linux-toradex.git/drivers/of/irq.c, branch v4.9.25

of/irq: Break out msi-map lookup (again)

2016-09-16T08:34:14+00:00

The PCI msi-map code is already doing double-duty translating IDs and
retrieving MSI parents, which unsurprisingly is the same functionality
we need for the identically-formatted PCI iommu-map property. Drag the
core parsing routine up yet another layer into the general OF-PCI code,
and further generalise it for either kind of lookup in either flavour
of map property.

Acked-by: Rob Herring 
Acked-by: Marc Zyngier 
Tested-by: Lorenzo Pieralisi 
Signed-off-by: Robin Murphy 
Signed-off-by: Will Deacon

of/irq: Mark interrupt controllers as populated before initialisation

2016-08-09T17:36:28+00:00

That way the init callback may clear the flag again, in case of drivers
split between early irq chip and a normal platform driver.

Fixes: 15cc2ed6dcf9 ("of/irq: Mark initialised interrupt controllers as populated")
Suggested-by: Rob Herring 
Signed-off-by: Philipp Zabel 
Acked-by: Jon Hunter 
Signed-off-by: Rob Herring

of: use pr_fmt prefix for all console printing

2016-07-18T21:57:42+00:00

Clean-up all the DT printk functions to use common pr_fmt prefix.

Some print statements such as kmalloc errors were redundant, so just
drop those.

Cc: Frank Rowand 
Cc: Pantelis Antoniou 
Reviewed-by: Frank Rowand 
Signed-off-by: Rob Herring

of/irq: Mark initialised interrupt controllers as populated

2016-06-24T20:17:02+00:00

For interrupt controllers successfully initialised early via device-tree,
mark these interrupt controllers as populated so we don't unnecessarily
create a device and populate any platform data later on in the boot
sequence when we populate all the various platform devices.

Signed-off-by: Jon Hunter 
Signed-off-by: Rob Herring

of: irq: fix of_irq_get[_byname]() kernel-doc

2016-06-03T12:22:01+00:00

The kernel-doc for the of_irq_get[_byname]()  is clearly inadequate in
describing the return values -- of_irq_get_byname() is documented better
than of_irq_get() but it  still doesn't mention that 0 is returned iff
irq_create_of_mapping() fails (it doesn't return an error code in this
case). Document all possible return value variants, making the writing
of the word "IRQ" consistent, while at it...

Fixes: 9ec36cafe43b ("of/irq: do irq resolution in platform_get_irq")
Fixes: ad69674e73a1 ("of/irq: do irq resolution in platform_get_irq_byname()")
Signed-off-by: Sergei Shtylyov 
CC: stable@vger.kernel.org
Signed-off-by: Rob Herring

of/irq: Fix msi-map calculation for nonzero rid-base

2016-02-11T22:51:41+00:00

The existing msi-map code is fine for shifting the entire RID space
upwards, but attempting finer-grained remapping reveals a bug. It turns
out that we are mistakenly treating the msi-base part as an offset, not
as a new base to remap onto, so things get squiffy when rid-base is
nonzero. Fix this, and at the same time add a sanity check against
having msi-map-mask clash with a nonzero rid-base, as that's another
thing one can easily get wrong.

CC: 
Signed-off-by: Robin Murphy 
Reviewed-by: Marc Zyngier 
Tested-by: Stuart Yoder 
Acked-by: Mark Rutland 
Acked-by: David Daney 
Signed-off-by: Rob Herring

Merge branch 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

2016-01-31T22:48:58+00:00

Pull IRQ fixes from Ingo Molnar:
 "Mostly irqchip driver fixes, but also an irq core crash fix and a
  build fix"

* 'irq-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  irqchip/mxs: Add missing set_handle_irq()
  irqchip/atmel-aic: Fix wrong bit operation for IRQ priority
  irqchip/gic-v3-its: Recompute the number of pages on page size change
  base: Export platform_msi_domain_[alloc,free]_irqs
  of: MSI: Simplify irqdomain lookup
  irqdomain: Allow domain lookup with DOMAIN_BUS_WIRED token
  irqchip: Fix dependencies for archs w/o HAS_IOMEM
  irqchip/s3c24xx: Mark init_eint as __maybe_unused
  genirq: Validate action before dereferencing it in handle_irq_event_percpu()

of: MSI: Simplify irqdomain lookup

2016-01-26T15:00:15+00:00

So far, when trying to associate a device with its MSI domain,
we first lookup the domain using a MSI token, and if this
doesn't return anything useful, we pick up any domain matching
the same node.

This logic is broken for two reasons:
1) Only the generic MSI code (PCI or platform) sets this token
   to PCI/MSI or platform MSI. So we're guaranteed that if there
   is something to be found, we will find it with the first call.
2) If we have a convoluted situation where:
   - a single node implements both wired and MSI interrupts
   - MSI support for that HW hasn't been compiled in
   we'll end up using the wired domain for MSIs anyway, and things
   break badly.

So let's just remove __of_get_msi_domain, and replace it by a direct
call to irq_find_matching_host, because that's what we really want.

Signed-off-by: Marc Zyngier 
Acked-by: Rob Herring 
Cc: Greg Kroah-Hartman 
Cc: Frank Rowand 
Cc: Grant Likely 
Cc: Thomas Petazzoni 
Cc: Jiang Liu 
Link: http://lkml.kernel.org/r/1453816347-32720-3-git-send-email-marc.zyngier@arm.com
Signed-off-by: Thomas Gleixner

Merge tag 'devicetree-for-4.5' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux

2016-01-14T19:13:28+00:00

Pull DeviceTree updates from Rob Herring:

 - Rework and export the changeset API to make it available to users
   other than DT overlays

 - ARM secure devices binding

 - OCTEON USB binding

 - Clean-up of various SRAM binding docs

 - Various other binding doc updates

* tag 'devicetree-for-4.5' of git://git.kernel.org/pub/scm/linux/kernel/git/robh/linux: (21 commits)
  drivers/of: Export OF changeset functions
  Fix documentation for adp1653 DT
  ARM: psci: Fix indentation in DT bindings
  of/platform: export of_default_bus_match_table
  of/unittest: Show broken behaviour in the platform bus
  of: fix declaration of of_io_request_and_map
  of/address: replace printk(KERN_ERR ...) with pr_err(...)
  of/irq: optimize device node matching loop in of_irq_init()
  dt-bindings: tda998x: Document the required 'port' node.
  net/macb: bindings doc: Merge cdns-emac to macb
  dt-bindings: Misc fix for the ATH79 DDR controllers
  dt-bindings: Misc fix for the ATH79 MISC interrupt controllers
  Documentation: dt: Add bindings for Secure-only devices
  dt-bindings: ARM: add arm,cortex-a72 compatible string
  ASoC: Atmel: ClassD: add GCK's parent clock in DT binding
  DT: add Olimex to vendor prefixes
  Documentation: fsl-quadspi: Add fsl,ls1021-qspi compatible string
  Documentation/devicetree: document OCTEON USB bindings
  usb: misc: usb3503: Describe better how to bind clock to the hub
  dt-bindings: Consolidate SRAM bindings from all vendors
  ...

of/irq: optimize device node matching loop in of_irq_init()

2015-12-09T21:30:57+00:00

Currently, of_irq_init() iterates over interrupt controller nodes
with for_each_matching_node(), and then gets each init function with
of_match_node() later.

This routine can be optimized with for_each_matching_node_and_match().
It allows to get the interrupt controller node and its init function
at the same time, saving __of_match_node() callings.

Signed-off-by: Masahiro Yamada 
Signed-off-by: Rob Herring