Move wakeup to after early console. This will allow us to display error
messages when cores are not woken up.  Also reduce the wait time for core
to come up.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/6303/

The early serial code is not needed because we already have early
printk support provided by common/earlycons.c

This change also fixes the following build error that occurs when
CONFIG_SERIAL_8250 is not configured for Netlogic XLR boards:

arch/mips/built-in.o: In function `nlm_early_serial_setup':
setup.c:(.init.text+0x274): undefined reference to `early_serial_setup'
make: *** [vmlinux] Error 1
Reported-by: NMarkos Chandras <markos.chandras@imgtec.com>
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/6083/

Add a default device tree fie for XLP9XX boards, and add code to use
this device tree if no DTB is passed to the kernel.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/6287/

Support for adding legacy IRQ domain for XLP9XX. The node id of the
PIC has to be calulated differently for XLP9XX.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/6286/

XLP9XX has a USB 3.0 controller on-chip with 2 xHCI ports. The USB
block is similar to the one on XLP2XX, so update usb-init-xlp2.c
to handle XLP9XX as well.
Signed-off-by: NGanesan Ramalingam <ganesanr@broadcom.com>
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/6285/

Add PCI support for Netlogic XLP9XX. The PCI registers and
SoC bus numbers have changed in XLP9XX.

Also skip a few (bus,dev,fn) combinations which have issues when
read.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/6284/

XLP9XX has 20 cores per node, opposed to 8 on earlier XLP8XX.
Update code that calculates node id from cpu id to handle this.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/6283/

Update bridge code. Add code to the XLP9XX registers for DRAM
size, limit and node when running on XLPXX
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/6282/

Update IO offset of the early console UART.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/6281/

Add the SYS block registers for XLP9XX, most of them have changed.
The wakeup sequence has been updated to set the coherent mode from
the main thread rather than the woken up thread.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/6280/

Functions for the XLP9XX interrupt table entry format and other PIC
register changes.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/6279/

Most IO block offsets have changed in XLP9XX. Update iomap.h to add the
new addresses of different SoC blocks like PIC, SYS, UART etc. that are
needed by the base code.

On XLP9xx, the SoC blocks of other nodes are seen on a PCI bus
corresponding to the node. Update iomap code to reflect this.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/6277/

Adds processor ID of XLP 9XX to asm/cpu.h.  Update netlogic/xlp-hal/xlp.h
to add cpu_is_xlp9xx() and to update cpu_is_xlpii() to support XLP 9XX.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/6274/

Use the FUSE register to get the list of active cores in the CPU
instead of using the CPU reset register, this is the recommended
method.

Also add code to mask the coremask with the default number of cores
for each processor series.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/6275/

On XLPII CPUs, the L1D cache has to be flushed with regular cache
operations before enabling threads in a core.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/6276/

No change in logic, the changes are:
* cleanup some whitespace and comments
* remove confusing argument of SYS_CPU_COHERENT_BASE macro
* make the numerical labels in macros consistent
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/6273/

Add MSI chip and MSIX chip definitions.

For MSI, we map the link interrupt to a MSI link IRQ which will
do a second level of dispatch based on the MSI status register.

The MSI chip definitions use the MSI enable register to enable
and disable the MSI irqs.

For MSI-X, we split the 32 available MSI-X vectors across the
four PCIe links (8 each). These PIC interrupts generate an IRQ
per link which uses a second level dispatch as well.

The MSI-X chip definition uses the standard functions to enable
and disable interrupts.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Signed-off-by: NJohn Crispin <blogic@openwrt.org>
Patchwork: http://patchwork.linux-mips.org/patch/6270/

The early_init_devtree() API was removed in linux-next for 3.13 with
commit "mips: use early_init_dt_scan". This causes Netlogic XLP compile
to fail:

arch/mips/netlogic/xlp/setup.c:101: undefined reference to `early_init_devtree'

Add xlp_early_init_devtree() which uses the __dt_setup_arch() to
handle early device tree related initialization to fix this.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Signed-off-by: NRob Herring <rob.herring@calxeda.com>

Panic() is going to add a \n itself and it's annoying if a panic message rolls
of the screen on a device with no scrollback.
Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Replace hardcoded CP0 PRId and CP1 FPIR register access masks throughout.
The change does not touch places that use shifted or partial masks.
Signed-off-by: NMaciej W. Rozycki <macro@linux-mips.org>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5838/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Add a default built-in device tree for XLP2xx SoC. The new file
xlp_fvp.dts has updated entries for I2C and memory.
Signed-off-by: NGanesan Ramalingam <ganesanr@broadcom.com>
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5705/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Add USB intialization code to handle the new XLP2XX USB 3.0 block.
This adds xlp/usb-init-xlp2.c to initialize the XLP2XX USB glue-logic
for XHCI and EHCI. Interrupt and IO offset code in xlp-hal/iomap.h
and xlp/nlm_hal.c is also updated for the new USB block.
Signed-off-by: NGanesan Ramalingam <ganesanr@broadcom.com>
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5744/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

XLP2xx has a new I2C controller which has 4 buses connected to
it. Update the IO offset and IRQ mapping code to reflect this.
Signed-off-by: NGanesan Ramalingam <ganesanr@broadcom.com>
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5707/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

On XLP2xx, wakeup code does not need to enable clock on the core.
Update xlp/wakeup.c to handle this.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5703/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Add code to calculate the CPU and PIC frequency for XLP2XX SoCs.

Since the PIC frequency on XLP2XX can be configured, add a new macro
pic_timer_freq() to be used in netlogic/common/time.c.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Cc: Ganesan Ramalingam <ganesanr@broadcom.com>
Patchwork: http://patchwork.linux-mips.org/patch/5701/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

The config7/config4 register has to be written on all the threads.
This does not cause any problems in XLP, but is needed for XLPII
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5700/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

XLP2XX is first in the series of 28nm XLPII processors.

The changes are to:
* Add processor ID for XLP2XX to asm/cpu.h and kernel/cpu-probe.c.
* Add a cpu_is_xlpii() function to check for XLPII processors.
* Update xlp_mmu_init() to use config4 to enable extended TLB.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5698/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

The flash chipselects can span 32MB, fix this in the built-in device
tree.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5704/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Do not pass a memory section in the built-in DTB, and let the
boot code use the values from the DRAM BARs.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5699/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Read the memory from the Bridge DRAM BARs, if it is not passed in
from the device tree. This will allow us to remove memory configuration
from built in device trees.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5743/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Removed parameters checked twice in logical OR operation.
Suggested by coccinelle and manually verified.
Signed-off-by: NAlexandru Juncu <alexj@rosedu.org>
Cc: jchandra@broadcom.com
Cc: linux-mips@linux-mips.org
Cc: linux-kernel@vger.kernel.org
Patchwork: https://patchwork.linux-mips.org/patch/5627/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Add a legacy irq domain for the XLP PIC interrupts. This will be used
when interrupts are assigned from the device tree. This change is required
after commit c5cdc67a "irqdomain: Remove temporary MIPS workaround code".
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Cc: Jayachandran C <jchandra@broadcom.com>
Patchwork: https://patchwork.linux-mips.org/patch/5597/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

The on-chip USB controller on Netlogic XLP does not suppport
DMA beyond 32-bit physical address. Set the coherent_dma_mask
of the USB in its PCI fixup to support this.
Signed-off-by: NGanesan Ramalingam <ganesanr@broadcom.com>
Signed-off-by: NJayachandran C. <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5596/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

commit 3747069b25e419f6b51395f48127e9812abc3596 upstream.

The __cpuinit type of throwaway sections might have made sense
some time ago when RAM was more constrained, but now the savings
do not offset the cost and complications.  For example, the fix in
commit 5e427ec2 ("x86: Fix bit corruption at CPU resume time")
is a good example of the nasty type of bugs that can be created
with improper use of the various __init prefixes.

After a discussion on LKML[1] it was decided that cpuinit should go
the way of devinit and be phased out.  Once all the users are gone,
we can then finally remove the macros themselves from linux/init.h.

Note that some harmless section mismatch warnings may result, since
notify_cpu_starting() and cpu_up() are arch independent (kernel/cpu.c)
and are flagged as __cpuinit  -- so if we remove the __cpuinit from
the arch specific callers, we will also get section mismatch warnings.
As an intermediate step, we intend to turn the linux/init.h cpuinit
related content into no-ops as early as possible, since that will get
rid of these warnings.  In any case, they are temporary and harmless.

Here, we remove all the MIPS __cpuinit from C code and __CPUINIT
from asm files.  MIPS is interesting in this respect, because there
are also uasm users hiding behind their own renamed versions of the
__cpuinit macros.

[1] https://lkml.org/lkml/2013/5/20/589

[ralf@linux-mips.org: Folded in Paul's followup fix.]
Signed-off-by: NPaul Gortmaker <paul.gortmaker@windriver.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5494/
Patchwork: https://patchwork.linux-mips.org/patch/5495/
Patchwork: https://patchwork.linux-mips.org/patch/5509/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Fix an issue in plat_irq_dispatch due to which it can call do_IRQ
with a PIC irq that is not mapped.

When a per-cpu interrupt and a PIC interrupt are both active, the
check 'eirr & PERCPU_IRQ_MASK' will be true, but the interrupt in 'i'
will be the number of the PIC interrupt. In this case, we will call
do_IRQ on the PIC interrupt without mapping it with nlm_irq_to_xirq().

Fix this by using __ffs64 instead of __ilog2_u64 and using the
interrupt number instead of mask to identify per-cpu interrupts.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5432/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

The XLPs in production do not need these workarounds. Remove the code and
the associated ifdef.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5430/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Fix a cache error found in stress test, caused by the prefetch instruction
going beyond valid memory when acessing the last page of a region. Add
the pref_backup logic similar to XLR in XLP too.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5431/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Earlier we just waited for the first thread of the CPU to come online
before proceeding to wake up others. Update it to wait for all the CPUs
in the core. This will be useful when the boot-up is slow, like while
debugging or when running in a simulator.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5429/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Fix an issue in the reset code. Since this code is copied to the
reset vector, using 'j' for looping is not correct. Use relative
branch 'b'.

Update the usage of 'j' in smpboot.S to be consistent although it
is not a bug there.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5427/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>

Move the nlm_cpu_ready[] array used by the cpu wakeup code to the
boot area, along with rest of the boot parameter code.
Signed-off-by: NJayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/5425/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>