Handling balloon hibernate / restore is tricky. If the balloon was
inflated before going into the hibernation state, upon resume, the host
will not have any memory of that. Any pages that were passed on to the
host earlier would most likely be invalid, and the host will have to
re-balloon to the previous value to get in the pre-hibernate state.
So the only sane thing for the guest to do here is to discard all the
pages that were put in the balloon. When to discard the pages is the
next question.
One solution is to deflate the balloon just before writing the image to
the disk (in the freeze() PM callback). However, asking for pages from
the host just to discard them immediately after seems wasteful of
resources. Hence, it makes sense to do this by just fudging our
counters soon after wakeup. This means we don't deflate the balloon
before sleep, and also don't put unnecessary pressure on the host.
This also helps in the thaw case: if the freeze fails for whatever
reason, the balloon should continue to remain in the inflated state.
This was tested by issuing 'swapoff -a' and trying to go into the S4
state. That fails, and the balloon stays inflated, as expected. Both
the host and the guest are happy.
Finally, in the restore() callback, we empty the list of pages that were
previously given off to the host, add the appropriate number of pages to
the totalram_pages counter, reset the num_pages counter to 0, and
all is fine.
As a last step, delete the vqs on the freeze callback to prepare for
hibernation, and re-create them in the restore and thaw callbacks to
resume normal operation.
The kthread doesn't race with any operations here, since it's frozen
before the freeze() call and is thawed after the thaw() and restore()
callbacks, so we're safe with that.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The probe and PM restore functions will share this code.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Remove all the vqs, disable napi and detach from the netdev on
hibernation.
Re-create vqs after restoring from a hibernated image, re-enable napi
and re-attach the netdev. This keeps networking working across
hibernation.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The remove and PM freeze functions will share this code.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The probe and PM restore functions will share this code.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Delete the vq and flush any pending requests from the block queue on the
freeze callback to prepare for hibernation.
Re-create the vq in the restore callback to resume normal function.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The probe and PM restore functions will share this code.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
To ensure we don't receive any more interrupts from the host after we
enter the freeze function, disable all vq interrupts.
There wasn't any problem seen due to this in tests, but applying this
patch makes the freeze case more robust.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Remove all vqs and associated buffers in the freeze callback which
prepares us to go into hibernation state. On restore, re-create all the
vqs and populate the input vqs with buffers to get to the pre-hibernate
state.
Note: Any outstanding unconsumed buffers are discarded; which means
there's a possibility of data loss in case the host or the guest didn't
consume any data already present in the vqs. This can be addressed in a
later patch series, perhaps in virtio common code.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This common code will be shared with the PM freeze function.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Handle thaw, restore and freeze notifications from the PM core. Expose
these to individual virtio drivers that can quiesce and resume vq
operations. For drivers not implementing the thaw() method, use the
restore method instead.
These functions also save device-specific data so that the device can be
put in pre-suspend state after resume, and disable and enable the PCI
device in the freeze and resume functions, respectively.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The older PM API doesn't have a way to get notifications on hibernate
events. Switch to the newer one that gives us those notifications.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Fix a theoretical race related to config work
handler: a config interrupt might happen
after we flush config work but before we
reset the device. It will then cause the
config work to run during or after reset.
Two problems with this:
- if this runs after device is gone we will get use after free
- access of config while reset is in progress is racy
(as layout is changing).
As a solution
1. flush after reset when we know there will be no more interrupts
2. add a flag to disable config access before reset
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Under the existing #ifdef DEBUG, check that they don't have more than
1/10 of a second between an add_buf() and a
virtqueue_notify()/virtqueue_kick_prepare() call.
We could get false positives on a really busy system, but good for
development.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
A virtio driver does virtqueue_add_buf() multiple times before finally
calling virtqueue_kick(); previously we only exposed the added buffers
in the virtqueue_kick() call. This means we don't need a memory
barrier in virtqueue_add_buf(), but it reduces concurrency as the
device (ie. host) can't see the buffers until the kick.
In the unusual (but now possible) case where a driver does add_buf()
and get_buf() without doing a kick, we do need to insert one before
our counter wraps. Otherwise we could wrap num_added, and later on
not realize that we have passed the marker where we should have
kicked.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Since we know vq->vring.num is a power of 2, modulus is lazy (it's asserted
in vring_new_virtqueue()).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Based on patch by Christoph for virtio_blk speedup:
Split virtqueue_kick to be able to do the actual notification
outside the lock protecting the virtqueue. This patch was
originally done by Stefan Hajnoczi, but I can't find the
original one anymore and had to recreated it from memory.
Pointers to the original or corrections for the commit message
are welcome.
Stefan's patch was here:
a6d06644e3http://www.spinics.net/lists/linux-virtualization/msg14616.html
Third time's the charm!
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Remove wrapper functions. This makes the allocation type explicit in
all callers; I used GPF_KERNEL where it seemed obvious, left it at
GFP_ATOMIC otherwise.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Reviewed-by: Christoph Hellwig <hch@lst.de>
The old documentation is left over from when we used a structure with
strategy pointers.
And move the documentation to the C file as per kernel practice.
Though I disagree...
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Trivial changes to remove forgotten junk, format comments, and correct names.
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: virtualization@lists.linux-foundation.org
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We were cheating with our barriers; using the smp ones rather than the
real device ones. That was fine, until rpmsg came along, which is
used to talk to a real device (a non-SMP CPU).
Unfortunately, just putting back the real barriers (reverting
d57ed95d) causes a performance regression on virtio-pci. In
particular, Amos reports netbench's TCP_RR over virtio_net CPU
utilization increased up to 35% while throughput went down by up to
14%.
By comparison, this branch is in the noise.
Reference: https://lkml.org/lkml/2011/12/11/22
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Host may now use MMC_CAP2_NO_SLEEP_CMD to disable the use
of eMMC sleep/awake command.
This option can be used when your platform has a buggy
kernel crash dump software, which is supposed to store
the dump on the eMMC, but is not able to wake up the eMMC
from sleep state.
In particular, failures have been seen with u-boot; even if
it is fixed there, platforms will be slow to update their
bootloader binaries.
Signed-off-by: Ulf Hansson <ulf.hansson@stericsson.com>
Reviewed-by: Hanumath Prasad <hanumath.prasad@stericsson.com>
Reviewed-by: Srinidhi Kasagar <srinidhi.kasagar@stericsson.com>
Acked-by: Subhash Jadavani <subhashj@codeaurora.org>
Signed-off-by: Chris Ball <cjb@laptop.org>
As the card-detect irq handler just schedules work to be done by a
thread, we can use request_threaded_irq to do much of the work for
us. This means that interrupts which arrive by handle_nested_irq
actually work.
Reviewed-by: Felipe Balbi <balbi@ti.com>
Tested-by: Grazvydas Ignotas <notasas@gmail.com>
Signed-off-by: NeilBrown <neilb@suse.de>
Acked-by: Kishore Kadiyala <kishorek.kadiyala@gmail.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
Runtime PM for SDIO is no longer enabled by default (see
5c7f0e083d) so it must now
be enabled per platform, in this case Medfield uses it.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
To allow the set_clock host op to disable the SDCLK source when not
needed, always call the host op when the requested clock speed is
zero. Do this even if host->clock already equals zero, because
the SDHCI driver may set that value (without calling the host op)
to force an update at the next (non-zero-speed) call.
Signed-off-by: Todd Poynor <toddpoynor@google.com>
Acked-by: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
Even if a driver provides separate card detection, an interrupt
is still needed to abort mmc requests that are in progress.
SDHCI_QUIRK2_OWN_CARD_DETECTION prevents that, so remove it.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
Retrieve the GPIO numbers for hardware reset and
card detect from platform data.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
Add a means of getting platform data for the SDHCI PCI
devices. The data is stored against the slot not the
device in order to support multi-slot devices.
The data allows platform-specific setup (such as getting
GPIO numbers from firmware or setting up wl12xx for SDIO)
to be done in platform support files instead of the
sdhci-pci driver.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
Do not enable card detection interrupts for non-removable cards.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
Presently the vmmc regulator is enabled when the host
controller is added and disabled when it is removed.
However, the vmmc regulator should be under the control
of the upper layers via ->set_ios(). Make it so.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
While calling mmc_cache_ctrl() a host is not claimed. This patch
adds the mmc_try_claim_host() for quick response in suspend.
Signed-off-by: Seungwon Jeon <tgih.jun@samsung.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
MMC master clock rate can vary for each instance of the MMC controller
on the device. Use clk_get_rate instead to get the value.
Signed-off-by: Balaji TK <balajitk@ti.com>
Reviewed-by: Venkatraman S <svenkatr@ti.com>
Tested-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
This patch converts the sh_mmcif MMC host driver to process requests
asynchronously instead of waiting in its .request() method for completion.
This is achieved by using threaded IRQs.
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Chris Ball <cjb@laptop.org>
This patch doesn't introduce any functional changes, it only simplifies
some code fragments, removes superfluous parameters, fixes typos.
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Chris Ball <cjb@laptop.org>
If an interrupt is coming with both error and data completion status bits
set, it has to be handled as an error interrupt, for which error interrupts
have to be processed first. The current version of the driver on the
contrary doesn't recognise such interrupts as an error event, which leads
to data corruption and breaks the error recovery.
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Chris Ball <cjb@laptop.org>
This patch adds a primitive helper to support card hotplug detection on
platforms, where a GPIO, capable of producing interrupts, is used for
detection of card-insertion and -removal events.
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Chris Ball <cjb@laptop.org>
SD/MMC controllers provide different card insertion and removal detection
methods. On some of them the controller itself issues an interrupt, on
others polling is used, on yet others auxiliary means are used for this
purpose, e.g., a GPIO IRQ. Further, on some systems one of those methods
can be chosen at driver probing time and configured in software. E.g., on
some systems the SD/MMC controller card hot-plug detection pin can be
configured either as a respective controller functions, or an IRQ-capable
GPIO. To support such flexible configurations a card hot-plug context
is added by this patch.
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Chris Ball <cjb@laptop.org>
This patch regroups the code slightly, adds documentation and allows
the rtpm counter of MMC_CAP_NEEDS_POLL devices to reach 0 again.
Signed-off-by: Bastian Hecht <hechtb@gmail.com>
[g.liakhovetski@gmx.de: restore pm_runtime_get_noresume()]
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Chris Ball <cjb@laptop.org>
This patch fixes the wrong comparison before setting the interface
voltage in DDR mode.
The assignment to the variable ddr before comaprison is either
ddr = MMC_1_2V_DDR_MODE; or ddr == MMC_1_8V_DDR_MODE. But the comparison
is done with the extended csd value if ddr == EXT_CSD_CARD_TYPE_DDR_1_2V.
Signed-off-by: Girish K S <girish.shivananjappa@linaro.org>
Acked-by: Subhash Jadavani <subhashj@codeaurora.org>
Acked-by: Philip Rakity <prakity@marvell.com>
Cc: <stable@kernel.org>
Signed-off-by: Chris Ball <cjb@laptop.org>
This patch adds the support for predefined multiple block r/w.
dw_mmc can support MMC_CAP_CMD23 capability.
Signed-off-by: Seungwon Jeon <tgih.jun@samsung.com>
Acked-by: Will Newton <will.newton@imgtec.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
This patch adds another capabilities field for MMC_CAPS2_XXX.
Signed-off-by: Seungwon Jeon <tgih.jun@samsung.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
This patch modifies dw_mmc to use dev_pm_ops.
Signed-off-by: Jaehoon Chung <jh80.chung@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Acked-by: James Hogan <james.hogan@imgtec.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
Kill block requests when the host realizes that the card is
removed from the slot and is sure that subsequent requests
are bound to fail. Do this silently so that the block
layer doesn't output unnecessary error messages.
Signed-off-by: Sujit Reddy Thumma <sthumma@codeaurora.org>
Acked-by: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
c31b50e (mmc: core: Use delayed work in clock gating framework,
2011-11-14) missed a few things during review:
o A useless pr_info()
o milliseconds was written as two words
o The sysfs file had units in its output
Fix all three problems.
Signed-off-by: Stephen Boyd <sboyd@codeaurora.org>
Cc: Sujit Reddy Thumma <sthumma@codeaurora.org>
Signed-off-by: Chris Ball <cjb@laptop.org>
Enable boot partitions to be read-only locked until next power on via
a sysfs entry. There will be one sysfs entry for each boot partition:
/sys/block/mmcblkXbootY/ro_lock_until_next_power_on
Each boot partition is locked by writing 1 to its file.
Signed-off-by: Johan Rudholm <johan.rudholm@stericsson.com>
Signed-off-by: John Beckett <john.beckett@stericsson.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
Add pm_caps into platform_data. This is power management, usually
for SDIO device such as SDIO WLAN.
Signed-off-by: Sangwook Lee <sangwook.lee@samsung.com>
Acked-by: Kukjin Kim <kgene.kim@samsung.com>
Signed-off-by: Chris Ball <cjb@laptop.org>
