This commit rearranges the HUB interrupt code by using MIPS_IRQ_CPU
interrupt handling code and modern Linux IRQ framework features to get
rid of global arrays. It also adds support for irq affinity setting.
Signed-off-by: Thomas Bogendoerfer <tbogendoerfer@suse.de>
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: James Hogan <jhogan@kernel.org>
Cc: linux-mips@vger.kernel.org
Cc: linux-kernel@vger.kernel.org

To make use of per_cpu variables in interrupt code per_cpu_init() must
be done after setup_per_cpu_areas(). This is achieved by calling it
in smp_prepare_boot_cpu() via a new smp_ops method.
Signed-off-by: Thomas Bogendoerfer <tbogendoerfer@suse.de>
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: James Hogan <jhogan@kernel.org>
Cc: linux-mips@vger.kernel.org
Cc: linux-kernel@vger.kernel.org

Move xtalk scanning to a later boot stage to be able using things like
kmalloc and friends.
Signed-off-by: Thomas Bogendoerfer <tbogendoerfer@suse.de>
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: James Hogan <jhogan@kernel.org>
Cc: linux-mips@vger.kernel.org
Cc: linux-kernel@vger.kernel.org

Topology and NMI output needs pr_cont() to look the way it was in the
old days of printk.
Signed-off-by: Thomas Bogendoerfer <tbogendoerfer@suse.de>
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: James Hogan <jhogan@kernel.org>
Cc: linux-mips@vger.kernel.org
Cc: linux-kernel@vger.kernel.org

Introduced bridge_read/bridge_write/bridge_set/bridge_clr for accessing
bridge register and get rid of volatile declarations. Also removed
all typedefs from arch/mips/include/asm/pci/bridge.h and cleaned up
language in arch/mips/pci/ops-bridge.c
Signed-off-by: Thomas Bogendoerfer <tbogendoerfer@suse.de>
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: James Hogan <jhogan@kernel.org>
Cc: linux-mips@vger.kernel.org
Cc: linux-kernel@vger.kernel.org

Replace hub register access with __raw_readq/__raw_writeq and get
rid of hubreg_t completely. Also remove no longer (probably never
used) used defines
Signed-off-by: Thomas Bogendoerfer <tbogendoerfer@suse.de>
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: James Hogan <jhogan@kernel.org>
Cc: linux-mips@vger.kernel.org
Cc: linux-kernel@vger.kernel.org

The irq_pages is the number of pages for irq stack, but not the
order which is needed by __get_free_pages().
We can use get_order() to calculate the accurate order.
Signed-off-by: Liu Xiang <liu.xiang6@zte.com.cn>
Signed-off-by: Paul Burton <paul.burton@mips.com>
Fixes: fe8bd18f ("MIPS: Introduce irq_stack")
Cc: linux-mips@vger.kernel.org
Cc: stable@vger.kernel.org # v4.11+

Commit e36863a5 ("MIPS: HIGHMEM DMA on noncoherent MIPS32
processors") introduced code which:

  1) Calculates an offset within a page, by ANDing an address
     with ~PAGE_MASK.

  2) Checks whether that offset is >= PAGE_SIZE.

This check can never evaluate true, making the code it guards
unreachable. smatch spots bogus arithmetic resulting from the
impossible condition, resulting in the following warning:

  arch/mips/mm/dma-noncoherent.c:125
    dma_sync_phys() warn: mask and shift to zero

Fix this by removing the impossible to satisfy condition & the
unreachable code it guards.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org
Cc: Christoph Hellwig <hch@lst.de>
Cc: Marek Szyprowski <m.szyprowski@samsung.com>
Cc: Robin Murphy <robin.murphy@arm.com>

REG_32BIT_ZERO_EX and REG_64BIT are always handled in exactly the same
way, and reg_val_propagate_range() never actually sets any register to
type REG_32BIT_ZERO_EX.

Remove the redundant & unused REG_32BIT_ZERO_EX.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Martin KaFai Lau <kafai@fb.com>
Cc: Song Liu <songliubraving@fb.com>
Cc: Yonghong Song <yhs@fb.com>
Cc: netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: Jiong Wang <jiong.wang@netronome.com>

The function prototype used to call JITed eBPF code (ie. the type of the
struct bpf_prog bpf_func field) returns an unsigned int. The MIPS n64
ABI that MIPS64 kernels target defines that 32 bit integers should
always be sign extended when passed in registers as either arguments or
return values.

This means that when returning any value which may not already be sign
extended (ie. of type REG_64BIT or REG_32BIT_ZERO_EX) we need to perform
that sign extension in order to comply with the n64 ABI. Without this we
see strange looking test failures from test_bpf.ko, such as:

  test_bpf: #65 ALU64_MOV_X:
    dst = 4294967295 jited:1 ret -1 != -1 FAIL (1 times)

Although the return value printed matches the expected value, this is
only because printf is only examining the least significant 32 bits of
the 64 bit register value we returned. The register holding the expected
value is sign extended whilst the v0 register was set to a zero extended
value by our JITed code, so when compared by a conditional branch
instruction the values are not equal.

We already handle this when the return value register is of type
REG_32BIT_ZERO_EX, so simply extend this to also cover REG_64BIT.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Fixes: b6bd53f9 ("MIPS: Add missing file for eBPF JIT.")
Cc: stable@vger.kernel.org # v4.13+
Cc: linux-mips@vger.kernel.org
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: Martin KaFai Lau <kafai@fb.com>
Cc: Song Liu <songliubraving@fb.com>
Cc: Yonghong Song <yhs@fb.com>
Cc: netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: Jiong Wang <jiong.wang@netronome.com>

mips_cm_error_report() contains a function call that's incorrectly
indented a level further than it ought to be. Remove a tab from the
start of both affected lines.
Signed-off-by: Paul Burton <paul.burton@mips.com>

1) Fix reset button support which is active *high*
2) Specify LEDs colors
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: James Hogan <jhogan@kernel.org>
Cc: Hauke Mehrtens <hauke@hauke-m.de>
Cc: linux-mips@linux-mips.org

Program rx/tx-delay always from DT.
Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

Delete board-specific link status. This info should now come from
the DT only.
Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

The fixed-link node in the DT should now take care of the link status,
so this hack can be deleted.
Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

Warn if deprecated link status is being used.
Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

Currently OCTEON ethernet falls back to phyless operation on
boards where we have no known PHY address or a fixed-link node.
Add fixed-link support for boards that need it, so we can clean up
the platform code and ethernet driver from some legacy code.
Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

Commit adcc81f1 ("MIPS: math-emu: Write-protect delay slot emulation
pages") left flush_cache_sigtramp() unused. Delete the dead code.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: linux-mips@vger.kernel.org

Commit 46011e6e ("MIPS: Make set_pte() SMP safe.") introduced an
open-coded version of cmpxchg() within set_pte(), that always operated
on a value the size of an unsigned long. That is, it used ll/sc
instructions when CONFIG_32BIT=y or lld/scd instructions when
CONFIG_64BIT=y.

This was broken for configurations in which pte_t is larger than an
unsigned long (with the exception of XPA configurations which have a
different implementation of set_pte()), because we no longer update the
whole PTE. Indeed commit 46011e6e ("MIPS: Make set_pte() SMP safe.")
notes:

> The case of CONFIG_64BIT_PHYS_ADDR && CONFIG_CPU_MIPS32 is *not*
> handled.

In practice this affects Netlogic XLR/XLS systems including
nlm_xlr_defconfig.

Commit 82f4f66d ("MIPS: Remove open-coded cmpxchg() in set_pte()")
then replaced this open-coded version of cmpxchg() with an actual call
to cmpxchg(). Unfortunately the configurations mentioned above then fail
to build because cmpxchg() can only operate on values 32 bits or smaller
in size, resulting in:

  arch/mips/include/asm/cmpxchg.h:166:11: error:
    call to '__cmpxchg_called_with_bad_pointer' declared with
    attribute error: Bad argument size for cmpxchg

One option that would fix the build failure & restore the previous
behaviour would be to cast the pte pointer to a pointer to unsigned
long, so that cmpxchg() would operate on just 32 bits of the PTE as it
has been since commit 46011e6e ("MIPS: Make set_pte() SMP safe.").
That feels like an ugly hack though, and the behaviour of set_pte() is
likely a little broken.

Instead we take advantage of the fact that the affected configurations
already know at compile time that the CPU will support 64 bits (ie. have
hardcoded cpu_has_64bits in cpu-feature-overrides.h) in order to allow
cmpxchg64() to be used in these configurations. set_pte() then makes use
of cmpxchg64() when necessary.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Fixes: 46011e6e ("MIPS: Make set_pte() SMP safe.")
Fixes: 82f4f66d ("MIPS: Remove open-coded cmpxchg() in set_pte()")

KVM makes use of check_switch_mmu_context(), check_mmu_context() &
get_new_mmu_context() which are no longer static inline functions in a
header. As such they need to be exported for KVM to successfully build
as a module, which was previously overlooked. Add the missing exports.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Fixes: 4ebea49c ("MIPS: mm: Un-inline get_new_mmu_context")
Fixes: 42d5b846 ("MIPS: mm: Unify ASID version checks")

Commit 7b3415f5 ("MIPS: Loongson32: Remove unused platform devices")
removed the definitions of platform devices which have no in tree
drivers from common Loongson32 code, but missed their removal from
Loongson 1B board code in arch/mips/loongson32/ls1b/board.c. This causes
build failures due to the missing declarations of ls1x_dma_pdev,
ls1x_nand_pdev & their associated *_set_platdata functions.

Remove the dead code from arch/mips/loongson32/ls1b/board.c to fix the
build.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Fixes: 7b3415f5 ("MIPS: Loongson32: Remove unused platform devices")

Commit a96d68ba ("MIPS: Loongson32: clarify we don't support MIPS16
and merge configs") attempted to reduce duplication in Kconfig by
consolidating some selects common to Loongson 1B & 1C CPUs under
CPU_LOONGSON1. Unfortunately it clearly wasn't tested because by
removing SYS_SUPPORTS_32BIT_KERNEL it prevented 32BIT from being enabled
leading to all sorts of strange build errors from a kernel configured to
build as neither 32 nor 64 bit.

Both loongson1b_defconfig & loongson1c_defconfig failed to build due to
this problem.

Revert the cleanup portions of commit a96d68ba ("MIPS: Loongson32:
clarify we don't support MIPS16 and merge configs"), keeping only its
removal of the selection of SYS_SUPPORTS_MIPS16.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Fixes: a96d68ba ("MIPS: Loongson32: clarify we don't support MIPS16 and merge configs")

Commit f263f2a2 ("MIPS: Compile post DMA flush only when needed")
pushed the selection of ARCH_HAS_SYNC_DMA_FOR_CPU down to various
SYS_HAS_CPU_* Kconfig entries corresponding to CPUs for which
cpu_needs_post_dma_flush() might return true, but unfortunately missed
the fact that some of these CPUs can be used in configurations with
DMA_NONCOHERENT=n. When this is the case the kernel build does not
include our definition of arch_sync_dma_for_cpu() from
arch/mips/mm/dma-noncoherent.c and the build fails with a link error.

One example of this problem is ip27_defconfig:

  kernel/dma/direct.o: In function `dma_direct_sync_single_for_cpu':
  direct.c:(.text+0x6c): undefined reference to `arch_sync_dma_for_cpu'
  kernel/dma/direct.o: In function `dma_direct_sync_sg_for_cpu':
  direct.c:(.text+0x1f0): undefined reference to `arch_sync_dma_for_cpu'
  kernel/dma/direct.o: In function `dma_direct_alloc':
  direct.c:(.text+0xc20): undefined reference to `arch_dma_alloc'
  kernel/dma/direct.o: In function `dma_direct_free':
  direct.c:(.text+0xc3c): undefined reference to `arch_dma_free'
  make[1]: *** [Makefile:1021: vmlinux] Error 1
  make: *** [Makefile:152: sub-make] Error 2

Fix this by selecting ARCH_HAS_SYNC_DMA_FOR_CPU only when
DMA_NONCOHERENT is also selected. The SYS_HAS_CPU_BMIPS5000 case is left
as-is because systems with that CPU always select DMA_NONCOHERENT
anyway.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Fixes: f263f2a2 ("MIPS: Compile post DMA flush only when needed")

Our hugepage support already exists for MIPS64 CPUs, and is already
enabled for older architecture revisions. There's nothing MIPSr6
specific involved, and our hugepage support already works fine for
MIPS64r6 CPUs such as the I6500, so allow it to be selected in Kconfig.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

set_pte() contains an open coded version of cmpxchg() - it atomically
replaces the buddy pte's value if it is currently zero. Simplify the
code considerably by just using cmpxchg() instead of reinventing it.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

Introduce support for using MemoryMapIDs (MMIDs) as an alternative to
Address Space IDs (ASIDs). The major difference between the two is that
MMIDs are global - ie. an MMID uniquely identifies an address space
across all coherent CPUs. In contrast ASIDs are non-global per-CPU IDs,
wherein each address space is allocated a separate ASID for each CPU
upon which it is used. This global namespace allows a new GINVT
instruction be used to globally invalidate TLB entries associated with a
particular MMID across all coherent CPUs in the system, removing the
need for IPIs to invalidate entries with separate ASIDs on each CPU.

The allocation scheme used here is largely borrowed from arm64 (see
arch/arm64/mm/context.c). In essence we maintain a bitmap to track
available MMIDs, and MMIDs in active use at the time of a rollover to a
new MMID version are preserved in the new version. The allocation scheme
requires efficient 64 bit atomics in order to perform reasonably, so
this support depends upon CONFIG_GENERIC_ATOMIC64=n (ie. currently it
will only be included in MIPS64 kernels).

The first, and currently only, available CPU with support for MMIDs is
the MIPS I6500. This CPU supports 16 bit MMIDs, and so for now we cap
our MMIDs to 16 bits wide in order to prevent the bitmap growing to
absurd sizes if any future CPU does implement 32 bit MMIDs as the
architecture manuals suggest is recommended.

When MMIDs are in use we also make use of GINVT instruction which is
available due to the global nature of MMIDs. By executing a sequence of
GINVT & SYNC 0x14 instructions we can avoid the overhead of an IPI to
each remote CPU in many cases. One complication is that GINVT will
invalidate wired entries (in all cases apart from type 0, which targets
the entire TLB). In order to avoid GINVT invalidating any wired TLB
entries we set up, we make sure to create those entries using a reserved
MMID (0) that we never associate with any address space.

Also of note is that KVM will require further work in order to support
MMIDs & GINVT, since KVM is involved in allocating IDs for guests & in
configuring the MMU. That work is not part of this patch, so for now
when MMIDs are in use KVM is disabled.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

Add a family of ginvt_* functions making it easy to emit a GINVT
instruction to globally invalidate TLB entries. We make use of the
_ASM_MACRO infrastructure to support emitting the instructions even if
the assembler isn't new enough to support them natively.

An associated STYPE_GINV definition & sync_ginv() function are added to
emit a sync instruction of type 0x14, which operates as a completion
barrier for these new GINVT (and GINVI) instructions.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

When we gain MMID support we'll be storing MMIDs as atomic64_t values
and accessing them via atomic64_* functions. This necessitates that we
don't use cpu_context() as the left hand side of an assignment, ie. as a
modifiable lvalue. In preparation for this introduce a new
set_cpu_context() function & replace all assignments with cpu_context()
on their left hand side with an equivalent call to set_cpu_context().

To enforce that cpu_context() should not be used for assignments, we
rewrite it as a static inline function.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

Introduce a new check_mmu_context() function to check an mm's ASID
version & get a new one if it's outdated, and a
check_switch_mmu_context() function which additionally sets up the new
ASID & page directory. Simplify switch_mm() & various
get_new_mmu_context() callsites in MIPS KVM by making use of the new
functions, which will help reduce the amount of code that requires
modification to gain MMID support.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

In preparation for adding MMID support to get_new_mmu_context() which
will increase the size of the function somewhat, move it from
asm/mmu_context.h into a C file.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

Split always-included objects to one per line in order to make it easier
to modify the list of included objects.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

All 3 variants of local_flush_tlb_mm() are now effectively simple calls
to drop_mmu_context(). Remove them and use drop_mmu_context() directly.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

The r4k variant of local_flush_tlb_mm() wraps its call to
drop_mmu_context() with a preempt_disable() & preempt_enable() pair, but
this is redundant since drop_mmu_context() disables interrupts and from
Documentation/preempt-locking.txt:

  Note that you do not need to explicitly prevent preemption if you are
  holding any locks or interrupts are disabled, since preemption is
  implicitly disabled in those cases.

Remove the redundant preempt_disable() & preempt_enable() calls.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

drop_mmu_context() is preceded by a comment indicating what happens if
the mm provided is currently active on the local CPU. Move that comment
into the block that executes in this case, adjusting slightly to reflect
its new location.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

If an mm does not have an ASID on the local CPU then drop_mmu_context()
is always redundant, since there's no context to "drop". Various callers
of drop_mmu_context() check whether the mm has been allocated an ASID
before making the call. Move that check into drop_mmu_context() and
remove it from callers to simplify them.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

If drop_mmu_context() is called with an mm that is not currently active
on the local CPU then there's no need for us to stop & start a hardware
page table walker because it can't be fetching entries for the ASID
corresponding to the mm we're operating on.

Move the htw_stop() & htw_start() calls into the block which we run only
if the mm is currently active, in order to avoid the redundant work.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

get_new_mmu_context() accepts a cpu argument, but implicitly assumes
that this is always equal to smp_processor_id() by operating on the
local CPU's TLB & icache.

Remove the cpu argument and have get_new_mmu_context() call
smp_processor_id() instead.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

The drop_mmu_context() function accepts a cpu argument, but it
implicitly expects that this is always equal to smp_processor_id() by
allocating & configuring an ASID on the local CPU when the mm is active
on the CPU indicated by the cpu argument.

All callers do provide the value of smp_processor_id() to the cpu
argument.

Remove the redundant argument and have drop_mmu_context() call
smp_processor_id() itself, making it clearer that the cpu variable
always represents the local CPU.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

MIPS has separate definitions of activate_mm() & switch_mm() which are
identical apart from switch_mm() checking that the ASID is valid before
acquiring a new one.

We know that when activate_mm() is called cpu_context(X, mm) will be
zero, and this will never be considered a valid ASID because we never
allow the ASID version number to be zero, instead beginning with version
1 using asid_first_version(). Therefore switch_mm() will always allocate
a new ASID when called for a new task, meaning that it will behave
identically to activate_mm().

Take advantage of this to remove the duplication & define activate_mm()
using switch_mm() just like many other architectures do.
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org

Delete unused cvmx-smix-defs.h.
Signed-off-by: Aaro Koskinen <aaro.koskinen@iki.fi>
Signed-off-by: Paul Burton <paul.burton@mips.com>
Cc: linux-mips@vger.kernel.org