| Age | Commit message (Collapse) | Author |
|---|
|
Remove an avoidable 3ms delay on some dm-raid1 and kcopyd I/O.
It is specified that any submitted bio without BIO_RW_SYNC flag may plug the
queue (i.e. block the requests from being dispatched to the physical device).
The queue is unplugged when the caller calls blk_unplug() function. Usually, the
sequence is that someone calls submit_bh to submit IO on a buffer. The IO plugs
the queue and waits (to be possibly joined with other adjacent bios). Then, when
the caller calls wait_on_buffer(), it unplugs the queue and submits the IOs to
the disk.
This was happenning:
When doing O_SYNC writes, function fsync_buffers_list() submits a list of
bios to dm_raid1, the bios are added to dm_raid1 write queue and kmirrord is
woken up.
fsync_buffers_list() calls wait_on_buffer(). That unplugs the queue, but
there are no bios on the device queue as they are still in the dm_raid1 queue.
wait_on_buffer() starts waiting until the IO is finished.
kmirrord is scheduled, kmirrord takes bios and submits them to the devices.
The submitted bio plugs the harddisk queue but there is no one to unplug it.
(The process that called wait_on_buffer() is already sleeping.)
So there is a 3ms timeout, after which the queues on the harddisks are
unplugged and requests are processed.
This 3ms timeout meant that in certain workloads (e.g. O_SYNC, 8kb writes),
dm-raid1 is 10 times slower than md raid1.
Every time we submit something asynchronously via dm_io, we must unplug the
queue actually to send the request to the device.
This patch adds an unplug call to kmirrord - while processing requests, it keeps
the queue plugged (so that adjacent bios can be merged); when it finishes
processing all the bios, it unplugs the queue to submit the bios.
It also fixes kcopyd which has the same potential problem. All kcopyd requests
are submitted with BIO_RW_SYNC.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Acked-by: Jens Axboe <jens.axboe@oracle.com>
|
|
This patch replaces the schedule() in the main kmirrord thread with a timer.
The schedule() could introduce an unwanted delay when work is ready to be
processed.
The code instead calls wake() when there's work to be done immediately, and
delayed_wake() after a failure to give a short delay before retrying.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
|
|
Publish the dm-io, dm-log and dm-kcopyd headers in include/linux.
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
|
|
Rename kcopyd.[ch] to dm-kcopyd.[ch].
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
|
|
Make dm.h macros and inlines available in include/linux/device-mapper.h
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
|
|
Remove client counting code that is no longer needed.
Initialization and destruction is made globally from dm_init and dm_exit and is
not based on client counts. Initialization allocates only one empty slab cache,
so there is no negative impact from performing the initialization always,
regardless of whether some client uses kcopyd or not.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
|
|
Change the global mempool in kcopyd into a per-device mempool to avoid
deadlock possibilities.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
|
|
Make one kcopyd thread per device.
The original shared kcopyd could deadlock.
Configuration:
|
|
Remove internal module reference fields from the interface.
Signed-off-by: Jonathan Brassow <jbrassow@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
|
|
Reorder a couple of functions in the file so the next patch is readable.
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
|
|
Clean up the dm-log interface to prepare for publishing it in include/linux.
Signed-off-by: Heinz Mauelshagen <hjm@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
|
|
Clean up the kcopyd interface to prepare for publishing it in include/linux.
Signed-off-by: Heinz Mauelshagen <hjm@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
|
|
Clean up the dm-io interface to prepare for publishing it in include/linux.
Signed-off-by: Heinz Mauelshagen <hjm@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
|
|
Rename 'error' to 'error_bits' for clarity.
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
|
|
Save pointer to dm_target in dm_snapshot structure.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
|
|
Move the dirty region log code into a separate module so
other targets can share the code.
Signed-off-by: Heinz Mauelshagen <hjm@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
|
|
Change dm-log.c messages from "mirror log" to "dirty region log" as
a new dm target wants to share this code.
Signed-off-by: Heinz Mauelshagen <hjm@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
|
|
Use shorter list_splice_init() for brevity.
Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
|
|
Limit the amount of memory allocated per snapshot on systems
with a large page size. (The larger default chunk size on
these systems compensates for the smaller number of pages reserved.)
Signed-off-by: Milan Broz <mbroz@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
|
|
If a snapshot has a smaller chunksize than the page size the
conversion to pages currently returns 0 instead of 1, causing:
kernel BUG in mempool_resize.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Milan Broz <mbroz@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Cc: stable@kernel.org
|
|
Otherwise all sorts of bad things can happen, including
spurious softlockup reports.
Other platforms have this same bug, in one form or
another, just don't see the issue because they
don't sleep as long as sparc64 can in NOHZ.
Thanks to some brilliant debugging by Peter Zijlstra.
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Second and third arguments were swapped for whatever reason.
Reported by Tom Callaway.
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
I found some places, that erroneously return the value obtained from
the copy_to_user() call: if some amount of bytes were not able to get
to the user (this is what this one returns) the proper behavior is to
return the -EFAULT error, not that number itself.
Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
|
|
Signed-off-by: Bryan Wu <cooloney@kernel.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
|
|
http://blackfin.uclinux.org/gf/project/uclinux-dist/tracker/?action=TrackerItemEdit&tracker_item_id=4053
Singed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
|
|
This driver did not control NCE signal during normal operations (only
enable NCE on probing and disable NCE on removing). This patch make
NCE signal inactive on idle state.
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
|
|
There is no guarantee that there is physical ram below 4GB, and in
fact many boxes don't have exactly that.
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
|
The sam926x docs allegedly don't list an "ECC_PARITY" field, and the
header files in the upstream kernel don't have it either.
Masking with it was useless anyway, so just remove it.
Signed-off-by: Richard Genoud <richard.genoud@gmail.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
|
|
master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6
|
|
In commit ba749ae98d5aa9d2ce9a7facde0deed454f92230 ([XFRM]: alg_key_len
should be unsigned to avoid integer divides
<http://git2.kernel.org/?p=linux/kernel/git/davem/net-2.6.git;a=commitdiff;h=ba749ae98d5aa9d2ce9a7facde0deed454f92230>)
alg_key_len field of struct xfrm_algo was converted to unsigned int to
avoid integer divides.
Then Herbert in commit 1a6509d991225ad210de54c63314fd9542922095
([IPSEC]: Add support for combined mode algorithms) added a new
structure xfrm_algo_aead, that resurrected a signed int for alg_key_len
and re-introduce integer divides.
This patch avoids these divides and saves 64 bytes of text on i386.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Noticed by davem.
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
|
|
Two is used in the wrong context here, as you are connecting to an
IPv6 network over IPv4; not connecting two IPv6 networks to an IPv4
one.
Signed-off-by: Michael Beasley <youvegotmoxie@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
- vlans were using a single CAM register (see mac_set_vlan_cam)
- setting the address filtering registers for vlans is not
needed when there is no vlan
The non-tagged interface is filtered out as soon as a tagged
(!= 0) interface is created. Its traffic appears again when an
zero-tagged interface is created.
Tested on Via Epia SN (VT6130 chipset) with several vlans whose
tag was above or beyond 255.
Signed-off-by: Séguier Régis <rseguier@e-teleport.net>
Acked-by: Francois Romieu <romieu@fr.zoreil.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
|
|
Fix the following sparse warning :
drivers/net/tg3.c:4025:3: warning: context imbalance in 'tg3_restart_hw'
- unexpected unlock
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
This patch fixes an issue seen with the realtek 8201 phy. This phy has a
problem with crossover detection and it needs to be disabled. The
problem only arises on certain switches. Therefore, a module parameter
has been added to allow enabling crossover detection if needed. The
default will be set to disabled.
Signed-off-by: Ayaz Abdulla <aabdulla@nvidia.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
|
|
This patch doubles the MDIO timeouts in EMAC as there are field
cases where they are two short to communicate with some PHYs.
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
|
|
gianfar was unable to handle failed skb allocation for rx buffers, so
we were spinning until it succeeded. Actually, it was worse--we were
spinning for a long time, and then silently failing. Instead, we take
Stephen Hemminger's suggestion to try the allocation earlier, and drop the
packet if it failed.
We also make a couple of tweaks to how buffer descriptors are set up.
Signed-off-by: Andy Fleming <afleming@freescale.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
|
|
Don't need to keep a struct netxen_new_user_info on the stack
when we only are interested in printing the serial_num. Change
to only reading the serial_num.
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
|
|
When timeout reaches 0 the postfix decrement still subtracts, so the test
fails.
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Roel Kluin <12o3l@tiscali.nl>
Acked-by: Peter Korsgaard <jacmet@sunsite.dk>
Cc: Jeff Garzik <jeff@garzik.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
|
|
Since 43cc71eed1250755986da4c0f9898f9a635cb3bf, the platform modalias is
prefixed with "platform:". Add MODULE_ALIAS() to the hotpluggable network
platform drivers, to re-enable auto loading.
NOTE: didn't change drivers/net/fs_enet/fs_enet-main.c "old binding" support.
That looks problematic in the first place (it even uses the ancient "struct
device_driver" binding scheme for platform_bus!) and I suspect it will vanish
soonish when arch/powerpc rules the world. Also, drivers/net/ne.c would have
needed more thought to sort out.
[akpm@linux-foundation.org: fix sgiseeq.c]
[dbrownell@users.sourceforge.net: more drivers, registration fixes]
Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Scott Wood <scottwood@freescale.com>
Cc: Vitaly Bordug <vitb@kernel.crashing.org>
Cc: Dale Farnsworth <dale@farnsworth.org>
Cc: Ben Dooks <ben-linux@fluff.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Andrew Victor <andrew@sanpeople.com>
Cc: Bryan Wu <bryan.wu@analog.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
|
|
This patch adds support for a new backoff algorithm for half duplex supported
in newer hardware. The old method is will be designated as legacy mode.
Re-seeding random values for the backoff algorithms are performed when a
transmit has failed due to a maximum retry count (1 to 15, where max is
considered the wraparound case of 0).
Signed-off-by: Ayaz Abdulla <aabdulla@nvidia.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
|
|
ehea_flush_sq() and ehea_purge_sq() should be static.
Cc: Jeff Garzik <jeff@garzik.org>
Cc: Thomas Klein <osstklei@de.ibm.com>
Cc: Thomas Klein <tklein@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
|
|
Sometimes the specific interaction between the platform and the PHY
requires special handling. For instance, to change where the PHY's
clock input is, or to add a delay to account for latency issues in the
data path. We add a mechanism for registering a callback with the PHY
Lib to be called on matching PHYs when they are brought up, or reset.
Signed-off-by: Andy Fleming <afleming@freescale.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
|
|
Move some code from atlx.c to atl1.c to prevent build conflict with
the upcoming atl2 code. No changes, just movement.
Signed-off-by: Chris Snook <csnook@redhat.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
|
|
There's no good reason to manually set the flash vendor in a module
parameter, outside of an Atheros hardware lab. Remove it, so nobody
accidentally bricks their board using it incorrectly.
Signed-off-by: Chris Snook <csnook@redhat.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
|
|
- useless initialization (korina_ope / korina_restart)
- use a single variable for the status code in korina_probe
and propagate the error status code from below
- useless checks in korina_remove : the variables are
necessarily set when korina_probe succeeds
Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
|
|
The driver takes the error unwind path without condition.
Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
|
|
Signed-off-by: Krzysztof Hałasa <khc@pm.waw.pl>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
|
|
The local variable "prefix" is never used anymore, and the content of
this string appears a bit later, directly in a call to "alloc_netdev"
after doing exactly the same if/else test. So there seems to be no
point keeping those 4 lines anymore.
Signed-off-by: Cédric Augonnet <cedric.augonnet@ens-lyon.org>
Signed-off-by: Krzysztof Hałasa <khc@pm.waw.pl>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
|
