Use the recently-added bio front_pad field to allocate struct dm_target_io.
Prior to this patch, dm_target_io was allocated from a mempool. For each
dm_target_io, there is exactly one bio allocated from a bioset.
This patch merges these two allocations into one allocation: we create a
bioset with front_pad equal to the size of dm_target_io so that every
bio allocated from the bioset has sizeof(struct dm_target_io) bytes
before it. We allocate a bio and use the bytes before the bio as
dm_target_io.
Mike Snitzer [Tue, 25 Sep 2012 00:31:01 +0000 (10:31 +1000)]
Support discards when the pool's block size is not a power of 2.
The block layer assumes discard_granularity is a power of 2 (in
blkdev_issue_discard), so we set this to the largest power of 2 that is
a divides into the number of sectors in each block, but never less than
DATA_DEV_BLOCK_SIZE_MIN_SECTORS.
This patch eliminates the "Discard support must be disabled when the
block size is not a power of 2" constraint that was imposed in commit 55f2b8b ("dm thin: support for non power of 2 pool blocksize"). That
commit was incomplete: using a block size that is not a power of 2
shouldn't mean disabling discard support on the device completely.
Signed-off-by: Mike Snitzer <snitzer@redhat.com> Signed-off-by: Joe Thornber <ejt@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Use the ACCESS_ONCE macro in dm-bufio and dm-verity where a variable
can be modified asynchronously (through sysfs) and we want to prevent
compiler optimizations that assume that the variable hasn't changed.
(See Documentation/atomic_ops.txt.)
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Mike Snitzer [Tue, 25 Sep 2012 00:31:00 +0000 (10:31 +1000)]
The access beyond the end of device BUG_ON that was introduced to
dm_request_fn via commit 29e4013de7ad950280e4b2208 ("dm: implement
REQ_FLUSH/FUA support for request-based dm") was an overly
drastic (but simple) response to this situation.
I have received a report that this BUG_ON was hit and now think
it would be better to use dm_kill_unmapped_request() to fail the clone
and original request with -EIO.
map_request() will assign the valid target returned by
dm_table_find_target to tio->ti. But when the target
isn't valid tio->ti is never assigned (because map_request isn't
called); so add a check for tio->ti != NULL to dm_done().
Reported-by: Mike Christie <michaelc@cs.wisc.edu> Signed-off-by: Mike Snitzer <snitzer@redhat.com> Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com> Cc: stable@vger.kernel.org # v2.6.37+ Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Mike Snitzer [Tue, 25 Sep 2012 00:31:00 +0000 (10:31 +1000)]
The discard limits that get established for a thin-pool or thin device
may be incompatible with the pool's data device. Avoid this by checking
the discard limits of the pool's data device. If an incompatibility is
found then the pool's 'discard passdown' feature is disabled.
Change thin_io_hints to ensure that a thin device always uses the same
queue limits as its pool device.
Introduce requested_pf to track whether or not the table line originally
contained the no_discard_passdown flag and use this directly for table
output. We prepare the correct setting for discard_passdown directly in
bind_control_target (called from pool_io_hints) and store it in
adjusted_pf rather than waiting until we have access to pool->pf in
pool_preresume.
Signed-off-by: Mike Snitzer <snitzer@redhat.com> Signed-off-by: Joe Thornber <ejt@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Mike Snitzer [Tue, 25 Sep 2012 00:30:59 +0000 (10:30 +1000)]
The dm thin pool target claims to support the zeroing of discarded
data areas. This turns out to be incorrect when processing discards
that do not exactly cover a complete number of blocks, so the target
must always set discard_zeroes_data_unsupported.
The thin pool target will zero blocks when they are allocated if the
skip_block_zeroing feature is not specified. The block layer
may send a discard that only partly covers a block. If a thin pool
block is partially discarded then there is no guarantee that the
discarded data will get zeroed before it is accessed again.
Due to this, thin devices cannot claim discards will always zero data.
Signed-off-by: Mike Snitzer <snitzer@redhat.com> Cc: stable@vger.kernel.org # 3.4+ Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Mike Snitzer [Tue, 25 Sep 2012 00:30:59 +0000 (10:30 +1000)]
Add a safety net that will re-use the DM device's existing limits in the
event that DM device has a temporary table that doesn't have any
component devices.
DM recalculates queue limits based only on devices which currently exist
in the table. This creates a problem in the event all devices are
temporarily removed such as all paths being lost in multipath. DM will
reset the limits to the maximum permissible, which can then assemble
requests which exceed the limits of the paths when the paths are
restored. The request will fail the blk_rq_check_limits() test when
sent to a path with lower limits, and will be retried without end by
multipath. This became a much bigger issue after commit fe86cdcef
("block: do not artificially constrain max_sectors for stacking
drivers").
Reported-by: David Jeffery <djeffery@redhat.com> Signed-off-by: Mike Snitzer <snitzer@redhat.com> Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Peter Huewe [Tue, 25 Sep 2012 00:30:54 +0000 (10:30 +1000)]
i2c: Make I2C available on UML
Remove the global dependency of the I2C subsystem on HAS_IOMEM and
move the dependency to the i2c/busses submenu, with an exception for
i2c-stub.
The generic I2C part does not need to have HAS_IOMEM set and thus now
becomes available in UML, so the I2C subsystem can now be used, e.g.
by the i2c-stub driver, for development of I2C device drivers.
[JD: Some adjustments.]
Signed-off-by: Peter Huewe <peterhuewe@gmx.de> Signed-off-by: Jean Delvare <khali@linux-fr.org>
Shubhrajyoti D [Tue, 25 Sep 2012 00:30:53 +0000 (10:30 +1000)]
i2c: Convert struct i2c_msg initialization to C99 format
Convert the struct i2c_msg initialization to C99 format. This makes
maintaining and editing the code simpler. Also helps once other fields
like transferred are added in future.
Signed-off-by: Shubhrajyoti D <shubhrajyoti@ti.com> Signed-off-by: Jean Delvare <khali@linux-fr.org>
Julia Lawall [Tue, 25 Sep 2012 00:30:53 +0000 (10:30 +1000)]
i2c-smbus: Convert kzalloc to devm_kzalloc
Converting kzalloc to devm_kzalloc simplifies the code and ensures that the
result, alert, is freed after the irq allocated by the subsequent
devm_request_irq. This in turn ensures that when an interrupt can be
triggered, the alert structure is still available.
The problem of a free after a devm_request_irq was found using the
following semantic match (http://coccinelle.lip6.fr/)
devm_request_irq@p1(e1,e2,...,x)
... when any
when != e2 = a
when != x = b
if (...) {
... when != e2 = c
when != x = d
free@p2(...,x,...);
...
return ...;
}
// </smpl>
Signed-off-by: Julia Lawall <Julia.Lawall@lip6.fr> Signed-off-by: Jean Delvare <khali@linux-fr.org>
Jean Delvare [Tue, 25 Sep 2012 00:30:53 +0000 (10:30 +1000)]
i2c-i801: Support SMBus multiplexing on Asus Z8 series
Add support for SMBus multiplexing on Asus Z8 motherboard series. On
these boards, the memory slots are behind a GPIO-controlled I2C
multiplexer. Models with 6 or 12 memory slots have 2 segments behind
the multiplexer, while models with 18 memory slots have 3 such
segments.
On these boards, only the memory slots are behind the multiplexer,
so it is possible to keep the autodetection mechanism.
The code is generic enough so it could work on other boards as long as
the multiplexer is controlled by GPIO pins. For other forms of
multiplexing (for example using an I2C device) additional code will be
needed.
Thanks to Asus for providing a board to develop and test this feature,
as well as all the technical information required.
At the moment, the GPIO driver must be loaded before the i2c-i801
driver, but I hope to solve this soon, using deferred probing on
the i2c-mux-gpio side.
Jean Delvare [Tue, 25 Sep 2012 00:30:52 +0000 (10:30 +1000)]
i2c-mux: Add support for device auto-detection
Let I2C bus segments behind multiplexers have a class. This allows for
device auto-detection on these segments. As long as parent segments
don't share the same class, it should be fine.
I implemented support in drivers i2c-mux-gpio and i2c-mux-pca954x. I
left i2c-mux-pca9541 and i2c-mux-pinctrl alone for the moment as I
don't know if this feature makes sense for the use cases of these
drivers.
Signed-off-by: Jean Delvare <khali@linux-fr.org> Cc: Peter Korsgaard <peter.korsgaard@barco.com> Cc: David Daney <david.daney@cavium.com> Cc: Michael Lawnick <ml.lawnick@gmx.de> Cc: Rodolfo Giometti <giometti@linux.it>
Pull tile gxio ABI fix from Chris Metcalf:
"This fixes a last-minute change in the Tilera hypervisor ABI for TRIO
(PCI root complex) support. We've locked in this ABI going forward
and will make sure no further ABI changes like this occur."
* git://git.kernel.org/pub/scm/linux/kernel/git/cmetcalf/linux-tile:
tile: gxio iorpc numbering change for TRIO interface
Merge tag 'stable/for-linus-3.6-rc7-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen
Pull a Xen fix from Konrad Rzeszutek Wilk:
"It is a bug-fix when we run the initial PV guest on a AMD K8 machine
and have CONFIG_AMD_NUMA enabled and detect the NUMA topology from the
Northbridge.
We end up in the situation where the initial domain gets too much
information and gets confused and crashes - the fix is to restrict the
domain to get the information - and we do it by just disabling NUMA on
the PV guest (the hypervisor is still able to do its proper NUMA
allocations of guests).
It is OK to disable the PV guest from accessing NUMA data as right now
we do not inject any NUMA node information to the PV guests. When we
do get to that point, then this patch will have to be reverted."
* Disable PV NUMA support as we do not do anything with it (yet) and it
can cause bootup crashes on certain AMD machines.
* tag 'stable/for-linus-3.6-rc7-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen:
xen/boot: Disable NUMA for PV guests.
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client
Pull two ceph fixes from Sage Weil:
"The first fixes a leak in the rbd setup error path, and the second
fixes a more serious problem with mismatched kmap/kunmap that surfaced
after the recent refactoring work."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/sage/ceph-client:
libceph: only kunmap kmapped pages
rbd: drop dev reference on error in rbd_open()
Jiang Liu [Wed, 22 Aug 2012 15:16:45 +0000 (23:16 +0800)]
PCI: acpiphp: Handle PCIe ports without native hotplug capability
Commit 0d52f54e2ef64c189dedc332e680b2eb4a34590a (PCI / ACPI: Make acpiphp
ignore root bridges using PCIe native hotplug) added code that made the
acpiphp driver completely ignore PCIe root complexes for which the kernel
had been granted control of the native PCIe hotplug feature by the BIOS
through _OSC. Later commit 619a5182d1f38a3d629ee48e04fa182ef9170052
"PCI hotplug: Always allow acpiphp to handle non-PCIe bridges" relaxed
the constraints to allow acpiphp driver handle non-PCIe bridges under
such a complex. The constraint needs to be relaxed further to allow
acpiphp driver to handle PCIe ports without native PCIe hotplug capability.
Some MR-IOV switch chipsets, such PLX8696, support multiple virtual PCIe
switches and may migrate downstream ports among virtual switches. To
migrate a downstream port from the source virtual switch to the target, the
port needs to be hot-removed from the source and hot-added into the target.
The pciehp driver can't be used here because there are no slots within the
virtual PCIe switch. So acpiphp driver is used to support downstream port
migration. A typical configuration is as below:
[Root without native PCIe HP]
[Upstream port of vswitch without native PCIe HP]
[Downstream port of vswitch with native PCIe HP]
[PCIe endpoint]
Here acpiphp driver will be used to handle root ports and upstream port
in the virtual switch, and pciehp driver will be used to handle downstream
ports in the virtual switch.
Signed-off-by: Jiang Liu <liuj97@gmail.com> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Acked-by: Rafael J. Wysocki <rjw@sisk.pl>
PCI/ACPI: Use acpi_driver_data() rather than searching acpi_pci_roots
This patch changes the implementation of acpi_pci_find_root().
We can access acpi_pci_root without scanning acpi_pci_roots list.
If hostbridge hotplug is supported, acpi_pci_roots list will be
protected by mutex. We should not access acpi_pci_roots list
if preventable to lessen deadlock risk.
PCI/ACPI: Use acpi_pci_root info rather than looking it up again
This patch changes the function signature of walk_root_bridge().
We can omit _STA, _SEG, and _BBN evaluation by passing not acpi_handle
but acpi_pci_root. Now that acpi_pci_slot_add() which is the only
caller of walk_root_bridge() gets acpi_pci_root structure, changing
signature of walk_root_bridge() is reasonable.
PCI/ACPI: Pass acpi_pci_root to acpi_pci_drivers' add/remove interface
This patch changes .add/.remove interfaces of acpi_pci_driver.
In the current implementation acpi_handle is passed as a parameter
of .add/.remove interface. However, the acpi_pci_root structure
contains more useful information than just the acpi_handle. This
enables us to avoid some useless lookups in each acpi_pci_driver.
Note: This changes interfaces used by acpi_pci_register_driver(), an
exported symbol. This patch updates all the in-kernel users, but any
out-of-kernel acpi_pci_register_driver() users will need updates.
projects / karo-tx-linux.git / log