Those two APIs were provided to optimize the calls of
tick_nohz_idle_enter() and rcu_idle_enter() into a single
irq disabled section. This way no interrupt happening in-between would
needlessly process any RCU job.
Now we are talking about an optimization for which benefits
have yet to be measured. Let's start simple and completely decouple
idle rcu and dyntick idle logics to simplify.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
It is assumed that rcu won't be used once we switch to tickless
mode and until we restart the tick. However this is not always
true, as in x86-64 where we dereference the idle notifiers after
the tick is stopped.
To prepare for fixing this, add two new APIs:
tick_nohz_idle_enter_norcu() and tick_nohz_idle_exit_norcu().
If no use of RCU is made in the idle loop between
tick_nohz_enter_idle() and tick_nohz_exit_idle() calls, the arch
must instead call the new *_norcu() version such that the arch doesn't
need to call rcu_idle_enter() and rcu_idle_exit().
Otherwise the arch must call tick_nohz_enter_idle() and
tick_nohz_exit_idle() and also call explicitly:
- rcu_idle_enter() after its last use of RCU before the CPU is put
to sleep.
- rcu_idle_exit() before the first use of RCU after the CPU is woken
up.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Frysinger <vapier@gentoo.org>
Cc: Guan Xuetao <gxt@mprc.pku.edu.cn>
Cc: David Miller <davem@davemloft.net>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Hans-Christian Egtvedt <hans-christian.egtvedt@atmel.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
The tick_nohz_stop_sched_tick() function, which tries to delay
the next timer tick as long as possible, can be called from two
places:
- From the idle loop to start the dytick idle mode
- From interrupt exit if we have interrupted the dyntick
idle mode, so that we reprogram the next tick event in
case the irq changed some internal state that requires this
action.
There are only few minor differences between both that
are handled by that function, driven by the ts->inidle
cpu variable and the inidle parameter. The whole guarantees
that we only update the dyntick mode on irq exit if we actually
interrupted the dyntick idle mode, and that we enter in RCU extended
quiescent state from idle loop entry only.
Split this function into:
- tick_nohz_idle_enter(), which sets ts->inidle to 1, enters
dynticks idle mode unconditionally if it can, and enters into RCU
extended quiescent state.
- tick_nohz_irq_exit() which only updates the dynticks idle mode
when ts->inidle is set (ie: if tick_nohz_idle_enter() has been called).
To maintain symmetry, tick_nohz_restart_sched_tick() has been renamed
into tick_nohz_idle_exit().
This simplifies the code and micro-optimize the irq exit path (no need
for local_irq_save there). This also prepares for the split between
dynticks and rcu extended quiescent state logics. We'll need this split to
further fix illegal uses of RCU in extended quiescent states in the idle
loop.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Mike Frysinger <vapier@gentoo.org>
Cc: Guan Xuetao <gxt@mprc.pku.edu.cn>
Cc: David Miller <davem@davemloft.net>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Hans-Christian Egtvedt <hans-christian.egtvedt@atmel.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Commit 8dc7a9c84 ("blackfin: Add export.h to files using
EXPORT_SYMBOL/THIS_MODULE") inserted some of the include statements into
sections protected by an unrelated #if CONFIG_... statement. This can cause,
depending on the configuration used, warnings like this one:
arch/blackfin/mach-bf537/boards/stamp.c:2940: warning: data definition has no type or storage class
arch/blackfin/mach-bf537/boards/stamp.c:2940: warning: type defaults to ‘int’ in declaration of ‘EXPORT_SYMBOL’
arch/blackfin/mach-bf537/boards/stamp.c:2940: warning: parameter names (without types) in function declaration
This patch fixes it by moving the includes out of the #if protected sections.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Acked-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Bob Liu <lliubbo@gmail.com>
The serial TX IRQ is not simply (RX IRQ + 1) on some Blackfin chips,
so move the values to the platform resources.
Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Bob Liu <lliubbo@gmail.com>
Its presence was implicit everywhere, but we are aiming to fix that,
so call out the users explicitly.
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
These particular files were just assuming that module.h was
somehow in the include paths. Give them the more minimalist
header file explicitly.
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
This flag is a NOOP and can be removed now.
Signed-off-by: Yong Zhang <yong.zhang0@gmail.com>
Acked-by: Bob Liu <lliubbo@kernel.org>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Remove i2c_board_info for driver that doesn't exist anymore.
Delete irq_flags for drivers that don't use them anymore.
Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
There is already an entry in the spi device table for the codec, but the
modalias was wrong. Also the config symbol name for the codec is wrong,
so this is fixed as well.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Scott Jiang <scott.jiang.linux@gmail.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
This header was being rewritten while the asm-generic kbuild support
was in flight, so it missed out on the update. Punt the stub and use
the kbuild now that everything has settled.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
SND_BF5XX_SOC is for machine drivers while SND_SOC is for codec drivers.
Signed-off-by: Scott Jiang <scott.jiang.linux@gmail.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The ASoC codec name is "ad1836" and not "ad183x" as the change to rename
things ultimately did not get merged.
Signed-off-by: Scott Jiang <scott.jiang.linux@gmail.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
__kfree_rcu() in rcupdate.h bugs when parameter offset is not a constant
at compile time. Since we build the kgdb_test module with -O0 and it
includes this header file, we hit the bug. So drop the -O0 and mark the
one func we need for the test as noinline (so we can set a breakpoint on
it and have it be hit).
Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Make sure our smp_send_reschedule() implementation matches the
scheduler_ipi() callback so that it can kick the idle cpu.
Signed-off-by: Steven Miao <realmz6@gmail.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
IRQF_SHARED is not part of the IORESOURCE_IRQ bits. It's expressed by
IORESOURCE_IRQ_SHAREABLE.
IORESOURCE_IRQ_HIGHEDGE and IRQF_TRIGGER_HIGH are contradicting
values, an interrupt can hardly be configured for both level and edge
at the same time. This was introduced in commit 45138439(Blackfin
arch: flash memory map and dm9000 resources updating) of course
without any hint in the changelog what the heck this is supposed to
do.
Acked-by: Javier Herrero <jherrero@hvsistemas.es>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Now that common code supports SMP systems, switch our SMP atomic logic
over to it to avoid code duplication.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The nfsservctl system call is now gone, so we should remove all
linkage for it.
Signed-off-by: NeilBrown <neilb@suse.de>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
After changing all consumers of atomics to include <linux/atomic.h>, we
ran into some compile time errors due to this dependency chain:
linux/atomic.h
-> asm/atomic.h
-> asm-generic/atomic-long.h
where atomic-long.h could use funcs defined later in linux/atomic.h
without a prototype. This patches moves the code that includes
asm-generic/atomic*.h to linux/atomic.h.
Archs that need <asm-generic/atomic64.h> need to select
CONFIG_GENERIC_ATOMIC64 from now on (some of them used to include it
unconditionally).
Compile tested on i386 and x86_64 with allnoconfig.
Signed-off-by: Arun Sharma <asharma@fb.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David Miller <davem@davemloft.net>
Acked-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is in preparation for more generic atomic primitives based on
__atomic_add_unless.
Signed-off-by: Arun Sharma <asharma@fb.com>
Signed-off-by: Hans-Christian Egtvedt <hans-christian.egtvedt@atmel.com>
Reviewed-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David Miller <davem@davemloft.net>
Acked-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This allows us to move duplicated code in <asm/atomic.h>
(atomic_inc_not_zero() for now) to <linux/atomic.h>
Signed-off-by: Arun Sharma <asharma@fb.com>
Reviewed-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: David Miller <davem@davemloft.net>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
[ poleg@redhat.com: no need to declare show_regs() in ptrace.h, sched.h does this ]
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch removes all the module loader hook implementations in the
architecture specific code where the functionality is the same as that
now provided by the recently added default hooks.
Signed-off-by: Jonas Bonn <jonas@southpole.se>
Acked-by: Mike Frysinger <vapier@gentoo.org>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
Tested-by: Michal Simek <monstr@monstr.eu>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The pcm driver name has been changed, but the device name has not.
Signed-off-by: Scott Jiang <scott.jiang@analog.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Do not trace arch_local_save_flags(), arch_local_irq_*() and friends.
Although they are marked inline, gcc may still make a function out of
them and add it to the pool of functions that are traced by the function
tracer. This can cause undesirable results (kernel panic, triple faults,
etc).
Add the notrace notation to prevent them from ever being traced.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The bug in the BF526 rom when doing a software reset exists only in older
silicon versions, so don't clear SWRST on newer parts.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The BF51x's alternative portmux Kconfig options were copy & pasted from
the BF52x, but never tweaked to reflect it. So drop the old options as
they were never used (and were simply wrong), and add the BF51x specific
pieces to the Kconfig and header.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Back in commit c03c2a8734, we fixed logic in the non-bf54x
GPIO resume code to set the data levels properly before the direction
to avoid spurious line glitches. But we missed the bf54x code paths.
So add the same fix there.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
No need to reload these registers constantly since they're always
available (we're not making any function calls in between).
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The Blackfin C ABI says we do not need to save/restore R0-R3 and P0-P2
as they are available as scratch registers. So don't bother.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Re-architect how we save/restore the gpio/port logic that only pertains
to bf538/bf539 parts by pulling it out of the core code paths and pushing
it out to bf538-specific locations.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The current save logic used in hibernation is to do a MMR load (base +
offset) into a register, and then push that onto the stack. Then when
restoring, pop off the stack into a register followed by a MMR store
(base + offset). These use plenty of 32bit insns rather than 16bit,
are pretty long winded, and full of pipeline bubbles.
So, by taking advantage of MMRs that are contiguous, the multi-register
push/pop insn, and register abuse, we can shrink this code considerably.
When saving, the new logic does a lot of loads into the data and pointer
registers before executing a single multi-register push insn. Then when
restoring, we do a single multi-register pop insn followed by a lot of
stores. Overall, this allows us to cut the insn count by ~30%, the code
size by ~45%, and drastically reduce the register hazards that trigger
bubbles in the pipeline.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
EVT0 is for emulation, EVT1 is for reset, and EVT4 is the "global int
disable" region. None of these are used by software (or even hardware),
so don't bother saving/restoring them when we hibernate since nothing
ever uses these in Linux (the only thing they would be useful for is
core-memory scratch, but that's just crazy talk).
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
This defines only get used in the hibernate code, so remove them from the
global dpmc header as no one else cares.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
The RETE/RETN registers are only used in emulation(JTAG) and NMI nodes,
or as scratch registers, neither of which need to be saved/restored as
this code doesn't execute at those core event levels.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
For parts with more than one SIC_IWR, we can optimize the writing
a little bit using better Blackfin insns.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Have the logic that uses peripheral interrupt blocks key off of pint
defines rather than CPU names so that things are generalized across
families.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Have the code work off of MMR names rather than CPU defines so there is
less code to tweak in the future with new parts.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
These defines don't accomplish much as GPIO_# is the same thing as #.
Each CPU already provides helpful symbolic defines like GPIO_<PIN>
which everyone uses, so just punt these # ones.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Not sure how these guys slipped in, but they're annoying me.
So bring these unicode space gremlins down to earth to normal
ascii spaces.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Frederic Weisbecker