It is often suggested that users set noop on SSDs and it turns out that
udev can do this for users.
Setting noop disables the IO priorization and IO reordering logic inside
the kernel, but leaves front/back merging in place. This reduction in
overhead should increase the number of requests sent to solid state
media to the maximum possible,which is said to improve performance on
SSDs. Unfortunately, few benchmarks try real world work loads with a
clear cache to measure the actual difference.
The benchmarks conducted by Daniel Nashed cleared the cache. They favor
noop, although the workload seems somewhat unrealistic:
http://blog.nashcom.de/nashcomblog.nsf/dx/linux-io-performance-tweek.htm
The BFQ developers' benchmarks on SSDs appear to account for both. They
show noop as being far better than CFQ and second only to BFQ, which is
out of tree:
https://lwn.net/Articles/600366/
In addition, I have experienced lockup-like effects on ext4 on an OCZ
Vertex 2 SSD with the discard mount option enabled when recursively
unlinking a subdirectory path that contains millions of files. The
system was useless for hours. Setting noop allowed the unlink to finish
in minutes. This is because the reordering from CFQ interleaved the
TRIM command with write IOs, effectively putting barriers between them
because because TRIM is a non-queued command prior to SATA 3.1.
A good default should perform well in general and have the property that
poor performance in the worst case scenarios is minimized. The
previous examples contradict CFQ's ability to achieve that on solid
state media.
I believe that we should implement a udev rule to set noop on solid
state media by default. It should be said that Milan Broz wrote it
first, although there is only one way to write this rule in a manner
consistent with the codebase:
http://permalink.gmane.org/gmane.linux.kernel.device-mapper.dm-crypt/6045
It should be said that this will be a regression for those that rely on
the "Block IO Controller" cgroup because it is only supported by CFQ
when CONFIG_CFQ_GROUP_IOSCHED=y. My experience as a ZoL developer is
that very few users rely on this behavior and consequently, I believe
that the benefit from enabling this far outweighs the harm to the few
that need it. Those that do need it should be able to disable this rule
themselves. Container management software that expects the Block IO
Controller to be supported should be modified to enable CFQ explicitly
if it does not already do that.
This has been tested against both a SATA mechanical drive and a SATA
solid state drive. It changes the elevator to noop on the solid state
drive, but does not touch it on the mechanical drive.
Signed-off-by: Richard Yao <ryao@gentoo.org>
mallinfo is not specified by POSIX or the C standards, therefore
it's not available for all libc libraries (musl).
Add the ability to disable mallinfo statistics.
Fixes:
selinux-util.c: In function ‘mac_selinux_init’:
selinux-util.c:70:25: error: storage size of ‘before_mallinfo’ isn’t known
struct mallinfo before_mallinfo, after_mallinfo;
Signed-off-by: Romain Naour <romain.naour@openwide.fr>
This allows a fd to be created and configured as part of one monitor, to be passed in
to create a second monitor without having to redo any of the configuration.
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
Unplugging and plugging in the cable will create various scancodes
on the keyboard controller.
Userspace within X should be able to interact with these to show
interesting messages. Assign them to generic prog1/prog2.
(David: add comment to hwdb explaining that these keycodes are reserved)
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
Since 3.19, the devices have the proper vid/pid and the model number in the
name.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
The main purpose of this hwdb was to tag touchpads that have the physical
trackstick buttons wired to the touchpad (Lenovo Carbon X1 3rd, Lenovo *50
series). This hwdb is not required on kernels 4.0 and above, the kernel now
re-routes button presses through the trackstick's device node. Userspace does
not need to do anything.
See kernel commit cdd9dc195916ef5644cfac079094c3c1d1616e4c.
This reverts commit 001a247324b44c0e0b8fdba41a6fc66e7465b8b6.
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
It is not udev's task to apply any of these setting that way, or
from udev rules files. Things need to be sortet out in the kernel,
or explicit whitelist can possibly be added to the hardware database.
Until that is sorted out, and general agreement, udev is not
willing to maintain any such lists or power management settings
in general.
"Thanks for digging this out! I thought my Kinesis keyboard got broken
and ordered a new one, only to find out that the new one doesn't work
as well. I'm not sure whether we should start collecting a blacklist
of keyboards which don't work with USB autosuspend, or rather a
whitelist? Or revert this wholesale?"
https://github.com/systemd/systemd/issues/340
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
The original commit (1aff206) doesn't explain why these were removed.
This adds them back since they are in fact needed.
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
When processing an event, the watch is disabled, make sure it is restorted after
a CHANGE event has been processed.
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
The intention was to turn this rule from using a blacklist to a whitelist, but
there was a stray '!'.
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
While usptream doesn't support a static libudev.a, we will try to
do so. However, mkdir_p() is used in lvm2 and util-linux, so
to avoid the collision, we rename it to udev_mkdir_p(). See:
https://bugs.gentoo.org/show_bug.cgi?id=520450
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
This reverts commit b2399d9b72.
This solves issue #108. While upstream also reverted this commit,
they did so using functions in terminal-util.c. We could import
that file and those functions but for such a small commit, its not
worth it. We may do so at some future time if there are further
gains. See:
40e749b59b
Signed-off-by: Anthony G. Basile <blueness@gentoo.org>
