When number of NPIV ports created are greater than the xids
allocated per pool -- for eg., creating 255 NPIV ports on a
system with nr_cpu_ids of 32, with each pool containing 128
xids -- and then generating a link event - for eg.,
shutdown/no shutdown -- on the switch port causes the hang
with the following stack trace.
Fix is to not cancel the disc_work if discovery is already
stopped, thus allowing lport state machine to restart and try
discovery again.
Signed-off-by: Bhanu Prakash Gollapudi <bprakash@broadcom.com> Acked-by: Joe Eykholt <jeykholt@cisco.com> Signed-off-by: Robert Love <robert.w.love@intel.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Vasu Dev [Sat, 9 Oct 2010 00:12:31 +0000 (17:12 -0700)]
[SCSI] libfc: possible race could panic system due to NULL fsp->cmd
It is unlikely but in case if it hits then it would cause panic
due to null cmd ptr, so far only one instance seen recently with
ESX though this was introduced long ago with this commit:-
commit c1ecb90a66c5afc7cc5c9349f9c3714eef4a5cfb
Author: Chris Leech <christopher.leech@intel.com>
Date: Thu Dec 10 09:59:26 2009 -0800
[SCSI] libfc: reduce hold time on SCSI host lock
Currently fsp->cmd is set to NULL w/o scsi_queue_lock before
dequeuing from scsi_pkt_queue and that could cause NULL
fsp->cmd in fc_fcp_cleanup_each_cmd for cmd completing
with fsp->cmd = NULL after fc_fcp_cleanup_each_cmd taken
reference. No need to set fsp->cmd to NULL as this is also
protected by fc_fcp_lock_pkt(), for above race the
fc_fcp_lock_pkt() in fc_fcp_cleanup_each_cmd() will fail
as that cmd is already done.
Mike mentioned same issue at
http://www.open-fcoe.org/pipermail/devel/2010-September/010533.html
Similarly moved sc_cmd->SCp.ptr = NULL under scsi_queue_lock so
that scsi abort error handler won't abort on completed cmds.
Signed-off-by: Vasu Dev <vasu.dev@intel.com> Signed-off-by: Robert Love <robert.w.love@intel.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Vasu Dev [Sat, 9 Oct 2010 00:12:25 +0000 (17:12 -0700)]
[SCSI] fcoe: set default FIP mode as FIP_MODE_FABRIC
Since sometimes current FIP_MODE_AUTO mode falls back to non-FIP
mode while DCB link still getting ready in fabric mode with
its peer switch, it falls back after few libfc flogi retries
and that is not we want while working with FIP enabled
switches in FABRIC mode, therefore sets default as FIP_MODE_FABRIC
as discussed and agreed before in this mail thread
http://www.open-fcoe.org/pipermail/devel/2010-August/010511.html
Signed-off-by: Vasu Dev <vasu.dev@intel.com> Signed-off-by: Robert Love <robert.w.love@intel.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Vasu Dev [Sat, 9 Oct 2010 00:12:20 +0000 (17:12 -0700)]
[SCSI] libfc: adds flogi retry in case DID is zero in RJT
Sometimes switch in NPV mode rejects flogi request with DID
zero and in that case flogi is not tried again and port
remains offline, so this patch validates DID for non zero
along with only ACC response to allow flogi retry
for RJT with DID=0 also succeed FLOGI in next try.
Signed-off-by: Vasu Dev <vasu.dev@intel.com> Signed-off-by: Robert Love <robert.w.love@intel.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Vasu Dev [Sat, 9 Oct 2010 00:12:15 +0000 (17:12 -0700)]
[SCSI] libfc: use DID_TRANSPORT_DISRUPTED while lport not ready
This is per Mile Christie feedback since in this case IO
could get retried for tape devices and therefore DID_REQUEUE
cannot be used, more details in this thread.
Signed-off-by: Vasu Dev <vasu.dev@intel.com> Signed-off-by: Robert Love <robert.w.love@intel.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Mike Christie [Sat, 9 Oct 2010 00:12:10 +0000 (17:12 -0700)]
[SCSI] libfc: fix setting of rport dev loss
There does not seem to be a reason why libfc adds a 5
second delay to the user requested value for the dev loss
tmo. There also does not seem to be a reason to allow
setting it to 0 (or really close).
This patch removes the extra 5 sec delay, and for 0 it
sets it to 1 like other fc drivers. We should actually
be able to set it to 0 since the queue_delayed_work API
will just call queue_work, but other drivers set it to 1 in
that case.
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu> Signed-off-by: Robert Love <robert.w.love@intel.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Dan Carpenter [Fri, 8 Oct 2010 07:03:07 +0000 (09:03 +0200)]
[SCSI] gdth: integer overflow in ioctl
gdth_ioctl_alloc() takes the size variable as an int.
copy_from_user() takes the size variable as an unsigned long.
gen.data_len and gen.sense_len are unsigned longs.
On x86_64 longs are 64 bit and ints are 32 bit.
We could pass in a very large number and the allocation would truncate
the size to 32 bits and allocate a small buffer. Then when we do the
copy_from_user(), it would result in a memory corruption.
CC: stable@kernel.org Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
The function __scsi_remove_target iterates through the SCSI devices on
the host, but it drops the host_lock before calling
scsi_remove_device. When the SCSI device is deleted from another
thread, the pointer to the SCSI device in scsi_remove_device can
become invalid. Fix this by getting a reference to the SCSI device
before dropping the host_lock to keep the SCSI device alive for the
call to scsi_remove_device.
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com> Cc: Stable Tree <stable@kernel.org> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
[SCSI] sd: Export effective protection mode in sysfs
Create a sysfs entry that reports the negotiated DIX/DIF protection mode
for a SCSI disk. This depends on the protection type the disk is
formatted with as well as the protection capabilities advertised by the
controller.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Mike Christie [Thu, 7 Oct 2010 05:51:17 +0000 (22:51 -0700)]
[SCSI] qla4xxx: grab hardware_lock in eh_abort before accessing srb
grab hardware_lock in eh_abort before accessing srb to avoid
race between command completion and get refcount on srb.
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu> Signed-off-by: Vikas Chaudhary <vikas.chaudhary@qlogic.com> Signed-off-by: Ravi Anand <ravi.anand@qlogic.com> Reviewed-by: Mike Christie <michaelc@cs.wisc.edu> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Vikas Chaudhary [Thu, 7 Oct 2010 05:50:56 +0000 (22:50 -0700)]
[SCSI] qla4xxx: update AER support for ISP82XX
* Cleanup qla4xxx_pci_mmio_enabled():
don't want to return PCI_ERS_NEED_RESET if firmware hung.
IDC will take care of it.
* Request irq after initialize_adapter() in qla82xx_error_recovery().
* Return all active commands from qla4xxx_pci_error_detected().
* Cleanup ql4_def.h
Signed-off-by: Vikas Chaudhary <vikas.chaudhary@qlogic.com> Signed-off-by: Ravi Anand <ravi.anand@qlogic.com> Reviewed-by: Mike Christie <michaelc@cs.wisc.edu> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Shyam Sundar [Thu, 7 Oct 2010 05:49:40 +0000 (22:49 -0700)]
[SCSI] qla4xxx: Add support for 8130/8131 AENs.
AEN 8130 Corresponds to an event representing the insertion (detection)
of a transceiver. It also reports the type of the SFP+.
AEN 8131 corresponds to the removal of a transceiver.
Signed-off-by: Vikas Chaudhary <vikas.chaudhary@qlogic.com> Signed-off-by: Shyam Sundar <shyam.sundar@qlogic.com> Signed-off-by: Ravi Anand <ravi.anand@qlogic.com> Reviewed-by: Mike Christie <michaelc@cs.wisc.edu> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
The seconds_since_last_heartbeat should be checked for consecutive
heartbeat checks. Currently it could happen that it gets set to
max (2 seconds) for non-consecutive heartbeat checks.
Signed-off-by: Vikas Chaudhary <vikas.chaudhary@qlogic.com> Signed-off-by: Lalit Chandivade <lalit.chandivade@qlogic.com> Signed-off-by: Ravi Anand <ravi.anand@qlogic.com> Reviewed-by: Mike Christie <michaelc@cs.wisc.edu> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Vikas Chaudhary [Thu, 7 Oct 2010 05:48:24 +0000 (22:48 -0700)]
[SCSI] qla4xxx: remove "ha->retry_reset_ha_cnt" from wait_for_hba_online
remove "ha->retry_reset_ha_cnt" from wait_for_hba_online as its
initialize to zero at driver init time so it could always return
QLA_ERROR from wait_for_hba_online() without waiting for hba to
come online.
Signed-off-by: Vikas Chaudhary <vikas.chaudhary@qlogic.com> Signed-off-by: Ravi Anand <ravi.anand@qlogic.com> Reviewed-by: Mike Christie <michaelc@cs.wisc.edu> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Vikas Chaudhary [Thu, 7 Oct 2010 05:47:48 +0000 (22:47 -0700)]
[SCSI] qla4xxx: Trivial cleanup
* cleanup function qla4xxx_recovery_timeout
- No need to wakeup dpc thread from function
qla4xxx_recovery_timeout() as we are not doing anything
in do_dpc() thread when wakeup from
qla4xxx_recovery_timeout()
* cleanup function qla4xxx_wait_for_hba_online
- Remove hard coded value from qla4xxx_wait_for_hba_online().
* cleanup function qla4xxx_start_firmware_from_flash
- display seconds
* cleanup function qla4_8xxx_load_risc
- Remove redundant code.
* cleanup function qla4xxx_get_firmware_status
- update debug statement
* cleanup function qla4_8xxx_try_start_fw
- update return status
Signed-off-by: Vikas Chaudhary <vikas.chaudhary@qlogic.com> Signed-off-by: Ravi Anand <ravi.anand@qlogic.com> Reviewed-by: Mike Christie <michaelc@cs.wisc.edu> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
During the delete, the SCSI device is in moved to SDEV_CANCEL. When
the FC transport class later calls scsi_target_unblock, this has no
effect, since scsi_internal_device_unblock ignores SCSI devics in this
state.
Fix by rejecting offline and cancel in the state transition.
Signed-off-by: Christof Schmitt <christof.schmitt@de.ibm.com>
[jejb: Original patch by Christof Schmitt, modified by Mike Christie] Cc: Stable Tree <stable@kernel.org> Signed-off-by: James Bottomley <James.Bottomley@suse.de>
[SCSI] zfcp: Call get_device on port before calling put_device
zfcp_unit_release calls put_device on the port. Ensure that get_device
has been called before possibly triggering the release function
through put_device or device_unregister.
Swen Schillig [Tue, 5 Oct 2010 15:12:53 +0000 (17:12 +0200)]
[SCSI] zfcp: Fix adapter activation on link down
If an exchange config is executed while the local link is down, the
request succeeds but the returned data is incomplete. Proceeding with
the adapter activation is leading to an unpredictable behaviour (e.g.
kernel panic) caused by invalid values. In such a scenario the
recommended ERP is to retry the action and wait for a link up event.
If the issue persists the activation has to fail.
Linus Torvalds [Sun, 24 Oct 2010 20:06:57 +0000 (13:06 -0700)]
Merge branch 'devel' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/edac
* 'devel' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/edac: (25 commits)
i7300_edac: Properly initialize per-csrow memory size
V4L/DVB: i7300_edac: better initialize page counts
MAINTAINERS: Add maintainer for i7300-edac driver
i7300-edac: CodingStyle cleanup
i7300_edac: Improve comments
i7300_edac: Cleanup: reorganize the file contents
i7300_edac: Properly detect channel on CE errors
i7300_edac: enrich FBD error info for corrected errors
i7300_edac: enrich FBD error info for fatal errors
i7300_edac: pre-allocate a buffer used to prepare err messages
i7300_edac: Fix MTR x4/x8 detection logic
i7300_edac: Make the debug messages coherent with the others
i7300_edac: Cleanup: remove get_error_info logic
i7300_edac: Add a code to cleanup error registers
i7300_edac: Add support for reporting FBD errors
i7300_edac: Properly detect the type of error correction
i7300_edac: Detect if the device is on single mode
i7300_edac: Adds detection for enhanced scrub mode on x8
i7300_edac: Clear the error bit after reading
i7300_edac: Add error detection code for global errors
...
Linus Torvalds [Sun, 24 Oct 2010 19:47:55 +0000 (12:47 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/penberg/slab-2.6
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/penberg/slab-2.6: (27 commits)
SLUB: Fix memory hotplug with !NUMA
slub: Move functions to reduce #ifdefs
slub: Enable sysfs support for !CONFIG_SLUB_DEBUG
SLUB: Optimize slab_free() debug check
slub: Move NUMA-related functions under CONFIG_NUMA
slub: Add lock release annotation
slub: Fix signedness warnings
slub: extract common code to remove objects from partial list without locking
SLUB: Pass active and inactive redzone flags instead of boolean to debug functions
slub: reduce differences between SMP and NUMA
Revert "Slub: UP bandaid"
percpu: clear memory allocated with the km allocator
percpu: use percpu allocator on UP too
percpu: reduce PCPU_MIN_UNIT_SIZE to 32k
vmalloc: pcpu_get/free_vm_areas() aren't needed on UP
SLUB: Fix merged slab cache names
Slub: UP bandaid
slub: fix SLUB_RESILIENCY_TEST for dynamic kmalloc caches
slub: Fix up missing kmalloc_cache -> kmem_cache_node case for memoryhotplug
slub: Add dummy functions for the !SLUB_DEBUG case
...
Linus Torvalds [Sun, 24 Oct 2010 19:47:25 +0000 (12:47 -0700)]
Merge branch 'kvm-updates/2.6.37' of git://git.kernel.org/pub/scm/virt/kvm/kvm
* 'kvm-updates/2.6.37' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (321 commits)
KVM: Drop CONFIG_DMAR dependency around kvm_iommu_map_pages
KVM: Fix signature of kvm_iommu_map_pages stub
KVM: MCE: Send SRAR SIGBUS directly
KVM: MCE: Add MCG_SER_P into KVM_MCE_CAP_SUPPORTED
KVM: fix typo in copyright notice
KVM: Disable interrupts around get_kernel_ns()
KVM: MMU: Avoid sign extension in mmu_alloc_direct_roots() pae root address
KVM: MMU: move access code parsing to FNAME(walk_addr) function
KVM: MMU: audit: check whether have unsync sps after root sync
KVM: MMU: audit: introduce audit_printk to cleanup audit code
KVM: MMU: audit: unregister audit tracepoints before module unloaded
KVM: MMU: audit: fix vcpu's spte walking
KVM: MMU: set access bit for direct mapping
KVM: MMU: cleanup for error mask set while walk guest page table
KVM: MMU: update 'root_hpa' out of loop in PAE shadow path
KVM: x86 emulator: Eliminate compilation warning in x86_decode_insn()
KVM: x86: Fix constant type in kvm_get_time_scale
KVM: VMX: Add AX to list of registers clobbered by guest switch
KVM guest: Move a printk that's using the clock before it's ready
KVM: x86: TSC catchup mode
...
Linus Torvalds [Sun, 24 Oct 2010 19:46:24 +0000 (12:46 -0700)]
Merge branch 'i2c-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jdelvare/staging
* 'i2c-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jdelvare/staging:
i2c-viapro: Don't log nacks
i2c/pca954x: Remove __devinit and __devexit from probe and remove functions
MAINTAINERS: Add maintainer for PCA9541 I2C bus master selector driver
i2c/mux: Driver for PCA9541 I2C Master Selector
i2c: Optimize function i2c_detect()
i2c: Discard warning message on device instantiation from user-space
i2c-amd8111: Add proper error handling
i2c: Change to new flag variable
i2c: Remove unneeded inclusions of <linux/i2c-id.h>
i2c: Let i2c_parent_is_i2c_adapter return the parent adapter
i2c: Simplify i2c_parent_is_i2c_adapter
i2c-pca-platform: Change device name of request_irq
i2c: Fix Kconfig dependencies
Linus Torvalds [Sun, 24 Oct 2010 19:44:59 +0000 (12:44 -0700)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid: (47 commits)
HID: fix mismerge in hid-lg
HID: hidraw: fix window in hidraw_release
HID: hid-sony: override usbhid_output_raw_report for Sixaxis
HID: add absolute axis resolution calculation
HID: force feedback support for Logitech RumblePad gamepad
HID: support STmicroelectronics and Sitronix with hid-stantuml driver
HID: magicmouse: Adjust major / minor axes to scale
HID: Fix for problems with eGalax/DWAV multi-touch-screen
HID: waltop: add support for Waltop Slim Tablet 12.1 inch
HID: add NOGET quirk for AXIS 295 Video Surveillance Joystick
HID: usbhid: remove unused hiddev_driver
HID: magicmouse: Use hid-input parsing rather than bypassing it
HID: trivial formatting fix
HID: Add support for Logitech Speed Force Wireless gaming wheel
HID: don't Send Feature Reports on Interrupt Endpoint
HID: 3m: Adjust major / minor axes to scale
HID: 3m: Correct touchscreen emulation
HID: 3m: Convert to MT slots
HID: 3m: Output proper orientation range
HID: 3m: Adjust to sequential MT HID protocol
...
* git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
crypto: Makefile - replace the use of <module>-objs with <module>-y
crypto: hifn_795x - use cancel_delayed_work_sync()
crypto: talitos - sparse check endian fixes
crypto: talitos - fix checkpatch warning
crypto: talitos - fix warning: 'alg' may be used uninitialized in this function
crypto: cryptd - Adding the AEAD interface type support to cryptd
crypto: n2_crypto - Niagara2 driver needs to depend upon CRYPTO_DES
crypto: Kconfig - update broken web addresses
crypto: omap-sham - Adjust DMA parameters
crypto: fips - FIPS requires algorithm self-tests
crypto: omap-aes - OMAP2/3 AES hw accelerator driver
crypto: updates to enable omap aes
padata: add missing __percpu markup in include/linux/padata.h
MAINTAINERS: Add maintainer entries for padata/pcrypt
Jean Delvare [Sun, 24 Oct 2010 16:16:59 +0000 (18:16 +0200)]
i2c-viapro: Don't log nacks
Transactions not acked can happen every now and then, in particular
during device detection, and various transaction types can be used for
this purpose. So stop logging this event, except when debugging is
enabled. This is what other similar drivers (e.g. i2c-i801 or
i2c-piix4) do.
Guenter Roeck [Sun, 24 Oct 2010 16:16:59 +0000 (18:16 +0200)]
i2c/pca954x: Remove __devinit and __devexit from probe and remove functions
The underlying I2C adapter may or may not be present when this driver
gets initialized, and may disappear later, so there is no safe time at
which the probe and remove functions can be discarded.
Signed-off-by: Guenter Roeck <guenter.roeck@ericsson.com> Signed-off-by: Jean Delvare <khali@linux-fr.org>
Guenter Roeck [Sun, 24 Oct 2010 16:16:58 +0000 (18:16 +0200)]
i2c/mux: Driver for PCA9541 I2C Master Selector
This patch adds support for PCA9541, an I2C Bus Master Selector.
The driver is modeled as single channel I2C Multiplexer to be able to utilize
the I2C multiplexer framework.
Signed-off-by: Guenter Roeck <guenter.roeck@ericsson.com> Reviewed-by: Tom Grennan <tom.grennan@ericsson.com> Acked-by: Jean Delvare <khali@linux-fr.org>
Jean Delvare [Sun, 24 Oct 2010 16:16:58 +0000 (18:16 +0200)]
i2c: Discard warning message on device instantiation from user-space
The "new_device" sysfs interface has been there for quite some time
now, nobody complained about it so it must be good enough. Time to
remove the warning and call it stable.
Signed-off-by: Jean Delvare <khali@linux-fr.org> Acked-by: Michael Lawnick <ml.lawnick@gmx.de>
Julia Lawall [Sun, 24 Oct 2010 16:16:58 +0000 (18:16 +0200)]
i2c-amd8111: Add proper error handling
The functions the functions amd_ec_wait_write and amd_ec_wait_read have an
unsigned return type, but return a negative constant to indicate an error
condition.
A sematic match that finds this problem is as follows:
(http://coccinelle.lip6.fr/)
Fixing amd_ec_wait_write and amd_ec_wait_read leads to the need to adjust
the return type of the functions amd_ec_write and amd_ec_read, which are
the only functions that call amd_ec_wait_write and amd_ec_wait_read.
amd_ec_write and amd_ec_read, in turn, are only called from within the
function amd8111_access, which already returns a signed typed value. Each
of the calls to amd_ec_write and amd_ec_read are updated using the
following semantic patch:
// <smpl>
@@
@@
+ status = amd_ec_write
- amd_ec_write
(...);
+ if (status) return status;
@@
@@
+ status = amd_ec_read
- amd_ec_read
(...);
+ if (status) return status;
// </smpl>
The patch also adds the declaration of the status variable.
Signed-off-by: Julia Lawall <julia@diku.dk> Signed-off-by: Jean Delvare <khali@linux-fr.org>
Jean Delvare [Sun, 24 Oct 2010 16:16:58 +0000 (18:16 +0200)]
i2c: Remove unneeded inclusions of <linux/i2c-id.h>
These drivers don't use anything which is defined in <linux/i2c-id.h>.
This header file was never meant to be included directly anyway, and
will be deleted soon.
Signed-off-by: Jean Delvare <khali@linux-fr.org> Acked-by: Ben Dooks <ben-linux@fluff.org> Acked-by: Dave Airlie <airlied@linux.ie> Cc: Hans Verkuil <hverkuil@xs4all.nl>
Jean Delvare [Sun, 24 Oct 2010 16:16:57 +0000 (18:16 +0200)]
i2c: Fix Kconfig dependencies
drivers/i2c/algos/Kconfig makes all the algorithms dependent on
!I2C_HELPER_AUTO, which triggers a Kconfig warning about broken
dependencies when some driver selects one of the algorithms. Ideally
we would make only the prompts dependent on !I2C_HELPER_AUTO, however
Kconfig doesn't currently support that. So we have to redefine the
symbols separately for the I2C_HELPER_AUTO=y case.
Signed-off-by: Jean Delvare <khali@linux-fr.org> Acked-by: Michal Marek <mmarek@suse.cz>
Jan Kiszka [Mon, 18 Oct 2010 13:38:40 +0000 (15:38 +0200)]
KVM: Drop CONFIG_DMAR dependency around kvm_iommu_map_pages
We also have to call kvm_iommu_map_pages for CONFIG_AMD_IOMMU. So drop
the dependency on Intel IOMMU, kvm_iommu_map_pages will be a nop anyway
if CONFIG_IOMMU_API is not defined.
KVM-Stable-Tag. Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Huang Ying [Fri, 8 Oct 2010 08:24:15 +0000 (16:24 +0800)]
KVM: MCE: Send SRAR SIGBUS directly
Originally, SRAR SIGBUS is sent to QEMU-KVM via touching the poisoned
page. But commit 96054569190bdec375fe824e48ca1f4e3b53dd36 prevents the
signal from being sent. So now the signal is sent via
force_sig_info_fault directly.
[marcelo: use send_sig_info instead]
Reported-by: Dean Nelson <dnelson@redhat.com> Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Huang Ying [Fri, 8 Oct 2010 08:24:14 +0000 (16:24 +0800)]
KVM: MCE: Add MCG_SER_P into KVM_MCE_CAP_SUPPORTED
Now we have MCG_SER_P (and corresponding SRAO/SRAR MCE) support in
kernel and QEMU-KVM, the MCG_SER_P should be added into
KVM_MCE_CAP_SUPPORTED to make all these code really works.
Reported-by: Dean Nelson <dnelson@redhat.com> Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Avi Kivity [Mon, 4 Oct 2010 10:55:49 +0000 (12:55 +0200)]
KVM: Disable interrupts around get_kernel_ns()
get_kernel_ns() wants preemption disabled. It doesn't make a lot of sense
during the get/set ioctls (no way to make them non-racy) but the callee wants
it.
Jan Kiszka [Sun, 26 Sep 2010 11:00:53 +0000 (13:00 +0200)]
KVM: x86: Fix constant type in kvm_get_time_scale
Older gcc versions complain about the improper type (for x86-32), 4.5
seems to fix this silently. However, we should better use the right type
initially.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by: Avi Kivity <avi@redhat.com>
Negate the effects of AN TYM spell while kvm thread is preempted by tracking
conversion factor to the highest TSC rate and catching the TSC up when it has
fallen behind the kernel view of time. Note that once triggered, we don't
turn off catchup mode.
A slightly more clever version of this is possible, which only does catchup
when TSC rate drops, and which specifically targets only CPUs with broken
TSC, but since these all are considered unstable_tsc(), this patch covers
all necessary cases.
The math in kvm_get_time_scale relies on the fact that
NSEC_PER_SEC < 2^32. To use the same function to compute
arbitrary time scales, we must extend the first reduction
step to shrink the base rate to a 32-bit value, and
possibly reduce the scaled rate into a 32-bit as well.
Note we must take care to avoid an arithmetic overflow
when scaling up the tps32 value (this could not happen
with the fixed scaled value of NSEC_PER_SEC, but can
happen with scaled rates above 2^31.
Avi Kivity [Tue, 21 Sep 2010 17:59:44 +0000 (19:59 +0200)]
KVM: cpu_relax() during spin waiting for reboot
It doesn't really matter, but if we spin, we should spin in a more relaxed
manner. This way, if something goes wrong at least it won't contribute to
global warming.
Signed-off-by: Avi Kivity <avi@redhat.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Mohammed Gamal [Sun, 19 Sep 2010 12:34:06 +0000 (14:34 +0200)]
KVM: Add kvm_inject_realmode_interrupt() wrapper
This adds a wrapper function kvm_inject_realmode_interrupt() around the
emulator function emulate_int_real() to allow real mode interrupt injection.
[avi: initialize operand and address sizes before emulating interrupts]
[avi: initialize rip for real mode interrupt injection]
[avi: clear interrupt pending flag after emulating interrupt injection]
Signed-off-by: Mohammed Gamal <m.gamal005@gmail.com> Signed-off-by: Avi Kivity <avi@redhat.com> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
KVM: SVM: do not generate "external interrupt exit" if other exit is pending
Nested SVM checks for external interrupt after injecting nested exception.
In case there is external interrupt pending the code generates "external
interrupt exit" and overwrites previous exit info. If previously injected
exception already generated exit it will be lost.
Avi Kivity [Sun, 19 Sep 2010 16:44:07 +0000 (18:44 +0200)]
KVM: Convert PIC lock from raw spinlock to ordinary spinlock
The PIC code used to be called from preempt_disable() context, which
wasn't very good for PREEMPT_RT. That is no longer the case, so move
back from raw_spinlock_t to spinlock_t.
Signed-off-by: Avi Kivity <avi@redhat.com> Acked-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
If preempted after kvmclock values are updated, but before hardware
virtualization is entered, the last tsc time as read by the guest is
never set. It underflows the next time kvmclock is updated if there
has not yet been a successful entry / exit into hardware virt.
Fix this by simply setting last_tsc to the newly read tsc value so
that any computed nsec advance of kvmclock is nulled.
Avi Kivity [Tue, 20 Jul 2010 12:06:17 +0000 (15:06 +0300)]
KVM: Non-atomic interrupt injection
Change the interrupt injection code to work from preemptible, interrupts
enabled context. This works by adding a ->cancel_injection() operation
that undoes an injection in case we were not able to actually enter the guest
(this condition could never happen with atomic injection).
Avi Kivity [Tue, 20 Jul 2010 11:43:23 +0000 (14:43 +0300)]
KVM: VMX: Parameterize vmx_complete_interrupts() for both exit and entry
Currently vmx_complete_interrupts() can decode event information from vmx
exit fields into the generic kvm event queues. Make it able to decode
the information from the entry fields as well by parametrizing it.
Avi Kivity [Tue, 20 Jul 2010 11:31:20 +0000 (14:31 +0300)]
KVM: VMX: Split up vmx_complete_interrupts()
vmx_complete_interrupts() does too much, split it up:
- vmx_vcpu_run() gets the "cache important vmcs fields" part
- a new vmx_complete_atomic_exit() gets the parts that must be done atomically
- a new vmx_recover_nmi_blocking() does what its name says
- vmx_complete_interrupts() retains the event injection recovery code
This helps in reducing the work done in atomic context.
Avi Kivity [Tue, 27 Jul 2010 09:30:24 +0000 (12:30 +0300)]
KVM: Check for pending events before attempting injection
Instead of blindly attempting to inject an event before each guest entry,
check for a possible event first in vcpu->requests. Sites that can trigger
event injection are modified to set KVM_REQ_EVENT:
- interrupt, nmi window opening
- ppr updates
- i8259 output changes
- local apic irr changes
- rflags updates
- gif flag set
- event set on exit
This improves non-injecting entry performance, and sets the stage for
non-atomic injection.
Avi Kivity [Mon, 13 Sep 2010 14:45:28 +0000 (16:45 +0200)]
KVM: MMU: Fix regression with ept memory types merged into non-ept page tables
Commit "KVM: MMU: Make tdp_enabled a mmu-context parameter" made real-mode
set ->direct_map, and changed the code that merges in the memory type depend
on direct_map instead of tdp_enabled. However, in this case what really
matters is tdp, not direct_map, since tdp changes the pte format regardless
of whether the mapping is direct or not.
As a result, real-mode shadow mappings got corrupted with ept memory types.
The result was a huge slowdown, likely due to the cache being disabled.
Change it back as the simplest fix for the regression (real fix is to move
all that to vmx code, and not use tdp_enabled as a synonym for ept).
KVM: X86: Report SVM bit to userspace only when supported
This patch fixes a bug in KVM where it _always_ reports the
support of the SVM feature to userspace. But KVM only
supports SVM on AMD hardware and only when it is enabled in
the kernel module. This patch fixes the wrong reporting.
Cc: stable@kernel.org Signed-off-by: Joerg Roedel <joerg.roedel@amd.com> Signed-off-by: Avi Kivity <avi@redhat.com>
KVM: SVM: Expect two more candiates for exit_int_info
This patch adds INTR and NMI intercepts to the list of
expected intercepts with an exit_int_info set. While this
can't happen on bare metal it is architectural legal and may
happen with KVMs SVM emulation.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com> Signed-off-by: Avi Kivity <avi@redhat.com>
KVM: SVM: Initialize Nested Nested MMU context on VMRUN
This patch adds code to initialize the Nested Nested Paging
MMU context when the L1 guest executes a VMRUN instruction
and has nested paging enabled in its VMCB.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com> Signed-off-by: Avi Kivity <avi@redhat.com>
With Nested Paging emulation the NX state between the two
MMU contexts may differ. To make sure that always the right
fault error code is recorded this patch moves the NX state
into struct kvm_mmu so that the code can distinguish between
L1 and L2 NX state.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com> Signed-off-by: Avi Kivity <avi@redhat.com>
KVM: MMU: Allow long mode shadows for legacy page tables
Currently the KVM softmmu implementation can not shadow a 32
bit legacy or PAE page table with a long mode page table.
This is a required feature for nested paging emulation
because the nested page table must alway be in host format.
So this patch implements the missing pieces to allow long
mode page tables for page table types.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com> Signed-off-by: Avi Kivity <avi@redhat.com>
This patch factors out the direct-mapping paths of the
mmu_alloc_roots function into a seperate function. This
makes it a lot easier to avoid all the unnecessary checks
done in the shadow path which may break when running direct.
In fact, this patch already fixes a problem when running PAE
guests on a PAE shadow page table.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com> Signed-off-by: Avi Kivity <avi@redhat.com>
This function is implemented to load the pdptr pointers of
the currently running guest (l1 or l2 guest). Therefore it
takes care about the current paging mode and can read pdptrs
out of l2 guest physical memory.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com> Signed-off-by: Avi Kivity <avi@redhat.com>