Sysbench thinks SD_BALANCE_WAKE is too agressive and kbuild doesn't
really mind too much, SD_BALANCE_NEWIDLE picks up most of the
slack.

On a dual socket, quad core, dual thread nehalem system:

sysbench (--num_threads=16):

 SD_BALANCE_WAKE-: 13982 tx/s
 SD_BALANCE_WAKE+: 15688 tx/s

kbuild (-j16):

 SD_BALANCE_WAKE-: 47.648295846  seconds time elapsed   ( +-   0.312% )
 SD_BALANCE_WAKE+: 47.608607360  seconds time elapsed   ( +-   0.026% )

(same within noise)
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

This is necessary to get ftrace syscall tracing working again.. a fairly
trivial and mechanical change. The one benefit is that this can also be
enabled on sh64, despite not having its own ftrace port.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

If we're looking to place a new task, we might as well find the
idlest position _now_, not 1 tick ago.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

Make the idle balancer more agressive, to improve a
x264 encoding workload provided by Jason Garrett-Glaser:

 NEXT_BUDDY NO_LB_BIAS
 encoded 600 frames, 252.82 fps, 22096.60 kb/s
 encoded 600 frames, 250.69 fps, 22096.60 kb/s
 encoded 600 frames, 245.76 fps, 22096.60 kb/s

 NO_NEXT_BUDDY LB_BIAS
 encoded 600 frames, 344.44 fps, 22096.60 kb/s
 encoded 600 frames, 346.66 fps, 22096.60 kb/s
 encoded 600 frames, 352.59 fps, 22096.60 kb/s

 NO_NEXT_BUDDY NO_LB_BIAS
 encoded 600 frames, 425.75 fps, 22096.60 kb/s
 encoded 600 frames, 425.45 fps, 22096.60 kb/s
 encoded 600 frames, 422.49 fps, 22096.60 kb/s

Peter pointed out that this is better done via newidle_idx,
not via LB_BIAS, newidle balancing should look for where
there is load _now_, not where there was load 2 ticks ago.

Worst-case latencies are improved as well as no buddies
means less vruntime spread. (as per prior lkml discussions)

This change improves kbuild-peak parallelism as well.
Reported-by: NJason Garrett-Glaser <darkshikari@gmail.com>
Signed-off-by: NMike Galbraith <efault@gmx.de>
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1253011667.9128.16.camel@marge.simson.net>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

When merging select_task_rq_fair() and sched_balance_self() we lost
the use of wake_idx, restore that and set them to 0 to make wake
balancing more aggressive.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

The problem with wake_idle() is that is doesn't respect things like
cpu_power, which means it doesn't deal well with SMT nor the recent
RT interaction.

To cure this, it needs to do what sched_balance_self() does, which
leads to the possibility of merging select_task_rq_fair() and
sched_balance_self().

Modify sched_balance_self() to:

  - update_shares() when walking up the domain tree,
    (it only called it for the top domain, but it should
     have done this anyway), which allows us to remove
    this ugly bit from try_to_wake_up().

  - do wake_affine() on the smallest domain that contains
    both this (the waking) and the prev (the wakee) cpu for
    WAKE invocations.

Then use the top-down balance steps it had to replace wake_idle().

This leads to the dissapearance of SD_WAKE_BALANCE and
SD_WAKE_IDLE_FAR, with SD_WAKE_IDLE replaced with SD_BALANCE_WAKE.

SD_WAKE_AFFINE needs SD_BALANCE_WAKE to be effective.

Touch all topology bits to replace the old with new SD flags --
platforms might need re-tuning, enabling SD_BALANCE_WAKE
conditionally on a NUMA distance seems like a good additional
feature, magny-core and small nehalem systems would want this
enabled, systems with slow interconnects would not.
Signed-off-by: NPeter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <new-submission>
Signed-off-by: NIngo Molnar <mingo@elte.hu>

After KYCR2 is set, udelay might become necessary if there are only a
small number of keys attached. This patch introduces an optional delay
through the platform data to address this problem.
Signed-off-by: NKuninori Morimoto <morimoto.kuninori@renesas.com>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

romimage macros which are used in kfr2r09 is very useful for other board.
This patch divides kfr2r09's romimage.h into
romimage-macros and partner-jet-setup.
Signed-off-by: NKuninori Morimoto <morimoto.kuninori@renesas.com>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This was #define'd as 0 on all platforms, so let's get rid of it.

This change makes pci_scan_slot() slightly easier to read.

Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: Tony Luck <tony.luck@intel.com>
Cc: David Howells <dhowells@redhat.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Reviewed-by: NMatthew Wilcox <willy@linux.intel.com>
Acked-by: NRussell King <linux@arm.linux.org.uk>
Acked-by: NRalf Baechle <ralf@linux-mips.org>
Acked-by: NKyle McMartin <kyle@mcmartin.ca>
Acked-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: NPaul Mundt <lethal@linux-sh.org>
Acked-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NAlex Chiang <achiang@hp.com>
Signed-off-by: NJesse Barnes <jbarnes@virtuousgeek.org>

This fixes up the kmap_coherent/kunmap_coherent() interface for recent
changes both in the page fault path and the shared cache flushers, as
well as adding in some optimizations.

One of the key things to note here is that the TLB flush itself is
deferred until the unmap, and the call in to update_mmu_cache() itself
goes away, relying on the regular page fault path to handle the lazy
dcache writeback if necessary.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

The kgdb stub has traditionally tied in to the NMI slot, and manually
handled debounce. Now that we have a generic way to do this instead, all
of the stub-specific debounce silliness can be killed off.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This implements support for NMI debugging that was shamelessly copied
from the avr32 port. A bit of special magic is needed in the interrupt
exception path given that the NMI exception handler is stubbed in to the
regular exception handling table despite being reported in INTEVT. So we
mangle the lookup and kick off an EXPEVT-style exception dispatch from
the INTEVT path for exceptions that do_IRQ() has no chance of handling.
As a result, we also drop the evt2irq() conversion from the do_IRQ() path
and just do it in assembly.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

A couple of these popped up on the sh2a build, causing build failures.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Some unaligned accesses are completely expected. For example, the
trapped_io code uses the unaligned access fixup code path so there's no
need to warn about having to fixup the unaligned access.
Signed-off-by: NMatt Fleming <matt@console-pimps.org>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

The method of ETHER_LINK pin is board dependence.
This patch adding paramters are:
 - no_ether_link          : If set to 1, do not use ETHER_LINK
 - ether_link_active_low  : If set to 1, ETHER_LINK is active low.
Signed-off-by: NYoshihiro Shimoda <shimoda.yoshihiro@renesas.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This is supposed to be the equivalent of __NR_syscalls, not
__NR_syscalls -1. The x86 code this was based on had simply fallen
out of sync at the time this was implemented. Fix it up now.

As a result, tracing of __NR_perf_counter_open works as advertised.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Needed by ftrace changes in -tip.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

sys_cacheflush should return with EINVAL if the cache parameter is not
one of ICACHE, DCACHE or BCACHE.
So, we need to include 0 in the first check.

It also adds the three definitions above as wrapper of the existent macros.

PS: ltp cacheflush01 test now passes.
Signed-off-by: NGiuseppe Cavallaro <peppe.cavallaro@st.com>
Signed-off-by: NStuart Menefy <stuart.menefy@st.com>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Adds a system call to allow user code to flush code from the cache.
You can use instructions for the data side, but the iside can
only be done by a flush ROM which really only works with a direct
mapped cache. The later SH4's have 2 way Iside, so this call allows
a portable way to flush the cache.
Signed-off-by: NStuart Menefy <stuart.menefy@st.com>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Reading from the ROM is not a good idea as it could disturb some
flash operation that it is in progress.
Signed-off-by: NStuart Menefy <stuart.menefy@st.com>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

The SH instruction set has several instructions which accept an 8 bit
immediate operand. For logical instructions this operand is zero extended,
for arithmetic instructions the operand is sign extended. After adding an
option to the assembler to check this, it was found that several pieces
of assembly code were assuming this behaviour, and in one case
getting it wrong.

So this patch explicitly sign extends any immediate operands, which makes
it obvious what is happening, and fixes the one case which got it wrong.
Signed-off-by: NStuart Menefy <stuart.menefy@st.com>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

The synopsys PCI cell used in the later STMicro chips requires code to
be run in order to do IO cycles, rather than just memory mapping the IO
space. Rather than extending the existing SH infrastructure to allow
this, use the GENERIC_IOMAP implmentation to save re-inventing the
wheel.

This set of changes allows the SH to be built with GENERIC_IOMAP
enabled, it just ifdef's out the functions provided by the GENERIC_IOMAP
implementation, and provides a few required missing functions.
Signed-off-by: NDavid McKay <david.mckay@st.com>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This patch is V3 of the SuperH Mobile Runtime PM platform bus
implentation matching Rafael's Runtime PM v16.

The code gets invoked from the SuperH specific Runtime PM
platform bus functions that override the weak symbols for:
 - platform_pm_runtime_suspend()
 - platform_pm_runtime_resume()
 - platform_pm_runtime_idle()

This Runtime PM implementation performs two levels of power
management. At the time of platform bus runtime suspend the
clock to the device is stopped instantly. Later on if all
devices within the power domain has their clocks stopped
then the device driver ->runtime_suspend() callbacks are
used to save hardware register state for each device.

Device driver ->runtime_suspend() calls are scheduled from
cpuidle context using platform_pm_runtime_suspend_idle().
When all devices have been fully suspended the processor
is allowed to enter deep sleep from cpuidle.

The runtime resume operation turns on clocks and also
restores registers if needed. It is worth noting that the
devices start in a suspended state and the device driver
is responsible for calling runtime resume before accessing
the actual hardware.

In this particular platform bus implementation runtime
resume is not allowed from interrupt context. Runtime
suspend is however allowed from interrupt context as
long as the synchronous functions are avoided.

[ updated for v17 -- PFM. ]
Signed-off-by: NMagnus Damm <damm@igel.co.jp>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

sh64 does not yet support GENERIC_BUG, but still wants unwinder support.
Alias UNWINDER_BUG and UNWINDER_BUG_ON to their BUG counterparts until
the conversion to GENERIC_BUG is completed.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This simplifies the unwinder trap handling, dropping the use of the
special trapa vector and simply piggybacking on top of the BUG support. A
new BUGFLAG_UNWINDER is added for flagging the unwinder fault, before
continuing on with regular BUG dispatch.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Allow a DWARF register to have an undefined value. When applied to the
DWARF return address register this lets lets us label a function as
having no direct caller, e.g. kernel_thread_helper().
Signed-off-by: NMatt Fleming <matt@console-pimps.org>

We can't assume that if we execute the unwinder code and the unwinder
was already running that it has faulted. Clearly two kernel threads can
invoke the unwinder at the same time and may be running simultaneously.

The previous approach used BUG() and BUG_ON() in the unwinder code to
detect whether the unwinder was incapable of unwinding the stack, and
that the next available unwinder should be used instead. A better
approach is to explicitly invoke a trap handler to switch unwinders when
the current unwinder cannot continue.
Signed-off-by: NMatt Fleming <matt@console-pimps.org>

The handling of DW_CFA_val_offset ops was incorrectly using the
DWARF_REG_OFFSET flag but the register's value cannot be calculated
using the DWARF_REG_OFFSET method. Create a new flag to indicate that a
different method must be used to calculate the register's value even
though there is no implementation for DWARF_VAL_OFFSET yet; it's mainly
just a place holder.
Signed-off-by: NMatt Fleming <matt@console-pimps.org>

Plug a memory leak in dwarf_unwinder_dump() where we didn't free the
memory that we had previously allocated for the DWARF frames and DWARF
registers.

Now is also a opportune time to implement our own mempool and kmem
cache. It's a good idea to have a certain number of frame and register
objects in reserve at all times, so that we are guaranteed to have our
allocation satisfied even when memory is scarce. Since we have pools to
allocate from we can implement the registers for each frame as a linked
list as opposed to a sparsely populated array. Whilst it's true that the
lookup time for a linked list is larger than for arrays, there's only
usually a maximum of 8 registers per frame. So the overhead isn't that
much of a concern.
Signed-off-by: NMatt Fleming <matt@console-pimps.org>

Signed-off-by: NYoshihiro Shimoda <shimoda.yoshihiro@renesas.com>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This does a bit of rework for making the cache flushers SMP-aware. The
function pointer-based flushers are renamed to local variants with the
exported interface being commonly implemented and wrapping as necessary.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

All CPU-specific overloads are done at runtime now, so this common header
can go away and simply be folded back in to asm/ version.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

When get_mmu_context() runs out of new ASIDs it flushes the TLB and
wraps around. Despite the fact the ASIDs are tracked per-CPU, a global
TLB flush was being used. Switch this over to a local one, as matches
the intent.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

As an excellent indicator of how much testing the DSP code gets, a couple
of rather glaring bugs in the DSP save/restore paths were found:

- In the DSP restore case a0 needs to be popped off before a0g,
  or the value of a0g is clobbered by the MSB of a0 in the case
  of sign extension.

- Beyond that, the save and restore orders were out of sync,
  so this fixes that up as well. At the same time, we switch over
  to using movs.l for both the save and restore of the general DSP
  registers as opposed to using sts.l (which was initially put in
  place to work around a bug in ancient binutils versions which
  the kernel no longer supports).
Reported-by: NChee Soon Yip <yip.cheesoon@renesas.com>
Cc: Chu Lih Kwek <kwek.chulih@renesas.com>,
Cc: General Lai <general.lai@renesas.com>,
Cc: Robert Cozens <Robert.Cozens@renesas.com>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Signed-off-by: NMichael Trimarchi <trimarchimichael@yahoo.it>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

While most platforms implement LED banks in sets of 8/16/32, some use
different configurations. This adds a LED mask to the heartbeat platform
data to allow platforms to constrain the bitmap, which is otherwise
derived from the register size.
Signed-off-by: NKuninori Morimoto <morimoto.kuninori@renesas.com>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This moves the initialization over to an early_initcall(). This fixes up
some lockdep interaction issues. At the same time, kill off some
superfluous locking in the init path.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Also, remove the "fix" to DW_CFA_def_cfa_register where we reset the
frame's cfa_offset to 0. This action is incorrect when handling
DW_CFA_def_cfa_register as the DWARF spec specifically states that the
previous contents of cfa_offset should be used with the new
register. The reason that I thought cfa_offset should be reset to 0 was
because it was being assigned a bogus value prior to executing the
DW_CFA_def_cfa_register op. It turns out that the bogus cfa_offset value
came from interpreting .cfi_escape pseudo-ops (those used by the GNU
extensions) as CFA_DW_def_cfa ops.
Signed-off-by: NMatt Fleming <matt@console-pimps.org>

Signed-off-by: NPaul Mundt <lethal@linux-sh.org>