Re: [PATCH] tracer for sys_open() - sreadahead
From: Frederic Weisbecker
Date: Tue Jan 27 2009 - 17:43:40 EST
On Tue, Jan 27, 2009 at 12:08:04PM -0800, Kok, Auke wrote:
>
> This tracer monitors regular file open() syscalls. This is a fast
> and low-overhead alternative to strace, and does not allow or
> require to be attached to every process.
>
> The tracer only logs succesfull calls, as those are the only ones we
> are currently interested in, and we can determine the absolute path
> of these files as we log.
>
> Signed-off-by: Auke Kok <auke-jan.h.kok@xxxxxxxxx>
Hi Auke,
Speaking about a global syscall tracer, I made a patch to trace only the syscalls
with the function-graph-tracer.
http://lkml.org/lkml/2008/12/30/267
Its approach and purpose is different than a tracer dedicated only to syscalls.
The function graph tracer traces execution graph of the functions and is more about
execution time spent and code flow whereas a syscall tracer can provide more specific
informations about syscalls.
So both are not overlaping.
But the low level part of my patch creates a thread flag _TIF_SYSCALL_TRACE which triggers
a ptrace hook when set.
This low-level part can easily be used by all tracers that would like to inspect syscalls.
Just a change is needed: Steven requested that the part inside syscall_trace_enter become
a tracepoint, making it totally shareable between tracers and easy to turn on and off.
And perhaps the parts that set/clear the flag on all tasks can be shared too.
So we can start with this low-level syscall tracing facility. If you want, I can adapt
this low-level part and submit a patch this week or the next one to give you this base
infrastructure.
Once we have it, I think a syscall tracer can be fed with new syscalls events through
several patch iterations, starting with the open and close one :-)
Are you ok with that?
Steven, Ingo, do you agree?
>
> diff --git a/fs/open.c b/fs/open.c
> index a3a78ce..8cf2a6b 100644
> --- a/fs/open.c
> +++ b/fs/open.c
> @@ -30,6 +30,10 @@
> #include <linux/audit.h>
> #include <linux/falloc.h>
>
> +#include <trace/fs.h>
> +
> +DEFINE_TRACE(do_sys_open);
> +
> int vfs_statfs(struct dentry *dentry, struct kstatfs *buf)
> {
> int retval = -ENODEV;
> @@ -1040,6 +1044,7 @@ long do_sys_open(int dfd, const char __user *filename, int
> flags, int mode)
> fsnotify_open(f->f_path.dentry);
> fd_install(fd, f);
> }
> + trace_do_sys_open(f, flags, mode, fd);
> }
> putname(tmp);
> }
> diff --git a/include/trace/fs.h b/include/trace/fs.h
> new file mode 100644
> index 0000000..870eec2
> --- /dev/null
> +++ b/include/trace/fs.h
> @@ -0,0 +1,11 @@
> +#ifndef _TRACE_FS_H
> +#define _TRACE_FS_H
> +
> +#include <linux/fs.h>
> +#include <linux/tracepoint.h>
> +
> +DECLARE_TRACE(do_sys_open,
> + TPPROTO(struct file *filp, int flags, int mode, long fd),
> + TPARGS(filp, flags, mode, fd));
> +
> +#endif
> diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
> index e2a4ff6..0400815 100644
> --- a/kernel/trace/Kconfig
> +++ b/kernel/trace/Kconfig
> @@ -149,6 +149,15 @@ config CONTEXT_SWITCH_TRACER
> This tracer gets called from the context switch and records
> all switching of tasks.
>
> +config OPEN_CLOSE_TRACER
> + bool "Trace open() calls"
> + depends on DEBUG_KERNEL
> + select TRACING
> + select MARKERS
> + help
> + This tracer records open() syscalls. These calls are made when
> + files are accessed on disk.
> +
> config BOOT_TRACER
> bool "Trace boot initcalls"
> depends on DEBUG_KERNEL
> diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile
> index 349d5a9..25cec6c 100644
> --- a/kernel/trace/Makefile
> +++ b/kernel/trace/Makefile
> @@ -20,6 +20,7 @@ obj-$(CONFIG_RING_BUFFER) += ring_buffer.o
>
> obj-$(CONFIG_TRACING) += trace.o
> obj-$(CONFIG_CONTEXT_SWITCH_TRACER) += trace_sched_switch.o
> +obj-$(CONFIG_OPEN_CLOSE_TRACER) += trace_open_close.o
> obj-$(CONFIG_SYSPROF_TRACER) += trace_sysprof.o
> obj-$(CONFIG_FUNCTION_TRACER) += trace_functions.o
> obj-$(CONFIG_IRQSOFF_TRACER) += trace_irqsoff.o
> diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
> index 4d3d381..24c17d2 100644
> --- a/kernel/trace/trace.h
> +++ b/kernel/trace/trace.h
> @@ -30,6 +30,7 @@ enum trace_type {
> TRACE_USER_STACK,
> TRACE_HW_BRANCHES,
> TRACE_POWER,
> + TRACE_OPEN,
>
> __TRACE_LAST_TYPE
> };
> diff --git a/kernel/trace/trace_open_close.c b/kernel/trace/trace_open_close.c
> new file mode 100644
> index 0000000..4250efc
> --- /dev/null
> +++ b/kernel/trace/trace_open_close.c
> @@ -0,0 +1,148 @@
> +/*
> + * trace open calls
> + * Copyright (C) 2009 Intel Corporation
> + *
> + * Based extensively on trace_sched_switch.c
> + * Copyright (C) 2007 Steven Rostedt <srostedt@xxxxxxxxxx>
> + *
> + */
> +
> +#include <linux/module.h>
> +#include <linux/fs.h>
> +#include <linux/debugfs.h>
> +#include <linux/kallsyms.h>
> +#include <linux/uaccess.h>
> +#include <linux/ftrace.h>
> +#include <trace/fs.h>
> +
> +#include "trace.h"
> +
> +
> +static struct trace_array *ctx_trace;
> +static int __read_mostly open_trace_enabled;
> +static atomic_t open_ref;
> +
> +static void probe_do_sys_open(struct file *filp, int flags, int mode, long fd)
> +{
> + char *buf;
> + char *fname;
> +
> + if (!atomic_read(&open_ref))
> + return;
> +
> + if (!open_trace_enabled)
> + return;
> +
> + buf = kzalloc(PAGE_SIZE, GFP_KERNEL);
> + if (!buf)
> + return;
> + fname = d_path(&filp->f_path, buf, PAGE_SIZE);
> +
> + if (IS_ERR(fname))
> + goto out;
> +
> + ftrace_printk("%s: open(\"%s\", %d, %d) = %ld\n",
> + current->comm, fname, flags, mode, fd);
> +out:
> + kfree(buf);
> +}
> +
> +static void open_trace_reset(struct trace_array *tr)
> +{
> + tr->time_start = ftrace_now(tr->cpu);
> + tracing_reset_online_cpus(tr);
> +}
> +
> +static int open_trace_register(void)
> +{
> + int ret;
> +
> + ret = register_trace_do_sys_open(probe_do_sys_open);
> + if (ret) {
> + pr_info("open trace: Could not activate tracepoint"
> + " probe to do_open\n");
> + }
> +
> + return ret;
> +}
> +
> +static void open_trace_unregister(void)
> +{
> + unregister_trace_do_sys_open(probe_do_sys_open);
> +}
> +
> +static void open_trace_start(void)
> +{
> + long ref;
> +
> + ref = atomic_inc_return(&open_ref);
> + if (ref == 1)
> + open_trace_register();
> +}
> +
> +static void open_trace_stop(void)
> +{
> + long ref;
> +
> + ref = atomic_dec_and_test(&open_ref);
> + if (ref)
> + open_trace_unregister();
> +}
> +
> +void open_trace_start_cmdline_record(void)
> +{
> + open_trace_start();
> +}
> +
> +void open_trace_stop_cmdline_record(void)
> +{
> + open_trace_stop();
> +}
> +
> +static void open_start_trace(struct trace_array *tr)
> +{
> + open_trace_reset(tr);
> + open_trace_start_cmdline_record();
> + open_trace_enabled = 1;
> +}
> +
> +static void open_stop_trace(struct trace_array *tr)
> +{
> + open_trace_enabled = 0;
> + open_trace_stop_cmdline_record();
> +}
> +
> +static int open_trace_init(struct trace_array *tr)
> +{
> + ctx_trace = tr;
> +
> + open_start_trace(tr);
> + return 0;
> +}
> +
> +static void reset_open_trace(struct trace_array *tr)
> +{
> + open_stop_trace(tr);
> +}
> +
> +static struct tracer open_trace __read_mostly =
> +{
> + .name = "open",
> + .init = open_trace_init,
> + .reset = reset_open_trace,
> +};
> +
> +__init static int init_open_trace(void)
> +{
> + int ret = 0;
> +
> + if (atomic_read(&open_ref))
> + ret = open_trace_register();
> + if (ret) {
> + pr_info("error registering open trace\n");
> + return ret;
> + }
> + return register_tracer(&open_trace);
> +}
> +device_initcall(init_open_trace);
> +
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/