It looks more reasonable to save inode->dirtied_when in the raw trace
output, and to show both the absolute raw value for easy scripting and
the relative age in seconds for easy human reading.
The trace event was newly added in 3.1-rc1, so we still have the chance
to change the format.
btw, are there ways for user space to query the HZ value of the
running kernel? Which is required for converting the raw dirtied_when
ticks to clock time.
CC: Jan Kara <jack@xxxxxxx>
Signed-off-by: Wu Fengguang <fengguang.wu@xxxxxxxxx>
---
include/trace/events/writeback.h | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)
--- linux-next.orig/include/trace/events/writeback.h 2011-08-29 09:51:12.000000000 +0800
+++ linux-next/include/trace/events/writeback.h 2011-08-29 09:52:19.000000000 +0800
@@ -433,7 +433,7 @@ DECLARE_EVENT_CLASS(writeback_single_ino
__array(char, name, 32)
__field(unsigned long, ino)
__field(unsigned long, state)
- __field(unsigned long, age)
+ __field(unsigned long, dirtied_when)
__field(unsigned long, writeback_index)
__field(long, nr_to_write)
__field(unsigned long, wrote)
@@ -444,19 +444,19 @@ DECLARE_EVENT_CLASS(writeback_single_ino
dev_name(inode->i_mapping->backing_dev_info->dev), 32);
__entry->ino = inode->i_ino;
__entry->state = inode->i_state;
- __entry->age = (jiffies - inode->dirtied_when) *
- 1000 / HZ;
+ __entry->age = (jiffies - inode->dirtied_when) / HZ;
__entry->writeback_index = inode->i_mapping->writeback_index;
__entry->nr_to_write = nr_to_write;
__entry->wrote = nr_to_write - wbc->nr_to_write;
),
- TP_printk("bdi %s: ino=%lu state=%s age=%lu "
+ TP_printk("bdi %s: ino=%lu state=%s dirtied_when=%lu age=%lu "
"index=%lu to_write=%ld wrote=%lu",
__entry->name,
__entry->ino,
show_inode_state(__entry->state),
- __entry->age,
+ __entry->dirtied_when,
+ (jiffies - __entry->dirtied_when) / HZ,
__entry->writeback_index,
__entry->nr_to_write,
__entry->wrote
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
