Merge branch 'topic/uek-4.1/aio-dio' of git://ca-git.us.oracle.com/linux-uek into uek/uek-4.1
* 'topic/uek-4.1/aio-dio' of git://ca-git.us.oracle.com/linux-uek:
block: loop: support DIO & AIO
block: loop: prepare for supporing direct IO
block: loop: use kthread_work
block: loop: set QUEUE_FLAG_NOMERGES for request queue of loop
fs: direct-io: don't dirtying pages for ITER_BVEC/ITER_KVEC direct read
nfs: don't dirty kernel pages read by direct-io
block: loop: avoiding too many pending per work I/O
block: loop: convert to per-device workqueue
Merge branch 'topic/uek-4.1/drivers' of git://ca-git.us.oracle.com/linux-uek into uek/uek-4.1
* 'topic/uek-4.1/drivers' of git://ca-git.us.oracle.com/linux-uek: (24 commits)
megaraid_sas: Permit large RAID0/1 requests
megaraid_sas : Modify return value of megasas_issue_blocked_cmd() and wait_and_poll() to consider command status returned by firmware
megaraid_sas : swap whole register in megasas_register_aen
megaraid_sas : fix megasas_fire_cmd_fusion calling convention
megaraid_sas : add missing byte swaps to the sriov code
megaraid_sas : bytewise or should be done on native endian variables
megaraid_sas : move endianness conversion into caller of megasas_get_seq_num
megaraid_sas : add endianness conversions for all ones
megaraid_sas : add endianness annotations
megaraid_sas : add missing __iomem annotations
megaraid_sas : megasas_complete_outstanding_ioctls() can be static
megaraid_sas : Support for Avago's Single server High Availability product
megaraid_sas : Add release date and update driver version
megaraid_sas : Modify driver's meta data to reflect Avago
megaraid_sas : Use Block layer tag support for internal command indexing
megaraid_sas : Enhanced few prints
megaraid_sas : Move controller's queue depth calculation in adapter specific function
megaraid_sas : Add separate functions for building sysPD IOs and non RW LDIOs
megaraid_sas : Add separate function for refiring MFI commands
megaraid_sas : Add separate function for setting up IRQs
...
Merge branch 'topic/uek-4.1/oracleasm' of git://ca-git.us.oracle.com/linux-uek into uek/uek-4.1
* 'topic/uek-4.1/oracleasm' of git://ca-git.us.oracle.com/linux-uek:
oracleasm: Fix trace output for warn_asm_ioc and check_asm_ioc
oracleasm: Fix occasional I/O stall due to merge error
oracleasm: Classify device connectivity issues as global errors
oracleasm: Deprecate mlog and implement support for tracepoints
oracleasm: Abolish mlog usage in integrity.c and clean up error printing.
oracleasm: Various code and whitespace cleanups.
oracleasm: 4.0 compat changes
oracleasm: Compat changes for 3.18
oracleasm: claim FMODE_EXCL access on disk during asm_open
oracleasm: Restrict logical block size reporting
oracleasm: Report logical block size
oracleasm: Compat changes for 3.10
oracleasm: Add support for new error return codes from block/SCSI
oracleasm: Compat changes for 3.8
oracleasm: Compat changes for 3.5
oracleasm: Introduce module parameter for block size selection
oracleasm: Data integrity support
oracleasm: Fix two merge errors
Oracle ASM Kernel Driver
Merge branch 'topic/uek-4.1/fuse' of git://ca-git.us.oracle.com/linux-uek into uek/uek-4.1
* 'topic/uek-4.1/fuse' of git://ca-git.us.oracle.com/linux-uek:
fuse: fix typo while displaying fuse numa mount option
fuse: add numa mount option
fuse: modify queues, allocation and locking for multiple nodes
fuse: add spinlock to protect fc reqctr
fuse: add fuse node struct
Merge branch 'topic/uek-4.1/ocfs2' of git://ca-git.us.oracle.com/linux-uek into uek/uek-4.1
* 'topic/uek-4.1/ocfs2' of git://ca-git.us.oracle.com/linux-uek:
ocfs2: call ocfs2_journal_access_di() before ocfs2_journal_dirty() in ocfs2_write_end_nolock()
ocfs2: avoid access invalid address when read o2dlm debug messages
ocfs2: make 'buffered' as the default coherency option
ocfs2: Suppress the error message from being printed in ocfs2_rename
ocfs2: Tighten free bit calculation in the global bitmap
ocfs2/trivial: Limit unaligned aio+dio write messages to once per day
ocfs2/trivial: Print message indicating unaligned aio+dio write
Merge branch 'topic/uek-4.1/uek-carry' of git://ca-git.us.oracle.com/linux-uek into uek/uek-4.1
* 'topic/uek-4.1/uek-carry' of git://ca-git.us.oracle.com/linux-uek:
cdc-acm: Increase number of devices to 64
ipmi: make kcs timeout parameters as module options
x86: perf: prevent spurious PMU NMIs on Haswell systems
x86/simplefb: simplefb was broken on Oracle and HP system, skip VIDEO_TYPE_EFI
x86, fpu: Avoid possible error in math_state_restore()
kernel: freezer: restore TIF_FREEZE
ksplice: Clear garbage data on the kernel stack when handling signals
sched: Disable default sched_autogroup to avoid the DBA performance regression
x86: add support for crashkernel=auto
Merge branch 'topic/uek-4.1/xen' of git://ca-git.us.oracle.com/linux-uek into uek/uek-4.1
* 'topic/uek-4.1/xen' of git://ca-git.us.oracle.com/linux-uek:
xen/microcode: Use dummy microcode_ops for non initial domain guest
xen/microcode: Fix compile warning.
microcode_xen: Add support for AMD family >= 15h
x86/microcode: check proper return code.
xen: add CPU microcode update driver
x86/xen: Disable APIC PM for Xen PV guests
xen/pvhvm: Support more than 32 VCPUs when migrating (v3).
yangwenfang [Fri, 30 Jan 2015 02:11:19 +0000 (13:11 +1100)]
ocfs2: call ocfs2_journal_access_di() before ocfs2_journal_dirty() in ocfs2_write_end_nolock()
After we call ocfs2_journal_access_di() in ocfs2_write_begin(),
jbd2_journal_restart() may also be called, in this function transaction
A's t_updates-- and obtains a new transaction B. If
jbd2_journal_commit_transaction() is happened to commit transaction A,
when t_updates==0, it will continue to complete commit and unfile buffer.
So when jbd2_journal_dirty_metadata(), the handle is pointed a new
transaction B, and the buffer head's journal head is already freed,
jh->b_transaction == NULL, jh->b_next_transaction == NULL, it returns
EINVAL, So it triggers the BUG_ON(status).
So I think we should put ocfs2_journal_access_di before
ocfs2_journal_dirty in the ocfs2_write_end. and it works well after my
modification.
Signed-off-by: vicky <vicky.yangwenfang@huawei.com> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <jlbec@evilplan.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
(cherry picked from commit 19bf7feab124221625b5c811b6192fff4e0cbb96)
jiangyiwen [Fri, 30 Jan 2015 02:11:19 +0000 (13:11 +1100)]
ocfs2: avoid access invalid address when read o2dlm debug messages
The following case will lead to a lockres is freed but is still in use.
cat /sys/kernel/debug/o2dlm/locking_state dlm_thread
lockres_seq_start
-> lock dlm->track_lock
-> get resA
resA->refs decrease to 0,
call dlm_lockres_release,
and wait for "cat" unlock.
Although resA->refs is already set to 0,
increase resA->refs, and then unlock
lock dlm->track_lock
-> list_del_init()
-> unlock
-> free resA
In such a race case, invalid address access may occurs. So we should
delete list res->tracking before resA->refs decrease to 0.
Signed-off-by: jiangyiwen <jiangyiwen@huawei.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Mark Fasheh <mfasheh@suse.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
(cherry picked from commit e87e805fe4a1cf38031ae0669e3a91c8a8251279)
Customers upgrading to uek2 and above will see the default coherency option
set to 'full' which impacts -ve performance. This patch changes coherence
option to buffered which keeps the default behaviour same as old(UEK1).
If an application that does direct i/o needs cache coherency then they can
use mount option 'coherency=full'
Signed-off-by: Wengang Wang <wen.gang.wang@oracle.com> Acked-by: Srinivas Eeda <srinivas.eeda@oracle.com>
(cherry picked from commit 020a20029508d2d7f36470bebd23f053de4b0dbe)
Currently non initial domain guest use Intel or AMD specific ops.
This will also slow up the startup on heavily overcommited guests (say 256VCPUs
on 20 PCPU), as there are many read and write to x86 MSR registers which will
trap to xen during microcode update. Finally it will fail and report errors.
A dummy ops could fix that and also make udevd silent (bug18379824)
by augmenting the commit c18a317f6892536851e5852b6aaa4ef42cbc11a2
"xen/microcode: Only load under initial domain." which fell short of its
intended fix.
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@oracle.com> Acked-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
(cherry picked from commit 62b84234f23c1020c690d162b7d8250042425e1e) Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Conflicts:
arch/x86/kernel/microcode_core.c
Konrad Rzeszutek Wilk [Fri, 6 Jun 2014 18:09:14 +0000 (14:09 -0400)]
xen/microcode: Fix compile warning.
We get a bunch of them. Might as well fix it.
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
(cherry picked from commit 3b99e95e07cee4bbe3ba2b511fc2ac38ff7769b9) Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Ian Campbell [Mon, 26 Nov 2012 09:41:02 +0000 (09:41 +0000)]
microcode_xen: Add support for AMD family >= 15h
Signed-off-by: Ian Campbell <ijc@hellion.org.uk> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
(cherry picked from commit 8b080aa43b95719d8981ba06f357abb6f0ba9d52) Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Ben Guthro [Thu, 3 Nov 2011 15:06:56 +0000 (11:06 -0400)]
x86/microcode: check proper return code.
After pulling in this change from your tree, I found the following bug,
when checking an enum value, which should be considered before inclusion:
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
(cherry picked from commit 33a4651e09e0d6fb6e9c1293810d8a66b734840a) Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Jeremy Fitzhardinge [Sat, 28 Mar 2009 00:39:15 +0000 (17:39 -0700)]
xen: add CPU microcode update driver
Xen does all the hard work for us, including choosing the right update
method for this cpu type and actually doing it for all cpus. We just
need to supply it with the firmware blob.
Because Xen updates all CPUs (and the kernel's virtual cpu numbers have
no fixed relationship with the underlying physical cpus), we only bother
doing anything for cpu "0".
[ Impact: allow CPU microcode update in Xen dom0 ] Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Conflicts:
arch/x86/xen/Kconfig
(cherry picked from commit da3d1c83399886c443cbf9e57455bcc2e5caf28c) Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Conflicts:
arch/x86/include/asm/microcode.h
arch/x86/kernel/Makefile
arch/x86/kernel/microcode_core.c
arch/x86/xen/Kconfig
Boris Ostrovsky [Tue, 4 Mar 2014 02:40:21 +0000 (21:40 -0500)]
x86/xen: Disable APIC PM for Xen PV guests
Xen PV guests support only few APIC registers and writes to
unsupported registers result in WARN_ONs. Most APIC accesses in these
guests have been eliminated; however, lapic_suspend/resume are still
called (on 32-bit kernels).
We can disable APIC power management in xen_smp_prepare_boot_cpu()
(which is called after APIC has been initialized).
Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
xen/pvhvm: Support more than 32 VCPUs when migrating (v3).
When Xen migrates an HVM guest, by default its shared_info can
only hold up to 32 CPUs. As such the hypercall
VCPUOP_register_vcpu_info was introduced which allowed us to
setup per-page areas for VCPUs. This means we can boot PVHVM
guest with more than 32 VCPUs. During migration the per-cpu
structure is allocated freshly by the hypervisor (vcpu_info_mfn
is set to INVALID_MFN) so that the newly migrated guest
can make an VCPUOP_register_vcpu_info hypercall.
Unfortunatly we end up triggering this condition in Xen:
/* Run this command on yourself or on other offline VCPUS. */
if ( (v != current) && !test_bit(_VPF_down, &v->pause_flags) )
which means we are unable to setup the per-cpu VCPU structures
for running vCPUS. The Linux PV code paths make this work by
iterating over every vCPU with:
1) is target CPU up (VCPUOP_is_up hypercall?)
2) if yes, then VCPUOP_down to pause it.
3) VCPUOP_register_vcpu_info
4) if it was down, then VCPUOP_up to bring it back up
But since VCPUOP_down, VCPUOP_is_up, and VCPUOP_up are
not allowed on HVM guests we can't do this. However with the
git commit XYZ ("hvm: Support more than 32 VCPUS when migrating.")
we can do this. As such first check if VCPUOP_is_up is actually
possible before trying this dance.
As most of this dance code is done already in 'xen_setup_vcpu'
lets make it callable on both PV and HVM. This means moving one
of the checks out to 'xen_setup_runstate_info'.
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Signed-off-by: Joe Jin <joe.jin@oracle.com> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com> Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
For slow or heavily used BMC contollers the default wait timeouts for IBF or OBF
bits in the driver may not be sufficient. This may cause problems during more
complicated oem operatoins on the BMC side.
These timeoutsare changed from hardcoded values in the code into kernel
module parameters. The default values are kept unchanged.
Signed-off-by: Chris Mason <chris.mason@oracle.com> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com> Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
When "perf" is run on Haswell-based systems under UEK3, we've
noticed that "extra" NMIs are being generated by the Performance
Monitoring Unit (PMU).
The PMU contains counters that can count occurrences of certain
kinds of events, such as branch misses or instructions retired.
These counters can be programmed to issue an interrupt when they
reach certain pre-set values. linux uses vector 2, the NMI vector,
for these interrupts, so the PMU interrupts behave just like other
sources of NMIs such as watchdog timers. Each consumer of NMIs
within the kernel is responsible for identifying the interrupts
it's interested in.
In the current case, the linux PMU-support code is failing to
"claim" certain of the NMIs that are originating in the PMU.
What happens when no piece of kernel code claims an NMI is that
an ugly kernel message gets generated and, if the sysctl variable
"unknown_nmi_panic" is set nonzero (as it is by default on Exadata
systems), the system panics.
The current UEK3 PMU handler attempts to determine whether a given
NMI belongs to it by scanning the PMU hardware's potential NMI
sources to find out whether any of them has triggered. Apparently,
Haswell has potential NMI sources that are indeed getting triggered,
but of which the PMU handler is not aware.
This commit contains two measures designed to prevent these
extra NMIs.
First, we've moved the write to the local APIC's APIC_LVTPC register
from near the beginning of the PMU NMI handler to near the end.
Upstream has discovered empirically that this helps elminate
the spurious NMIs. See:
Second, this change takes advantage of a bit in the APIC_LVTPC
register that gets set when (and only when) a PMU-originated NMI
is being delivered to the CPU core. This bit is a "mask" bit,
which when set, disables delivery of these NMIs to the core.
Having processed an NMI, system software must clear this bit in
order to enable delivery of the next one.
The fix involves sampling this bit and claiming the NMI if it's a
PMU NMI, even if its origin has not been otherwise determined.
Note that this change also helps render the PMU NMI handler immune
to the addition of more sources to the PMUs on future CPUs.
Signed-off-by: Dan Duval <dan.duval@oracle.com> Reviewed-by: Santosh Shilimkar <santosh.shilimkar@oracle.com> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com> Signed-off-by: Dan Duval <dan.duval@oracle.com>
(cherry picked from commit ed921c01bcd2cad94dbd659ad2031a877e85acb8)
As descriped in https://bugzilla.kernel.org/show_bug.cgi?id=98721
When kernel 4.0.4 was tested on Oracle and HP system with UEFI mode, no output and
login on console.
Simplefb was broken on these systems when orig_video_isVGA is VIDEO_TYPE_EFI, so
skip it.
This patch was tested on Oracle Sun server X5-2 series and HP ProLiant DL380 Gen9
with kernel 4.0.4
Annie Li [Wed, 14 Jan 2015 04:53:18 +0000 (23:53 -0500)]
x86, fpu: Avoid possible error in math_state_restore()
For non-eager fpu mode, thread's fpu state is allocated during the first
fpu usage (in the context of device not available exception). This can be
a blocking call and hence we enable interrupts (which were originally
disabled when the exception happened), allocate memory and disable
interrupts etc.
Math_state_restore() is called from multiple places
and it is error pone if the caller expects interrupts to be disabled
throughout the execution of math_state_restore(). Can lead to subtle
bugs like Ubuntu bug #1265841. So simplifying the code which cause subtle
bugs.
The patch has one known problem when the machine is running baremetal
(or PVHVM) and when there is low amount of memory. The problem is that the
applications won't get SIGKILL when the FPU area can't be allocated and
instead they will continue on running - without any FPU context
allocated for them. The 'init_fpu(tsk)' can return -ENOMEM and that
patch does not check that condition. This update will be tracked
in another bug since the patch already fixes two known issues related
to corruption.
Sasha Levin [Thu, 15 Jan 2015 01:41:18 +0000 (20:41 -0500)]
ksplice: Clear garbage data on the kernel stack when handling signals
The garbage data can give false-positives for the Ksplice safety checks
making it difficult (or sometimes impossible) to apply the rebootless
updates. Clear the garbage with 0-words to avoid this.
Santosh Shilimkar [Mon, 9 Feb 2015 23:08:30 +0000 (15:08 -0800)]
sched: Disable default sched_autogroup to avoid the DBA performance regression
SCHED_AUTOGROUP optimizes the scheduler for common desktop workloads by
automatically creating and populating task groups. Though it helps desktop CPU
hungry workloads(linke build jobs), we found that it crteates 10% regerssion
on DBA perfromance.
Signed-off-by: Brian Maly <brian.maly@oracle.com> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com> Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Santosh Shilimkar [Wed, 3 Dec 2014 22:44:26 +0000 (14:44 -0800)]
uek-rpm: configs: Create baseline config for uek4[ol6/ol7]
The UEK4 kernel is based on kernel.org v4.1. Brief summary about
UEK4(4.1) new/added relevant Kernel Features w.r.t UEK3(v3.8)
For more details please refer the attachment associated with bugs
and the config files from UEK git tree.
- Block Multi-Queue support
- Low Latency Socket Poll
- Full Dynamic Tick
- SLUB Memory Allocator as a default
- Networking performance Improvements [Bulk network packet transmission]
- XEN Related updates: (PV Ticket locks, pvSCSI, Xen PVH guest and Xen-netback)
- SPARC architecture updates
- Automatic NUMA balancing turned ON default.
- CGROUP improvements
- CONFIG_SCHED_STATS is turned OFF from production kernels
- TRANSPARENT_HUGEPAGE enabled with IO problem being addressed in v3.18
- CMA allows to allocate big physically-contiguous blocks of memory.
- Data Center TCP (DCTCP)though it needs support of RFC3168(ECN)
- Open vSwitch support with GRE, VXLAN and GENEVE tunneling support
- nftables, the successor of iptables
- NFSv4.2 client support, NFSv4.1 client support for migration and Ceph client caching support.
- FOO over UDP and Virtual (secure) IPv6: tunneling
- Zswap and Bcache support.
- BTRFS improvements
- X86_INTEL_MPX
- X86_VSYSCALL_EMULATION
- NET_DSA_HWMON
- NET_FOU_IP_TUNNELS
- NET_SWITCHDEV
- IPVLAN
- BLK_DEV_RAM_DAX and FS_DAX
- I40E_FCOE
- NDSD_PNFS
- DM Multipath and MQ conversions
- Very basic support for PMEM
Martin K. Petersen [Wed, 27 May 2015 01:27:47 +0000 (21:27 -0400)]
oracleasm: Fix trace output for warn_asm_ioc and check_asm_ioc
The trace logic transposed the warn_asm_ioc and check_asm_ioc values. We
would treat the former as a flag and the latter as an integer. Fix this
so we print warning error codes and integrity buffer presence correctly.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Martin K. Petersen [Wed, 27 May 2015 01:15:34 +0000 (21:15 -0400)]
oracleasm: Fix occasional I/O stall due to merge error
Commit c05b6f12aae5 (oracleasm: Deprecate mlog and implement support for
tracepoints) inadvertently changed the maybe_wait_io logic so that we
would occasionally hang while waiting for I/O completion. Make sure we
only return when there is an actual error.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Martin K. Petersen [Thu, 21 May 2015 21:42:15 +0000 (17:42 -0400)]
oracleasm: Classify device connectivity issues as global errors
We used to set the ASM_LOCAL_ERROR qualifier when we got ENOLINK, EBADE
or ENODEV status from the storage stack. The assumption was that the
error could be caused by a pulled cable or a bad switch port and that
other nodes in a cluster might still have access to the storage.
The ASM team would prefer these types of errors to be treated as global,
however, as this would be consistent with database behavior when ASMLIB
is not in the picture.
Remove the ASM_LOCAL_ERROR flag from the device connectivity error code
path.
Martin K. Petersen [Tue, 5 May 2015 13:44:45 +0000 (09:44 -0400)]
oracleasm: Deprecate mlog and implement support for tracepoints
- Replace mlog_bug_on_msg() with BUG_ON()
- Remove mlog_entry() and mlog_exit() calls
- Introduce tracing to replace the important mlog() calls. The
following tracepoints are available:
oracleasm:disk Disk setup and teardown
oracleasm:req Internal request setup and teardown
oracleasm:bio Bios submitted to the I/O stack
oracleasm:ioc I/O descriptors from the RDBMS
oracleasm:integrity Data integrity payload setup
oracleasm:querydisk Disk properties
- Remove masklog.* and proc filesystem registration code
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
asm_open_disk should take exclusive access on asm disk during open to
prevent it from getting deleted while in use.
Signed-off-by: Srinivas Eeda <srinivas.eeda@oracle.com> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Martin K. Petersen [Thu, 24 Apr 2014 19:36:42 +0000 (15:36 -0400)]
oracleasm: Restrict logical block size reporting
Hanlin pointed out that we should only report the additional logical
block size when we are in physical block size mode. When the
"use_logical_block_size" module parameter is in use we should revert to
the old behavior.
Reported-by: Hanlin Chien <hanlin.qian@oracle.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Martin K. Petersen [Fri, 21 Mar 2014 19:20:19 +0000 (15:20 -0400)]
oracleasm: Report logical block size
Report the device's logical block size in the qd_feature variable. This
allows ASM to determine whether a disk group can be imported should a
storage device change its physical block size reporting.
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Martin K. Petersen [Wed, 28 Nov 2012 14:36:23 +0000 (06:36 -0800)]
oracleasm: Introduce module parameter for block size selection
Orabug: 15924773
We have encountered a few devices which after a firmware update
communicate different characteristics to the OS. In particular, some
devices begin to report their physical block size. This in turn will
cause oracleasm to report a different block size to ASM and mounting the
disk group will fail.
Introduce a module parameter which permits the logical block size to be
reported instead of the physical.
Ming Lei [Sun, 28 Jun 2015 06:06:32 +0000 (23:06 -0700)]
block: loop: support DIO & AIO
There are about 3 advantages to use direct I/O and AIO on
read/write loop's backing file:
1) double cache can be avoided, then memory usage gets
decreased a lot
2) not like user space direct I/O, there isn't cost of
pinning pages
3) avoid context switch for obtaining good throughput
- in buffered file read, random I/O top throughput is often obtained
only if they are submitted concurrently from lots of tasks; but for
sequential I/O, most of times they can be hit from page cache, so
concurrent submissions often introduce unnecessary context switch
and can't improve throughput much. There was such discussion[1]
to use non-blocking I/O to improve the problem for application.
- with direct I/O and AIO, concurrent submissions can be
avoided and random read throughput can't be affected meantime
Follows my fio test result:
1. 16 jobs fio test inside ext4 file system over loop block
1) How to run
- linux kernel: 4.1.0-rc2-next-20150506 with the patchset
- the loop block is over one image on HDD.
- linux psync, 16 jobs, size 400M, ext4 over loop block
- test result: IOPS from fio output
2) Throughput result:
-------------------------------------------------------------
test cases |randread |read |randwrite |write |
-------------------------------------------------------------
base |240 |8705 |3763 |20914
-------------------------------------------------------------
base+loop aio |242 |9258 |4577 |21451
-------------------------------------------------------------
3) context switch
- context switch decreased by ~16% with loop aio for randread,
and decreased by ~33% for read
4) memory usage
- After these four tests with loop aio: ~10% memory becomes used
- After these four tests without loop aio: more than 55% memory
becomes used
2. single job fio test inside ext4 file system over loop block(for Maxim Patlasov)
1) How to run
- linux kernel: 4.1.0-rc2-next-20150506 with the patchset
- the loop block is over one image on HDD.
- linux psync, 1 job, size 4000M, ext4 over loop block
- test result: IOPS from fio output
2) Throughput result:
-------------------------------------------------------------
test cases |randread |read |randwrite |write |
-------------------------------------------------------------
base |109 |21180 |4192 |22782
-------------------------------------------------------------
base+loop aio |114 |21018 |5404 |22670
-------------------------------------------------------------
3) context switch
- context switch decreased by ~10% with loop aio for randread,
and decreased by ~50% for read
4) memory usage
- After these four tests with loop aio: ~10% memory becomes used
- After these four tests without loop aio: more than 55% memory
becomes used
Both 'context switch' and 'memory usage' data are got from sar.
[1] https://lwn.net/Articles/612483/
[2] sar graph when running fio over loop without the patchset
http://kernel.ubuntu.com/~ming/block/loop-aio/v3/lo-nonaio.pdf
[3] sar graph when running fio over loop with the patchset
http://kernel.ubuntu.com/~ming/block/loop-aio/v3/lo-aio.pdf
[4] sar graph when running fio over loop without the patchset
http://kernel.ubuntu.com/~ming/block/loop-aio/v3/lo-nonaio-1job.pdf
[5] sar graph when running fio over loop with the patchset
http://kernel.ubuntu.com/~ming/block/loop-aio/v3/lo-aio-1job.pdf
Signed-off-by: Ming Lei <ming.lei@canonical.com> Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
Ming Lei [Sun, 28 Jun 2015 06:06:32 +0000 (23:06 -0700)]
block: loop: use kthread_work
The following patch will use dio/aio to submit IO to backing file,
then it needn't to schedule IO concurrently from work, so
use kthread_work for decreasing context switch cost a lot.
For non-AIO case, single thread has been used for long long time,
and it was just converted to work in v4.0, which has caused performance
regression for fedora live booting already. In discussion[1], even
though submitting I/O via work concurrently can improve random read IO
throughput, meantime it might hurt sequential read IO performance, so
better to restore to single thread behaviour.
For the following AIO support, it is better to use multi hw-queue
with per-hwq kthread than current work approach suppose there is so
high performance requirement for loop.
[1] http://marc.info/?t=143082678400002&r=1&w=2 Signed-off-by: Ming Lei <ming.lei@canonical.com> Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
Ming Lei [Sun, 28 Jun 2015 06:06:32 +0000 (23:06 -0700)]
fs: direct-io: don't dirtying pages for ITER_BVEC/ITER_KVEC direct read
When direct read IO is submitted from kernel, it is often
unnecessary to dirty pages, for example of loop, dirtying pages
have been considered in the upper filesystem(over loop) side
already, and they don't need to be dirtied again.
So this patch doesn't dirtying pages for ITER_BVEC/ITER_KVEC
direct read, and loop should be the 1st case to use ITER_BVEC/ITER_KVEC
for direct read I/O.
The patch is based on previous Dave's patch.
Cc: Dave Kleikamp <dave.kleikamp@oracle.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Ming Lei <ming.lei@canonical.com> Signed-off-by: Santosh Shilimkar <ssantosh@kernel.org>
Ming Lei [Tue, 5 May 2015 11:49:55 +0000 (19:49 +0800)]
block: loop: avoiding too many pending per work I/O
If there are too many pending per work I/O, too many
high priority work thread can be generated so that
system performance can be effected.
This patch limits the max_active parameter of workqueue as 16.
This patch fixes Fedora 22 live booting performance
regression when it is booted from squashfs over dm
based on loop, and looks the following reasons are
related with the problem:
- not like other filesyststems(such as ext4), squashfs
is a bit special, and I observed that increasing I/O jobs
to access file in squashfs only improve I/O performance a
little, but it can make big difference for ext4
- nested loop: both squashfs.img and ext3fs.img are mounted
as loop block, and ext3fs.img is inside the squashfs
- during booting, lots of tasks may run concurrently
Fixes: b5dd2f6047ca108001328aac0e8588edd15f1778 Cc: stable@vger.kernel.org (v4.0) Cc: Justin M. Forbes <jforbes@fedoraproject.org> Signed-off-by: Ming Lei <ming.lei@canonical.com> Acked-by: Tejun Heo <tj@kernel.org> Signed-off-by: Jens Axboe <axboe@fb.com> Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Santosh Shilimkar [Tue, 19 May 2015 17:08:01 +0000 (10:08 -0700)]
block: loop: convert to per-device workqueue
Documentation/workqueue.txt:
If there is dependency among multiple work items used
during memory reclaim, they should be queued to separate
wq each with WQ_MEM_RECLAIM.
Loop devices can be stacked, so we have to convert to per-device
workqueue. One example is Fedora live CD.
Fixes: b5dd2f6047ca108001328aac0e8588edd15f1778 Cc: stable@vger.kernel.org (v4.0) Cc: Justin M. Forbes <jforbes@fedoraproject.org> Signed-off-by: Ming Lei <ming.lei@canonical.com> Acked-by: Tejun Heo <tj@kernel.org> Signed-off-by: Jens Axboe <axboe@fb.com>
Conflicts:
Allow max_sectors to be tuned to enable I/O request sizes of 1MB. This
is not supported for RAID5/6 volumes but the UEK[23] kernels lack an
infrastructure for communicating per-LUN request size limits. This has
been remedied by upstream commit bcdb247c6b6a.
In the meantime allow setting the max_sectors module parameter to 1MB
for Invader and Fury cards.
Signed-off-by: Kashyap Desai <kashyap.desai@avagotech.com> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Guangyu Sun <guangyu.sun@oracle.com>
(cherry picked from commit 54e01aad1ef694d7ec4026d2efb5c8d19f981513) Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Sumit.Saxena@avagotech.com [Wed, 6 May 2015 13:31:02 +0000 (19:01 +0530)]
megaraid_sas : Modify return value of megasas_issue_blocked_cmd() and wait_and_poll() to consider command status returned by firmware
This patch is rebased on top of recently sent 18 patches(submitted by me) for
megaraid_sas driver.
Change the return value of wait_and_poll() and megsas_issue_blocked_cmd()
based on MFI_STAT returned by firmware for that command. Earlier driver always
send return type based on command completion (but never check MFI_STAT_OK for
that command), so even if command is failed by firmware still driver will
return SUCCESS status from these functions wait_and_poll() and
megsas_issue_blocked_cmd() and if caller of these functions does not check
command status (MFI_STAT), then it may endup using invalid data returned in
DMA buffers(one of the example is megasas_ld_list_query DCMD). Best thing to
avoid this type of issue is do error handling and set proper return type from
caller function wait_and_poll() and megsas_issue_blocked_cmd().
The change proposed in this patch will fix the regression introduced in patch-
"90dc9d9 megaraid_sas : MFI MPT linked list corruption fix" inside function
megasas_ld_list_query(). Prior to this MFI MPT linked list corruption fix
patch, megasas_ld_list_query() function used to check DCMD status(returned by
firmware) but with this linked list corruption fix patch, DCMD status will not
be checked inside function megasas_ld_list_query() and introduced this issue
of wrong data being used by function megasas_ld_list_query().
Cc: <stable@vger.kernel.org> Signed-off-by: Kashyap Desai <kashyap.desai@avagotech.com> Signed-off-by: Sumit Saxena <sumit.saxena@avagotech.com> Reviewed-by: Tomas Henzl <thenzl@redhat.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
Christoph Hellwig [Thu, 23 Apr 2015 11:04:24 +0000 (16:34 +0530)]
megaraid_sas : swap whole register in megasas_register_aen
Swap the whole 32 bits we read from the hardware instead of swapping
just the 16bits we care about in place later.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Sumit Saxena <sumit.saxena@avagotech.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Tomas Henzl <thenzl@redhat.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
The fusion HBAs don't really use the instance template like the other
variants, as it branches off at a much higher level. So instead of
trying to squeeze megasas_fire_cmd_fusion into the wrong calling
convention call it locally with argument data types that match what
is passed.
[jejb: fix up 32 bit compile failure] Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Sumit Saxena <sumit.saxena@avagotech.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Tomas Henzl <thenzl@redhat.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
Christoph Hellwig [Thu, 23 Apr 2015 11:03:54 +0000 (16:33 +0530)]
megaraid_sas : add missing byte swaps to the sriov code
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Sumit Saxena <sumit.saxena@avagotech.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Tomas Henzl <thenzl@redhat.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
Christoph Hellwig [Thu, 23 Apr 2015 11:03:39 +0000 (16:33 +0530)]
megaraid_sas : bytewise or should be done on native endian variables
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Sumit Saxena <sumit.saxena@avagotech.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Tomas Henzl <thenzl@redhat.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
Christoph Hellwig [Thu, 23 Apr 2015 11:03:24 +0000 (16:33 +0530)]
megaraid_sas : move endianness conversion into caller of megasas_get_seq_num
Converting structure fields in place is always a bad idea, and in this case
by moving it into the only caller we also only have to do a single byte
swap as most fields of this structure are never used.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Sumit Saxena <sumit.saxena@avagotech.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Tomas Henzl <thenzl@redhat.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
Christoph Hellwig [Thu, 23 Apr 2015 11:03:09 +0000 (16:33 +0530)]
megaraid_sas : add endianness conversions for all ones
Add noop conversions for all ones to make sparse happy.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Sumit Saxena <sumit.saxena@avagotech.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Tomas Henzl <thenzl@redhat.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
Christoph Hellwig [Thu, 23 Apr 2015 11:02:54 +0000 (16:32 +0530)]
megaraid_sas : add endianness annotations
This adds endianness annotations to all data structures, and a few
variables directly referencing them.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Sumit Saxena <sumit.saxena@avagotech.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Tomas Henzl <thenzl@redhat.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
Christoph Hellwig [Thu, 23 Apr 2015 11:02:39 +0000 (16:32 +0530)]
megaraid_sas : add missing __iomem annotations
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Sumit Saxena <sumit.saxena@avagotech.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Tomas Henzl <thenzl@redhat.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
megaraid_sas : megasas_complete_outstanding_ioctls() can be static
drivers/scsi/megaraid/megaraid_sas_base.c:1701:6: sparse: symbol 'megasas_complete_outstanding_ioctls' was not declared. Should it be static?
From: Christoph Hellwig <hch@lst.de>
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Sumit Saxena <sumit.saxena@avagotech.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Tomas Henzl <thenzl@redhat.com> Signed-off-by: James Bottomley <JBottomley@Odin.com>
megaraid_sas : Support for Avago's Single server High Availability product
This patch will add support for Single Server High Availability(SSHA) cluster
support. Here is the short decsription of changes done to add support for
SSHA-
1) Host will send system's Unique ID based on DMI_PRODUCT_UUID to firmware.
2) Toggle the devhandle in LDIO path for Remote LDs.
Signed-off-by: Kashyap Desai <kashyap.desai@avagotech.com> Signed-off-by: Sumit Saxena <sumit.saxena@avagotech.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Tomas Henzl <thenzl@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <JBottomley@Odin.com>
megaraid_sas : Add release date and update driver version
This patch will upgrade the driver version and add back the release date and
sysfs hook for the same. Some internal applications uses sysfs parameter for
release date, so they were broken because of removal of release date from
sysfs.
Signed-off-by: Sumit Saxena <sumit.saxena@avagotech.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Tomas Henzl <thenzl@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <JBottomley@Odin.com>
megaraid_sas : Modify driver's meta data to reflect Avago
Signed-off-by: Sumit Saxena <sumit.saxena@avagotech.com> Reviewed-by: Hannes Reinecke <hare@suse.de> Reviewed-by: Tomas Henzl <thenzl@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <JBottomley@Odin.com>
megaraid_sas : Use Block layer tag support for internal command indexing
megaraid_sas driver will use block layer provided tag for indexing internal
MPT frames to get any unique MPT frame tied with tag. Each IO request
submitted from SCSI mid layer will get associated MPT frame from MPT framepool
(retrieved and return back using spinlock inside megaraid_sas driver's
submission/completion call back). Getting MPT frame from MPT Frame pool is
very expensive operation because of associated spin lock operation (spinlock
overhead increase on multi NUMA node). This type of locking in driver is very
expensive call considering each IO request need - Acquire and Release of the
same lock.
With this support, in IO path driver will directly provide the unique command
index(which is based on block layer tag) and will get the MPT frame tied to
the tag and this way driver can get rid off lock, which synchronizes the
access to MPT frame pool while fetching and returning MPT frame from the pool.
This support in driver provides siginificant performance improvement(on multi
NUMA node system)on latest upstream with SCSI.MQ as well as on existing linux
distributions.
Here is the data for test executed at Avago-
- IO Tool- FIO
- 4 Socket SMC server. (4 NUMA node server)
- 12 SSDs in JBOD mode .
- 4K Rand READ, QD=32
- SCSI MQ x86_64 (Latest Upstream kernel)
- upto 300% Performance Improvement.
If IOs are running on single Node, perfromance gain is less, but as soon as
increase number of nodes, performance improvement is significant. IOs running
on all 4 NUMA nodes, with this patch applied IOPs observed was 1170K vs 344K
IOPs seen without this patch.
Logically, there are two parts of this patch- 1) Block layer tag support 2)
changes in calling convention of return_cmd. part 2 will revert the changes
done by patch- 90dc9d9 megaraid_sas : MFI MPT linked list corruption fix
because changes done in part 1 has fixed the problem of MFI MPT linked list
corruption. part 2 is very much dependent on part 1, so we decided to have
single patch for these two logical changes.
[jejb: remove chatty printk pointed out by hch] Signed-off-by: Kashyap Desai <kashyap.desai@avagotech.com> Signed-off-by: Sumit Saxena <sumit.saxena@avagotech.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <JBottomley@Odin.com>
megaraid_sas : Add separate function for setting up IRQs
This patch will create separate functions for- 1) setting up IRQs for MSI-x
interrupts 2) setting up IRQs for legacy interrupts 3) freeing up IRQs. and
enable interrupts after adapter's initialization. The reason behind
initialising adapter earlier is: by that time firmware is operational and can
send interrupts, so better to use interrupt based interface to send internal
DCMD to firmware instead of using polling method, since MFI frames' pool size
is reduced and polling method does not free up MFI frame for fusion adapters,
so sending more DCMDs with polled method may cause MFI frames's pool go out of
frames and end up failing DCMD.
Signed-off-by: Kashyap Desai <kashyap.desai@avagotech.com> Signed-off-by: Sumit Saxena <sumit.saxena@avagotech.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: James Bottomley <JBottomley@Odin.com>
This new firmware fixes several minor bugs:
1. In switch dependent mode, DCB priority was used to override inner vlan
priority.
2. In switch dependent mode, inner vlan was added in case of DCB priority
even if outer vlan was present.
3. In switch dependent mode, outer vlan was overridden by DCB priority when
working in STATIC COS mode while inner vlan was present.
4. iSCSI - under heavy iSCSI traffic, when TCP out-of-order condition
occurred, it was possible for the connection to close and recover.
5. iSCSI - connections on-chip TCP establishment might have failed.
6. iSCSI - out-of-order isles might have caused on-chip TCP connections
to fail in their graceful termination.
7. iSCSI - there was a theoretical race in which an RST packet sent from
pure-ack queue in specific timing could cause a credit-return overflow.
8. iSCSI - not all packets were completed on a forward channel.
9. DCB - fixed for 4-port devices; Until now, wrong credit counters were
used, causing dcb to fail.
10. Fixed false parity reported in CAM memories when operating near -5% on
the 1.0V core supply.
11. ETS default settings are set to fairness between traffic classes (rather
than strict priority), and uses the same chip receive buffer configuration
for both PFC and pause.
Joe Jin [Wed, 16 May 2012 05:48:22 +0000 (13:48 +0800)]
bnx2: Update driver to use new mips firmware.
bnx2-mips-06-6.2.3 and bnx2-mips-09-6.2.1.b
New firmware fixes iSCSI problems with some LeftHand targets that don't
set TTT=0xffffffff for Data-In according to spec. Firmware generates
exception warnings for this condition and becomes very slow. This is
fixed by suppressing these warnings when using default error mask.
(cherry picked from commit(UEK3) c2c20ef43d00b1439631e603f8dcee9a803cd8b3) Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Joe Jin <joe.jin@oracle.com> Signed-off-by: Maxim Uvarov <maxim.uvarov@oracle.com>
Conflicts:
drivers/net/ethernet/broadcom/bnx2.c
firmware/Makefile Signed-off-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
James Puthukattukaran [Wed, 22 Apr 2015 14:57:05 +0000 (10:57 -0400)]
net: Adding the hxge driver
The hxge driver
supports the Sun Blade 6000 Virtualized Multi-Fabric 10GbE M2 Network Express
Module (NEM). This NEM provides virtualized 10GbE access to 10 server module
(blades) in a Sun 6000 Blade Chassis, each blade sharing the 10GbE bandwidth.
From the perspective of each server module, it looks like it owns a Oracle
10Gb NIC interface. For more information, see
http://www.oracle.com/us/products/servers-storage/sun-blade-6000-10gbe-m2-nem-ds-080640.pdf
Signed-off-by: James Puthukattukaran <james.puthukattukaran@oracle.com>
Ashish Samant [Wed, 20 May 2015 02:10:36 +0000 (19:10 -0700)]
fuse: fix typo while displaying fuse numa mount option
The mount command output on FUSE filesystem output incorrectly shows 'numa'
option together with the previous option.
This patch adds a comma separator and fixes the issue.
This patch adds numa mount option. When this option is enabled, FUSE groups
all queues and creates one set per numa node. Users of /dev/fuse should listen
on /dev/fuse from all numa nodes.
fuse: modify queues, allocation and locking for multiple nodes
This patch makes provision for creating multiple instances of fuse_node
structure, which can be used in cases where a separate fuse_node has to be
created per numa node. It also introduces a new spinlock fn->lock
to synchronize elements within the fuse_node struct thus reducing contention
on fc->lock.
fc->lock protects other members along with sequence counter. Since the
change introduced by next few patches increases parallelism, it increases
contention on fc->lock. Having a new seq_lock spinlock to protect unique
sequence counter will reduce contention and also makes code simpler.
This patch introduces new structure fuse_node, which groups some fields
from fuse_conn structure. In the next few patches, an instance of this
fuse_node struct is created per NUMA node to improve performance.
Sunil Mushran [Tue, 8 Nov 2011 21:00:19 +0000 (13:00 -0800)]
ocfs2: Tighten free bit calculation in the global bitmap
When clearing bits in the global bitmap, we do not test the current bit value.
This patch tightens the code by considering the possiblity that the bit being
cleared was already cleared.
Now this should not happen. But we are seeing stray instances in which free
bit count in the global bitmap exceeds the total bit count. In each instance
the bitmap is correct. Only the free bit count is incorrect.
This patch checks the current bit value and increments the free bit count
only if the bit was previously set. It also prints information to allow
us to debug further.
Linus Torvalds [Sat, 20 Jun 2015 20:54:22 +0000 (13:54 -0700)]
Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux
Pull drm fixes from Dave Airlie:
"A smattering of fixes,
mgag200:
don't accept modes that aren't aligned properly as hw can't do it
i915:
two regression fixes
radeon:
one query to allow userspace fixes
one oops fixer for older hw with new options enabled"
* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux:
drm/radeon: don't probe MST on hw we don't support it on
drm/radeon: Add RADEON_INFO_VA_UNMAP_WORKING query
drm/mgag200: Reject non-character-cell-aligned mode widths
Revert "drm/i915: Don't skip request retirement if the active list is empty"
drm/i915: Always reset vma->ggtt_view.pages cache on unbinding
Linus Torvalds [Fri, 19 Jun 2015 17:34:14 +0000 (07:34 -1000)]
Merge tag 'sound-4.1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
Pull sound fixes from Takashi Iwai:
"Nothing looks scary, just a few usual HD-audio regression fixes and
fixup, in addition to a minor Kconfig dependency fix for the old MIPS
drivers"
* tag 'sound-4.1' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ALSA: hda - Fix unused label skip_i915
ALSA: hda - Fix noisy outputs on Dell XPS13 (2015 model)
ALSA: mips: let SND_SGI_O2 select SND_PCM
ALSA: hda - Fix audio crackles on Dell Latitude E7x40
ALSA: hda - adding a DAC/pin preference map for a HP Envy TS machine
projects / users / jedix / linux-maple.git / log