))
+ tty_write_msg(tty, "Error");
+ tty_write_msg(tty, ": ");
+ tty_write_msg(tty, msg);
+ tty_kref_put(tty);
}
-void
-trace_call_on_all_cpus(void (*fn)(void *arg), void *arg)
+void cfs_print_to_console(struct ptldebug_header *hdr, int mask,
+ const char *buf, int len, const char *file,
+ const char *fn)
{
- cpumask_t cpus_allowed = current->cpus_allowed;
- /* use cpus_allowed to quiet 2.4 UP kernel warning only */
- cpumask_t m = cpus_allowed;
- int cpu;
-
- /* Run the given routine on every CPU in thread context */
- for (cpu = 0; cpu < num_possible_cpus(); cpu++) {
- if (!cpu_online(cpu))
- continue;
+ char *prefix = "Lustre";
+
+ if (hdr->ph_subsys == S_LND || hdr->ph_subsys == S_LNET)
+ prefix = "LNet";
+
+ if (mask & D_CONSOLE) {
+ if (mask & D_EMERG)
+ pr_emerg("%sError: %.*s", prefix, len, buf);
+ else if (mask & D_ERROR)
+ pr_err("%sError: %.*s", prefix, len, buf);
+ else if (mask & D_WARNING)
+ pr_warn("%s: %.*s", prefix, len, buf);
+ else if (mask & libcfs_printk)
+ pr_info("%s: %.*s", prefix, len, buf);
+ } else {
+ if (mask & D_EMERG)
+ pr_emerg("%sError: %d:%d:(%s:%d:%s()) %.*s", prefix,
+ hdr->ph_pid, hdr->ph_extern_pid, file,
+ hdr->ph_line_num, fn, len, buf);
+ else if (mask & D_ERROR)
+ pr_err("%sError: %d:%d:(%s:%d:%s()) %.*s", prefix,
+ hdr->ph_pid, hdr->ph_extern_pid, file,
+ hdr->ph_line_num, fn, len, buf);
+ else if (mask & D_WARNING)
+ pr_warn("%s: %d:%d:(%s:%d:%s()) %.*s", prefix,
+ hdr->ph_pid, hdr->ph_extern_pid, file,
+ hdr->ph_line_num, fn, len, buf);
+ else if (mask & (D_CONSOLE | libcfs_printk))
+ pr_info("%s: %.*s", prefix, len, buf);
+ }
- cpus_clear(m);
- cpu_set(cpu, m);
- set_cpus_allowed(current, m);
+ if (mask & D_TTY)
+ cfs_tty_write_message(prefix, mask, buf);
+}
- fn(arg);
+int cfs_trace_max_debug_mb(void)
+{
+ int total_mb = (cfs_totalram_pages() >> (20 - PAGE_SHIFT));
- set_cpus_allowed(current, cpus_allowed);
- }
+ return max(512, (total_mb * 80) / 100);
}