* git://git.kernel.org/pub/scm/linux/kernel/git/lethal/sh-2.6: (132 commits)
sh: oprofile: Fix up the module build.
sh: add UIO support for JPU on SH7722.
serial: sh-sci: Fix up port pinmux for SH7366.
sh: mach-rsk: Use uImage generation by default for rsk7201/7203.
sh: mach-sh03: Fix up pata_platform build breakage.
sh: enable deferred io LCDC on Migo-R
video: sh_mobile_lcdcfb deferred io support
video: deferred io with physically contiguous memory
video: deferred io cleanup
video: fix deferred io fsync()
sh: add LCDC interrupt configuration to AP325 and Migo-R
sh_mobile_lcdc: use FB_SYS helpers instead of FB_CFB
sh: split coherent pages
sh: dma: Kill off ISA DMA wrapper.
sh: Conditionalize the code dumper on CONFIG_DUMP_CODE.
sh: Kill off the unused SH_ALPHANUMERIC debug option.
sh: Enable skipping of bss on debug platforms for sh32 also.
doc: Update sh cpufreq documentation.
sh: mrshpc_setup_windows() needs to be inline.
serial: sh-sci: sci_poll_get_char() is only used by CONFIG_CONSOLE_POLL.
...
and are queued up for v2.6.29. This shows that the facility is still not
tested well enough to release into a stable kernel - disable it for now and
reactivate in .29. In .29 the hardware-branch-tracer will use the DS/BTS
facilities too - hopefully resulting in better code.
Kyle McMartin [Tue, 23 Dec 2008 13:44:30 +0000 (08:44 -0500)]
parisc: disable UP-optimized flush_tlb_mm
flush_tlb_mm's "optimized" uniprocessor case of allocating a new
context for userspace is exposing a race where we can suddely return
to a syscall with the protection id and space id out of sync, trapping
on the next userspace access.
Harry Ciao [Tue, 23 Dec 2008 21:57:16 +0000 (13:57 -0800)]
edac: fix edac core deadlock when removing a device
When deleting an edac device, we have to wait for its edac_dev.work to be
completed before deleting the whole edac_dev structure. Since we have no
idea which work in current edac_poller's workqueue is the work we are
conerned about, we wait for all work in the edac_poller's workqueue to be
proceseed. This is done via flush_cpu_workqueue() which inserts a
wq_barrier into the tail of the workqueue and then sleeping on the
completion of this wq_barrier. The edac_poller will wake up sleepers when
it is found.
EDAC core creates only one kernel worker thread, edac_poller, to run the
works of all current edac devices. They share the same callback function
of edac_device_workq_function(), which would grab the mutex of
device_ctls_mutex first before it checks the device. This is exactly
where edac_poller and rmmod would have a great chance to deadlock.
In below call trace of rmmod > ... >
edac_device_del_device >
edac_device_workq_teardown > flush_workqueue > flush_cpu_workqueue,
device_ctls_mutex would have already been grabbed by
edac_device_del_device(). So, on one hand rmmod would sleep on the
completion of a wq_barrier, holding device_ctls_mutex; on the other hand
edac_poller would be blocked on the same mutex when it's running any one
of works of existing edac evices(Note, this edac_dev.work is likely to be
totally irrelevant to the one that is being removed right now)and never
would have a chance to run the work of above wq_barrier to wake rmmod up.
edac_device_workq_teardown() should not be called within the critical
region of device_ctls_mutex. Just like is done in edac_pci_del_device()
and edac_mc_del_mc(), where edac_pci_workq_teardown() and
edac_mc_workq_teardown() are called after related mutex are released.
Moreover, an edac_dev.work should check first if it is being removed. If
this is the case, then it should bail out immediately. Since not all of
existing edac devices are to be removed, this "shutting flag" should be
contained to edac device being removed. The current edac_dev.op_state can
be used to serve this purpose.
The original deadlock problem and the solution have been witnessed and
tested on actual hardware. Without the solution, rmmod an edac driver
would result in below deadlock:
Evgeniy Polyakov [Tue, 23 Dec 2008 21:57:12 +0000 (13:57 -0800)]
w1: fix slave selection on big-endian systems
During test of the w1-gpio driver i found that in "w1.c:679
w1_slave_found()" the device id is converted to little-endian with
"cpu_to_le64()", but its not converted back to cpu format in "w1_io.c:293
w1_reset_select_slave()".
Based on a patch created by Andreas Hummel.
[akpm@linux-foundation.org: remove unneeded cast] Reported-by: Andreas Hummel <andi_hummel@gmx.de> Signed-off-by: Evgeniy Polyakov <zbr@ioremap.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
V4L/DVB (9920): em28xx: fix NULL pointer dereference in call to VIDIOC_INT_RESET command
Fix a NULL pointer dereference that would occur if the video decoder tied to
the em28xx supports the VIDIOC_INT_RESET call (for example: the cx25840 driver)
Jarkko Nikula [Mon, 22 Dec 2008 08:57:33 +0000 (10:57 +0200)]
ASoC: Fix DSP formats in SSM2602 audio codec
Thanks to Troy Kisky <troy.kisky@boundarydevices.com> for noticing.
- DSP_A format has 1-bit data delay which corresponds to SSM6202 submode 2
- DSP_B has 0-bit data delay which corresponds to submode 1
- Currently driver sets them opposite so swap the submode setting
Signed-off-by: Jarkko Nikula <jarkko.nikula@nokia.com> Cc: Cliff Cai <cliff.cai@analog.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Jarkko Nikula [Mon, 22 Dec 2008 08:21:36 +0000 (10:21 +0200)]
ASoC: Fix incorrect DSP format in OMAP McBSP DAI and affected drivers
- OMAP McBSP DAI driver claims to support DSP_A format which has 1-bit data
delay but configures link for 0-bit data delay which is in fact DSP_B
- Fix this by changing format from DSP_A to DSP_B
- Fix also TLV320AIC23 codec and OSK5912 machine drivers since the same
error is populated also there
Signed-off-by: Jarkko Nikula <jarkko.nikula@nokia.com> Acked-by: Arun KS <arunks@mistralsolutions.com> Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Takanari Hayama [Mon, 22 Dec 2008 02:05:03 +0000 (11:05 +0900)]
sh: add UIO support for JPU on SH7722.
Add JPU support on Migo-R via UIO. This make use of Magnus's
generic UIO platform driver. Chunk of contiguous memory to hold
intermediate image and compressed data during encode and decode.
Signed-off-by: Takanari Hayama <taki@igel.co.jp> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Magnus Damm [Fri, 19 Dec 2008 06:34:41 +0000 (15:34 +0900)]
video: sh_mobile_lcdcfb deferred io support
This patch adds sh_mobile_lcdcfb deferred io support for SYS panels.
The LCDC hardware block managed by the sh_mobile_lcdcfb driver supports
RGB or SYS panel configurations. SYS panels come with an external display
controller that is resposible for refreshing the actual LCD panel. RGB
panels are controlled directly by the LCDC and they need to be refreshed
by the LCDC hardware.
In the case of SYS panels we can save some power by configuring the LCDC
hardware block in one-shot mode. In this one-shot mode panel refresh is
managed by software. This works well together with deferred io since it
allows us to stop clocks for most of the time and only enable clocks when
we actually want to trigger an update. When there is no fbdev activity
the clocks are kept stopped which allows us to deep sleep.
The refresh rate in deferred io mode is set using platform data. The same
platform data can also be used to disable deferred io mode.
As with other deferred io frame buffers user space code should use fsync()
on the frame buffer device to trigger an update.
Signed-off-by: Magnus Damm <damm@igel.co.jp> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Magnus Damm [Fri, 19 Dec 2008 06:34:32 +0000 (15:34 +0900)]
video: deferred io with physically contiguous memory
Extend the deferred io code from only supporting vmalloc()ed frame
buffer memory to support both vmalloc()ed and physically contiguous
frame buffer memory.
The sh_mobile_lcdcfb hardware does not support scatter gather so
we need physically contiguous memory to back our frame buffer.
Signed-off-by: Magnus Damm <damm@igel.co.jp> Acked-by: Jaya Kumar <jayakumar.lkml@gmail.com> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Magnus Damm [Fri, 19 Dec 2008 06:34:09 +0000 (15:34 +0900)]
video: fix deferred io fsync()
If CONFIG_FB_DEFERRED_IO is set, but there are framebuffers
registered that does not make use of deferred io, then fsync()
on those framebuffers will result in a crash. Fix that.
This is needed for sh_mobile_lcdcfb since we always enable
deferred io at compile time but we may disable deferred io
for some types of hardware configurations.
Signed-off-by: Magnus Damm <damm@igel.co.jp> Acked-by: Jaya Kumar <jayakumar.lkml@gmail.com> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Magnus Damm [Fri, 19 Dec 2008 03:02:16 +0000 (12:02 +0900)]
sh: add LCDC interrupt configuration to AP325 and Migo-R
Add LCDC interrupt resources for AP325 and Migo-R. The LCDC driver does
not require interrupts at this point, but changes such as one-shot SYS
mode using deferred io, and wait-for-vblank will both need this.
Signed-off-by: Magnus Damm <damm@igel.co.jp> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Paul Mundt [Wed, 17 Dec 2008 06:33:43 +0000 (15:33 +0900)]
sh: Enable skipping of bss on debug platforms for sh32 also.
This enables the same functionality that sh64 has for sh32. When running
on simulated hardware or via remote memory via the debug interface,
memory is gauranteed to be zero on boot already, and skipping the zeroing
of BSS has measurable boot time benefits.
Paul Mundt [Wed, 17 Dec 2008 05:53:24 +0000 (14:53 +0900)]
serial: sh-sci: sci_poll_get_char() is only used by CONFIG_CONSOLE_POLL.
sci_poll_put_char() happens to also be used by the serial console,
while sci_poll_get_char() is only used by CONFIG_CONSOLE_POLL. Add
another gnarly ifdef to shut up the compiler.
Paul Mundt [Wed, 17 Dec 2008 05:41:54 +0000 (14:41 +0900)]
sh: Kill off cf-enabler with extreme prejudice.
Now that the rest of the boards that were using cf-enabler "generically"
have switched to setting up their mappings on their own, only the mach-se
boards were left using it. All of the cf-enabler using mach-se boards
use a special initialization of the MRSHPC windows rather than going
through the special PTE as other SH-4 platforms do. This consolidates
the MRSHPC setup logic, hooks it up on the boards that care, and gets rid
of any and all remaining references to cf-enabler.
This has been long overdue, as cf-enabler has been the bane of
arch/sh/kernel for the last 7 years. Good riddance.
Paul Mundt [Wed, 17 Dec 2008 03:48:41 +0000 (12:48 +0900)]
sh: mach-sh03: Move off of cf_enabler to pata_platform, as per landisk.
This gets rid of the cf enabler use on mach-sh03 and switches to use
pata_platform with the proper address directly. cf_enabler is
subsequently disabled for mach-sh03.
Paul Mundt [Wed, 17 Dec 2008 03:20:15 +0000 (12:20 +0900)]
sh: Simplify kernel_thread_helper() for sh32.
This can use the same implementation as sh64, the generated assembly is
the same between the new and old version, so there is not much point in
leaving it open coded in inline assembly.
This is preparatory work for future consolidation of the _32/_64
variants.
Paul Mundt [Wed, 17 Dec 2008 02:47:59 +0000 (11:47 +0900)]
sh: Kill off sh_bios_in_gdb_mode().
With the reworked kgdb support, we always detach and reinitialize the
stub. This was mostly a feature for handoffs between sh-ipl+g and the
kgdb stub, but virtually no sh-ipl+g versions ever had this working
right in the first place.
Given that the sh-ipl+g stubs in general use today don't even support
the GDB stub, and we have already killed off the special casing in the
sh-sci serial driver, kill off this now unused symbol too.
Dave Peverley [Tue, 16 Dec 2008 00:35:40 +0000 (09:35 +0900)]
sh: oprofile: Backtrace support.
This patch improves the oprofile support on sh and adds backtrace
support.
Signed-off-by: Dave Peverley <dpeverley@mpc-data.co.uk> Signed-off-by: Chris Smith <chris.smith@st.com> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Paul Mundt [Tue, 16 Dec 2008 00:33:53 +0000 (09:33 +0900)]
sh: oprofile: Refactor common setup code for multiple driver support.
This re-implements the old op_model_null code in to something more
generic, where multiple drivers, backtrace, etc. can all be interfaced.
Based largely on arch/mips/oprofile/common.c.
Matt Fleming [Sun, 14 Dec 2008 12:02:27 +0000 (12:02 +0000)]
sh: Convert Cayman boards from hw_interrupt_type to irq_chip
I've been unable to even compile-test this change because I don't have
an sh5 toolchain. All uses of hw_interrupt_type for SuperH boards have
now been converted to use irq_chip.
Signed-off-by: Matt Fleming <mjf@gentoo.org> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Matt Fleming [Sun, 14 Dec 2008 12:02:26 +0000 (12:02 +0000)]
sh: Convert Dreamcast support from hw_interrupt_type to irq_chip
Switch the dreamcast IRQ code over to the irq_chip way of doing things,
so that we can set GENERIC_HARDIRQS_NO__DO_IRQ for all SuperH boards.
Also, whilst I'm here change some things to make checkpatch.pl happy:
- Indent with tabs, not with spaces
- Include <linux/io.h>, not <asm/io.h>
- Fix the multi-line comment style
- Fix some typos in the comments
Tested-by: Adrian McMenamin <adrian@newgolddream.dyndns.info> Signed-off-by: Matt Fleming <mjf@gentoo.org> Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Paul Mundt [Thu, 11 Dec 2008 09:46:46 +0000 (18:46 +0900)]
sh: Generic kgdb stub support.
This migrates from the old bitrotted kgdb stub implementation and moves
to the generic stub. In the process support for SH-2/SH-2A is also added,
which the old stub never provided.
Paul Mundt [Wed, 10 Dec 2008 10:26:44 +0000 (19:26 +0900)]
sh: Handle cases where setup{_rt,}_frame() fail on SH-5 signal delivery.
Presently these cases are not handled properly due to the return value
not being passed back. This needs to be correct to get proper behaviour
out of things like the tracehook signal notifier, amongst others.
Paul Mundt [Wed, 10 Dec 2008 08:16:09 +0000 (17:16 +0900)]
sh: Kill off sh64's unused alloc/free_task_struct() definitions.
These were left over from some time ago, sh64 never got around to
defining __HAVE_ARCH_TASK_STRUCT_ALLOCATOR during the conversion, and it
has no need to. Kill these off and use the generic versions instead.
Paul Mundt [Wed, 10 Dec 2008 08:09:36 +0000 (17:09 +0900)]
sh: mach-migor: Kill off unused Migo-R machvec.
This kills off the special Migo-R machvec, as nothing is using it. By
default this will switch to using the generic machvec, which provides the
same functionality. This saves us a bit of space in the machvec section.
projects / linux-2.6-omap-h63xx.git / log