Commit graph

26 commits

Author SHA1 Message Date
Takashi YOSHII
65ada547d6 clocksource: sh_cmt: Remove nested spinlock fix
There are control flow that sh_cmt_set_next() does double
spin-lock. The callers sh_cmt_{start,stop}() already have
lock. But another callers sh_cmt_clock_event_{start,next}()
does not.

Now sh_cmt_set_next() does not lock by itself. All the
callers should hold spin-lock before calling it.

[damm@opensource.se: use __sh_cmt_set_next() to simplify code]
[damm@opensource.se: added stable, suitable for v2.6.35 + v2.6.36]
Cc: stable@kernel.org
Signed-off-by: Takashi YOSHII <takashi.yoshii.zj@renesas.com>
Signed-off-by: Magnus Damm <damm@opensource.se>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2010-12-17 19:38:33 +09:00
Magnus Damm
03ff858c09 ARM: shmobile: remove sh_timer_config clk member
Now when the SH-Mobile ARM platforms have been converted
to use device name it is possible to remove "clk" from
struct sh_timer_config.

Signed-off-by: Magnus Damm <damm@opensource.se>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2010-10-31 10:40:39 -04:00
Magnus Damm
38409d726b clocksource: sh_cmt: Rate calculation fix
Fix the rate calculation in the CMT driver.
Without this fix the clocksource runs way
too fast and we get a divide-by-zero error.

Signed-off-by: Magnus Damm <damm@opensource.se>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2010-08-04 16:01:29 +09:00
Magnus Damm
43809473b9 clocksource: sh_cmt: One-off clockevent fix V2
Fix a one-off error in the CMT driver V2. The match register
should be programmed with the period minus one.

Many thanks to Eiraku-san for tracking down this issue.

Signed-off-by: Magnus Damm <damm@opensource.se>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2010-08-04 16:01:24 +09:00
Paul Mundt
5967d33ce8 clocksource: sh_cmt: Fix up bogus shift value.
The previous CMT fixup accidentally copied in the TMU shift value, reset
this back to its original value while preserving the TMU fix.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2010-06-22 16:41:31 +09:00
Paul Mundt
f4d7c3565c clocksource: sh_cmt: compute mult and shift before registration
Based on the sh_tmu change in 66f49121ff
("clocksource: sh_tmu: compute mult and shift before registration").
The same issues impact the sh_cmt driver, so we take the same approach
here.

Cc: stable@kernel.org
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2010-06-02 17:10:44 +09:00
Paul Mundt
e19553427c Merge branch 'sh/stable-updates'
Conflicts:
	arch/sh/kernel/dwarf.c
	drivers/dma/shdma.c

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2010-04-26 16:08:27 +09:00
Paul Mundt
fecf066c2d sh: Disable IRQ balancing for timer and IPI IRQs.
Make sure that the timer IRQs and IPIs aren't enabled for IRQ balancing.
IPIs are disabled as a result of being percpu while the timers simply
disable balancing outright.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2010-04-15 11:59:28 +09:00
Tejun Heo
5a0e3ad6af include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files.  percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.

percpu.h -> slab.h dependency is about to be removed.  Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability.  As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.

  http://userweb.kernel.org/~tj/misc/slabh-sweep.py

The script does the followings.

* Scan files for gfp and slab usages and update includes such that
  only the necessary includes are there.  ie. if only gfp is used,
  gfp.h, if slab is used, slab.h.

* When the script inserts a new include, it looks at the include
  blocks and try to put the new include such that its order conforms
  to its surrounding.  It's put in the include block which contains
  core kernel includes, in the same order that the rest are ordered -
  alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
  doesn't seem to be any matching order.

* If the script can't find a place to put a new include (mostly
  because the file doesn't have fitting include block), it prints out
  an error message indicating which .h file needs to be added to the
  file.

The conversion was done in the following steps.

1. The initial automatic conversion of all .c files updated slightly
   over 4000 files, deleting around 700 includes and adding ~480 gfp.h
   and ~3000 slab.h inclusions.  The script emitted errors for ~400
   files.

2. Each error was manually checked.  Some didn't need the inclusion,
   some needed manual addition while adding it to implementation .h or
   embedding .c file was more appropriate for others.  This step added
   inclusions to around 150 files.

3. The script was run again and the output was compared to the edits
   from #2 to make sure no file was left behind.

4. Several build tests were done and a couple of problems were fixed.
   e.g. lib/decompress_*.c used malloc/free() wrappers around slab
   APIs requiring slab.h to be added manually.

5. The script was run on all .h files but without automatically
   editing them as sprinkling gfp.h and slab.h inclusions around .h
   files could easily lead to inclusion dependency hell.  Most gfp.h
   inclusion directives were ignored as stuff from gfp.h was usually
   wildly available and often used in preprocessor macros.  Each
   slab.h inclusion directive was examined and added manually as
   necessary.

6. percpu.h was updated not to include slab.h.

7. Build test were done on the following configurations and failures
   were fixed.  CONFIG_GCOV_KERNEL was turned off for all tests (as my
   distributed build env didn't work with gcov compiles) and a few
   more options had to be turned off depending on archs to make things
   build (like ipr on powerpc/64 which failed due to missing writeq).

   * x86 and x86_64 UP and SMP allmodconfig and a custom test config.
   * powerpc and powerpc64 SMP allmodconfig
   * sparc and sparc64 SMP allmodconfig
   * ia64 SMP allmodconfig
   * s390 SMP allmodconfig
   * alpha SMP allmodconfig
   * um on x86_64 SMP allmodconfig

8. percpu.h modifications were reverted so that it could be applied as
   a separate patch and serve as bisection point.

Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.

Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-30 22:02:32 +09:00
Paul Mundt
c2a25e8197 clocksource: Deprecate clock string across the SH drivers.
We want to get rid of the clock string from platform data entirely,
depending on the clkdev-based clock lookup to do the right thing for us
instead.

This converts all of the SH drivers to request their associated function
clocks directly, and if there is no match for that then we fall back on
the legacy lookup while warning about it. After all of the outstanding
CPUs have been converted to clkdev lookups the clock string will be
killed off completely.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2010-03-29 16:55:43 +09:00
Paul Mundt
214a607a4f clocksource: Use dev_name() universally across the SH drivers.
There is no need to copy in the name from the sh timer config now that
dev_name() is available early. We prefer the dev_name() variant for
consistent naming.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2010-03-10 16:26:25 +09:00
Linus Torvalds
e56425b135 Merge branch 'timers-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip
* 'timers-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  posix-timers.c: Don't export local functions
  clocksource: start CMT at clocksource resume
  clocksource: add suspend callback
  clocksource: add argument to resume callback
  ntp: Cleanup xtime references in ntp.c
  ntp: Make time_esterror and time_maxerror static
2010-03-01 08:48:25 -08:00
Paul Mundt
da64c2a8de clocksource: Fix up a registration/IRQ race in the sh drivers.
All of the SH clocksource drivers follow the scheme that the IRQ is setup
prior to registering the clockevent. The interrupt handler in the
clockevent cases looks to the event handler function pointer being filled
in by the registration code, permitting us to get in to situations where
asserted IRQs step in to the handler before registration has had a chance
to complete and hitting a NULL pointer deref.

In practice this is not an issue for most platforms, but some of them
with fairly special loaders (or that are chain-loading from another
kernel) may enter in to this situation. This fixes up the oops reported
by Rafael on hp6xx.

Reported-and-tested-by: Rafael Ignacio Zurita <rafaelignacio.zurita@gmail.com>
Cc: stable@kernel.org
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2010-02-25 16:37:46 +09:00
Magnus Damm
c81628848a clocksource: start CMT at clocksource resume
Add code to start the CMT timer on clocksource resume.  While at it handle
the suspend case as well.  Remove the platform device specific suspend
calls.

This makes sure the timer is started during sysdev_resume().  Without this
patch the clocksource may be read as suspended, this after sysdev resume
but before platform device resume.

Signed-off-by: Magnus Damm <damm@opensource.se>
Cc: john stultz <johnstul@us.ibm.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
2010-02-05 14:54:10 +01:00
Magnus Damm
f6431732f1 sh: CMT suspend/resume
This patch updates the SuperH CMT driver with suspend and resume
callbacks for the suspend-to-ram case. This patch stops the CMT
channel at suspend time to avoid unwanted wake up events.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2009-08-15 12:58:45 +09:00
Magnus Damm
be890a1a95 sh: turn off irqs when disabling CMT/TMU timers
Modify the CMT and TMU drivers to disable interrupts when
disabling the timer. Only using start/stop bits is not
enough.

This fixes a bootup hang on Migo-R when the CMT is replaced
by TMU for clockevents but the CMT keeps on delivering irqs
even though the timer start bit is off.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2009-06-17 15:39:56 +09:00
Paul Mundt
e7fad451f0 clocksource: Drop unused irqaction.mask from SH drivers.
The irqaction.mask is legacy code that is wholly unused and going away,
so simply drop its use in the SH drivers completely.

Fixes up build failures in -next.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2009-06-14 21:18:24 +09:00
Paul Mundt
d1fcc0a8db clocksource: sh_mtu2/cmt_register() should be static.
Neither of these need to be exported, so just make them static.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2009-05-03 18:05:42 +09:00
Paul Mundt
46a12f7426 sh: Consolidate MTU2/CMT/TMU timer platform data.
All of the SH timers use a roughly identical structure for platform data,
which presently is broken out for each block. Consolidate all of these
definitions, as there is no reason for them to be broken out in the first
place.

Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2009-05-03 17:57:17 +09:00
Magnus Damm
3014f47460 clocksource: sh_cmt 16-bit fixes
This patch contains various fixes for 16-bit cmt hardware.
With this applied periodic clockevents work fine on sh7203.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2009-04-30 12:57:30 +09:00
Magnus Damm
8e0b842948 sh: setup timers in late_time_init()
This patch moves the SuperH timer setup code from time_init()
to late_time_init(). Good things about this change:
 - interrupts: they are enabled at late_time_init()
 - mm: regular kmalloc() can be used at late_time_init()

Together with moving to late_time_init() this patch changes
the sh_cmt driver to always allocate with kmalloc(). This
simplifies the code a bit and also fixes section mismatches.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2009-04-28 18:01:25 +09:00
Magnus Damm
5b644c7a21 clocksource: improve sh_cmt clocksource overflow handling
This patch improves the sh_cmt clocksource handling.

Currently the counter value is ignored in the case of
overflow. With this patch the overflow flag is read
before and after reading the counter, removing any
counter value and overflow flag mismatch issues.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2009-04-28 18:01:21 +09:00
Magnus Damm
19bdc9d061 clocksource: sh_cmt clocksource support
Add clocksource support to the sh_cmt driver. With this in
place we can do tickless with a single CMT channel.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2009-04-22 09:31:04 +09:00
Magnus Damm
e475eedb09 clocksource: sh_cmt earlytimer support
Add Early Platform Driver support to the sh_cmt driver
using the earlytimer class.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2009-04-19 13:06:22 +09:00
Magnus Damm
3093e78eba clocksource: sh_cmt: use remove_irq() and remove clockevent workaround
Update the sh_cmt driver to make use of recent irq and clockevent changes:
 - use remove_irq() together with setup_irq()
 - remove mult workaround since WARN_ON() now has been moved

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2009-04-02 11:18:48 +09:00
Magnus Damm
3fb1b6ad06 sh: CMT clockevent platform driver
SuperH CMT clockevent driver.

Both 16-bit and 32-bit CMT versions are supported, but only 32-bit
is tested. This driver contains support for both clockevents and
clocksources, but no unregistration is supported at this point.

Works fine as clock source and/or event in periodic or oneshot mode.
Tested on sh7722 and sh7723, but should work with any cpu/architecture.

This version is lacking clocksource and early platform driver support
for now - this to minimize the amount of dependencies.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
2009-01-29 16:56:32 +09:00