Gavin Shan [Wed, 1 Oct 2014 04:34:51 +0000 (14:34 +1000)]
powerpc/eeh: Show hex prefix for PE state sysfs
As Michael suggested, the hex prefix for the output of EEH PE
state sysfs entry (/sys/bus/pci/devices/xxx/eeh_pe_state) is
always informative to users.
Suggested-by: Michael Ellerman <mpe@ellerman.id.au> Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The dma_get_required_mask() function is used by some drivers to
query the platform about what DMA mask is needed to cover all of
memory. This is a bit of a strange semantic when we have to choose
between IOMMU translation or bypass, but essentially what it means
is "what DMA mask will give best performances".
Currently, our IOMMU backend always returns a 32-bit mask here, we
don't do anything special to it when we have bypass available. This
causes some drivers to choose a 32-bit mask, thus losing the ability
to use the bypass window, thinking this is more efficient. The problem
was reported from the driver of following device:
This patch adds an override of that function in order to, instead,
return a 64-bit mask whenever a bypass window is available in order
for drivers to prefer this configuration.
Reported-by: Murali N. Iyer <mniyer@us.ibm.com> Suggested-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
It should have been part of commit 1ad7a72c5 ("powerpc/eeh: Report
frozen parent PE prior to child PE"). There are 2 ways to report
EEH errors: proactively polling because of 0xFF's returned from
PCI config or IO read, or interrupt driven event. We missed to
report and handle parent frozen PE prior to child frozen PE for
the later case on PowerNV platform.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
powerpc/eeh: Dump PCI config space for all child devices
The PEs can be organized as nested. Current implementation doesn't
dump PCI config space for subordinate devices of child PEs. However,
the frozen PE could be caused by those subordinate devices of its
child PEs.
The patch dumps PCI config space for all subordinate devices of the
problematic PE.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
powerpc/eeh: Emulate EEH recovery for VFIO devices
When enabling EEH functionality on passed through devices (PE)
with VFIO, the devices in the PE would be removed permanently
from guest side. In that case, the PE remains frozen state.
When returning PE to host, or restarting the guest again, we
had mechanism unfreezing the PE by clearing PESTA/B frozen
bits. However, that's not enough for some adapters, which are
indicated as following "lspci" shows. Those adapters require
hot reset on the parent bus to bring their firmware back to
workable state. Otherwise, those adaptrs won't be operative
and the host (for returning case) or the guest will fail to
load the drivers for those adapters without exception.
The patch adds mechanism to emulate EEH recovery (for hot reset
on parent PCI bus) on 3 gates to fix the issue: open/release one
adapter of the PE, enable EEH functionality on one adapter of the
PE.
PE would be owned by userland, which probably request PE reset
done in host side. During the reset, we should drop the PCI
config accesses to the PE with help of flag EEH_PE_RESET.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
powerpc/pseries: Decrease message level on EEH initialization
As Anton suggested, the patch decreases the message level on EEH
initialization to avoid unnecessary messages if required. Also,
we have unified hint if any of needful RTAS calls is missed, and
then we can check /proc/device-tree to figure out the missed RTAS
calls.
Suggested-by: Anton Blanchard <anton@samba.org> Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Function pcibios_set_pcie_reset_state() can be used to do PCI
reset. PCI config access during the reset usually causes EEH
errors unexpectedly. In order to avoid the EEH error, the patch
blocks PCI config access during reset with the help of flag
EEH_PE_RESET, which is similar to what we did in EEH PE reset
path.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
powerpc/eeh: Unfreeze PE on enabling EEH functionality
When passing through PE to guest, that's possibly in frozen
state. The driver for the pass-through devices on guest side
can't be loaded successfully as reported. We already had one
gate in eeh_dev_open() to clear PE frozen state accordingly,
but that's not enough because the function is only called at
QEMU startup for once.
The patch adds another gate in eeh_pe_set_option() so that the
PE frozen state can be cleared at QEMU restart time.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
powerpc/eeh: Fix improper condition in eeh_pci_enable()
The function eeh_pci_enable() is called to apply various requests
to one particular PE: Enabling EEH, Disabling EEH, Enabling IO,
Enabling DMA, Freezing PE. When enabling IO or DMA on one specific
PE, we need check that IO or DMA isn't enabled previously. But
the condition used to do the check isn't completely correct because
one PE would be in DMA frozen state with workable IO path, or vice
versa.
The patch fixes the improper condition.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The problem was reported by Carol: In the scenario of passing mlx4
adapter to guest, EEH error could be recovered successfully. When
returning the device back to host, the driver (mlx4_core.ko)
couldn't be loaded successfully because of error number -5 (-EIO)
returned from mlx4_get_ownership(), which hits offlined PCI device.
The root cause is that we missed to put the affected devices into
normal state on clearing PE isolated state right after PE reset.
The patch fixes above issue by putting the affected devices to
normal state when clearing PE isolated state in eeh_pe_state_clear().
Cc: stable@vger.kernel.org Reported-by: Carol L. Soto <clsoto@us.ibm.com> Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The frozen state on one specific PE is probably caused by error
injection, which is done with help of PAPR error injection registers.
According to the hardware spec, those registers should be cleared
automatically after one-shot frozen PE. However, that's not always
true, at least on P7IOC of Firebird-L. So we have to clear them
before doing PE reset to avoid recursive EEH errors at recovery
stage.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The patch adds debugfs file (/sys/kernel/debug/powerpc/PCIxxxx/
err_injct), which accepts following formated string, to support
error injection. It will be used to support userland utility
"errinjct" in future.
The patch introduces eeh_ops::err_inject(), which allows to inject
specified errors to indicated PE for testing purpose. The functionality
isn't support on pSeries platform. On PowerNV, the functionality
relies on OPAL API opal_pci_err_inject().
Signed-off-by: Mike Qiu <qiudayu@linux.vnet.ibm.com> Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
When passing through device, its PE might have been put into frozen
state. One obvious example would be: the passed PE is forced to be
offline because of hitting maximal allowed EEH errors in userland.
In that case, the frozen state won't be cleared and then the PE is
returned back to host, which might not have chance detecting and
recovering from it.
The patch adds more check when passing through device and clear the
PE frozen state if necessary.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The PCI devices that have been passed through are enabled before
reset, we need restore to the enabled state after reset. Otherwise,
MMIO access might be issued to disabled devices after reset and
causes exceptional recursive EEH error.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The patch adds one more option (EEH_OPT_FREEZE_PE) to set_option()
method to proactively freeze PE, which will be issued before resetting
pass-throughed PE to drop MMIO access during reset because it's
always contributing to recursive EEH error.
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
The patch adds sysfs entry "eeh_pe_state". Reading on it returns
the PE's state while writing to it clears the frozen state. It's
used to check or clear the PE frozen state from userland for
debugging purpose.
The patch also replaces printk(KERN_WARNING ...) with pr_warn() in
eeh_sysfs.c
Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
powerpc/eeh: Drop unused argument in eeh_check_failure()
eeh_check_failure() is used to check frozen state of the PE which
owns the indicated I/O address. The argument "val" of the function
isn't used. The patch drops it and return the frozen state of the
PE as expected.
Cc: Vishal Mansur <vmansur@linux.vnet.ibm.com> Signed-off-by: Gavin Shan <gwshan@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
It uses mprotect to force a SEGV across a page boundary, and a SEGV
handler to lookup the exception tables and run the fixup routine.
It also compares the result against a normal load.
Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
This patch introduces a macro to convert eeh_dev to eeh_pe. By doing so, it
will prevent converting with NULL pointer.
Signed-off-by: Wei Yang <weiyang@linux.vnet.ibm.com> Acked-by: Gavin Shan <gwshan@linux.vnet.ibm.com> CC: Michael Ellerman <mpe@ellerman.id.au>
V3 -> V4:
1. move the macro definition from include/linux/pci.h to
arch/powerpc/include/asm/eeh.h
V2 -> V3:
1. rebased on 3.17-rc4
2. introduce a macro
3. use this macro in several other places
V1 -> V2:
1. code style and patch subject adjustment
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
We can unindent the bulk of htab_dt_scan_page_sizes() by returning early
if the property is not found. That is nice in and of itself, but also
has the advantage of making it clear that we always return success once
we have found the ibm,segment-page-sizes property.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
powerpc/ppc64: Clean up the boot-time settings display
At boot we display a bunch of low level settings which can be useful to
know, and can help to spot bugs when things are fundamentally
misconfigured.
At the moment they are very widely spaced, so that we can accommodate
the line:
ppc64_caches.dcache_line_size = 0xYY
But we only print that line when the cache line size is not 128, ie.
almost never, so it just makes the display look odd usually.
The ppc64_caches prefix is redundant so remove it, which means we can
align things a bit closer for the common case. While we're there
replace the last use of camelCase (physicalMemorySize), and use
phys_mem_size.
Li Zhong [Wed, 27 Aug 2014 09:34:01 +0000 (17:34 +0800)]
powerpc: some changes in numa_setup_cpu()
this patches changes some error handling logics in numa_setup_cpu(),
when cpu node is not found, so:
if the cpu is possible, but not present, -1 is kept in numa_cpu_lookup_table,
so later, if the cpu is added, we could set correct numa information for it.
if the cpu is present, then we set the first online node to
numa_cpu_lookup_table instead of 0 ( in case 0 might not be an online node? )
Cc: Nishanth Aravamudan <nacc@linux.vnet.ibm.com> Cc: Nathan Fontenot <nfont@linux.vnet.ibm.com> Signed-off-by: Li Zhong <zhong@linux.vnet.ibm.com> Acked-by: Nishanth Aravamudan <nacc@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Li Zhong [Wed, 27 Aug 2014 09:34:00 +0000 (17:34 +0800)]
powerpc: Only set numa node information for present cpus at boottime
As Nish suggested, it makes more sense to init the numa node informatiion
for present cpus at boottime, which could also avoid WARN_ON(1) in
numa_setup_cpu().
With this change, we also need to change the smp_prepare_cpus() to set up
numa information only on present cpus.
For those possible, but not present cpus, their numa information
will be set up after they are started, as the original code did before commit 2fabf084b6ad.
Li Zhong [Wed, 27 Aug 2014 09:33:59 +0000 (17:33 +0800)]
powerpc: Fix warning reported by verify_cpu_node_mapping()
With commit 2fabf084b6ad ("powerpc: reorder per-cpu NUMA information's
initialization"), during boottime, cpu_numa_callback() is called
earlier(before their online) for each cpu, and verify_cpu_node_mapping()
uses cpu_to_node() to check whether siblings are in the same node.
It skips the checking for siblings that are not online yet. So the only
check done here is for the bootcpu, which is online at that time. But
the per-cpu numa_node cpu_to_node() uses hasn't been set up yet (which
will be set up in smp_prepare_cpus()).
So I saw something like following reported:
[ 0.000000] CPU thread siblings 1/2/3 and 0 don't belong to the same
node!
As we don't actually do the checking during this early stage, so maybe
we could directly call numa_setup_cpu() in do_init_bootmem().
Cc: Nishanth Aravamudan <nacc@linux.vnet.ibm.com> Cc: Nathan Fontenot <nfont@linux.vnet.ibm.com> Signed-off-by: Li Zhong <zhong@linux.vnet.ibm.com> Acked-by: Nishanth Aravamudan <nacc@linux.vnet.ibm.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Paul Mackerras [Tue, 2 Sep 2014 04:35:07 +0000 (14:35 +1000)]
powerpc: Split out instruction analysis part of emulate_step()
This splits out the instruction analysis part of emulate_step() into
a separate analyse_instr() function, which decodes the instruction,
but doesn't execute any load or store instructions. It does execute
integer instructions and branches which can be executed purely by
updating register values in the pt_regs struct. For other instructions,
it returns the instruction type and other details in a new
instruction_op struct. emulate_step() then uses that information
to execute loads, stores, cache operations, mfmsr, mtmsr[d], and
(on 64-bit) sc instructions.
The reason for doing this is so that the KVM code can use it instead
of having its own separate instruction emulation code. Possibly the
alignment interrupt handler could also use this.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Michael Ellerman [Thu, 28 Aug 2014 08:40:47 +0000 (18:40 +1000)]
powerpc: Check flat device tree version at boot
In commit e6a6928c3ea1 "of/fdt: Convert FDT functions to use libfdt",
the kernel stopped supporting old flat device tree formats. The minimum
supported version is now 0x10.
There was a checking function added, early_init_dt_verify(), but it's
not called on powerpc.
The result is, if you boot with an old flat device tree, the kernel will
fail to parse it correctly, think you have no memory etc. and hilarity
ensues.
We can't really fix it, but we can at least catch the fact that the
device tree is in an unsupported format and panic(). We can't call
BUG(), it's too early.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Paul Mackerras [Tue, 2 Sep 2014 04:23:16 +0000 (14:23 +1000)]
powerpc/powernv: Don't call generic code on offline cpus
On PowerNV platforms, when a CPU is offline, we put it into nap mode.
It's possible that the CPU wakes up from nap mode while it is still
offline due to a stray IPI. A misdirected device interrupt could also
potentially cause it to wake up. In that circumstance, we need to clear
the interrupt so that the CPU can go back to nap mode.
In the past the clearing of the interrupt was accomplished by briefly
enabling interrupts and allowing the normal interrupt handling code
(do_IRQ() etc.) to handle the interrupt. This has the problem that
this code calls irq_enter() and irq_exit(), which call functions such
as account_system_vtime() which use RCU internally. Use of RCU is not
permitted on offline CPUs and will trigger errors if RCU checking is
enabled.
To avoid calling into any generic code which might use RCU, we adopt
a different method of clearing interrupts on offline CPUs. Since we
are on the PowerNV platform, we know that the system interrupt
controller is a XICS being driven directly (i.e. not via hcalls) by
the kernel. Hence this adds a new icp_native_flush_interrupt()
function to the native-mode XICS driver and arranges to call that
when an offline CPU is woken from nap. This new function reads the
interrupt from the XICS. If it is an IPI, it clears the IPI; if it
is a device interrupt, it prints a warning and disables the source.
Then it does the end-of-interrupt processing for the interrupt.
The other thing that briefly enabling interrupts did was to check and
clear the irq_happened flag in this CPU's PACA. Therefore, after
flushing the interrupt from the XICS, we also clear all bits except
the PACA_IRQ_HARD_DIS (interrupts are hard disabled) bit from the
irq_happened flag. The PACA_IRQ_HARD_DIS flag is set by power7_nap()
and is left set to indicate that interrupts are hard disabled. This
means we then have to ignore that flag in power7_nap(), which is
reasonable since it doesn't indicate that any interrupt event needs
servicing.
Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
of_device_ids (i.e. compatible strings and the respective data) are not
supposed to change at runtime. All functions working with of_device_ids
provided by <linux/of.h> work with const of_device_ids. This allows to
mark all struct of_device_id const, too.
While touching these line also put the __init annotation at the right
position where necessary.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Tony Breeds [Thu, 14 Aug 2014 07:05:04 +0000 (17:05 +1000)]
powerpc/boot: Don't install zImage.* from make install
in commit 29f1aff2c (powerpc: Copy bootable images in the default
install script) we changed to copying all the built boot targets based
on the assumption that it's backwards compatible. It turns out that
debian devived installkernel scripts will barf if not given exactly 4
args.
This change reverts make install to just install the vmlinux (we can
change the dfault in a seperate patch) and introduces a new make
zInstall which works with a more flexible installkernel script.
Cc: Grant Likely <grant.likely@secretlab.ca> Signed-off-by: Tony Breeds <tony@bakeyournoodle.com> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull vfs fixes from Al Viro:
"double iput() on failure exit in lustre, racy removal of spliced
dentries from ->s_anon in __d_materialise_dentry() plus a bunch of
assorted RCU pathwalk fixes"
The RCU pathwalk fixes end up fixing a couple of cases where we
incorrectly dropped out of RCU walking, due to incorrect initialization
and testing of the sequence locks in some corner cases. Since dropping
out of RCU walk mode forces the slow locked accesses, those corner cases
slowed down quite dramatically.
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
be careful with nd->inode in path_init() and follow_dotdot_rcu()
don't bugger nd->seq on set_root_rcu() from follow_dotdot_rcu()
fix bogus read_seqretry() checks introduced in b37199e
move the call of __d_drop(anon) into __d_materialise_unique(dentry, anon)
[fix] lustre: d_make_root() does iput() on dentry allocation failure
vfs: avoid non-forwarding large load after small store in path lookup
The performance regression that Josef Bacik reported in the pathname
lookup (see commit 99d263d4c5b2 "vfs: fix bad hashing of dentries") made
me look at performance stability of the dcache code, just to verify that
the problem was actually fixed. That turned up a few other problems in
this area.
There are a few cases where we exit RCU lookup mode and go to the slow
serializing case when we shouldn't, Al has fixed those and they'll come
in with the next VFS pull.
But my performance verification also shows that link_path_walk() turns
out to have a very unfortunate 32-bit store of the length and hash of
the name we look up, followed by a 64-bit read of the combined hash_len
field. That screws up the processor store to load forwarding, causing
an unnecessary hickup in this critical routine.
It's caused by the ugly calling convention for the "hash_name()"
function, and easily fixed by just making hash_name() fill in the whole
'struct qstr' rather than passing it a pointer to just the hash value.
With that, the profile for this function looks much smoother.
Merge branch 'parisc-3.17-1' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux
Pull parisc updates from Helge Deller:
"The most important patch is a new Light Weigth Syscall (LWS) for 8,
16, 32 and 64 bit atomic CAS operations which is required in order to
be able to implement the atomic gcc builtins on our platform.
Other than that, we wire up the seccomp, getrandom and memfd_create
syscalls, fixes a minor off-by-one bug and a wrong printk string"
* 'parisc-3.17-1' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
parisc: Implement new LWS CAS supporting 64 bit operations.
parisc: Wire up seccomp, getrandom and memfd_create syscalls
parisc: dino: fix %d confusingly prefixed with 0x in format string
parisc: sys_hpux: NUL terminator is one past the end
Al Viro [Sun, 14 Sep 2014 01:59:43 +0000 (21:59 -0400)]
be careful with nd->inode in path_init() and follow_dotdot_rcu()
in the former we simply check if dentry is still valid after picking
its ->d_inode; in the latter we fetch ->d_inode in the same places
where we fetch dentry and its ->d_seq, under the same checks.
Cc: stable@vger.kernel.org # 2.6.38+ Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Sun, 14 Sep 2014 01:55:46 +0000 (21:55 -0400)]
don't bugger nd->seq on set_root_rcu() from follow_dotdot_rcu()
return the value instead, and have path_init() do the assignment. Broken by
"vfs: Fix absolute RCU path walk failures due to uninitialized seq number",
which was Cc-stable with 2.6.38+ as destination. This one should go where
it went.
To avoid dummy value returned in case when root is already set (it would do
no harm, actually, since the only caller that doesn't ignore the return value
is guaranteed to have nd->root *not* set, but it's more obvious that way),
lift the check into callers. And do the same to set_root(), to keep them
in sync.
Cc: stable@vger.kernel.org # 2.6.38+ Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Merge tag 'ntb-3.17' of git://github.com/jonmason/ntb
Pull ntb driver bugfixes from Jon Mason:
"NTB driver fixes for queue spread and buffer alignment. Also, update
to MAINTAINERS to reflect new e-mail address"
* tag 'ntb-3.17' of git://github.com/jonmason/ntb:
ntb: Add alignment check to meet hardware requirement
MAINTAINERS: update NTB info
NTB: correct the spread of queues over mw's
Merge branch 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull ARM irq chip fixes from Thomas Gleixner:
"Another pile of ARM specific irq chip fixlets:
- off by one bugs in the crossbar driver
- missing annotations
- a bunch of "make it compile" updates
I pulled the lot today from Jason, but it has been in -next for at
least a week"
* 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
irqchip: gic-v3: Declare rdist as __percpu pointer to __iomem pointer
irqchip: gic: Make gic_default_routable_irq_domain_ops static
irqchip: exynos-combiner: Fix compilation error on ARM64
irqchip: crossbar: Off by one bugs in init
irqchip: gic-v3: Tag all low level accessors __maybe_unused
irqchip: gic-v3: Only define gic_peek_irq() when building SMP
Dave Jiang [Thu, 28 Aug 2014 20:53:02 +0000 (13:53 -0700)]
ntb: Add alignment check to meet hardware requirement
The NTB translate register must have the value to be BAR size aligned.
This alignment check make sure that the DMA memory allocated has the
proper alignment. Another requirement for NTB to function properly with
memory window BAR size greater or equal to 4M is to use the CMA feature
in 3.16 kernel with the appropriate CONFIG_CMA_ALIGNMENT and
CONFIG_CMA_SIZE_MBYTES set.
Signed-off-by: Dave Jiang <dave.jiang@intel.com> Signed-off-by: Jon Mason <jdmason@kudzu.us>
Jon Mason [Thu, 19 Jun 2014 17:11:13 +0000 (10:11 -0700)]
NTB: correct the spread of queues over mw's
The detection of an uneven number of queues on the given memory windows
was not correct. The mw_num is zero based and the mod should be
division to spread them evenly over the mw's.
Merge branches 'locking-urgent-for-linus' and 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull futex and timer fixes from Thomas Gleixner:
"A oneliner bugfix for the jinxed futex code:
- Drop hash bucket lock in the error exit path. I really could slap
myself for intruducing that bug while fixing all the other horror
in that code three month ago ...
and the timer department is not too proud about the following fixes:
- Deal with a long standing rounding bug in the timeval to jiffies
conversion. It's a real issue and this fix fell through the cracks
for quite some time.
- Another round of alarmtimer fixes. Finally this code gets used
more widely and the subtle issues hidden for quite some time are
noticed and fixed. Nothing really exciting, just the itty bitty
details which bite the serious users here and there"
* 'locking-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
futex: Unlock hb->lock in futex_wait_requeue_pi() error path
* 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
alarmtimer: Lock k_itimer during timer callback
alarmtimer: Do not signal SIGEV_NONE timers
alarmtimer: Return relative times in timer_gettime
jiffies: Fix timeval conversion to jiffies
Josef Bacik found a performance regression between 3.2 and 3.10 and
narrowed it down to commit bfcfaa77bdf0 ("vfs: use 'unsigned long'
accesses for dcache name comparison and hashing"). He reports:
On xfs on a fio card this goes at about 20k dir/sec with 3.2, and 12k
dir/sec with 3.10. This is because we spend waaaaay more time in
__d_lookup on 3.10 than in 3.2.
The new hashing function for strings is suboptimal for <
sizeof(unsigned long) string names (and hell even > sizeof(unsigned
long) string names that I've tested). I broke out the old hashing
function and the new one into a userspace helper to get real numbers
and this is what I'm getting:
Old hash table had 1000000 entries, 0 dupes, 0 max dupes
New hash table had 12628 entries, 987372 dupes, 900 max dupes
We had 11400 buckets with a p50 of 30 dupes, p90 of 240 dupes, p99 of 567 dupes for the new hash
My test does the hash, and then does the d_hash into a integer pointer
array the same size as the dentry hash table on my system, and then
just increments the value at the address we got to see how many
entries we overlap with.
As you can see the old hash function ended up with all 1 million
entries in their own bucket, whereas the new one they are only
distributed among ~12.5k buckets, which is why we're using so much
more CPU in __d_lookup".
The reason for this hash regression is two-fold:
- On 64-bit architectures the down-mixing of the original 64-bit
word-at-a-time hash into the final 32-bit hash value is very
simplistic and suboptimal, and just adds the two 32-bit parts
together.
In particular, because there is no bit shuffling and the mixing
boundary is also a byte boundary, similar character patterns in the
low and high word easily end up just canceling each other out.
- the old byte-at-a-time hash mixed each byte into the final hash as it
hashed the path component name, resulting in the low bits of the hash
generally being a good source of hash data. That is not true for the
word-at-a-time case, and the hash data is distributed among all the
bits.
The fix is the same in both cases: do a better job of mixing the bits up
and using as much of the hash data as possible. We already have the
"hash_32|64()" functions to do that.
Reported-by: Josef Bacik <jbacik@fb.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Christoph Hellwig <hch@infradead.org> Cc: Chris Mason <clm@fb.com> Cc: linux-fsdevel@vger.kernel.org Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Make hash_64() use a 64-bit multiply when appropriate
The hash_64() function historically does the multiply by the
GOLDEN_RATIO_PRIME_64 number with explicit shifts and adds, because
unlike the 32-bit case, gcc seems unable to turn the constant multiply
into the more appropriate shift and adds when required.
However, that means that we generate those shifts and adds even when the
architecture has a fast multiplier, and could just do it better in
hardware.
Use the now-cleaned-up CONFIG_ARCH_HAS_FAST_MULTIPLIER (together with
"is it a 64-bit architecture") to decide whether to use an integer
multiply or the explicit sequence of shift/add instructions.
Make ARCH_HAS_FAST_MULTIPLIER a real config variable
It used to be an ad-hoc hack defined by the x86 version of
<asm/bitops.h> that enabled a couple of library routines to know whether
an integer multiply is faster than repeated shifts and additions.
This just makes it use the real Kconfig system instead, and makes x86
(which was the only architecture that did this) select the option.
NOTE! Even for x86, this really is kind of wrong. If we cared, we would
probably not enable this for builds optimized for netburst (P4), where
shifts-and-adds are generally faster than multiplies. This patch does
*not* change that kind of logic, though, it is purely a syntactic change
with no code changes.
This was triggered by the fact that we have other places that really
want to know "do I want to expand multiples by constants by hand or
not", particularly the hash generation code.
Merge tag 'dm-3.17-fix2' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm
Pull device mapper fix from Mike Snitzer:
"Fix a race in the DM cache target that caused dirty blocks to be
marked as clean. This could cause no writeback to occur or spurious
dirty block counts"
* tag 'dm-3.17-fix2' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
dm cache: fix race causing dirty blocks to be marked as clean
Merge branch 'for-linus' of git://git.kernel.dk/linux-block
Pull block fixes from Jens Axboe:
"A small collection of fixes for the current rc series. This contains:
- Two small blk-mq patches from Rob Elliott, cleaning up error case
at init time.
- A fix from Ming Lei, fixing SG merging for blk-mq where
QUEUE_FLAG_SG_NO_MERGE is the default.
- A dev_t minor lifetime fix from Keith, fixing an issue where a
minor might be reused before all references to it were gone.
- Fix from Alan Stern where an unbalanced queue bypass caused SCSI
some headaches when it does a series of add/del on devices without
fully registrering the queue.
- A fix from me for improving the scaling of tag depth in blk-mq if
we are short on memory"
* 'for-linus' of git://git.kernel.dk/linux-block:
blk-mq: scale depth and rq map appropriate if low on memory
Block: fix unbalanced bypass-disable in blk_register_queue
block: Fix dev_t minor allocation lifetime
blk-mq: cleanup after blk_mq_init_rq_map failures
blk-mq: pass along blk_mq_alloc_tag_set return values
blk-merge: fix blk_recount_segments
Merge tag 'stable/for-linus-3.17-b-rc4-arm-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip
Pull Xen ARM bugfix from Stefano Stabellini:
"The patches fix the "xen_add_mach_to_phys_entry: cannot add" bug that
has been affecting xen on arm and arm64 guests since 3.16. They
require a few hypervisor side changes that just went in xen-unstable.
A couple of days ago David sent out a pull request with a few other
Xen fixes (it is already in master). Sorry we didn't synchronized
better among us"
* tag 'stable/for-linus-3.17-b-rc4-arm-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
xen/arm: remove mach_to_phys rbtree
xen/arm: reimplement xen_dma_unmap_page & friends
xen/arm: introduce XENFEAT_grant_map_identity
Richard Larocque [Wed, 10 Sep 2014 01:31:05 +0000 (18:31 -0700)]
alarmtimer: Lock k_itimer during timer callback
Locks the k_itimer's it_lock member when handling the alarm timer's
expiry callback.
The regular posix timers defined in posix-timers.c have this lock held
during timout processing because their callbacks are routed through
posix_timer_fn(). The alarm timers follow a different path, so they
ought to grab the lock somewhere else.
Cc: stable@vger.kernel.org Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@kernel.org> Cc: Richard Cochran <richardcochran@gmail.com> Cc: Prarit Bhargava <prarit@redhat.com> Cc: Sharvil Nanavati <sharvil@google.com> Signed-off-by: Richard Larocque <rlarocque@google.com> Signed-off-by: John Stultz <john.stultz@linaro.org>
Richard Larocque [Wed, 10 Sep 2014 01:31:04 +0000 (18:31 -0700)]
alarmtimer: Do not signal SIGEV_NONE timers
Avoids sending a signal to alarm timers created with sigev_notify set to
SIGEV_NONE by checking for that special case in the timeout callback.
The regular posix timers avoid sending signals to SIGEV_NONE timers by
not scheduling any callbacks for them in the first place. Although it
would be possible to do something similar for alarm timers, it's simpler
to handle this as a special case in the timeout.
Prior to this patch, the alarm timer would ignore the sigev_notify value
and try to deliver signals to the process anyway. Even worse, the
sanity check for the value of sigev_signo is skipped when SIGEV_NONE was
specified, so the signal number could be bogus. If sigev_signo was an
unitialized value (as it often would be if SIGEV_NONE is used), then
it's hard to predict which signal will be sent.
Cc: stable@vger.kernel.org Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@kernel.org> Cc: Richard Cochran <richardcochran@gmail.com> Cc: Prarit Bhargava <prarit@redhat.com> Cc: Sharvil Nanavati <sharvil@google.com> Signed-off-by: Richard Larocque <rlarocque@google.com> Signed-off-by: John Stultz <john.stultz@linaro.org>
Richard Larocque [Wed, 10 Sep 2014 01:31:03 +0000 (18:31 -0700)]
alarmtimer: Return relative times in timer_gettime
Returns the time remaining for an alarm timer, rather than the time at
which it is scheduled to expire. If the timer has already expired or it
is not currently scheduled, the it_value's members are set to zero.
This new behavior matches that of the other posix-timers and the POSIX
specifications.
This is a change in user-visible behavior, and may break existing
applications. Hopefully, few users rely on the old incorrect behavior.
Cc: stable@vger.kernel.org Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@kernel.org> Cc: Richard Cochran <richardcochran@gmail.com> Cc: Prarit Bhargava <prarit@redhat.com> Cc: Sharvil Nanavati <sharvil@google.com> Signed-off-by: Richard Larocque <rlarocque@google.com>
[jstultz: minor style tweak] Signed-off-by: John Stultz <john.stultz@linaro.org>
Andrew Hunter [Thu, 4 Sep 2014 21:17:16 +0000 (14:17 -0700)]
jiffies: Fix timeval conversion to jiffies
timeval_to_jiffies tried to round a timeval up to an integral number
of jiffies, but the logic for doing so was incorrect: intervals
corresponding to exactly N jiffies would become N+1. This manifested
itself particularly repeatedly stopping/starting an itimer:
would add a full tick to val, _even if it was exactly representable in
terms of jiffies_ (say, the result of a previous rounding.) Doing
this repeatedly would cause unbounded growth in val. So fix the math.
Here's what was wrong with the conversion: we essentially computed
(eliding seconds)
jiffies = usec * (NSEC_PER_USEC/TICK_NSEC)
by using scaling arithmetic, which took the best approximation of
NSEC_PER_USEC/TICK_NSEC with denominator of 2^USEC_JIFFIE_SC =
x/(2^USEC_JIFFIE_SC), and computed:
jiffies = (usec * x) >> USEC_JIFFIE_SC
and rounded this calculation up in the intermediate form (since we
can't necessarily exactly represent TICK_NSEC in usec.) But the
scaling arithmetic is a (very slight) *over*approximation of the true
value; that is, instead of dividing by (1 usec/ 1 jiffie), we
effectively divided by (1 usec/1 jiffie)-epsilon (rounding
down). This would normally be fine, but we want to round timeouts up,
and we did so by adding 2^USEC_JIFFIE_SC - 1 before the shift; this
would be fine if our division was exact, but dividing this by the
slightly smaller factor was equivalent to adding just _over_ 1 to the
final result (instead of just _under_ 1, as desired.)
In particular, with HZ=1000, we consistently computed that 10000 usec
was 11 jiffies; the same was true for any exact multiple of
TICK_NSEC.
We could possibly still round in the intermediate form, adding
something less than 2^USEC_JIFFIE_SC - 1, but easier still is to
convert usec->nsec, round in nanoseconds, and then convert using
time*spec*_to_jiffies. This adds one constant multiplication, and is
not observably slower in microbenchmarks on recent x86 hardware.
Tested: the following program:
int main() {
struct itimerval zero = {{0, 0}, {0, 0}};
/* Initially set to 10 ms. */
struct itimerval initial = zero;
initial.it_interval.tv_usec = 10000;
setitimer(ITIMER_PROF, &initial, NULL);
/* Save and restore several times. */
for (size_t i = 0; i < 10; ++i) {
struct itimerval prev;
setitimer(ITIMER_PROF, &zero, &prev);
/* on old kernels, this goes up by TICK_USEC every iteration */
printf("previous value: %ld %ld %ld %ld\n",
prev.it_interval.tv_sec, prev.it_interval.tv_usec,
prev.it_value.tv_sec, prev.it_value.tv_usec);
setitimer(ITIMER_PROF, &prev, NULL);
}
return 0;
}
Cc: stable@vger.kernel.org Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: Paul Turner <pjt@google.com> Cc: Richard Cochran <richardcochran@gmail.com> Cc: Prarit Bhargava <prarit@redhat.com> Reviewed-by: Paul Turner <pjt@google.com> Reported-by: Aaron Jacobs <jacobsa@google.com> Signed-off-by: Andrew Hunter <ahh@google.com>
[jstultz: Tweaked to apply to 3.17-rc] Signed-off-by: John Stultz <john.stultz@linaro.org>
Thomas Gleixner [Thu, 11 Sep 2014 21:44:35 +0000 (23:44 +0200)]
futex: Unlock hb->lock in futex_wait_requeue_pi() error path
futex_wait_requeue_pi() calls futex_wait_setup(). If
futex_wait_setup() succeeds it returns with hb->lock held and
preemption disabled. Now the sanity check after this does:
if (match_futex(&q.key, &key2)) {
ret = -EINVAL;
goto out_put_keys;
}
which releases the keys but does not release hb->lock.
So we happily return to user space with hb->lock held and therefor
preemption disabled.
Unlock hb->lock before taking the exit route.
Reported-by: Dave "Trinity" Jones <davej@redhat.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Darren Hart <dvhart@linux.intel.com> Reviewed-by: Davidlohr Bueso <dave@stgolabs.net> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: stable@vger.kernel.org Link: http://lkml.kernel.org/r/alpine.DEB.2.10.1409112318500.4178@nanos Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Merge tag 'char-misc-3.17-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc
Pull char/misc driver fix from Greg KH:
"Here is one misc driver fix for 3.17-rc5. It resolves a kernel oops
that can happen in the lattice FPGA driver if the firmware isn't
present on the system.
It's been in the linux-next tree for a while now"
* tag 'char-misc-3.17-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
Lattice ECP3 FPGA: Check firmware pointer
Merge tag 'staging-3.17-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging
Pull staging driver fixes from Greg KH:
"Here are 3 tiny staging driver fixes for 3.17-rc5.
Two are fixes for the imx-drm driver, resolving issues that have been
reported. The other is a memory leak fix for the Android sync driver,
due to changes that went into 3.17-rc1.
All have been in linux-next for a while"
* tag 'staging-3.17-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
android: fix reference leak in sync_fence_create
imx-drm: imx-ldb: fix NULL pointer in imx_ldb_unbind()
imx-drm: ipuv3-plane: fix ipu_plane_dpms()
Merge tag 'tty-3.17-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty
Pull tty/serial fixes from Greg KH:
"Here are 3 patches for 3.17-rc5. Two serial driver fixes that resolve
some reported issues, and one new device id.
All have been in linux-next just fine"
* tag 'tty-3.17-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
tty: xuartps: Fix tx_emtpy() callback
tty/serial: at91: BUG: disable interrupts when !UART_ENABLE_MS()
serial: 8250_dw: Add ACPI ID for Intel Braswell
Merge tag 'usb-3.17-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb
Pull USB fixes from Greg KH:
"Here are some USB and PHY fixes for 3.17-rc5.
Nothing major here, just a number of tiny fixes for reported issues,
and some new device ids as well.
All have been tested in linux-next"
* tag 'usb-3.17-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (46 commits)
xhci: fix oops when xhci resumes from hibernate with hw lpm capable devices
usb: xhci: Fix OOPS in xhci error handling code
xhci: Fix null pointer dereference if xhci initialization fails
storage: Add single-LUN quirk for Jaz USB Adapter
uas: Add missing le16_to_cpu calls to asm1051 / asm1053 usb-id check
usb: chipidea: msm: Initialize PHY on reset event
usb: chipidea: msm: Use USB PHY API to control PHY state
usb: hub: take hub->hdev reference when processing from eventlist
uas: Disable uas on ASM1051 devices
usb: dwc2/gadget: avoid disabling ep0
usb: dwc2/gadget: delay enabling irq once hardware is configured properly
usb: dwc2/gadget: do not call disconnect method in pullup
usb: dwc2/gadget: break infinite loop in endpoint disable code
usb: dwc2/gadget: fix phy initialization sequence
usb: dwc2/gadget: fix phy disable sequence
uwb: init beacon cache entry before registering uwb device
USB: ftdi_sio: Add support for GE Healthcare Nemo Tracker device
USB: document the 'u' flag for usb-storage quirks parameter
usb: host: xhci: fix compliance mode workaround
usb: dwc3: fix TRB completion when multiple TRBs are started
...
projects / mirror_ubuntu-eoan-kernel.git / log