10 May, 2007
40 commits
-
Since nonboot CPUs are now disabled after tasks and devices have been
frozen and the CPU hotplug infrastructure is used for this purpose, we need
special CPU hotplug notifications that will help the CPU-hotplug-aware
subsystems distinguish normal CPU hotplug events from CPU hotplug events
related to a system-wide suspend or resume operation in progress. This
patch introduces such notifications and causes them to be used during
suspend and resume transitions. It also changes all of the
CPU-hotplug-aware subsystems to take these notifications into consideration
(for now they are handled in the same way as the corresponding "normal"
ones).[oleg@tv-sign.ru: cleanups]
Signed-off-by: Rafael J. Wysocki
Cc: Gautham R Shenoy
Cc: Pavel Machek
Signed-off-by: Oleg Nesterov
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
It's very common for file systems to need to zero part or all of a page,
the simplist way is just to use kmap_atomic() and memset(). There's
actually a library function in include/linux/highmem.h that does exactly
that, but it's confusingly named memclear_highpage_flush(), which is
descriptive of *how* it does the work rather than what the *purpose* is.
So this patchset renames the function to zero_user_page(), and calls it
from the various places that currently open code it.This first patch introduces the new function call, and converts all the
core kernel callsites, both the open-coded ones and the old
memclear_highpage_flush() ones. Following this patch is a series of
conversions for each file system individually, per AKPM, and finally a
patch deprecating the old call. The diffstat below shows the entire
patchset.[akpm@linux-foundation.org: fix a few things]
Signed-off-by: Nate Diller
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
usbatm_do_heavy_init() calls allow_signal() which plays with parent process's
->sighand.Signed-off-by: Oleg Nesterov
Acked-by: Duncan Sands
Cc: Greg KH
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
flush_work(wq, work) doesn't need the first parameter, we can use cwq->wq
(this was possible from the very beginnig, I missed this). So we can unify
flush_work_keventd and flush_work.Also, rename flush_work() to cancel_work_sync() and fix all callers.
Perhaps this is not the best name, but "flush_work" is really bad.(akpm: this is why the earlier patches bypassed maintainers)
Signed-off-by: Oleg Nesterov
Cc: Jeff Garzik
Cc: "David S. Miller"
Cc: Jens Axboe
Cc: Tejun Heo
Cc: Auke Kok ,
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
(akpm: bypassed maintainers, sorry. There are other patches which depend on
this)Cc: "Maciej W. Rozycki"
Cc: David Howells
Cc: Jeff Garzik
Cc: Oleg Nesterov
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
(akpm: bypassed maintainers, sorry. There are other patches which depend on
this)Cc: "Maciej W. Rozycki"
Cc: David Howells
Cc: Jeff Garzik
Cc: Oleg Nesterov
Cc: Tejun Heo
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Switch e1000 over to flush_work_keventd(). This probably fixes a netdev-close
versus linkwatch rtnl_lock() deadlock which nobody knew about.(akpm: bypassed maintainers, sorry. There are other patches which depend on
this)Cc: "Maciej W. Rozycki"
Cc: David Howells
Cc: "David S. Miller"
Cc: Jeff Garzik
Acked-by: Auke Kok
Cc: Oleg Nesterov
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Convert tg3 over to flush_work_keventd(). Remove nasty now-unneeded deadlock
avoidance logic.(akpm: bypassed maintainers, sorry. There are other patches which depend on
this)Cc: "Maciej W. Rozycki"
Cc: David Howells
Cc: "David S. Miller"
Cc: Michael Chan
Cc: Jeff Garzik
Cc: Oleg Nesterov
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Now that the cpu_is_xxx() macros are available both on AVR32 and AT91, we can
remove a couple of #ifdefs from this driver. One of them is actually wrong --
new_1 should be set on AVR32 but isn't. This causes the bus clock to run at
twice the speed it is configured to.Signed-off-by: Haavard Skinnemoen
Cc: David Brownell
Acked-by: Andrew Victor
Cc: Nicolas Ferre
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Fix tty_set_ldisc in tty_io.c so that tty->receive_room is only cleared if
actually changing line disciplines.Without this fix a problem occurs when requesting the line discipline to
change to the same line discipline. In this case tty->receive_room is
cleared but ldisc->open() is not called to set tty->receive_room back to a
sane value. The result is that tty->receive_room is stuck at 0 preventing
the tty flip buffer from passing receive data to the line discipline.For example: a switch from N_TTY to N_TTY followed by a select() call for
read input results in data never being received because tty->receive_room
is stuck at zero.A switch from N_TTY to N_TTY followed by a read() call works because the
read() call itself sets tty->receive_room correctly (but select does not).Previously (< 2.6.18) this was not a problem because the tty flip buffer
pushed data to the line discipline without regard for tty->receive room.Signed-off-by: Paul Fulghum
Acked-by: Alan Cox
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
[akpm@linux-foundation.org: fix it]
Signed-off-by: WANG Cong
Cc: Paul Clements
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Driver for the on-chip hardware random number generator on PA Semi
PA6T-1682M.Signed-off-by: Egor Martovetsky
Signed-off-by: Olof Johansson
Signed-off-by: Michael Buesch
Cc: Paul Mackerras
Cc: Kumar Gala
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Cleanup using bitrev8 in cm4000_cs driver.
Cc: Harald Welte
Signed-off-by: Akinobu Mita
Cc: Dominik Brodowski
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This patch fixes error paths in module_init and probe functions in cm4000_cs
and cm4040_cs drivers.Cc: Harald Welte
Signed-off-by: Akinobu Mita
Cc: Dominik Brodowski
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
The MCA bus has a few "integrated" functions, which are effectively virtual
slots on the bus. The problem is that these special functions don't have
dedicated pos IDs, so we have to manufacture ids for them outside the pos
space ... and these ids can't be matched by the standard matching function,
so add a special registration that requests a list of pos ids or a particular
integrated function.Signed-off-by: James Bottomley
Signed-off-by: Linus Torvalds -
There's a bug in the MCA bus matching algorithm in that it promotes from
signed short to int before comparing with the actual id and does sign
extension on anything > 0x7fff (which means that pos ids > 0x7fff never get
correctly matched).Signed-off-by: James Bottomley
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
[ With Johannes Berg ]
Separate the hibernation (aka suspend to disk code) from the other suspend
code. In particular:* Remove the definitions related to hibernation from include/linux/pm.h
* Introduce struct hibernation_ops and a new hibernate() function to hibernate
the system, defined in include/linux/suspend.h
* Separate suspend code in kernel/power/main.c from hibernation-related code
in kernel/power/disk.c and kernel/power/user.c (with the help of
hibernation_ops)
* Switch ACPI (the only user of pm_ops.pm_disk_mode) to hibernation_opsSigned-off-by: Rafael J. Wysocki
Cc: Greg KH
Cc: Pavel Machek
Cc: Nigel Cunningham
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This patch removes the possibility of having uninitialized log state if the
log device has failed.When a mirror resumes operation, it calls 'resume' on the logging module. If
disk based logging is being used, the log device is read to fill in the log
state. If the log device has failed, we cannot simply return, because this
would leave the in-memory log state uninitialized. Instead, we assume all
regions are out-of-sync and reset the log state. Failure to do this could
result in the logging code reporting a region as in-sync, even though it
isn't; which could result in a corrupted mirror.Signed-off-by: Jonathan Brassow
Signed-off-by: Alasdair G Kergon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
The call to rh_in_sync() in do_reads() should be allowed to block. It is in
the mirror worker thread which already permits blocking operations. This will
be needed to support clustered mirroring which will perform network
operations.Signed-off-by: Jonathan Brassow
Signed-off-by: Alasdair G Kergon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
With the code as it is, it is possible for oustanding clear region requests
never to get flushed when a mirror is deactivated or suspended. This means
there will always be some resync work required when a mirror is activated,
even though it may very well be in-sync.Always requesting the flush doesn't hurt us. This is because the log tracks
whether any changes occurred and, if not, no flush is performed.Signed-off-by: Jonathan Brassow
Signed-off-by: Alasdair G Kergon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
New device-mapper target that can delay I/O (for testing). Reads can be
separated from writes, redirected to different underlying devices and delayed
by differing amounts of time.Signed-off-by: Heinz Mauelshagen
Signed-off-by: Milan Broz
Signed-off-by: Alasdair G Kergon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
More bio_list helper functions for new targets (including dm-delay and
dm-loop) to manipulate lists of bios.Signed-off-by: Heinz Mauelshagen
Signed-off-by: Bryn Reeves
Signed-off-by: Milan Broz
Signed-off-by: Alasdair G Kergon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Remove old dm-io interface.
Signed-off-by: Milan Broz
Signed-off-by: Alasdair G Kergon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This patch ports dm-raid1.c to the new dm-io interface.
Signed-off-by: Milan Broz
Signed-off-by: Alasdair G Kergon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This patch ports dm-log.c to the new dm-io interface in order to make it
scalable to have a large number of persistent dirty logs active in parallel.Signed-off-by: Heinz Mauelshagen
Cc: Milan Broz
Signed-off-by: Alasdair G Kergon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This patch ports dm-exception-store.c to the new, scalable dm_io() interface.
It replaces dm_io_get()/dm_io_put() by
dm_io_client_create()/dm_io_client_destroy() calls and
dm_io_sync_vm() by dm_io() to achive this.Signed-off-by: Heinz Mauelshagen
Cc: Milan Broz
Signed-off-by: Alasdair G Kergon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This patch ports kcopyd.c to the new, scalable dm_io() interface.
Signed-off-by: Milan Broz
Signed-off-by: Heinz Mauelshagen
Signed-off-by: Alasdair G Kergon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Add a new API to dm-io.c that uses a private mempool and bio_set for each
client.The new functions to use are dm_io_client_create(), dm_io_client_destroy(),
dm_io_client_resize() and dm_io().Signed-off-by: Heinz Mauelshagen
Signed-off-by: Alasdair G Kergon
Cc: Milan Broz
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Introduce struct dm_io_client to prepare for per-client mempools and bio_sets.
Temporary functions bios() and io_pool() choose between the per-client
structures and the global ones so the old and new interfaces can co-exist.Make error_bits optional.
Signed-off-by: Heinz Mauelshagen
Signed-off-by: Alasdair G Kergon
Cc: Milan Broz
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Delay decrementing the 'struct io' reference count until after the bio has
been freed so that a bio destructor function may reference it. Required by a
later patch.Signed-off-by: Heinz Mauelshagen
Signed-off-by: Alasdair G Kergon
Cc: Milan Broz
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This patch adds the ability to specify desired features in the mirror
constructor/mapping table.The first feature of interest is "handle_errors". Currently, mirroring will
ignore any I/O errors from the devices. Subsequent patches will check for
this flag and handle the errors. If flag/feature is not present, mirror will
do nothing - maintaining backwards compatibility.Signed-off-by: Jonathan E Brassow
Signed-off-by: Alasdair G Kergon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This patch reports the status of the log device so that userspace can detect
the error and take appropriate action.Signed-off-by: Jonathan E Brassow
Signed-off-by: Alasdair G Kergon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This patch gives the disk logging code the ability to store the fact that an
error occured on the log device. In addition, an event is raised when an
error is encountered during I/O to the log device.Signed-off-by: Jonathan E Brassow
Signed-off-by: Alasdair G Kergon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Allow check_device_area to succeed if a device has an i_size of zero. This
addresses an issue seen on DASD devices setting up a multipath table for paths
in online and offline state.Signed-off-by: Mike Anderson
Signed-off-by: Alasdair G Kergon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Make the mapped device structure accessible to hardware handlers so error
messages can include the device name.Signed-off-by: Edward Goggin
Signed-off-by: Alasdair G Kergon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Add a new IV generation method 'null' to read old filesystem images created
with SuSE's loop_fish2 module.Signed-off-by: Ludwig Nussel
Acked-By: Christophe Saout
Signed-off-by: Alasdair G Kergon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Allocate smaller clones
With the previous dm-crypt fixes, there is no need for the clone bios to have
the same bvec size as the original - we just need to make them big enough for
the remaining number of pages. The only requirement is that we clear the
"out" index in convert_context, so that crypt_convert starts storing data at
the right position within the clone bio.Signed-off-by: Olaf Kirch
Signed-off-by: Alasdair G Kergon
Cc: Jens Axboe
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Get rid of first_clone in dm-crypt
This gets rid of first_clone, which is not really needed. Apparently, cloned
bios used to share their bvec some time way in the past - this is no longer
the case. Contrarily, this even hurts us if we try to create a clone off
first_clone after it has completed, and crypt_endio has destroyed its bvec.Signed-off-by: Olaf Kirch
Signed-off-by: Alasdair G Kergon
Cc: Jens Axboe
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Do not access the bio after generic_make_request
We should never access a bio after generic_make_request - there's no guarantee
it still exists.Signed-off-by: Olaf Kirch
Signed-off-by: Alasdair G Kergon
Cc: Jens Axboe
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Call clone_init early
We need to call clone_init as early as possible - at least before call
bio_put(clone) in any error path. Otherwise, the destructor will try to
dereference bi_private, which may still be NULL.Signed-off-by: Olaf Kirch
Signed-off-by: Alasdair G Kergon
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
