5 files changed, 97 insertions, 0 deletions

diff --git a/include/linux/sched.h b/include/linux/sched.h
index ad9c46071ff8..1208feab46e0 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -82,6 +82,7 @@ struct sched_param {
 #include <linux/resource.h>
 #include <linux/timer.h>
 #include <linux/hrtimer.h>
+#include <linux/task_io_accounting.h>
 
 #include <asm/processor.h>
 
@@ -1013,6 +1014,7 @@ struct task_struct {
         wait_queue_t *io_wait;
 /* i/o counters(bytes read/written, #syscalls */
         u64 rchar, wchar, syscr, syscw;
+        struct task_io_accounting ioac;
 #if defined(CONFIG_TASK_XACCT)
         u64 acct_rss_mem1;      /* accumulated rss usage */
         u64 acct_vm_mem1;       /* accumulated virtual memory usage */
diff --git a/include/linux/task_io_accounting.h b/include/linux/task_io_accounting.h
new file mode 100644
index 000000000000..44d00e9cceea
--- /dev/null
+++ b/include/linux/task_io_accounting.h
@@ -0,0 +1,37 @@
+/*
+ * task_io_accounting: a structure which is used for recording a single task's
+ * IO statistics.
+ *
+ * Don't include this header file directly - it is designed to be dragged in via
+ * sched.h.
+ *
+ * Blame akpm@osdl.org for all this.
+ */
+
+#ifdef CONFIG_TASK_IO_ACCOUNTING
+struct task_io_accounting {
+        /*
+         * The number of bytes which this task has caused to be read from
+         * storage.
+         */
+        u64 read_bytes;
+
+        /*
+         * The number of bytes which this task has caused, or shall cause to be
+         * written to disk.
+         */
+        u64 write_bytes;
+
+        /*
+         * A task can cause "negative" IO too.  If this task truncates some
+         * dirty pagecache, some IO which another task has been accounted for
+         * (in its write_bytes) will not be happening.  We _could_ just
+         * subtract that from the truncating task's write_bytes, but there is
+         * information loss in doing that.
+         */
+        u64 cancelled_write_bytes;
+};
+#else
+struct task_io_accounting {
+};
+#endif
diff --git a/include/linux/task_io_accounting_ops.h b/include/linux/task_io_accounting_ops.h
new file mode 100644
index 000000000000..df2a319106b2
--- /dev/null
+++ b/include/linux/task_io_accounting_ops.h
@@ -0,0 +1,47 @@
+/*
+ * Task I/O accounting operations
+ */
+#ifndef __TASK_IO_ACCOUNTING_OPS_INCLUDED
+#define __TASK_IO_ACCOUNTING_OPS_INCLUDED
+
+#ifdef CONFIG_TASK_IO_ACCOUNTING
+static inline void task_io_account_read(size_t bytes)
+{
+        current->ioac.read_bytes += bytes;
+}
+
+static inline void task_io_account_write(size_t bytes)
+{
+        current->ioac.write_bytes += bytes;
+}
+
+static inline void task_io_account_cancelled_write(size_t bytes)
+{
+        current->ioac.cancelled_write_bytes += bytes;
+}
+
+static inline void task_io_accounting_init(struct task_struct *tsk)
+{
+        memset(&tsk->ioac, 0, sizeof(tsk->ioac));
+}
+
+#else
+
+static inline void task_io_account_read(size_t bytes)
+{
+}
+
+static inline void task_io_account_write(size_t bytes)
+{
+}
+
+static inline void task_io_account_cancelled_write(size_t bytes)
+{
+}
+
+static inline void task_io_accounting_init(struct task_struct *tsk)
+{
+}
+
+#endif          /* CONFIG_TASK_IO_ACCOUNTING */
+#endif          /* __TASK_IO_ACCOUNTING_OPS_INCLUDED */
diff --git a/init/Kconfig b/init/Kconfig
index 14d484606fab..9edf103b3ec3 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -304,6 +304,15 @@ config TASK_XACCT
 
           Say N if unsure.
 
+config TASK_IO_ACCOUNTING
+        bool "Enable per-task storage I/O accounting (EXPERIMENTAL)"
+        depends on TASK_XACCT
+        help
+          Collect information on the number of bytes of storage I/O which this
+          task has caused.
+
+          Say N if unsure.
+
 config SYSCTL
         bool
 
diff --git a/kernel/fork.c b/kernel/fork.c
index 8c859eef8e6a..086e172d0d3d 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -36,6 +36,7 @@
 #include <linux/syscalls.h>
 #include <linux/jiffies.h>
 #include <linux/futex.h>
+#include <linux/task_io_accounting_ops.h>
 #include <linux/rcupdate.h>
 #include <linux/ptrace.h>
 #include <linux/mount.h>
@@ -1055,6 +1056,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
         p->wchar = 0;           /* I/O counter: bytes written */
         p->syscr = 0;           /* I/O counter: read syscalls */
         p->syscw = 0;           /* I/O counter: write syscalls */
+        task_io_accounting_init(p);
         acct_clear_integrals(p);
 
         p->it_virt_expires = cputime_zero;