6 years agoMerge branch 'next/kvm' into mips-for-linux-next
Ralf Baechle [Thu, 9 May 2013 15:56:40 +0000]
Merge branch 'next/kvm' into mips-for-linux-next

6 years agoKVM/MIPS32: Binary patching of select privileged instructions.
Sanjay Lal [Thu, 22 Nov 2012 02:34:16 +0000]
KVM/MIPS32: Binary patching of select privileged instructions.

Currently, the following instructions are translated:
- CACHE (indexed)
- CACHE (va based): translated to a SYNCI, overkill on D-CACHE operations,
  but still much faster than a trap.
- mfc0/mtc0: the virtual COP0 registers for the guest are implemented as
  2-D array.
  [COP#][SEL] and this is mapped into the guest kernel address space @ VA 0x0.
  mfc0/mtc0 operations are transformed to load/stores.

Signed-off-by: Sanjay Lal <sanjayl@kymasys.com>
Cc: kvm@vger.kernel.org
Cc: linux-mips@linux-mips.org
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoKVM/MIPS32: Do not call vcpu_load when injecting interrupts.
Sanjay Lal [Thu, 22 Nov 2012 02:34:15 +0000]
KVM/MIPS32: Do not call vcpu_load when injecting interrupts.

Signed-off-by: Sanjay Lal <sanjayl@kymasys.com>
Cc: kvm@vger.kernel.org
Cc: linux-mips@linux-mips.org
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoMIPS: Export symbols used by KVM/MIPS module
Sanjay Lal [Thu, 22 Nov 2012 02:34:14 +0000]
MIPS: Export symbols used by KVM/MIPS module

Signed-off-by: Sanjay Lal <sanjayl@kymasys.com>
Cc: kvm@vger.kernel.org
Cc: linux-mips@linux-mips.org
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoMIPS: ASM offsets for VCPU arch specific fields.
Sanjay Lal [Thu, 22 Nov 2012 02:34:12 +0000]
MIPS: ASM offsets for VCPU arch specific fields.

Signed-off-by: Sanjay Lal <sanjayl@kymasys.com>
Cc: kvm@vger.kernel.org
Cc: linux-mips@linux-mips.org
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoMIPS: If KVM is enabled then use the KVM specific routine to flush the TLBs on a...
Sanjay Lal [Thu, 22 Nov 2012 02:34:11 +0000]
MIPS: If KVM is enabled then use the KVM specific routine to flush the TLBs on a ASID wrap.

Signed-off-by: Sanjay Lal <sanjayl@kymasys.com>
Cc: kvm@vger.kernel.org
Cc: linux-mips@linux-mips.org
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoMIPS: Export routines needed by the KVM module.
Sanjay Lal [Thu, 22 Nov 2012 02:34:10 +0000]
MIPS: Export routines needed by the KVM module.

Signed-off-by: Sanjay Lal <sanjayl@kymasys.com>
Cc: kvm@vger.kernel.org
Cc: linux-mips@linux-mips.org
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoKVM/MIPS32: Routines to handle specific traps/exceptions while executing the guest.
Sanjay Lal [Thu, 22 Nov 2012 02:34:09 +0000]
KVM/MIPS32: Routines to handle specific traps/exceptions while executing the guest.

Signed-off-by: Sanjay Lal <sanjayl@kymasys.com>
Cc: kvm@vger.kernel.org
Cc: linux-mips@linux-mips.org
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoKVM/MIPS32: Guest interrupt delivery.
Sanjay Lal [Thu, 22 Nov 2012 02:34:08 +0000]
KVM/MIPS32: Guest interrupt delivery.

Signed-off-by: Sanjay Lal <sanjayl@kymasys.com>
Cc: kvm@vger.kernel.org
Cc: linux-mips@linux-mips.org
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoKVM/MIPS32: COP0 accesses profiling.
Sanjay Lal [Thu, 22 Nov 2012 02:34:07 +0000]
KVM/MIPS32: COP0 accesses profiling.

Signed-off-by: Sanjay Lal <sanjayl@kymasys.com>
Cc: kvm@vger.kernel.org
Cc: linux-mips@linux-mips.org
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoKVM/MIPS32: Release notes and KVM module Makefile
Sanjay Lal [Thu, 22 Nov 2012 02:34:06 +0000]
KVM/MIPS32: Release notes and KVM module Makefile

Signed-off-by: Sanjay Lal <sanjayl@kymasys.com>
Cc: kvm@vger.kernel.org
Cc: linux-mips@linux-mips.org
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoKVM/MIPS32: MMU/TLB operations for the Guest.
Sanjay Lal [Thu, 22 Nov 2012 02:34:05 +0000]
KVM/MIPS32: MMU/TLB operations for the Guest.

- Note that this file is statically linked with the rest of the host kernel (KSEG0). This is because kernel modules are
loaded into mapped space on MIPS and we want to make sure that we don't get any host kernel TLB faults while
manipulating TLBs.
- Virtual Guest TLBs are implemented as 64 entry array regardless of the number of host TLB entries.
- Shadow TLBs map Guest virtual addresses to Host physical addresses.

    - TLB miss handling details:
        Guest KSEG0 TLBMISS (0x40000000 – 0x60000000): Transparent to the Guest.
        Guest KSEG2/3 (0x60000000 – 0x80000000) & Guest UM TLBMISS (0x00000000 – 0x40000000)
            Lookup in Guest/Virtual TLB
            If an entry doesn’t match
                deliver appropriate TLBMISS LD/ST exception to the guest
            If entry does exist in the Guest TLB and is NOT Valid
                Deliver TLB invalid exception to the guest
            If entry does exist in the Guest TLB and is VALID
                Inject the TLB entry into the Shadow TLB

Signed-off-by: Sanjay Lal <sanjayl@kymasys.com>
Cc: kvm@vger.kernel.org
Cc: linux-mips@linux-mips.org
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoKVM/MIPS32: Privileged instruction/target branch emulation.
Sanjay Lal [Thu, 22 Nov 2012 02:34:04 +0000]
KVM/MIPS32: Privileged instruction/target branch emulation.

- The Guest kernel is run in UM and privileged instructions cause a trap.
- If the instruction causing the trap is in a branch delay slot, the branch
  needs to be emulated to figure out the PC @ which the guest will resume
  execution.

Signed-off-by: Sanjay Lal <sanjayl@kymasys.com>
Cc: kvm@vger.kernel.org
Cc: linux-mips@linux-mips.org
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoKVM/MIPS32: KVM Guest kernel support.
Sanjay Lal [Thu, 22 Nov 2012 02:34:03 +0000]
KVM/MIPS32: KVM Guest kernel support.

Both Guest kernel and Guest Userspace execute in UM. The memory map is as follows:
Guest User address space:   0x00000000 -> 0x40000000
Guest Kernel Unmapped:      0x40000000 -> 0x60000000
Guest Kernel Mapped:        0x60000000 -> 0x80000000
- Guest Usermode virtual memory is limited to 1GB.

Signed-off-by: Sanjay Lal <sanjayl@kymasys.com>
Cc: kvm@vger.kernel.org
Cc: linux-mips@linux-mips.org
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoKVM/MIPS32: MIPS arch specific APIs for KVM
Sanjay Lal [Thu, 22 Nov 2012 02:34:02 +0000]
KVM/MIPS32: MIPS arch specific APIs for KVM

- Implements the arch specific APIs for KVM, some are stubs for MIPS
- kvm_mips_handle_exit(): Main 'C' distpatch routine for handling exceptions while in "Guest" mode.
- Also implements in-kernel timer interrupt support for the guest.

Signed-off-by: Sanjay Lal <sanjayl@kymasys.com>
Cc: kvm@vger.kernel.org
Cc: linux-mips@linux-mips.org
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoKVM/MIPS32: Entry point for trampolining to the guest and trap handlers.
Sanjay Lal [Thu, 22 Nov 2012 02:34:01 +0000]
KVM/MIPS32: Entry point for trampolining to the guest and trap handlers.

- __kvm_mips_vcpu_run: main entry point to enter guest, we save kernel context, load
  up guest context from and ERET to guest context.
- mips32_exception: L1 exception handler(s), save k0/k1 and jump to main handlers.
- mips32_GuestException: Generic exception handlers for exceptions/interrupts while in
  guest context.  Save guest context, restore some kernel context and jump to
  main 'C' handler: kvm_mips_handle_exit()

Signed-off-by: Sanjay Lal <sanjayl@kymasys.com>
Cc: kvm@vger.kernel.org
Cc: linux-mips@linux-mips.org
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoKVM/MIPS32: Arch specific KVM data structures.
Sanjay Lal [Thu, 22 Nov 2012 02:34:00 +0000]
KVM/MIPS32: Arch specific KVM data structures.

Signed-off-by: Sanjay Lal <sanjayl@kymasys.com>
Cc: kvm@vger.kernel.org
Cc: linux-mips@linux-mips.org
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoKVM/MIPS32: Infrastructure/build files.
Sanjay Lal [Thu, 22 Nov 2012 02:33:59 +0000]
KVM/MIPS32: Infrastructure/build files.

- Add the KVM option to MIPS build files.
- Add default config files for KVM host/guest kernels.
- Change the link address for the Malta KVM Guest kernel to UM (0x40100000).
- Add KVM Kconfig file with KVM/MIPS specific options

Signed-off-by: Sanjay Lal <sanjayl@kymasys.com>
Cc: kvm@vger.kernel.org
Cc: linux-mips@linux-mips.org
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoMIPS: IP27: Remove pfn_t.
Ralf Baechle [Wed, 8 May 2013 01:51:58 +0000]
MIPS: IP27: Remove pfn_t.

In the Linux kernel traditionally pfns are represented by an unsigned long.
However a few bits of the SGI IP27 platform code that were ported from
IRIX are using pfn_t for historic reasons.  This is conflicting with
KVM's use of pfn_t.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoMerge branch 'mips-next-3.10' of git://git.linux-mips.org/pub/scm/john/linux-john...
Ralf Baechle [Tue, 7 May 2013 23:27:46 +0000]
Merge branch 'mips-next-3.10' of git://git.linux-mips.org/john/linux-john into mips-for-linux-next

6 years agoMIPS: BCM63XX: add missing clocks for BCM6328 and BCM6362
Jonas Gorski [Mon, 22 Apr 2013 10:57:06 +0000]
MIPS: BCM63XX: add missing clocks for BCM6328 and BCM6362

Add some mosty unused, but missing clocks for BCM6328 and BCM6362.
This also fixes PCIe init on BCM6362.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5200/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: ath79: make use of the new memory detection code
John Crispin [Mon, 15 Apr 2013 09:45:09 +0000]
MIPS: ath79: make use of the new memory detection code

There is now a generic function for detecting memory size. Use this instead of
the one found in the ath79 support.

Signed-off-by: John Crispin <blogic@openwrt.org>
Acked-by: Gabor Juhos <juhosg@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5149/

6 years agoMIPS: ralink: make use of the new memory detection code
John Crispin [Sat, 13 Apr 2013 13:15:51 +0000]
MIPS: ralink: make use of the new memory detection code

Call detect_memory_region() from plat_mem_setup() unless the size was already
read from the system controller.

Signed-off-by: John Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5184/

6 years agoMIPS: ralink: add memory definition for MT7620
John Crispin [Sun, 14 Apr 2013 07:55:29 +0000]
MIPS: ralink: add memory definition for MT7620

Populate struct soc_info with the data that describes our RAM window.

Signed-off-by: John Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5183/

6 years agoMIPS: ralink: add memory definition for RT3883
John Crispin [Sat, 13 Apr 2013 18:23:19 +0000]
MIPS: ralink: add memory definition for RT3883

Populate struct soc_info with the data that describes our RAM window.

Signed-off-by: John Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5182/

6 years agoMIPS: ralink: add memory definition for RT2880
John Crispin [Sat, 13 Apr 2013 13:37:37 +0000]
MIPS: ralink: add memory definition for RT2880

Populate struct soc_info with the data that describes our RAM window.

Signed-off-by: John Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5181/

6 years agoMIPS: ralink: add memory definition for RT305x
John Crispin [Sat, 13 Apr 2013 13:13:40 +0000]
MIPS: ralink: add memory definition for RT305x

Populate struct soc_info with the data that describes our RAM window.

As memory detection fails on RT5350 we read the amount of available memory
from the system controller.

Signed-off-by: John Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5180/

6 years agoMIPS: ralink: add memory definition to struct ralink_soc_info
John Crispin [Sat, 13 Apr 2013 13:10:14 +0000]
MIPS: ralink: add memory definition to struct ralink_soc_info

Depending on the actual SoC we have a different base address as well as minimum
and maximum size for RAM. Add these fields to the per SoC structure.

Signed-off-by: John Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5179/

6 years agoMIPS: add detect_memory_region()
John Crispin [Sat, 13 Apr 2013 11:15:47 +0000]
MIPS: add detect_memory_region()

Add a generic way of detecting the available RAM. This function is based on the
implementation already used by ath79.

Signed-off-by: John Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5178/

6 years agoDT: MIPS: ralink: add MT7620A dts files
John Crispin [Fri, 12 Apr 2013 06:27:41 +0000]
DT: MIPS: ralink: add MT7620A dts files

Add a dtsi file for MT7620A SoC and a sample dts file.

Signed-off-by: John Crispin <blogic@openwrt.org>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Patchwork: http://patchwork.linux-mips.org/patch/5190/

6 years agoDT: MIPS: ralink: add RT3883 dts files
John Crispin [Fri, 12 Apr 2013 06:27:39 +0000]
DT: MIPS: ralink: add RT3883 dts files

Add a dtsi file for RT3883 SoC and a sample dts file.

Signed-off-by: John Crispin <blogic@openwrt.org>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Patchwork: http://patchwork.linux-mips.org/patch/5189/

6 years agoDT: MIPS: ralink: add RT2880 dts files
John Crispin [Fri, 12 Apr 2013 06:27:37 +0000]
DT: MIPS: ralink: add RT2880 dts files

Add a dtsi file for RT2880 SoC and a sample dts file.

Signed-off-by: John Crispin <blogic@openwrt.org>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Patchwork: http://patchwork.linux-mips.org/patch/5188/

6 years agoDT: MIPS: ralink: clean up RT3050 dtsi and dts file
John Crispin [Thu, 21 Mar 2013 16:47:07 +0000]
DT: MIPS: ralink: clean up RT3050 dtsi and dts file

* remove nodes for cores whose drivers are not upstream yet
* add compat string for an additional soc
* fix a whitespace error

Signed-off-by: John Crispin <blogic@openwrt.org>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Patchwork: http://patchwork.linux-mips.org/patch/5186/

6 years agoDT: add documentation for the Ralink MIPS SoCs
Gabor Juhos [Sat, 13 Apr 2013 07:02:40 +0000]
DT: add documentation for the Ralink MIPS SoCs

This patch adds binding documentation for the
compatible values of the Ralink MIPS SoCs.

Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
Signed-off-by: John Crispin <blogic@openwrt.org>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Patchwork: http://patchwork.linux-mips.org/patch/5187/

6 years agoDT: add vendor prefixes for Ralink
John Crispin [Sat, 13 Apr 2013 08:11:51 +0000]
DT: add vendor prefixes for Ralink

Signed-off-by: John Crispin <blogic@openwrt.org>
Acked-by: Grant Likely <grant.likely@secretlab.ca>

6 years agoMIPS: ralink: add cpu-feature-overrides.h
Gabor Juhos [Wed, 10 Apr 2013 07:19:07 +0000]
MIPS: ralink: add cpu-feature-overrides.h

Add cpu-feature-overrides.h for RT288x, RT305x and RT3883.

Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
Signed-off-by: John Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5175/

6 years agoMIPS: ralink: adds support for MT7620 SoC family
John Crispin [Thu, 21 Mar 2013 16:49:02 +0000]
MIPS: ralink: adds support for MT7620 SoC family

Add support code for mt7620 SOC.

The code detects the SoC and registers the clk / pinmux settings.

Signed-off-by: John Crispin <blogic@openwrt.org>
Acked-by: Gabor Juhos <juhosg@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5177/

6 years agoMIPS: ralink: adds support for RT3883 SoC family
John Crispin [Sun, 27 Jan 2013 08:39:02 +0000]
MIPS: ralink: adds support for RT3883 SoC family

Add support code for rt3883 SOC.

The code detects the SoC and registers the clk / pinmux settings.

Signed-off-by: John Crispin <blogic@openwrt.org>
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5185/

6 years agoMIPS: ralink: adds support for RT2880 SoC family
John Crispin [Sun, 27 Jan 2013 08:17:20 +0000]
MIPS: ralink: adds support for RT2880 SoC family

Add support code for rt2880 SOC.

The code detects the SoC and registers the clk / pinmux settings.

Signed-off-by: John Crispin <blogic@openwrt.org>
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5176/

6 years agoMIPS: ralink: add uart mask to struct ralink_pinmux
John Crispin [Fri, 12 Apr 2013 10:45:27 +0000]
MIPS: ralink: add uart mask to struct ralink_pinmux

Add a field for the uart muxing mask and set it inside the rt305x setup code.

Signed-off-by: John Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5744/

6 years agoMIPS: ralink: add pci group to struct ralink_pinmux
Gabor Juhos [Fri, 12 Apr 2013 10:40:23 +0000]
MIPS: ralink: add pci group to struct ralink_pinmux

This will be used for RT3662/RT3883.

Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
Acked-by: John Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5173/

6 years agoMIPS: ralink: make the RT305x pinmuxing structure static
John Crispin [Fri, 12 Apr 2013 20:16:12 +0000]
MIPS: ralink: make the RT305x pinmuxing structure static

These structures are exported via struct ralink_pinmux rt_gpio_pinmux and can
hence be static.

Signed-off-by: John Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5172/

6 years agoMIPS: ralink: rename gpio_pinmux to rt_gpio_pinmux
John Crispin [Fri, 12 Apr 2013 20:12:09 +0000]
MIPS: ralink: rename gpio_pinmux to rt_gpio_pinmux

Add proper namespacing to the variable.

Signed-off-by: John Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5171/

6 years agoMIPS: ralink: make early_printk work on RT2880
John Crispin [Tue, 9 Apr 2013 16:31:15 +0000]
MIPS: ralink: make early_printk work on RT2880

RT2880 has a different location for the early serial port.

Signed-off-by: John Crispin <blogic@openwrt.org>
Acked-by: Gabor Juhos <juhosg@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5170/

6 years agoMIPS: ralink: add RT5350 sdram register defines
John Crispin [Mon, 25 Mar 2013 10:19:58 +0000]
MIPS: ralink: add RT5350 sdram register defines

Add a few missing defines that are needed to make memory detection work on the
RT5350.

Signed-off-by: John Crispin <blogic@openwrt.org>
Acked-by: Gabor Juhos <juhosg@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5169/

6 years agoMIPS: ralink: add missing comment in irq driver
John Crispin [Sat, 16 Mar 2013 15:28:54 +0000]
MIPS: ralink: add missing comment in irq driver

Trivial patch that adds a comment that makes the code more readable.

Signed-off-by: John Crispin <blogic@openwrt.org>
Acked-by: Gabor Juhos <juhosg@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5168/

6 years agoMIPS: ralink: fix RT305x clock setup
John Crispin [Fri, 22 Mar 2013 18:25:59 +0000]
MIPS: ralink: fix RT305x clock setup

Add a few missing clocks.

Signed-off-by: John Crispin <blogic@openwrt.org>
Acked-by: Gabor Juhos <juhosg@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5167/

6 years agoMIPS: ralink: add RT3352 register defines
John Crispin [Thu, 21 Mar 2013 18:01:49 +0000]
MIPS: ralink: add RT3352 register defines

Add a few missing defines that are needed to make USB and clock detection work
on the RT3352.

Signed-off-by: John Crispin <blogic@openwrt.org>
Acked-by: Gabor Juhos <juhosg@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5166/

6 years agoMIPS: ralink: add PCI IRQ handling
Gabor Juhos [Wed, 10 Apr 2013 07:07:27 +0000]
MIPS: ralink: add PCI IRQ handling

The Ralink IRQ code was not handling the PCI IRQ yet. Add this functionaility
to make PCI work on rt3883.

Signed-off-by: John Crispin <blogic@openwrt.org>
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5165/

6 years agoMIPS: move mips_{set,get}_machine_name() to a more generic place
John Crispin [Thu, 11 Apr 2013 05:34:59 +0000]
MIPS: move mips_{set,get}_machine_name() to a more generic place

Previously this functionality was only available to users of the mips_machine
api. Moving the code to prom.c allows us to also add a OF wrapper.

Signed-off-by: John Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5164/

6 years agoMIPS: octeon: Fix GPIO number in IRQ chip private data
Alexander Sverdlin [Thu, 11 Apr 2013 15:29:39 +0000]
MIPS: octeon: Fix GPIO number in IRQ chip private data

Current GPIO chip implementation in octeon-irq is still broken, even after upstream
commit 87161ccdc61862c8b49e75c21209d7f79dc758e9 (MIPS: Octeon: Fix broken interrupt
controller code). It works for GPIO IRQs that have reset-default configuration, but
not for edge-triggered ones.

The problem is in octeon_irq_gpio_map_common(), which passes modified "hw" variable
(which has range of possible values 16..31) as "gpio_line" parameter to
octeon_irq_set_ciu_mapping(), which saves it in private data of the IRQ chip. Later,
neither octeon_irq_gpio_setup() is able to re-configure GPIOs (cvmx_write_csr() is
writing to non-existent CVMX_GPIO_BIT_CFGX), nor octeon_irq_ciu_gpio_ack() is able
to acknowledge such IRQ, because "mask" is incorrect.

Fix is trivial and has been tested on Cavium Octeon II -based board, including
both level-triggered and edge-triggered GPIO IRQs.

Signed-off-by: Alexander Sverdlin <alexander.sverdlin.ext@nsn.com>
Cc: David Daney <david.daney@cavium.com>
Acked-by: David Daney <david.daney@cavium.com>
Patchwork: http://patchwork.linux-mips.org/patch/4980/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: pci: convert to devm_ioremap_resource()
Silviu-Mihai Popescu [Tue, 12 Mar 2013 10:30:05 +0000]
MIPS: pci: convert to devm_ioremap_resource()

Convert all uses of devm_request_and_ioremap() to the newly introduced
devm_ioremap_resource() which provides more consistent error handling.

Signed-off-by: Silviu-Mihai Popescu <silviupopescu1990@gmail.com>
Patchwork: http://patchwork.linux-mips.org/patch/4986/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: Remove unneeded volatile from arch/mips/lib/bitops.c
David Daney [Tue, 26 Feb 2013 22:22:33 +0000]
MIPS: Remove unneeded volatile from arch/mips/lib/bitops.c

The operations on the bitmap pointers are protected by "memory"
clobbering raw_local_irq_{save,restore}(), so there is no need for
volatile here.  By removing the volatile we get better code generation
out of the compiler.

Signed-off-by: David Daney <david.daney@cavium.com>
Patchwork: http://patchwork.linux-mips.org/patch/4966/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: Init new mmu_context for each possible CPU to avoid memory corruption
Huacai Chen [Sun, 17 Mar 2013 11:50:14 +0000]
MIPS: Init new mmu_context for each possible CPU to avoid memory corruption

Currently, init_new_context() only for each online CPU, this may cause
memory corruption when CPU hotplug and fork() happens at the same time.
To avoid this, we make init_new_context() cover each possible CPU.

Scenario:
1, CPU#1 is being offline;
2, On CPU#0, do_fork() call dup_mm() and copy a mm_struct to the child;
3, On CPU#0, dup_mm() call init_new_context(), since CPU#1 is offline
   and init_new_context() only covers the online CPUs, child has the
   same asid as its parent on CPU#1 (however, child's asid should be 0);
4, CPU#1 is being online;
5, Now, if both parent and child run on CPU#1, memory corruption (e.g.
   segfault, bus error, etc.) will occur.

Signed-off-by: Huacai Chen <chenhc@lemote.com>
Acked-by: David Daney <david.daney@cavium.com>
Patchwork: http://patchwork.linux-mips.org/patch/4995/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: Build uasm-generated code only once to avoid CPU Hotplug problem
Huacai Chen [Sun, 17 Mar 2013 11:49:38 +0000]
MIPS: Build uasm-generated code only once to avoid CPU Hotplug problem

This and the next patch resolve memory corruption problems while CPU
hotplug. Without these patches, memory corruption can triggered easily
as below:

On a quad-core MIPS platform, use "spawn" of UnixBench-5.1.3 (http://
code.google.com/p/byte-unixbench/) and a CPU hotplug script like this
(hotplug.sh):
while true; do
echo 0 >/sys/devices/system/cpu/cpu1/online
echo 0 >/sys/devices/system/cpu/cpu2/online
echo 0 >/sys/devices/system/cpu/cpu3/online
sleep 1
echo 1 >/sys/devices/system/cpu/cpu1/online
echo 1 >/sys/devices/system/cpu/cpu2/online
echo 1 >/sys/devices/system/cpu/cpu3/online
sleep 1
done

Run "hotplug.sh" and then run "spawn 10000", spawn will get segfault
after a few minutes.

This patch:
Currently, clear_page()/copy_page() are generated by Micro-assembler
dynamically. But they are unavailable until uasm_resolve_relocs() has
finished because jump labels are illegal before that. Since these
functions are shared by every CPU, we only call build_clear_page()/
build_copy_page() only once at boot time. Without this patch, programs
will get random memory corruption (segmentation fault, bus error, etc.)
while CPU Hotplug (e.g. one CPU is using clear_page() while another is
generating it in cpu_cache_init()).

For similar reasons we modify build_tlb_refill_handler()'s invocation.

V2:
1, Rework the code to make CPU#0 can be online/offline.
2, Introduce cpu_has_local_ebase feature since some types of MIPS CPU
   need a per-CPU tlb_refill_handler().

Signed-off-by: Huacai Chen <chenhc@lemote.com>
Signed-off-by: Hongbing Hu <huhb@lemote.com>
Acked-by: David Daney <david.daney@cavium.com>
Patchwork: http://patchwork.linux-mips.org/patch/4994/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: Kconfig: remove "config MIPS_DISABLE_OBSOLETE_IDE"
Paul Bolle [Mon, 8 Apr 2013 17:51:13 +0000]
MIPS: Kconfig: remove "config MIPS_DISABLE_OBSOLETE_IDE"

The Kconfig symbol MIPS_DISABLE_OBSOLETE_IDE was added in v2.6.10. It
has never been used. Let's remove it.

The symbol was originally introduced by the following commit

commit 2bfa662b64a7ee593f3039c1d3fd81a7766a63cd
Author: Pete Popov <ppopov@embeddedalley.com>
Date:   Tue Oct 12 06:24:19 2004 +0000
    - Db1550 bug fixes
    - updated defconfig
    - updated Kconfig to use DMA_COHERENT since new silicon is coherent

Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
Patchwork: http://patchwork.linux-mips.org/patch/5064/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: Kconfig: remove "config MIPS_BOARDS_GEN"
Paul Bolle [Mon, 8 Apr 2013 17:06:31 +0000]
MIPS: Kconfig: remove "config MIPS_BOARDS_GEN"

The Kconfig symbol MIPS_BOARDS_GEN is unused since v2.6.27. It should
now be removed.

Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
Patchwork: http://patchwork.linux-mips.org/patch/5063/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: remove obsolete Kconfig macros
Paul Bolle [Mon, 25 Mar 2013 20:27:19 +0000]
MIPS: remove obsolete Kconfig macros

The support for PB1100, PB1500, and PB1550 got merged into the code for
DB1000 and DB1550 code in v3.7. When that was done the three related
Kconfig symbols were dropped. But not all related Kconfig macros were
removed. Do so now.

Note that the PB1100 code in the Au1100 LCD driver is removed entirely
and not converted to use its current Kconfig macro. That is done because
the macros it uses (PB1100_G_CONTROL, PB1100_G_CONTROL_BL, and
PB1100_G_CONTROL_VDD) are never defined. Actually only one of these was
ever defined (PB1100_G_CONTROL) but that define was removed in v2.6.34.
So, as far as I can tell, this code could have never compiled.

Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
Patchwork: http://patchwork.linux-mips.org/patch/5040/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: Netlogic: Fix oprofile compile on XLR uniprocessor
Jayachandran C [Mon, 25 Mar 2013 06:51:52 +0000]
MIPS: Netlogic: Fix oprofile compile on XLR uniprocessor

The commit c783390a0ecef08df5c804f8c5f647431a04f502 [MIPS: oprofile:
Support for XLR/XLS processors] causes a compilation failure when
oprofile is enabled and SMP is not configured.

arch/mips/oprofile/op_model_mipsxx.c: In function 'mipsxx_cpu_setup':
arch/mips/oprofile/op_model_mipsxx.c:181:2: error: implicit declaration of function 'cpu_logical_map'

To fix this, update oprofile_skip_cpu to not call cpu_logical_map when
CONFIG_SMP is not defined.

Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/5037/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: Netlogic: Merge platform usb.h to usb-init.c
Jayachandran C [Sat, 23 Mar 2013 17:28:00 +0000]
MIPS: Netlogic: Merge platform usb.h to usb-init.c

The definitions are not used anywhere else, and merging it will
make adding the new USB definitions for XLPII series easier.
While there, cleanup some whitespace in usb-init.c. There is no
change to logic due to this commit.

Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/5027/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: Netlogic: Support for multiple built-in device trees
Jayachandran C [Sat, 23 Mar 2013 17:27:59 +0000]
MIPS: Netlogic: Support for multiple built-in device trees

This enables us to have a default device tree per SoC family to be built
into the kernel. The default device tree for XLP3xx has been added as part
of this change. Later this can be used to provide support default boards
for XLP2xx and XLP9xx SoCs.

Kconfig options are provided for each default device tree so that just the
needed ones can be selected to be built into the kernel.

Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/5023/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: Netlogic: Remove unused code
Jayachandran C [Sat, 23 Mar 2013 17:27:58 +0000]
MIPS: Netlogic: Remove unused code

Remove unused functions and redundant comments from
arch/mips/include/asm/netlogic/haldefs.h

Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/5029/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: Netlogic: Add 32-bit support for XLP
Jayachandran C [Sat, 23 Mar 2013 17:27:57 +0000]
MIPS: Netlogic: Add 32-bit support for XLP

Update asm/netlogic/haldefs.h to extend register access functions
nlm_{read,write}_reg64() for 32-bit compilation. When compiled for 32-bit
the functions will read 64 IO registers with interrupts disabled.

Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/5026/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: Netlogic: Avoid using fixed PIC IRT index
Jayachandran C [Sat, 23 Mar 2013 17:27:56 +0000]
MIPS: Netlogic: Avoid using fixed PIC IRT index

The index for a device interrupt in the PIC interrupt routing table
changes for different chips in the XLP family.  Avoid using the fixed
entries and derive the index value from the SoC device header.

Add workarounds for some devices which do not report the IRT index
correctly.

Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/5025/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: Netlogic: print cpumask with cpumask_scnprintf
Jayachandran C [Sat, 23 Mar 2013 17:27:55 +0000]
MIPS: Netlogic: print cpumask with cpumask_scnprintf

Use standard function to print cpumask. Also fixup the name of the
variable used and make it static.

Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/5024/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: Netlogic: Remove unused EIMR/EIRR functions
Jayachandran C [Sat, 23 Mar 2013 17:27:54 +0000]
MIPS: Netlogic: Remove unused EIMR/EIRR functions

Remove the definitions of {read,write}_c0_{eirr,eimr}. These functions
are now unused after the PIC and IRQ code has been updated to use
optimized EIMR/EIRR functions which work on both 32-bit and 64-bit.

Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/5021/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: Netlogic: Optimize and fix write_c0_eimr()
Jayachandran C [Sat, 23 Mar 2013 17:27:53 +0000]
MIPS: Netlogic: Optimize and fix write_c0_eimr()

Remove the irq save/restore from write_c0_eimr(), as it is always called
with interrupts off.

This allows us to remove workaround in write_c0_eimr() to fix up the
flags used by local_irq_save. This fixup worked on XLR, but will break
when 32-bit support is added to r2 cpus like XLP.

Signed-off-by: Jayachandran C <jchandra@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/5022/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: BCM63XX: merge bcm63xx_clk.h into bcm63xx/clk.c
Jonas Gorski [Sat, 6 Apr 2013 10:31:02 +0000]
MIPS: BCM63XX: merge bcm63xx_clk.h into bcm63xx/clk.c

All the header file does is provide the internal structure of clk,
which shouldn't be used by anyone except clk.c itself anyway.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Patchwork: http://patchwork.linux-mips.org/patch/5055/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: BCM63XX: add flash detection for BCM6362
Jonas Gorski [Thu, 21 Mar 2013 14:03:20 +0000]
MIPS: BCM63XX: add flash detection for BCM6362

BCM6362 support booting from SPI flash and NAND.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5012/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: BCM63XX: enable pcie for BCM6362
Jonas Gorski [Thu, 21 Mar 2013 14:03:19 +0000]
MIPS: BCM63XX: enable pcie for BCM6362

The PCIe controller is almost the same as the BCM6328 one, with only
the SERDES register being at a different location.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5011/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: BCM63XX: enable SPI controller for BCM6362
Jonas Gorski [Thu, 21 Mar 2013 14:03:18 +0000]
MIPS: BCM63XX: enable SPI controller for BCM6362

The SPI controller shares the same register layout as the 6358 one.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5010/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: BCM63XX: add basic BCM6362 support
Jonas Gorski [Thu, 21 Mar 2013 14:03:17 +0000]
MIPS: BCM63XX: add basic BCM6362 support

Add basic support for detecting and booting the BCM6362.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5009/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: BCM63XX: rework chip detection
Jonas Gorski [Thu, 21 Mar 2013 14:03:16 +0000]
MIPS: BCM63XX: rework chip detection

Instead of trying to use a correlation of cpu prid and chip id and
hoping they will always be unique, use the cpu prid to determine the
chip id register location and just read out the chip id.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5008/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: BCM63XX: fix revision ID width
Jonas Gorski [Thu, 21 Mar 2013 14:03:15 +0000]
MIPS: BCM63XX: fix revision ID width

The REVID is only 8 bit wide.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5007/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: BCM63XX: remove duplicate spi register definitions
Jonas Gorski [Thu, 21 Mar 2013 14:03:14 +0000]
MIPS: BCM63XX: remove duplicate spi register definitions

BCM6338 and BCM6348, and BCM6358 and everything after that share the
same register layout. To not have to redefine them for each new chip
and keep the code size small, only use the definitions for the first
chip with the certain layout.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/5006/
Acked-by: John Crispin <blogic@openwrt.org>

6 years agoMIPS: Remove redundant instructions from arch_spin_{,try}lock.
David Daney [Thu, 25 Apr 2013 15:48:53 +0000]
MIPS: Remove redundant instructions from arch_spin_{,try}lock.

We were doing:
   SRL  $r,$?,16
   ANDI $r,$r,0xffff

The logical right shift by 16 leaves the upper 16 bits clear, so the
subsequent masking out of those bits is redundant, and can safely be
removed.

Signed-off-by: David Daney <david.daney@cavium.com>
Cc: linux-mips@linux-mips.org
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

6 years agoLinux 3.9-rc7
Linus Torvalds [Mon, 15 Apr 2013 00:45:16 +0000]
Linux 3.9-rc7

6 years agoMerge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 14 Apr 2013 18:13:24 +0000]
Merge branch 'x86-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull x86 fixes from Ingo Molnar:
 "Misc fixes"

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/mm: Flush lazy MMU when DEBUG_PAGEALLOC is set
  x86/mm/cpa/selftest: Fix false positive in CPA self test
  x86/mm/cpa: Convert noop to functional fix
  x86, mm: Patch out arch_flush_lazy_mmu_mode() when running on bare metal
  x86, mm, paravirt: Fix vmalloc_fault oops during lazy MMU updates

6 years agoMerge branch 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 14 Apr 2013 18:12:17 +0000]
Merge branch 'sched-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull scheduler fixes from Ingo Molnar:
 "Misc fixlets"

* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched/cputime: Fix accounting on multi-threaded processes
  sched/debug: Fix sd->*_idx limit range avoiding overflow
  sched_clock: Prevent 64bit inatomicity on 32bit systems
  sched: Convert BUG_ON()s in try_to_wake_up_local() to WARN_ON_ONCE()s

6 years agoMerge branch 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 14 Apr 2013 18:10:44 +0000]
Merge branch 'perf-urgent-for-linus' of git://git./linux/kernel/git/tip/tip

Pull perf fixes from Ingo Molnar:
 "Misc fixlets"

* 'perf-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  perf: Fix error return code
  ftrace: Fix strncpy() use, use strlcpy() instead of strncpy()
  perf: Fix strncpy() use, use strlcpy() instead of strncpy()
  perf: Fix strncpy() use, always make sure it's NUL terminated
  perf: Fix ring_buffer perf_output_space() boundary calculation
  perf/x86: Fix uninitialized pt_regs in intel_pmu_drain_bts_buffer()

6 years agoMerge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux
Linus Torvalds [Sun, 14 Apr 2013 17:55:20 +0000]
Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux

Pull drm fixes from Dave Airlie:
 "One fix for a hotplug locking regressions, and one fix for an oops if
  you unplug the monitor at an inopportune moment on the udl device."

* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux:
  drm/fb-helper: Fix locking in drm_fb_helper_hotplug_event
  udl: handle EDID failure properly.

6 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/gerg/m68knommu
Linus Torvalds [Sun, 14 Apr 2013 17:54:40 +0000]
Merge branch 'for-linus' of git://git./linux/kernel/git/gerg/m68knommu

Pull m68knommu fix from Greg Ungerer:
 "This contains only a single compilation fix for ColdFire m68k targets
  that use local non-GPIOLIB support."

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/gerg/m68knommu:
  m68k: define a local gpio_request_one() function

6 years agoMerge git://www.linux-watchdog.org/linux-watchdog
Linus Torvalds [Sun, 14 Apr 2013 17:53:54 +0000]
Merge git://www.linux-watchdog.org/linux-watchdog

Pull watchdog fix from Wim Van Sebroeck:
 "It will fix compile errors for the at91rm9200_wdt driver"

* git://www.linux-watchdog.org/linux-watchdog:
  watchdog: Revert the AT91RM9200_WATCHDOG dependency

6 years agoMerge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux...
Linus Torvalds [Sun, 14 Apr 2013 17:52:54 +0000]
Merge branch 'for-linus' of git://git./linux/kernel/git/mason/linux-btrfs

Pull one more btrfs fix from Chris Mason:
 "This has a recent fix from Josef for our tree log replay code.  It
  fixes problems where the inode counter for the number of bytes in the
  file wasn't getting updated properly during fsync replay.

  The commit did get rebased this morning, but it was only to clean up
  the subject line.  The code hasn't changed."

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
  Btrfs: make sure nbytes are right after log replay

6 years agoMerge tag 'trace-fixes-v3.9-rc-v3' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sun, 14 Apr 2013 17:50:55 +0000]
Merge tag 'trace-fixes-v3.9-rc-v3' of git://git./linux/kernel/git/rostedt/linux-trace

Pull ftrace fixes from Steven Rostedt:
 "Namhyung Kim found and fixed a bug that can crash the kernel by simply
  doing: echo 1234 | tee -a /sys/kernel/debug/tracing/set_ftrace_pid

  Luckily, this can only be done by root, but still is a nasty bug."

* tag 'trace-fixes-v3.9-rc-v3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
  ftrace: Move ftrace_filter_lseek out of CONFIG_DYNAMIC_FTRACE section
  tracing: Fix possible NULL pointer dereferences

6 years agoAdd file_ns_capable() helper function for open-time capability checking
Linus Torvalds [Sun, 14 Apr 2013 17:06:31 +0000]
Add file_ns_capable() helper function for open-time capability checking

Nothing is using it yet, but this will allow us to delay the open-time
checks to use time, without breaking the normal UNIX permission
semantics where permissions are determined by the opener (and the file
descriptor can then be passed to a different process, or the process can
drop capabilities).

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

6 years agowatchdog: Revert the AT91RM9200_WATCHDOG dependency
Nicolas Ferre [Wed, 10 Apr 2013 12:36:22 +0000]
watchdog: Revert the AT91RM9200_WATCHDOG dependency

Compiling the at91rm9200_wdt.c driver without at91rm9200
support was leading to several errors:

drivers/built-in.o: In function `at91_wdt_close':
at91_adc.c:(.text+0xc9fe4): undefined reference to `at91_st_base'
drivers/built-in.o: In function `at91_wdt_write':
at91_adc.c:(.text+0xca004): undefined reference to `at91_st_base'
drivers/built-in.o: In function `at91wdt_shutdown':
at91_adc.c:(.text+0xca01c): undefined reference to `at91_st_base'
drivers/built-in.o: In function `at91wdt_suspend':
at91_adc.c:(.text+0xca038): undefined reference to `at91_st_base'
drivers/built-in.o: In function `at91_wdt_open':
at91_adc.c:(.text+0xca0cc): undefined reference to `at91_st_base'
drivers/built-in.o:at91_adc.c:(.text+0xca2c8): more undefined references to
`at91_st_base' follow

So, reverting the modification of the "depends" Kconfig line
introduced by patch a6a1bcd37 (watchdog: at91rm9200: add DT support)
seems to be the good solution.

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Acked-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>

6 years agovfs: Revert spurious fix to spinning prevention in prune_icache_sb
Suleiman Souhlal [Sat, 13 Apr 2013 23:03:06 +0000]
vfs: Revert spurious fix to spinning prevention in prune_icache_sb

Revert commit 62a3ddef6181 ("vfs: fix spinning prevention in prune_icache_sb").

This commit doesn't look right: since we are looking at the tail of the
list (sb->s_inode_lru.prev) if we want to skip an inode, we should put
it back at the head of the list instead of the tail, otherwise we will
keep spinning on it.

Discovered when investigating why prune_icache_sb came top in perf
reports of a swapping load.

Signed-off-by: Suleiman Souhlal <suleiman@google.com>
Signed-off-by: Hugh Dickins <hughd@google.com>
Cc: stable@vger.kernel.org # v3.2+
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

6 years agokobject: fix kset_find_obj() race with concurrent last kobject_put()
Linus Torvalds [Sat, 13 Apr 2013 22:15:30 +0000]
kobject: fix kset_find_obj() race with concurrent last kobject_put()

Anatol Pomozov identified a race condition that hits module unloading
and re-loading.  To quote Anatol:

 "This is a race codition that exists between kset_find_obj() and
  kobject_put().  kset_find_obj() might return kobject that has refcount
  equal to 0 if this kobject is freeing by kobject_put() in other
  thread.

  Here is timeline for the crash in case if kset_find_obj() searches for
  an object tht nobody holds and other thread is doing kobject_put() on
  the same kobject:

    THREAD A (calls kset_find_obj())     THREAD B (calls kobject_put())
    splin_lock()
                                         atomic_dec_return(kobj->kref), counter gets zero here
                                         ... starts kobject cleanup ....
                                         spin_lock() // WAIT thread A in kobj_kset_leave()
    iterate over kset->list
    atomic_inc(kobj->kref) (counter becomes 1)
    spin_unlock()
                                         spin_lock() // taken
                                         // it does not know that thread A increased counter so it
                                         remove obj from list
                                         spin_unlock()
                                         vfree(module) // frees module object with containing kobj

    // kobj points to freed memory area!!
    kobject_put(kobj) // OOPS!!!!

  The race above happens because module.c tries to use kset_find_obj()
  when somebody unloads module.  The module.c code was introduced in
  commit 6494a93d55fa"

Anatol supplied a patch specific for module.c that worked around the
problem by simply not using kset_find_obj() at all, but rather than make
a local band-aid, this just fixes kset_find_obj() to be thread-safe
using the proper model of refusing the get a new reference if the
refcount has already dropped to zero.

See examples of this proper refcount handling not only in the kref
documentation, but in various other equivalent uses of this pattern by
grepping for atomic_inc_not_zero().

[ Side note: the module race does indicate that module loading and
  unloading is not properly serialized wrt sysfs information using the
  module mutex.  That may require further thought, but this is the
  correct fix at the kobject layer regardless. ]

Reported-analyzed-and-tested-by: Anatol Pomozov <anatol.pomozov@gmail.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: stable@vger.kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

6 years agoBtrfs: make sure nbytes are right after log replay
Josef Bacik [Fri, 5 Apr 2013 20:50:09 +0000]
Btrfs: make sure nbytes are right after log replay

While trying to track down a tree log replay bug I noticed that fsck was always
complaining about nbytes not being right for our fsynced file.  That is because
the new fsync stuff doesn't wait for ordered extents to complete, so the inodes
nbytes are not necessarily updated properly when we log it.  So to fix this we
need to set nbytes to whatever it is on the inode that is on disk, so when we
replay the extents we can just add the bytes that are being added as we replay
the extent.  This makes it work for the case that we have the wrong nbytes or
the case that we logged everything and nbytes is actually correct.  With this
I'm no longer getting nbytes errors out of btrfsck.

Cc: stable@vger.kernel.org
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>

6 years agox86-32: Fix possible incomplete TLB invalidate with PAE pagetables
Dave Hansen [Fri, 12 Apr 2013 23:23:54 +0000]
x86-32: Fix possible incomplete TLB invalidate with PAE pagetables

This patch attempts to fix:

https://bugzilla.kernel.org/show_bug.cgi?id=56461

The symptom is a crash and messages like this:

chrome: Corrupted page table at address 34a03000
*pdpt = 0000000000000000 *pde = 0000000000000000
Bad pagetable: 000f [#1] PREEMPT SMP

Ingo guesses this got introduced by commit 611ae8e3f520 ("x86/tlb:
enable tlb flush range support for x86") since that code started to free
unused pagetables.

On x86-32 PAE kernels, that new code has the potential to free an entire
PMD page and will clear one of the four page-directory-pointer-table
(aka pgd_t entries).

The hardware aggressively "caches" these top-level entries and invlpg
does not actually affect the CPU's copy.  If we clear one we *HAVE* to
do a full TLB flush, otherwise we might continue using a freed pmd page.
(note, we do this properly on the population side in pud_populate()).

This patch tracks whenever we clear one of these entries in the 'struct
mmu_gather', and ensures that we follow up with a full tlb flush.

BTW, I disassembled and checked that:

if (tlb->fullmm == 0)
and
if (!tlb->fullmm && !tlb->need_flush_all)

generate essentially the same code, so there should be zero impact there
to the !PAE case.

Signed-off-by: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Peter Anvin <hpa@zytor.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Artem S Tashkinov <t.artem@mailcity.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

6 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending
Linus Torvalds [Fri, 12 Apr 2013 22:26:42 +0000]
Merge git://git./linux/kernel/git/nab/target-pending

Pull SCSI target fixes from Nicholas Bellinger:
 "Here are remaining target-pending items for v3.9-rc7 code.

  The tcm_vhost patches are more than I'd usually include in a -rc7
  pull, but are changes required for v3.9 to work correctly with the
  pending vhost-scsi-pci QEMU upstream series merge.  (Paolo CC'ed)

  Plus Asias's conversion to use vhost_virtqueue->private_data + RCU for
  managing vhost-scsi endpoints has gotten alot of review + testing over
  the past weeks, and MST has ACKed the full series.

  Also, there is a target patch to fix a long-standing bug within
  control CDB handling with Standby/Offline/Transition ALUA port access
  states, that had been incorrectly rejecting the control CDBs required
  for LUN scan to work during these port group states.  CC'ing to
  stable."

* git://git.kernel.org/pub/scm/linux/kernel/git/nab/target-pending:
  target: Fix incorrect fallthrough of ALUA Standby/Offline/Transition CDBs
  tcm_vhost: Send bad target to guest when cmd fails
  tcm_vhost: Add vhost_scsi_send_bad_target() helper
  tcm_vhost: Fix tv_cmd leak in vhost_scsi_handle_vq
  tcm_vhost: Remove double check of response
  tcm_vhost: Initialize vq->last_used_idx when set endpoint
  tcm_vhost: Use vq->private_data to indicate if the endpoint is setup
  tcm_vhost: Use ACCESS_ONCE for vs->vs_tpg[target] access

6 years agoMerge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Linus Torvalds [Fri, 12 Apr 2013 22:18:56 +0000]
Merge tag 'scsi-fixes' of git://git./linux/kernel/git/jejb/scsi

Pull SCSI fixes from James Bottomley:
 "This is a set of ten bug fixes (and two consisting of copyright year
  update and version number change) pretty much all of which involve
  either a crash or a hang except the removal of the random sleep from
  the qla2xxx driver (which is a coding error so bad, we want it gone
  before anyone has a chance to copy it)."

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  [SCSI] lpfc: fix potential NULL pointer dereference in lpfc_sli4_rq_put()
  [SCSI] libsas: fix handling vacant phy in sas_set_ex_phy()
  [SCSI] ibmvscsi: Fix slave_configure deadlock
  [SCSI] qla2xxx: Update the driver version to 8.04.00.13-k.
  [SCSI] qla2xxx: Remove debug code that msleeps for random duration.
  [SCSI] qla2xxx: Update copyright dates information in LICENSE.qla2xxx file.
  [SCSI] qla2xxx: Fix crash during firmware dump procedure.
  [SCSI] Revert "qla2xxx: Add setting of driver version string for vendor application."
  [SCSI] ipr: dlpar failed when adding an adapter back
  [SCSI] ipr: fix addition of abort command to HRRQ free queue
  [SCSI] st: Take additional queue ref in st_probe
  [SCSI] libsas: use right function to alloc smp response
  [SCSI] ipr: ipr_test_msi() fails when running with msi-x enabled adapter

6 years agoMerge branch 'for-next' of git://git.samba.org/sfrench/cifs-2.6
Linus Torvalds [Fri, 12 Apr 2013 22:18:20 +0000]
Merge branch 'for-next' of git://git.samba.org/sfrench/cifs-2.6

Pull CIFS fix from Steve French:
 "Fixes a regression in cifs in which a password which begins with a
  comma is parsed incorrectly as a blank password"

* 'for-next' of git://git.samba.org/sfrench/cifs-2.6:
  cifs: Allow passwords which begin with a delimitor

6 years agoftrace: Move ftrace_filter_lseek out of CONFIG_DYNAMIC_FTRACE section
Steven Rostedt (Red Hat) [Fri, 12 Apr 2013 20:40:13 +0000]
ftrace: Move ftrace_filter_lseek out of CONFIG_DYNAMIC_FTRACE section

As ftrace_filter_lseek is now used with ftrace_pid_fops, it needs to
be moved out of the #ifdef CONFIG_DYNAMIC_FTRACE section as the
ftrace_pid_fops is defined when DYNAMIC_FTRACE is not.

Cc: stable@vger.kernel.org
Cc: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

6 years agotracing: Fix possible NULL pointer dereferences
Namhyung Kim [Thu, 11 Apr 2013 06:55:01 +0000]
tracing: Fix possible NULL pointer dereferences

Currently set_ftrace_pid and set_graph_function files use seq_lseek
for their fops.  However seq_open() is called only for FMODE_READ in
the fops->open() so that if an user tries to seek one of those file
when she open it for writing, it sees NULL seq_file and then panic.

It can be easily reproduced with following command:

  $ cd /sys/kernel/debug/tracing
  $ echo 1234 | sudo tee -a set_ftrace_pid

In this example, GNU coreutils' tee opens the file with fopen(, "a")
and then the fopen() internally calls lseek().

Link: http://lkml.kernel.org/r/1365663302-2170-1-git-send-email-namhyung@kernel.org

Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: stable@vger.kernel.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

6 years agoMerge tag 'sound-3.9' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
Linus Torvalds [Fri, 12 Apr 2013 14:45:17 +0000]
Merge tag 'sound-3.9' of git://git./linux/kernel/git/tiwai/sound

Pull sound fixes from Takashi Iwai:
 "This contains a few small ASoC fixes (wm8903, wm5102, samsung-i2s,
  tegra, and soc-compress) and an endian fix for NI USB-audio devices,
  update for Mark's e-mail address.

  No scary changes, AFAIS."

* tag 'sound-3.9' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  MAINTAINERS: Update e-mail address
  ASoC: wm5102: Correct lookup of arizona struct in SYSCLK event
  ASoC: wm8903: Fix the bypass to HP/LINEOUT when no DAC or ADC is running
  ALSA: usb-audio: fix endianness bug in snd_nativeinstruments_*
  ASoC: tegra: Don't claim to support PCM pause and resume
  ASoC: Samsung: set drvdata before adding secondary device
  ASoC: Samsung: return error if drvdata is not set
  ASoC: compress: Cancel delayed power down if needed
  ASoC: core: Fix to check return value of snd_soc_update_bits_locked()

6 years agoMerge tag 'asoc-maintainers-v3.9-rc6' of git://git.kernel.org/pub/scm/linux/kernel...
Takashi Iwai [Fri, 12 Apr 2013 11:53:35 +0000]
Merge tag 'asoc-maintainers-v3.9-rc6' of git://git./linux/kernel/git/broonie/sound into for-linus

MAINTAINERS: Update e-mail address

Update the e-mail address I use for subsystems.

6 years agoMAINTAINERS: Update e-mail address
Mark Brown [Fri, 12 Apr 2013 10:39:57 +0000]
MAINTAINERS: Update e-mail address

Update the e-mail address I use for subsystems.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

6 years agoMerge tag 'asoc-v3.9-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie...
Takashi Iwai [Fri, 12 Apr 2013 08:27:39 +0000]
Merge tag 'asoc-v3.9-rc6' of git://git./linux/kernel/git/broonie/sound into for-linus

ASoC: Updates for v3.9

A few updates, more than I'd like, fixing some relatively small issues
but mostly driver specific ones.  Nothing wildly exciting so if it
doesn't make v3.9 it won't be the end of the world but it'd be nice.