Now that the DWARF unwinder is being used to provide perf callstacks
unwinding speed is an issue. It is no longer being used in exceptional
circumstances where we don't care about runtime performance, e.g. when
panicing, so it makes sense improve performance is possible.

With this patch I saw a 42% improvement in unwind time when calling
return_address(1). Greater improvements will be seen as the number of
levels unwound increases as each unwind is now cheaper.

Note that insertion time has doubled but that's just the price we pay
for keeping the trees balanced. However, this is a one-time cost for
kernel boot/module load and so the improvements in lookup time dominate
the extra time we spend keeping the trees balanced.
Signed-off-by: NMatt Fleming <matt@console-pimps.org>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

There's no need to setup the frame pointer again in
call_handle_tlbmiss. The frame pointer will already have been setup in
handle_interrupt.
Signed-off-by: NMatt Fleming <matt@console-pimps.org>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Unfortunately, due to poor DWARF info in current toolchains, unwinding
through interrutps cannot be done reliably. The problem is that the
DWARF info for function epilogues is wrong.

Take this standard epilogue sequence,

80003cc4:       e3 6f           mov     r14,r15
80003cc6:       26 4f           lds.l   @r15+,pr
80003cc8:       f6 6e           mov.l   @r15+,r14
						<---- interrupt here
80003cca:       f6 6b           mov.l   @r15+,r11
80003ccc:       f6 6a           mov.l   @r15+,r10
80003cce:       f6 69           mov.l   @r15+,r9
80003cd0:       0b 00           rts

If we take an interrupt at the highlighted point, the DWARF info will
bogusly claim that the return address can be found at some offset from
the frame pointer, even though the frame pointer was just restored. The
worst part is if the unwinder finds a text address at the bogus stack
address - unwinding will continue, for a bit, until it finally comes
across an unexpected address on the stack and blows up.

The only solution is to stop unwinding once we've calculated the
function that was executing when the interrupt occurred. This PC can be
easily calculated from pt_regs->pc.
Signed-off-by: NMatt Fleming <matt@console-pimps.org>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

In order to allow the DWARF unwinder to unwind through exceptions we
need to setup the frame pointer register (r14).
Signed-off-by: NMatt Fleming <matt@console-pimps.org>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

The address that ret_from_exception and ret_from_irq will return to is
found in the stack slot for SPC, not PR. This error was causing the
DWARF unwinder to pick up the wrong return address on the stack and then
unwind using the unwind tables for the wrong function.

While I'm here I might as well add CFI annotations for the other
registers since they could be useful when unwinding.
Signed-off-by: NMatt Fleming <matt@console-pimps.org>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This improves power management for the SIUA controller on sh7722. Similar
patches might be desired for other SIU-enabled SH platforms.
Signed-off-by: NGuennadi Liakhovetski <g.liakhovetski@gmx.de>
Acked-by: NMagnus Damm <damm@opensource.se>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Signed-off-by: NMarek Skuczynski <mareksk7@gmail.com>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Signed-off-by: NMarek Skuczynski <mareksk7@gmail.com>
Acked-by: NMatt Fleming <matt@console-pimps.org>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

'flush_old_exec()' is the point of no return when doing an execve(), and
it is pretty badly misnamed.  It doesn't just flush the old executable
environment, it also starts up the new one.

Which is very inconvenient for things like setting up the new
personality, because we want the new personality to affect the starting
of the new environment, but at the same time we do _not_ want the new
personality to take effect if flushing the old one fails.

As a result, the x86-64 '32-bit' personality is actually done using this
insane "I'm going to change the ABI, but I haven't done it yet" bit
(TIF_ABI_PENDING), with SET_PERSONALITY() not actually setting the
personality, but just the "pending" bit, so that "flush_thread()" can do
the actual personality magic.

This patch in no way changes any of that insanity, but it does split the
'flush_old_exec()' function up into a preparatory part that can fail
(still called flush_old_exec()), and a new part that will actually set
up the new exec environment (setup_new_exec()).  All callers are changed
to trivially comply with the new world order.
Signed-off-by: NH. Peter Anvin <hpa@zytor.com>
Cc: stable@kernel.org
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Enables support for kmemleak on sh.
Signed-off-by: NChris Smith <chris.smith@st.com>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Presently trapped I/O is only registered if it's not explicitly disabled
for the platforms that select it openly. From the fault path this runs
through an address lookup before figuring out that nothing matches and
falls back through the error path, but we can forego the lookup
completely by testing if it's been explicitly disabled. This provides a
measurable speedup for things like qemu that rely on runtime disabling.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

The old ctrl in/out routines are non-portable and unsuitable for
cross-platform use. While drivers/sh has already been sanitized, there
is still quite a lot of code that is not. This converts the arch/sh/ bits
over, which permits us to flag the routines as deprecated whilst still
building with -Werror for the architecture code, and to ensure that
future users are not added.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Now that cached_to_uncached works as advertized in 32-bit mode and we're
never going to be able to map < 16MB anyways, there's no need for the
special uncached section. Kill it off.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This provides a variable for tracking the uncached mapping size, and uses
it for pretty printing the uncached lowmem range. Beyond this, we'll also
be building on top of this for figuring out from where the remainder of
P2 becomes usable when constructing unrelated mappings.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This effectively neutralizes P2 by getting rid of P1 identity mapping
for all available memory and instead only establishes a single unbuffered
PMB entry (16MB -- the smallest available) that covers the kernel.

As using segmentation for abusing caching attributes in drivers is no
longer supported (and there are no drivers that can be enabled in 32-bit
mode that do this), this provides us with all of the uncached access
needs by the kernel itself.

Drivers and their ilk need to specify their caching attributes when
remapping through page tables, as usual.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

All of the cached/uncached mapping setup is duplicated for each size, and
also misses out on the 16MB case. Rather than duplicating the same iter
code for that we just consolidate it in to a helper macro that builds an
iter for each size. The 16MB case is then trivially bolted on at the end.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This provides a machine_ops-based reboot interface loosely cloned from
x86, and converts the native sh32 and sh64 cases over to it.

Necessary both for tying in SMP support and also enabling platforms like
SDK7786 to add support for their microcontroller-based power managers.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Usually we can look to the CVR to work out whether we have an FPU or not.
Unfortunately not all parts comply with this, so just set the flag
manually for all SH-4 parts and clear it on the only SH-4 that doesn't
have one (SH4-501).
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Implement .set_rate() for all SH "div4 clocks," .enable(), .disable(), and
.set_parent() for those, that support them. This allows, among other uses,
reparenting of SIU clocks to the external source, and enabling and
disabling of the IrDA clock on sh7722.
Signed-off-by: NGuennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This rewrites the SH7786 clock framework support completely. It's
reworked to provide all of the DIV4 and MSTP function clocks. This brings
it in line with the current clock framework code and lets us drop SH7786
from the list of CPUs that require legacy CPG handling.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

sh64 on the other hand provides both direct broken out syscalls as well
as socketcall access. As there are binaries that use both socketcall has
to stay around. The current ABI prefers direct syscalls.

It was pointed out that when sys_recvmmsg was added in, sys_accept4 was
overlooked. This takes care of wiring it up.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

sh32 at the moment only uses sys_socketcall to reach these, so unwire
recvmmsg for now. While we're at it, add it to the ignore list, as per
the s390 change.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

The breakpoint handler was renamed on sh32, but sh64 was overlooked in
the conversion. Fix it up now.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This kills off the sh64-specific state restorer and switches over to
the generic one.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This updates the sh64 processor info with the sh32 changes in order to
tie in to the generic task_xstate management code.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This tosses in a local_irq_enable()/disable() pair around the init_fpu()
callsite in the FPU state restore exception handler. Fixes up a slab BUG
triggered by making a slab cache allocation that can sleep whilst
irqs_disabled(). This follows the behaviour undertaken by the x86
implementation.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

More and more boards are going to start shipping that boot with the MMU
in 32BIT mode by default. Previously we relied on the bootloader to
setup PMB mappings for use by the kernel but we also need to cater for
boards whose bootloaders don't set them up.

If CONFIG_PMB_LEGACY is not enabled we have full control over our PMB
mappings and can compress our address space. Usually, the distance
between the the cached and uncached mappings of RAM is always 512MB,
however we can compress the distance to be the amount of RAM on the
board.

pmb_init() now becomes much simpler. It no longer has to calculate any
mappings, it just has to synchronise the software PMB table with the
hardware.

Tested on SDK7786 and SH7785LCR.
Signed-off-by: NMatt Fleming <matt@console-pimps.org>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Some devices need to be ioremap'd and accessed very early in the boot
process. It is not possible to use the standard ioremap() function in
this case because that requires kmalloc()'ing some virtual address space
and kmalloc() may not be available so early in boot.

This patch provides fixmap mappings that allow physical address ranges
to be remapped into the kernel address space during the early boot
stages.
Signed-off-by: NMatt Fleming <matt@console-pimps.org>

Presently the secondary CPU entry point is only aimed at 29bit phys mode,
causing it to point to a stray virtual address in 32bit mode. Fix it up
after consulting with our shiny new __in_29bit_mode().
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Valid sizes include 256kB, not 258kB.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This introduces some much overdue chainsawing of the fixed PMB support.
fixed PMB was introduced initially to work around the fact that dynamic
PMB mode was relatively broken, though they were never intended to
converge. The main areas where there are differences are whether the
system is booted in 29-bit mode or 32-bit mode, and whether legacy
mappings are to be preserved. Any system booting in true 32-bit mode will
not care about legacy mappings, so these are roughly decoupled.

Regardless of the entry point, PMB and 32BIT are directly related as far
as the kernel is concerned, so we also switch back to having one select
the other.

With legacy mappings iterated through and applied in the initialization
path it's now possible to finally merge the two implementations and
permit dynamic remapping overtop of remaining entries regardless of
whether boot mappings are crafted by hand or inherited from the boot
loader.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

The mass produced cuts use an updated PVR value, add them to the list.
Signed-off-by: NMatt Fleming <matt@console-pimps.org>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This follows the x86 xstate changes and implements a task_xstate slab
cache that is dynamically sized to match one of hard FP/soft FP/FPU-less.

This also tidies up and consolidates some of the SH-2A/SH-4 FPU
fragmentation. Now fpu state restorers are commonly defined, with the
init_fpu()/fpu_init() mess reworked to follow the x86 convention.
The fpu_init() register initialization has been replaced by xstate setup
followed by writing out to hardware via the standard restore path.

As init_fpu() now performs a slab allocation a secondary lighterweight
restorer is also introduced for the context switch.

In the future the DSP state will be rolled in here, too.

More work remains for math emulation and the SH-5 FPU, which presently
uses its own special (UP-only) interfaces.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Presently this has a BUG_ON() for failure cases, as powerpc does. Switch
this over to a SLAB_PANIC instead.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Presently the thread_info allocators are special cased, depending on
THREAD_SHIFT < PAGE_SHIFT. This provides a sensible definition for them
regardless of configuration, in preparation for extended CPU state.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

start_thread() will become a bit heavier with the xstate freeing to be
added in, so move it out-of-line in preparation.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This adds some VBR sanity checks in the sh_bios code to ensure that the
BIOS VBR is in range before blindly trapping in to it. This permits
boards with varying boot loader configurations to always leave support
for sh-bios enabled and it will just be disabled at run-time if not
found.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This splits out the unaligned access counters and userspace bits in to
their own generic interface, which will allow them to be wired up on sh64
too.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Now that the sh-sci earlyprintk is taken care of by the sh-sci driver
directly, there's no longer any reason for having a split-out
early_printk framework. sh_bios is the only other thing that uses it, so
we just migrate the leftovers in to there. As it's possible to have
multiple early_param()'s for the same string, there's not much point in
having this split out anymore anyways, particularly since the sh_bios
dependencies are still special-cased within sh-sci itself.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

sh_bios_char_out() is not used by anything in-tree these days, so just
get rid of it.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>