Hi Greg, hi Kay, and all other interested people.
This series aims at cleaning up and fixing some bugs around printk().
Patches 9 and 11 might require some discussion, see below.
(This is not to be confused with the printk.c reorganization that Joe
Perches currently does [1]. This series is really focused around printk().)
Let's get straight to the individual patches (series is currently against
v3.7-rc5, but applies to next-20121113 as well):
1. printk: drop ambiguous LOG_CONT flag
This is the only patch that was previously posted [2]. Compared to
the previous version, a small bug got fixed. With this patch,
independent continuation records are no longer incorrectly merged.
2. printk: use saved timestamp for temporarily buffered message
3. printk: reclaim cont buffer immediately for fully printed messages
4. printk: reuse reclaimed continuation buffer immediately
5. printk: move wake_klogd-check out of the loop
These four patches fall into the optimization category. The only
measurable effect is probably that continuation records are more often
merged than before (patches 3+4).
6. printk: reorganize storage of continuation buffer
7. printk: let cont_print_text() reuse existing code
These two reduce replicated code within printk.c and fix an actual bug
in the console output of logged messages.
8. printk: refactor locking in console_unlock()
Another bugfix. And due to the refactored locking, I hopefully closed
all the smaller time-windows where the same race could have happened.
(This patch does not fix all issues within console_unlock() that I am
aware of, see patch description.)
9. printk: avoid glitches in console output
This one has a notable effect on the console output when there are
concurrent printk()s. Personally, I think it improves the readability,
but there could be a reason I'm not aware of why such a change is
not acceptable. Please discuss.
10. printk: retain unknown log-level until log_store()
11. printk: track previously logged message in log_store()
12. printk: drop now useless tracking of previous message flags
The change due to these three patches is only noticeable for a user in
a few corner cases. OTOH, it improves source readability by removing
some of the historical growth (patch 12) that only makes sense when
you have actually read the history.
Regards
Jan
[1] http://lkml.org/lkml/2012/10/24/748
[2] http://lkml.org/lkml/2012/9/26/392
Jan H. Schönherr (12):
printk: drop ambiguous LOG_CONT flag
printk: use saved timestamp for temporarily buffered message
printk: reclaim cont buffer immediately for fully printed messages
printk: reuse reclaimed continuation buffer immediately
printk: move wake_klogd-check out of the loop
printk: reorganize storage of continuation buffer
printk: let cont_print_text() reuse existing code
printk: refactor locking in console_unlock()
printk: avoid glitches in console output
printk: retain unknown log-level until log_store()
printk: track previously logged message in log_store()
printk: drop now useless tracking of previous message flags
kernel/printk.c | 368 ++++++++++++++++++++++++++++++++------------------------
1 file changed, 209 insertions(+), 159 deletions(-)
--
1.8.0.316.g291341c.dirty
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
