XFS and UBIFS can pass long lists to list_sort(); this alternative
implementation scales better, reaching ~3x performance gain when list
length exceeds the L2 cache size.
Stand-alone program timings were run on a Core 2 duo L1=32KB L2=4MB,
gcc-4.4, with flags extracted from an Ubuntu kernel build. Object size is
581 bytes compared to 455 for Mark J. Roberts' code.
Worst case for either implementation is a list length just over a power of
two, and to roughly the same degree, so here are timing results for a
range of 2^N+1 lengths. List elements were 16 bytes each including malloc
overhead; initial order was random.
time (msec)
Tatham-Roberts
| generic-Mullis-v2
loop_count length | | ratio
4000000 2 206 294 1.427
2000000 3 176 227 1.289
1000000 5 199 172 0.864
500000 9 235 178 0.757
250000 17 243 182 0.748
125000 33 261 196 0.750
62500 65 277 209 0.754
31250 129 292 219 0.75
15625 257 317 235 0.741
7812 513 340 252 0.741
3906 1025 362 267 0.737
1953 2049 388 283 0.729 ~ L1 size
976 4097 556 323 0.580
488 8193 678 361 0.532
244 16385 773 395 0.510
122 32769 844 418 0.495
61 65537 917 454 0.495
30 131073 1128 543 0.481
15 262145 2355 869 0.369 ~ L2 size
7 524289 5597 1714 0.306
3 1048577 6218 2022 0.325
Mark's code does not actually implement the usual or generic mergesort,
but rather a variant from Simon Tatham described here:
http://www.chiark.greenend.org.uk/~sgtatham/algorithms/listsort.html
Simon's algorithm performs O(log N) passes over the entire input list,
doing merges of sublists that double in size on each pass. The generic
algorithm instead merges pairs of equal length lists as early as possible,
in recursive order. For either algorithm, the elements that extend the
list beyond power-of-two length are a special case, handled as nearly as
possible as a "rounding-up" to a full POT.
Some intuition for the locality of reference implications of merge order
may be gotten by watching this animation:
http://www.sorting-algorithms.com/merge-sort
Simon's algorithm requires only O(1) extra space rather than the generic
algorithm's O(log N), but in my non-recursive implementation the actual
O(log N) data is merely a vector of ~20 pointers, which I've put on the
stack.
Long-running list_sort() calls: If the list passed in may be long, or the
client's cmp() callback function is slow, the client's cmp() may
periodically invoke cond_resched() to voluntarily yield the CPU. All
inner loops of list_sort() call back to cmp().
Stability of the sort: distinct elements that compare equal emerge from
the sort in the same order as with Mark's code, for simple test cases. A
boot-time test is provided to verify this and other correctness
requirements.
A kernel that uses drm.ko appears to run normally with this change; I have
no suitable hardware to similarly test the use by UBIFS.
[akpm@linux-foundation.org: style tweaks, fix comment, make list_sort_test __init]
Signed-off-by: Don Mullis <don.mullis@gmail.com>
Cc: Dave Airlie <airlied@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Artem Bityutskiy <dedekind@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Patchwork queues show the acceptance/rejection state of submitted patches
for various MAINTAINER trees. Document their existence.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If MAINTAINERS section entries are misformatted, it was possible to have
an infinite loop.
Correct the defect by always moving the index to the end of section + 1
Also, exit check for exclude as soon as possible.
Signed-off-by: Joe Perches <joe@perches.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
perlcritic is a standard checker for Perl Best Practices. This patch
fixes most of the warnings in the get_maintainer script. If kernel
programmers are going to have checkpatch they should write clean scripts
as well...
Bareword file handle opened at line 176, column 1. See pages 202,204 of PBP. (Severity: 5)
Two-argument "open" used at line 176, column 1. See page 207 of PBP. (Severity: 5)
Bareword file handle opened at line 207, column 5. See pages 202,204 of PBP. (Severity: 5)
Two-argument "open" used at line 207, column 5. See page 207 of PBP. (Severity: 5)
Bareword file handle opened at line 246, column 6. See pages 202,204 of PBP. (Severity: 5)
Two-argument "open" used at line 246, column 6. See page 207 of PBP. (Severity: 5)
Bareword file handle opened at line 258, column 2. See pages 202,204 of PBP. (Severity: 5)
Two-argument "open" used at line 258, column 2. See page 207 of PBP. (Severity: 5)
Expression form of "eval" at line 983, column 17. See page 161 of PBP. (Severity: 5)
Expression form of "eval" at line 985, column 17. See page 161 of PBP. (Severity: 5)
Subroutine prototypes used at line 1186, column 1. See page 194 of PBP. (Severity: 5)
Subroutine prototypes used at line 1206, column 1. See page 194 of PBP. (Severity: 5)
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: Joe Perches <joe@perches.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Doesn't need or accept '-' as a trailing option to read stdin. Doesn't
print usage() after bad options. Adds --usage as command line equivalent
of --help
Suggested-by: Borislav Petkov <petkovbb@googlemail.com>
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Add an imperfect option to search a source file for email addresses.
New option: --file-emails or --fe
email addresses in files are freeform text and are nearly impossible to
parse. Still, might as well try to do a somewhat acceptable job of
finding them. This code should find all addresses that are in the form
addr@domain.tld
The code assumes that up to 3 alphabetic words along with dashes, commas,
and periods that preceed the email address are a name.
If 3 words are found for the name, and one of the first two words are a
single letter and period, or just a single letter then the 3 words are use
as name otherwise the last 2 words are used.
Some variants that are shown correctly:
John Smith <jksmith@domain.org>
Random J. Developer <rjd@tld.com>
Random J. Developer (rjd@tld.com)
J. Random Developer rjd@tld.com
Variants that are shown nominally correctly:
Written by First Last (funny-addr@somecompany.com)
is shown as:
First Last <funny-addr@somecompany.com>
Variants that are shown incorrectly:
Some Really Long Name <srln@foo.bar>
MontaVista Software, Inc. <source@mvista.com>
are returned as:
Long Name <srln@foo.bar>
"Software, Inc" <source@mvista.com>
--roles and --rolestats show "(in file)" for matches.
For instance:
Without -file-emails:
$ ./scripts/get_maintainer.pl -f -nogit -roles net/core/netpoll.c
David S. Miller <davem@davemloft.net> (maintainer:NETWORKING [GENERAL])
linux-kernel@vger.kernel.org (open list)
With -fe:
$ ./scripts/get_maintainer.pl -f -fe -nogit -roles net/core/netpoll.c
David S. Miller <davem@davemloft.net> (maintainer:NETWORKING [GENERAL])
Matt Mackall <mpm@selenic.com> (in file)
Ingo Molnar <mingo@redhat.com> (in file)
linux-kernel@vger.kernel.org (open list)
netdev@vger.kernel.org (open list:NETWORKING [GENERAL])
The number of email addresses in the file in not limited. Neither is the
number of returned email addresses.
Signed-off-by: Joe Perches <joe@perches.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
tasklist_lock does protect the task and its pid, it can't go away. The
problem is that find_pid_ns() itself is unsafe without rcu lock, it can
race with copy_process()->free_pid(any_pid).
Protecting copy_process()->free_pid(any_pid) with tasklist_lock would make
it possible to call find_task_by_pid_ns() under tasklist safely, but we
don't do so because we are trying to get rid of the read_lock sites of
tasklist_lock.
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
I've had some complaints about panic_timeout being wildly innacurate on
shared processor PowerPC partitions (a 3 minute panic_timeout taking 30
minutes).
The problem is we loop on mdelay(1) and with a 1ms in 10ms hypervisor
timeslice each of these will take 10ms (ie 10x) longer. I expect other
platforms with shared processor hypervisors will see the same issue.
This patch keeps the old behaviour if we have a panic_blink (only keyboard
LEDs right now) and does 1 second mdelays if we don't.
Signed-off-by: Anton Blanchard <anton@samba.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Make sure compiler won't do weird things with limits. E.g. fetching them
twice may return 2 different values after writable limits are implemented.
I.e. either use rlimit helpers added in commit 3e10e716ab ("resource:
add helpers for fetching rlimits") or ACCESS_ONCE if not applicable.
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: john stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Don Mullis
André Goddard Rosa
Joe Perches
Stephen Hemminger
Gustavo F. Padovan
Michael Neuling
Tetsuo Handa
Anton Blanchard
Jiri Slaby
Thiago Farina
Jaswinder Singh Rajput